Datasets:

MAdAiLab

/

lex_glue_scotus

like 0

456 U.S. 605 102 S.Ct. 2049 72 L.Ed.2d 367 Joseph S. HOPPER, Commissioner, Alabama Department of Corrections and James D. White, Warden, Petitionersv.John Louis EVANS, III. No. 80-1714. Argued March 24, 1982. Decided May 24, 1982. Syllabus Respondent was convicted in an Alabama state court of the capital offense of an intentional killing during a robbery, and was sentenced to death. At the time of respondent's trial, an Alabama statute precluded jury instructions on lesser included offenses in capital cases. The conviction and sentence were affirmed on automatic appeal. Subsequently, habeas corpus proceedings were brought in Federal District Court seeking to have the conviction set aside on the ground, inter alia, that respondent had been convicted and sentenced under a statute that unconstitutionally precluded consideration of lesser included offenses. The District Court denied relief. Pending an appeal, the Alabama statute precluding lesser included offense instructions in capital cases was invalidated in Beck v. Alabama, 447 U.S. 625, 100 S.Ct. 2382, 65 L.Ed.2d 392. The Court of Appeals then reversed the District Court, concluding that Beck v. Alabama meant that the Alabama preclusion clause so "infected" respondent's trial that he must be retried so that he might have the opportunity to introduce evidence of some lesser included offense. Held : The Alabama preclusion clause did not prejudice respondent in any way, and he is not entitled to a new trial, where his own evidence negates the possibility that a lesser included offense instruction might have been warranted. The Court of Appeals misread Beck v. Alabama, which held that due process requires that a lesser included offense instruction be given only when the evidence warrants such an instruction. Here, the evidence not only supported the claim that respondent intended to kill the victim but affirmatively negated any claim that he did not intend to kill the victim. Accordingly, an instruction on the offense of unintentional killing was not warranted. Pp. 2052-2054. 628 F.2d 400 and 639 F.2d 221, reversed. Edward E. Carnes, Montgomery, Ala., for petitioners. John L. Carroll, Montgomery, Ala., for respondent. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to determine whether, after invalidation of a state law which precluded instructions on lesser included offenses in capital cases, a new trial is required in a capital case in which the defendant's own evidence negates the possibility that such an instruction might have been warranted. 2 * A. 3 Shortly after respondent was released on parole from an Indiana prison in 1976, he and Wayne Ritter, who had been a fellow inmate, embarked on what respondent himself described as a cross-country crime "spree." App. 9. According to respondent's testimony, they committed about 30 armed robberies, 9 kidnapings, and 2 extortion schemes in seven different States during a 2-month period. Respondent testified that on January 5, 1977, he and Ritter entered a pawnshop in Mobile, Ala., intending to rob it. Ritter asked the pawnshop owner, Edward Nassar, to show him a gun. When Nassar handed the gun to Ritter, respondent pulled his own gun and announced that he intended to rob him. Nassar dropped to his hands and knees and crawled toward his office. Respondent then shot him in the back, killing him. Nassar's two daughters, aged seven and nine, were in the pawnshop at the time of the murder. 4 Respondent and Ritter were captured by the Federal Bureau of Investigation in Little Rock, Ark., on March 7, 1977. A gun, which was identified by ballistics tests as the weapon used to kill Nassar, was found in their motel room and the gun Nassar showed Ritter at the pawnshop was found in their car. After being fully advised of his constitutional rights, respondent signed a detailed written confession on March 8, 1977, admitting that he shot Nassar in the back. He repeated and elaborated on his confession before a grand jury in Mobile on April 4, 1977. He told the grand jury that Nassar was not the only person he had ever killed, that he felt no remorse because of that murder, that he would kill again in similar circumstances, and that he intended to return to a life of crime if he was ever freed. Since he doubted that he ever would be freed, he told the grand jury that he wanted to be executed as soon as possible. The grand jury indicted him under Ala.Code § 13-11-2(a)(2) (1975), which makes "[r]obbery or attempts thereof when the victim is intentionally killed by the defendant" a capital offense. B 5 Under Alabama law, capital punishment may be imposed only after conviction by a jury. Prothro v. State, 370 So.2d 740, 746-747 (Ala.Crim.App.1979). The prosecution, therefore, declined to accept respondent's guilty plea. A psychiatrist, appointed by the court, concluded that respondent was competent to stand trial. Respondent and Ritter were tried together. The evidence against respondent included his confession to the Federal Bureau of Investigation, two eyewitnesses who identified him, and ballistic evidence matching the bullet that killed Nassar with respondent's gun. 6 Against his attorneys' advice, respondent testified in his own behalf. He told the jury he had shot Nassar, and informed it that he had "no intention whatsoever of ever reforming in any way" and would return to a life of crime if released. App. 38. Release from prison in the near future appeared unlikely since he was wanted for a number of crimes in different States as a result of the armed robbery spree. Respondent told the jury: "I would rather die by electrocution than spend the rest of my life in the penitentiary. So, I'm asking very sincerely that you come back with a positive verdict for the State." Ibid. 7 The judge instructed the jury that it could not convict respondent merely on the basis of his confession, but must consider all the evidence, and could find him guilty only if the State had proved its case beyond a reasonable doubt. Prior to this Court's judgment in Beck v. Alabama, 447 U.S. 625, 100 S.Ct. 2382, 65 L.Ed.2d 392 (1980), a jury hearing a capital case in Alabama was precluded by statute from considering lesser included offenses. Alabama required a jury to convict the defendant of the capital offense charged or return a verdict of not guilty. The jurors were instructed to impose the death sentence if they concluded that the defendant was guilty, and they were not told that the trial judge could reduce the sentence to a sentence of life imprisonment without possibility of parole. Id., at 639, n. 15, 100 S.Ct., at 2390, n. 15. The jury in this case returned its verdict of guilty in less than 15 minutes. 8 The trial judge sentenced respondent to death and entered written findings that the aggravating circumstances in his case far outweighed any mitigating circumstances. The conviction and sentence were subject to automatic appeal and were affirmed on review. Evans v. State, 361 So.2d 654 (Ala.Crim.App.1977), aff'd, 361 So.2d 666 (Ala.1978), cert. denied, 440 U.S. 930, 99 S.Ct. 1267, 59 L.Ed.2d 486 (1979). C 9 Respondent's mother initiated habeas corpus proceedings under 28 U.S.C. § 2254. Respondent then changed his previous attitude of desiring execution. His habeas corpus petition to the District Court for the Southern District of Alabama challenged his conviction on a number of grounds, including an allegation that he had been convicted and sentenced under a statute which unconstitutionally precluded consideration of lesser included offenses. He did not allege that he had been prejudiced by the Alabama death penalty statute's preclusion clause, but instead argued that the statute was unconstitutional on its face and that his conviction therefore must be set aside. The District Court held a hearing, and subsequently rejected respondent's arguments, noting that respondent had confessed at least four times to shooting Nassar. Evans v. Britton, 472 F.Supp. 707, 711-712 (1979). 10 Subsequently, in Beck v. Alabama, supra, we held that the sentence of death could not be imposed after a jury verdict of guilt of a capital offense, when the jury was not permitted to consider a verdict of guilt of a lesser included noncapital offense, provided that the evidence would have supported such a verdict. The petitioner in Beck was also involved in a robbery in the course of which a murder occurred. He contended, however, that he did not kill the victim or intend his death. Instead he claimed that while he was attempting to tie up the victim, an 80-year-old man, his accomplice unexpectedly struck and killed the man. The State conceded that, on the evidence in that case, Beck would have been entitled to an instruction on the lesser included, noncapital offense of felony murder except for the preclusion clause. Id., at 629-630, 100 S.Ct., at 2385-2386. 11 Our opinion in Beck stressed that the jury was faced with a situation in which its choices were only to convict the defendant and sentence him to death or find him not guilty. The jury could not take a third option of finding that although the defendant had committed a grave crime, it was not so grave as to warrant capital punishment. We concluded that a jury might have convicted Beck but also might have rejected capital punishment if it believed Beck's testimony. On the facts shown in Beck, we held that the defendant was entitled to a lesser included offense instruction as a matter of due process. Id., at 637, 100 S.Ct., at 2389. 12 In the instant case, the Court of Appeals for the Fifth Circuit, purporting to rely on Beck, reversed the District Court's denial of relief. Evans v. Britton, 628 F.2d 400 (1980), modified, 639 F.2d 221 (1981). We granted certiorari, 452 U.S. 960, 101 S.Ct. 3107, 69 L.Ed.2d 970 (1981), and we now reverse. II A. 13 The Court of Appeals misread our opinion in Beck. The Beck opinion considered the alternatives open to a jury which is constrained by a preclusion clause and therefore unable to convict a defendant of a lesser included offense when there was evidence which, if believed, could reasonably have led to a verdict of guilt of a lesser offense. In such a situation, we concluded, a jury might convict a defendant of a capital offense because it found that the defendant was guilty of a serious crime. 447 U.S., at 642, 100 S.Ct., at 2391. Or a jury might acquit because it does not think the crime warrants death, even if it concludes that the defendant is guilty of a lesser offense. Id., at 642-643, 100 S.Ct., at 2391-2392. While in some cases a defendant might profit from the preclusion clause, we concluded that "in every case [it] introduce[s] a level of uncertainty and unreliability into the factfinding process that cannot be tolerated in a capital case." Id., at 643, 100 S.Ct., at 2392. 14 The Court of Appeals, quoting this statement from our Beck opinion, repeatedly stressed the words "in every case." 639 F.2d, at 223-224; 628 F.2d, at 401. It concluded that we meant that the Alabama preclusion clause was a "brooding omnipresence" which might "infect virtually every aspect of any capital defendant's trial from beginning to end." Ibid. It is important to note that our holding in Beck was limited to the question submitted on certiorari, and we expressly pointed out that we granted the writ in that case to decide whether a jury must be permitted to convict a defendant of a lesser included offense "when the evidence would have supported such a verdict." 447 U.S., at 627, 100 S.Ct., at 2384. Thus, our holding was that the jury must be permitted to consider a verdict of guilt of a noncapital offense "in every case" in which "the evidence would have supported such a verdict." Our holding in Beck, like our other Eighth Amendment decisions in the past decade, was concerned with insuring that sentencing discretion in capital cases is channelled so that arbitrary and capricious results are avoided. See, e.g., Roberts v. Louisiana, 428 U.S. 325, 334, 96 S.Ct. 3001, 3006, 49 L.Ed.2d 974 (1976) (plurality opinion); Woodson v. North Carolina, 428 U.S. 280, 303, 96 S.Ct. 2978, 2990, 49 L.Ed.2d 944 (1976) (plurality opinion); Gregg v. Georgia, 428 U.S. 153, 188, 96 S.Ct. 2909, 2932, 49 L.Ed.2d 859 (1976) (principal opinion); Furman v. Georgia, 408 U.S. 238, 313, 92 S.Ct. 2726, 2764, 33 L.Ed.2d 346 (1972), (WHITE, J., concurring); id., at 309-310, 92 S.Ct. at 2762-2763 (STEWART, J., concurring); and id., at 398-399, 92 S.Ct., at 2808-2809 (BURGER, C. J., dissenting). 15 In Roberts v. Louisiana, supra, the Court considered a Louisiana statute which was the obverse of the Alabama preclusion clause. In Louisiana, prior to Roberts, every jury in a capital murder case was permitted to return a verdict of guilty of the noncapital crimes of second-degree murder and manslaughter, "even if there [was] not a scintilla of evidence to support the lesser verdicts." Id., 428 U.S., at 334, 96 S.Ct., at 3006 (plurality opinion). Such a practice was impermissible, a plurality of the Court concluded, because it invited the jurors to disregard their oaths and convict a defendant of a lesser offense when the evidence warranted a conviction of first-degree murder, inevitably leading to arbitrary results. Id., at 335, 96 S.Ct., at 3007. The analysis in Roberts thus suggests that an instruction on a lesser offense in this case would have been impermissible absent evidence supporting a conviction of a lesser offense. 16 Beck held that due process requires that a lesser included offense instruction be given when the evidence warrants such an instruction. But due process requires that a lesser included offense instruction be given only when the evidence warrants such an instruction. The jury's discretion is thus channelled so that it may convict a defendant of any crime fairly supported by the evidence. Under Alabama law, the rule in noncapital cases is that a lesser included offense instruction should be given if "there is any reasonable theory from the evidence which would support the position." Fulghum v. State, 291 Ala. 71, 75, 277 So.2d 886, 890 (1973). The federal rule is that a lesser included offense instruction should be given "if the evidence would permit a jury rationally to find [a defendant] guilty of the lesser offense and acquit him of the greater." Keeble v. United States, 412 U.S. 205, 208, 93 S.Ct. 1993, 1995, 36 L.Ed.2d 844 (1973). The Alabama rule clearly does not offend federal constitutional standards, and no reason has been advanced why it should not apply in capital cases. B 17 The uniqueness of respondent's claims has been outlined in the statement of facts, but those facts merit emphasis for they bear on the key issue of whether there was any evidentiary basis to support a conviction of a lesser included offense. From the outset, beginning with his appearance before the grand jury, respondent made it crystal clear that he had killed the victim, that he intended to kill him, and that he would do the same thing again in similar circumstances. At trial, he testified that he always tried to choose places to rob so that he could avoid killing people. However, he also testified that, if necessary, he was always prepared to kill. App. 19-21. Respondent was convicted, under Ala.Code § 13-11-2(a)(2) (1975), of robbery when the victim was intentionally killed. 18 In this Court, respondent contends that he could have been convicted under Ala.Code § 13-1-70 (1975), which makes a "homicide . . . committed in the perpetration of, or the attempt to perpetrate, any . . . robbery" a noncapital offense. Respondent concedes that a conviction is warranted under this section only when a defendant lacks intent to kill. Brief for Respondent 26. Respondent's current claim is a curious—even cynical—new version of the claim of self-defense. His testimony given before the grand jury was: 19 "I was going to shoot him if he reached for a—a firearm, yeah. Uh, of course, our intention always, you know, never to hurt anybody, if you don't have to. That's— that's stupidity, you know. But if it ever came down to a case of, you know, of me or somebody else, well that's—that's pure instinct. That's self-preservation. I'm going to fire ; I'm not going to waste any time . . . ." 20 App. 19 (emphasis supplied). 21 On the basis of this testimony, he implies that he had no malice toward the victim nor intent to kill him. Of course, it can be argued that this case is not one of a killer with affirmative, purposeful malice; his claim bears some resemblance to that of a hired killer who, bearing no ill will or malice toward his victim, simply engages in the pursuit of his chosen occupation. Respondent thus blandly—even boldly—proclaims that, although he will try not to kill his victims, he will do it if he finds it to be an occupational necessity. 22 It would be an extraordinary perversion of the law to say that intent to kill is not established when a felon, engaged in an armed robbery, admits to shooting his victim in the back in the circumstances shown here. The evidence not only supported the claim that respondent intended to kill the victim, but affirmatively negated any claim that he did not intend to kill the victim. An instruction on the offense of unintentional killing during this robbery was therefore not warranted. See Fulghum, supra. 23 Finally, the Court of Appeals stated, and respondent argues, that the mere existence of the preclusion clause so "infected" respondent's trial that he must be retried so that he may have the opportunity to introduce evidence of some lesser included offense. Respondent suggests no plausible claim which he might conceivably have made, had there been no preclusion clause, that is not contradicted by his own testimony at trial.* The preclusion clause did not prejudice respondent in any way, and a new trial is not warranted. See Chapman v. California, 386 U.S. 18, 24, 87 S.Ct. 824, 828, 17 L.Ed.2d 705 (1967). 24 Reversed. 25 Justice BRENNAN and Justice MARSHALL, concurring in part and dissenting in part. 26 We join the opinion of the Court to the extent that it reverses the judgment of the Court of Appeals invalidating respondent's conviction. But we adhere to our view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. See Gregg v. Georgia, 428 U.S. 153, 227, 231, 96 S.Ct. 2909, 2971, 2973, 49 L.Ed.2d 859 (1976) (BRENNAN, J., dissenting); Furman v. Georgia, 408 U.S. 238, 314, 92 S.Ct. 2726, 2765, 33 L.Ed.2d 346 (1972) (MARSHALL, J., concurring). Consequently, we would affirm the judgment of the Court of Appeals to the extent that it invalidates the sentence of death imposed upon respondent. * In another case with different facts, a defendant might make a plausible claim that he would have employed different trial tactics—for example, that he would have introduced certain evidence or requested certain jury instructions—but for the preclusion clause. However, that is not this case, since the defendant here confessed that he shot the victim and then pleaded guilty to capital murder.

01

456 U.S. 645 102 S.Ct. 2071 72 L.Ed.2d 398 WOELKE & ROMERO FRAMING, INC., Petitioner,v.NATIONAL LABOR RELATIONS BOARD et al. PACIFIC NORTHWEST CHAPTER OF the ASSOCIATED BUILDERS & CONTRACTORS, INC., Petitioner, v. NATIONAL LABOR RELATIONS BOARD et al. OREGON-COLUMBIA CHAPTER, the ASSOCIATED GENERAL CONTRACTORS OF AMERICA, INC., Petitioner, v. NATIONAL LABOR RELATIONS BOARD et al. Nos. 80-1798, 80-1808 and 81-91. Argued March 3, 1982. Decided May 24, 1982. Rehearing Denied Oct. 4, 1982. See 459 U.S. 899, 103 S.Ct. 198. Syllabus Section 8(e) of the National Labor Relations Act (Act) proscribes secondary agreements between unions and employers—that is, agreements that require an employer to cease doing business with another party. However, § 8(e) contains a proviso which exempts from the proscription of § 8(e) agreements between a union and an employer in the construction industry concerning the contracting or subcontracting of work to be performed at a construction jobsite. In No. 80-1798, petitioner Woelke & Romero Framing, Inc. (Woelke), and respondent union, in negotiating a new collective-bargaining agreement, reached an impasse over the union's demand for a clause that would prohibit Woelke from subcontracting work at any construction jobsite "except to a person, firm or corporation, party to an appropriate, current labor agreement with the appropriate Union, or subordinate body signatory to this Agreement." When Woelke's construction sites were picketed in support of the union's demand for the subcontracting clause, Woelke filed unfair labor practice charges with the National Labor Relations Board, asserting that subcontracting clauses are sheltered by § 8(e)'s construction industry proviso only if they are limited in application to particular jobsites at which both union and nonunion workers are employed. Woelke argued that because the clause sought by the union violated § 8(e), the picketing violated § 8(b)(4)(A) of the Act, which prohibits coercing an employer "to enter into any agreement which is prohibited by" § 8(e). The Board held that subcontracting clauses are lawful whenever they are sought or negotiated in the context of collective-bargaining relationships, and that therefore picketing to obtain such a clause was permitted under § 8(b)(4)(A). In Nos. 80-1808 and 81-91, a labor dispute resulted in unfair labor practice charges being filed against respondent union by a member (petitioner in No. 80-1808) of an association of construction industry employers (petitioner in No. 81-91), challenging the validity under § 8(e) of a subcontracting clause which was substantially similar to the clause involved in No. 80-1798, and which was included in a collective-bargaining agreement between the union and the association. The Board held that such clause was protected by the construction industry proviso. The Court of Appeals consolidated the petitioners' requests for review of the Board's orders and ultimately decided to enforce the orders, holding that union signatory subcontracting clauses are protected so long as they are negotiated in the context of a collective-bargaining relationship and that picketing may be used to obtain such a clause. Held: 1. The construction industry proviso to § 8(e) ordinarily shelters union signatory subcontracting clauses that are sought or negotiated in the context of a collective-bargaining relationship, even when not limited in application to particular jobsites at which both union and nonunion workers are employed. The subcontracting clauses at issue here are protected by the proviso. Pp. 2076-2083. (a) The plain language and the legislative history of § 8(e) and the construction industry proviso clearly indicate that Congress intended to protect subcontracting clauses like those at issue here. Pp. 2076-2080. (b) The legislative history does not support petitioners' argument that the proviso was intended primarily as a response to the decision in NLRB v. Denver Building & Construction Trades Council, 341 U.S. 675, 71 S.Ct. 943, 95 L.Ed. 1284—which held that picketing a general contractor's entire project in order to protest the presence of a nonunion subcontractor is an illegal secondary boycott—and thus should be interpreted as permitting only those subcontracting agreements that are designed to reduce friction at jobsites where union workers are forced to work alongside nonunion workers. The proviso serves a variety of purposes unrelated to that case's holding, and even as a response to that case is only partly concerned with jobsite friction. Pp. 2080-2081. (c) While subcontracting clauses like those at issue here create "top-down" pressure for unionization—requiring subcontractors, in order to obtain work, to force their employees to become union members—such pressure is implicit in the construction industry proviso and Congress thus decided to accept whatever top-down pressure such clauses might entail. Moreover, the top-down organizing effect of such clauses is limited by other provisions of the Act. Pp. 2081-2083. 2. The Court of Appeals was without jurisdiction to decide that unions do not violate § 8(b)(4)(A) when they picket to obtain a subcontracting clause sheltered by the construction industry proviso. Neither Woelke nor the Board's General Counsel raised that issue during the proceedings before the Board in No. 80-1798, and thus judicial review is barred by § 10(e) of the Act, which provides that "no objection that has not been urged before the Board . . . shall be considered by the court." The § 10(e) bar applies even though the Board held that the picketing was not banned by § 8(b)(4)(A). Woelke's failure to object to the Board's decision in a petition for reconsideration or rehearing prevents consideration of the question by the courts. P. 2083. 654 F.2d 1301, affirmed in part, vacated in part, and remanded. 1 John W. Prager, Jr., Los Angeles, Cal., for Woelke & Romero Framing, Inc. 2 Lewis K. Scott, Portland, Or., for Pacific Northwest Chapter and Oregon-Columbia Chapter of Associated Builders. 3 Norton J. Come, Washington, D.C., for respondent NLRB. 4 Laurence Gold, Washington, D.C., for respondent unions. 5 Justice MARSHALL delivered the opinion of the Court. 6 In these consolidated cases, petitioners ask us to decide whether union signatory subcontracting clauses that are sought or negotiated in the context of a collective-bargaining relationship are protected by the construction industry proviso to § 8(e) of the National Labor Relations Act (Act), 29 U.S.C. § 158(e). Such clauses bar subcontracting except to subcontractors who are signatories to agreements with particular unions. Petitioners also ask us to decide whether a union violates § 8(b)(4)(A) of the Act, 29 U.S.C. § 158(b)(4)(A), when it pickets to obtain a lawful subcontracting clause. 7 The United States Court of Appeals for the Ninth Circuit held that subcontracting clauses sought or negotiated in the context of a collective-bargaining relationship are protected by the construction industry proviso even when not limited in application to particular jobsites at which both union and nonunion workers are employed. It further held that picketing to obtain such clauses does not violate § 8(b)(4)(A). See 654 F.2d 1301 (1981) (en banc). We affirm the holding that the subcontracting clauses at issue here are protected by the construction industry proviso. Because we conclude that the Court of Appeals did not have jurisdiction to consider the picketing question, we do not review that portion of its decision. 8 * A. 9 These cases arise out of two separate labor disputes. The first involves petitioner Woelke & Romero Framing, Inc. (Woelke), a framing subcontractor in the construction industry in southern California. From July 1974 to June 1977, Woelke was party to a collective-bargaining agreement with respondent United Brotherhood of Carpenters and Joiners of America (Carpenters). Shortly before this agreement was to expire, Woelke and Carpenters commenced bargaining for the purpose of negotiating a successor agreement. In August 1977, however, the parties reached an impasse over Carpenters' demand for a union signatory subcontracting clause. This clause would have prohibited Woelke from subcontracting work at any construction jobsite "except to a person, firm or corporation, party to an appropriate, current labor agreement with the appropriate Union, or subordinate body signatory to this Agreement." 1 App. 86.1 10 In support of Carpenters' demand for a subcontracting clause, two Carpenters locals picketed Woelke's construction sites, causing some work stoppages. Woelke filed unfair labor practice charges with the National Labor Relations Board, asserting that the subcontracting clause violated § 8(e) of the Act, which proscribes secondary agreements between unions and employers—that is, agreements that require an employer to cease doing business with another party, in order to influence the labor relations of that party. Woelke argued that because the clause violated § 8(e), Carpenters' picketing in support of that restriction violated § 8(b)(4)(A), 29 U.S.C. § 158(b)(4)(A).2 11 The Board agreed that the union signatory subcontracting clauses at issue were secondary in thrust. It ruled, however, that they were saved by the construction industry proviso to § 8(e), which exempts agreements between a union and employer concerning work to be performed at a construction jobsite. The Board rejected Woelke's contention that subcontracting clauses are sheltered by the proviso only if they are limited in application to particular jobsites at which both union and nonunion workers are employed. According to the Board, such clauses are lawful whenever they are sought or negotiated "in the context of a collective bargaining relationship." Carpenters Local No. 944 (Woelke & Romero Framing, Inc.), 239 N.L.R.B. 241, 250 (1978), citing Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S. 616, 95 S.Ct. 1830, 44 L.Ed.2d 418 (1975). The Board further indicated that since the subcontracting clauses were lawful, picketing to obtain a subcontracting proposal was permitted under § 8(b)(4)(A). Carpenters Local No. 944, supra, at 251. B 12 The second dispute concerns a collective-bargaining agreement between petitioner Oregon-Columbia Chapter of the Associated General Contractors of America, Inc. (Oregon AGC), and respondent Local 701 of the International Union of Operating Engineers, AFL-CIO (Engineers). Oregon AGC is an association of approximately 200 construction industry employers in Oregon and southwest Washington. Since 1960, the contract between Oregon AGC and the Engineers has contained a subcontracting clause prohibiting Oregon AGC from subcontracting construction jobsite work to "any person, firm or company who does not have an existing labor agreement with the [Engineers] Union covering such work." 2 App. 9-10; see id., at 12.3 In addition, the agreement authorized Engineers to take "such action as they deem necessary," including strikes and other economic self-help, to enforce awards obtained through the grievance and arbitration process on matters covered by the agreement. Id., at 10. 13 In April 1977, petitioner Pacific Northwest Chapter of the Associated Builders and Contractors, Inc. (Pacific Northwest), a member of Oregon AGC, filed unfair labor practice charges, asserting that the contract between the Oregon AGC and the Engineers violated § 8(e). Relying on the same reasoning employed in Carpenters Local No. 944, the Board held that the union signatory subcontracting clauses, standing alone, would be protected by the construction industry proviso. International Union of Operating Engineers, Local No. 701 (Pacific Northwest Chapter of Associated Builders & Contractors, Inc.), 239 N.L.R.B. 274, 277 (1978). With one member dissenting, however, it decided that the provision of the contract permitting the union to enforce the subcontracting clause was not protected by the proviso. Id., at 276.4 C 14 Woelke, Oregon AGC, and Pacific Northwest all sought review of the Board's orders in the Court of Appeals.5 The Ninth Circuit panel consolidated the cases and reversed the Board's decisions. It reasoned that the proviso was designed solely to minimize friction between union and nonunion workers employed at the same jobsite. Thus, the proviso shelters subcontracting clauses "only where a collective bargaining relationship exists and even then only when the employer or his subcontractor has employees who are members of the signatory union at work at some time at the jobsite at which the employer wishes to engage a nonunion subcontractor." 609 F.2d 1341, 1347 (1979) (three-judge panel). Because it found that the clauses were unlawful, it did not reach the questions whether picketing or striking either to obtain or enforce a valid subcontracting clause was lawful. Id., at 1351. 15 At respondents' request, the cases were reheard en banc. The en banc panel decided to enforce the Board's orders in their entirety. 654 F.2d 1301 (1981). The majority held that union signatory subcontracting clauses are protected so long as they are sought or negotiated in the context of a collective-bargaining relationship.6 Id., at 1322. It further held that economic pressure may be used to obtain a subcontracting agreement, but that it may not be employed to enforce a subcontracting agreement. Id., at 1323-1324. 16 Woelke, Oregon AGC, and Pacific Northwest asked this Court to review the conclusion that the subcontracting agreements sought by respondents are protected by the construction industry proviso. Woelke also asked this Court to decide whether unions violate § 8(b)(4)(A) when they picket to obtain lawful subcontracting clauses.7 We granted certiorari. 454 U.S. 814, 102 S.Ct. 90, 70 L.Ed.2d 83 (1981). II A. 17 Section 8(e), which was added to the Act by the 1959 Landrum-Griffin Act, Pub.L. 86-257, 73 Stat. 543-544, 29 U.S.C. § 158(e), states: 18 "It shall be an unfair labor practice for any labor organization and any employer to enter into any contract or agreement, express or implied, whereby such employer ceases or refrains or agrees to cease or refrain from handling, using, selling, transporting or otherwise dealing in any of the products of any other employer, or to cease doing business with any other person, and any contract or agreement entered into heretofore or hereafter containing such an agreement shall be to such extent unenforceable and void." 19 The union subcontracting clauses at issue here fall within the general prohibition of § 8(e); they require the general contractor to boycott the services of nonunion subcontractors in order to influence the labor relations policies of the subcontractor. The construction industry proviso to § 8(e) states, however, that 20 "nothing in this subsection (e) shall apply to an agreement between a labor organization and an employer in the construction industry relating to the contracting or subcontracting of work to be done at the site of the construction, alteration, painting, or repair of a building, structure, or other work." 21 Thus, the question we must answer here is whether the union signatory subcontracting clauses sought or obtained by respondent unions are protected by this proviso. 22 Read literally, the proviso would seem to shelter the subcontracting agreements—it expressly states that § 8(e) does not apply to agreements that limit the contracting of construction site work. In Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S., at 628, 95 S.Ct., at 1837, however, this Court warned that § 8(e) "must be interpreted in light of the statutory setting and the circumstances surrounding its enactment." In that case, the Court decided that the proviso did not exempt subcontracting agreements that were not sought or obtained in the context of a collective-bargaining relationship, even though they were covered by the plain language of the statute. The Court reasoned that Congress did not intend to authorize such agreements.8 23 The subcontracting clauses at issue here were sought or negotiated in the context of collective-bargaining relationships. Petitioners argue, however, that we should further confine the scope of the proviso. They contend that Congress designed the proviso to solve the problems that arise when union and nonunion workers are employed at the same jobsite. Thus, it should be interpreted to protect only those agreements that are limited in application to construction projects where both union and nonunion workers are employed. 24 After examining the construction industry proviso "in light of the statutory setting and the circumstances surrounding its enactment," Connell Construction Co., supra, at 628, 95 S.Ct., at 1637, we conclude that it should not be confined as petitioners suggest. The legislative history of § 8(e) and the construction industry proviso clearly indicates that Congress intended to protect subcontracting clauses like those at issue here. B 25 Prior to 1959, there were gaps in the existing protections against secondary boycotts. In Carpenters v. NLRB, 357 U.S. 93, 78 S.Ct. 1011, 2 L.Ed.2d 1186 (1958) (Sand Door), this Court held that a union could not engage in strikes or other concerted activity to enforce "hot cargo" agreements—agreements that required employers to boycott the goods or services of another party with whom the union had a dispute. However, Sand Door indicated that employers and unions were free to enter into hot cargo agreements, and that compliance was lawful so long as it was voluntary. Id., at 108, 78 S.Ct., at 1020. 26 Section 8(e), which prohibits hot cargo agreements, was designed to eliminate the loophole created by the Sand Door decision. See Connell Construction Co., supra, 421 U.S., at 628, 95 S.Ct., at 1837. See also National Woodwork Manufacturers Assn. v. NLRB, 386 U.S. 612, 634, 87 S.Ct. 1250, 1262, 18 L.Ed.2d 357 (1967). The provision represents a compromise between bills reported out by the Senate and House. The Senate bill would have outlawed hot cargo agreements only in the trucking industry. 105 Cong.Rec. 6556 (1959), 2 NLRB, Legislative History of the Labor-Management Reporting and Disclosure Act of 1959, pp. 1161-1162 (1959) (Leg.Hist.). The legislation proposed by the House—the Landrum-Griffin bill—was much broader. It made it an unfair labor practice for any labor organization and any employer to enter into an agreement whereby the employer agrees to "cease doing business with any other person." H.R. 8400, 86th Cong., 1st Sess., § 705(b)(1) (1959), 1 Leg.Hist. 683. The Conference Committee decided to adopt the House bill. However, the Senate conferees insisted on a proviso that exempted hot cargo agreements in the garment industry, and also agreements relating to work to be done at the site of a construction project. 105 Cong.Rec. 17899 (1959), 2 Leg.Hist. 1432. 27 The legislative history contains several references to the construction industry proviso. After noting that the proviso extends only to work to be performed at the site of the construction, the Conference Report states: 28 "The committee of conference does not intend that this proviso should be construed so as to change the present state of the law with respect to the validity of this specific type of agreement relating to work to be done at the site of the construction project or to remove the limitations which the present law imposes with respect to such agreements. Picketing to enforce such contracts would be illegal under the Sand Door case (Local 1976, United Brotherhood of Carpenters v. NLRB, 357 U.S. 93 [78 S.Ct. 1011, 2 L.Ed.2d 1186] (1958)). To the extent that such agreements are legal today under section 8(b)(4) of the National Labor Relations Act, as amended, the proviso would prevent such legality from being affected by section 8(e)." H.R.Conf.Rep.No.1147, 86th Cong., 1st Sess., 39 (1959), 1 Leg.Hist. 943 (emphasis added). 29 Senator John F. Kennedy, who was chairman of the Conference Committee, provided a similar explanation during subsequent congressional debate.9 30 "The first proviso under new section 8(e) of the National Labor Relations Act is intended to preserve the present state of the law with respect to picketing at the site of a construction project and with respect to the validity of agreements relating to the contracting of work to be done at the site of the construction project." 105 Cong.Rec. 17900 (1959), 2 Leg.Hist. 1433. Senator Kennedy also said: 31 "The Landrum-Griffin bill extended the 'hot cargo' provisions of the Senate bill, which we applied only to Teamsters, to all agreements between an employer and a labor union by which the employer agrees not to do business with another concern. The Senate insisted upon a qualification for the clothing and apparel industries and for agreements relating to work to be done at the site of a construction project. Both changes were necessary to avoid serious damage to the pattern of collective bargaining in these industries." 105 Cong.Rec. 17899 (1959), 2 Leg.Hist. 1432. 32 Other legislators expressed similar views. They emphasized that the final bill would not change the law with respect to construction site subcontracting agreements. See 105 Cong.Rec. 18128 (1959), 2 Leg.Hist. 1715 (remarks of Rep. Barden); 105 Cong.Rec. 18135 (1959), 2 Leg.Hist. 1721 (remarks of Rep. Thompson); 105 Cong.Rec. 19849 (1959), 2 Leg.Hist. 1823 (post-enactment memorandum by Sen. Dirksen); 105 Cong.Rec. 19772 (1959), 2 Leg.Hist. 1858 (postenactment memorandum by Sen. Goldwater). 33 These statements reveal that Congress wished "to preserve the status quo" regarding agreements between unions and contractors in the construction industry. National Woodwork Manufacturers Assn., supra, at 637, 87 S.Ct., at 1264. To the extent that subcontracting agreements were part of the pattern of collective bargaining in the construction industry, and lawful, Congress wanted to ensure that they remained lawful. Given this expression of legislative intent, we can determine whether the clauses challenged in these cases are within the scope of the proviso—or whether petitioners' narrow interpretation of the proviso is appropriate—by examining Congress' perceptions regarding the status quo in the construction industry. 34 There is ample evidence that Congress believed that union signatory contract clauses of the type at issue here were part of the pattern of collective bargaining in the construction industry. Comments made by Senator Kennedy clearly indicate that he believed broad subcontracting agreements were legal in 1959: 35 "Agreements by which a contractor in the construction industry promises not to subcontract work on a construction site to a nonunion contractor appear to be legal today. They will not be unlawful under section 8(e). The proviso is also applicable to all other agreements involving undertakings not to do work on a construction project site with other contractors or subcontractors regardless of the precise relation between them." 105 Cong.Rec. 17900 (1959), 2 Leg.Hist. 1433. 36 Senator Kennedy's views were shared by other legislators. Senator Curtis, testifying before the Senate Labor Committee, stated that broad subcontracting agreements were not illegal, and were used "extensively" by the building trades unions. Labor-Management Reform Legislation: Hearings on S. 505, S. 748, S. 76, S. 1002, S. 1137, and S. 1311 before the Subcommittee on Labor of the Senate Committee on Labor and Public Welfare, 86th Cong., 1st Sess., 752 (1959) (Senate Hearings). The House Labor Committee heard similar testimony. Representatives of an employer and an independent union complained that employers and unions could lawfully enter into subcontracting clauses, and that, as a result, employers whose employees had selected another union were denied any opportunity to compete for construction jobs. They described agreements very similar to those at issue here. Labor-Management Reform Legislation: Hearings on H.R. 3540, H.R. 3302, H.R. 4473, and H.R. 4474 before a Joint Subcommittee of the House Committee on Education and Labor, 86th Cong., 1st Sess., 2363 (1959) (statement of Howard Lane) (House Hearings); id., at 2365-2366 (statement of Edward M. Carlton).10 37 Petitioners argue that Congress' perception of the status quo was inaccurate. According to petitioners, subcontracting clauses were not extensively used in the construction industry prior to 1959, and neither the Board nor the courts had ruled that such clauses were lawful. However, "the relevant inquiry is not whether Congress correctly perceived the then state of the law, but rather what its perception of the state of the law was." Brown v. GSA, 425 U.S. 820, 828, 96 S.Ct. 1961, 1965, 48 L.Ed.2d 402 (1976). In any event, Congress' belief that subcontracting agreements were common and lawful was accurate. 38 The Board and the United States Court of Appeals for the District of Columbia Circuit had upheld broad subcontracting clauses. See Associated General Contractors of America, Inc. (St. Maurice, Helmkamp & Musser ), 119 N.L.R.B. 1026 (1957), review denied and enf'd sub nom. Operating Engineers Local Union No. 3 v. NLRB, 105 U.S.App.D.C. 307, 266 F.2d 905, cert. denied sub nom. St. Maurice, Helmkamp & Musser v. NLRB, 361 U.S. 834, 80 S.Ct. 86, 4 L.Ed.2d 76 (1959).11 Significantly, petitioners are unable to point to any pre-1959 cases in which a subcontracting agreement was found to be unlawful because it was not limited to particular jobsites at which the signatory union workers were employed. 39 A report published in 1961, which examined "the prevalence and characteristics of subcontracting provisions in effect in 1959 in the construction industry," indicates that broad subcontracting agreements were quite common. Lunden, Subcontracting Clauses in Major Contracts, 84 Monthly Lab.Rev. 579 (1961). The study examined 155 construction contracts, covering 700,000 construction workers, and found that 444,000 of those workers were employed under contracts with subcontracting provisions. Id., at 582. The most frequent requirement, found in more than 50 major contracts, obligated contractors to subcontract work only to subcontractors who would apply all the "terms and conditions" of the master agreement. Id., at 715-716. The Lunden report does not describe a single agreement that limited the applicability of a subcontracting restriction to jobsites at which both union and nonunion workers were employed.12 40 In short, Congress believed that broad subcontracting clauses similar to those at issue here were part of the pattern of collective bargaining prior to 1959, and that the Board and the courts had found them to be lawful. This perception was apparently accurate. Thus, endorsing the clauses at issue here is fully consistent with the legislative history of § 8(e) and the construction industry proviso. III 41 Petitioners attach little significance to the legislative history we have just described. Instead, they focus on congressional references to this Court's decision in NLRB v. Denver Building & Construction Trades Council, 341 U.S. 675, 71 S.Ct. 943, 95 L.Ed. 1284 (1951) (Denver Building Trades), which they believe support their narrow interpretation of the proviso. They also contend that if the clause is interpreted to protect any subcontracting agreements sought or obtained in the context of a collective-bargaining agreement, unions will have a potent organizational weapon. However, neither of these arguments compels the adoption of a restricted interpretation of the proviso. 42 * Petitioners contend that Congress adopted the construction industry proviso primarily because it wanted to overrule this Court's decision in Denver Building Trades, supra. That case held that picketing a general contractor's entire project in order to protest the presence of a nonunion subcontractor is an illegal secondary boycott. According to petitioners, Congress disliked the Denver Building Trades rule because it might lead to uneasy employee relationships on the jobsite: if union workers were forced to work alongside nonunion workers, friction might result. Given this congressional purpose, the proviso should be interpreted as permitting only those subcontracting agreements that are designed to reduce friction at particular jobsites. 43 Petitioners are correct in suggesting that the decision in Denver Building Trades contributed to Congress' decision to adopt the construction industry proviso. See Connell Construction Co., 421 U.S., at 629, 95 S.Ct., at 1838. At the time Congress was considering the proper scope of § 8(e), it had before it several proposals that would have effectively overruled the Denver Building Trades decision. See, e.g., § 702(d) of H.R. 8342, 86th Cong., 1st Sess. (1959), 1 Leg.Hist. 752-753 (Elliot bill). "It was partly in this frame of reference that the [construction industry] proviso to Section 8(e) was written." 105 Cong.Rec. 20005 (1959), 2 Leg.Hist. 1861 (remarks of Rep. Kearns). 44 It is clear, however, that those who wished to overrule Denver Building Trades were concerned about more than the possibility of jobsite friction. Critics of Denver Building Trades complained that contractors and subcontractors working together on a single construction project are not the sort of neutral parties that the secondary boycott provisions were designed to protect. They pointed out that the Denver Building Trades rule denied construction workers the right to engage in economic picketing at their place of employment. And they emphasized that the employees of various subcontractors have a close community of interest, and that the wages and working conditions of one set of employees may affect others.13 In fact, as the Court of Appeals noted, the problem of jobsite friction between union and nonunion workers received relatively little emphasis. See 654 F.2d, at 1319. 45 The proviso helps mitigate the impact of the Denver Building Trades decision: although it does not overrule the ban on picketing, it confirms that construction industry unions may enter into agreements that would prohibit the subcontracting of jobsite work to nonunion firms. However, petitioners' argument—that the proviso was intended primarily as a response to Denver Building Trades, and that it should therefore be interpreted as protecting only those clauses designed to prevent jobsite friction—rests on faulty premises. As we have already shown, see supra, at 2077-2080 the proviso was not designed solely as a response to the Denver Building Trades problem. And even as a response to Denver Building Trades, the proviso is only partly concerned with jobsite friction.14 B 46 Petitioners further contend that if the subcontracting clauses at issue here are approved, the unions will have a powerful organizing tool. Subcontractors will not be able to obtain work unless their employees are represented by the union. Thus, they will force their employees to become members of the union. In effect, the subcontracting clauses will create a "top-down" pressure for unionization; they will take the representation decision out of the hands of the employees and place it in the hands of the employers. 47 It is undoubtedly true that one of the central aims of the 1959 amendments to the Act was to restrict the ability of unions to engage in top-down organizing campaigns. See Connell Construction Co., supra, at 632, 95 S.Ct., at 1839 (discussing legislative history). It is also true that secondary subcontracting agreements like those at issue here create top-down organizing pressure. However, even if the agreements were limited in application to jobsites at which both union and nonunion workers were employed, there would be some top-down organizing effect. Such pressure is implicit in the construction industry proviso. The bare assertion that a particular subcontracting agreement encourages top-down organizing pressure does not resolve the issue we confront in these cases: how much top-down pressure did Congress intend to tolerate when it decided to exempt construction site projects from § 8(e)? As we have already explained, we believe that Congress endorsed subcontracting agreements obtained in the context of a collective-bargaining relationship—and decided to accept whatever top-down pressure such clauses might entail. Congress concluded that the community of interests on the construction jobsite justified the top-down organizational consequences that might attend the protection of legitimate collective-bargaining objectives.15 48 The top-down organizing effect of subcontracting clauses sought or obtained in the context of a collective-bargaining relationship is limited in a number of ways by other provisions of the National Labor Relations Act.16 A subcontractor cannot be subjected to unlimited picketing to force it into a union agreement without regard to the wishes of its employees. See 29 U.S.C. § 158(b)(7)(C).17 An additional safeguard is provided by § 8(f), 29 U.S.C. § 158(f), which authorizes unions and employers in the construction industry to enter into collective-bargaining agreements, even though the employees of that employer have not designated the union as their lawful bargaining representative. When a union obtains a subcontracting clause from a general contractor, subcontractors frequently attempt to ensure that they remain eligible for work by entering into a § 8(f) agreement—known as a prehire agreement—with the union. If they do so, however, § 8(f) expressly states that their employees may challenge the union's representative status by filing an election petition with the Board. And the subcontractors themselves, if they do not have a stable work force among whom the union has secured a majority, may be free to repudiate the agreement at any project on which the union has not demonstrated that it represents a majority of their employees. See NLRB v. Iron Workers, 434 U.S. 335, 98 S.Ct. 651, 54 L.Ed.2d 586 (1978); Giordano Construction Co., 256 N.L.R.B. 47, 47-48, 107 LRRM 1164, 1165-1166 (1981). 49 Despite petitioners' assertions to the contrary, nonunion employees are not frozen out of the job market by subcontracting agreements. Even where construction unions successfully negotiate collective-bargaining agreements that require both general contractors and subcontractors to obtain their labor from union hiring halls, the union must refer both members and nonmembers to available jobs. 29 U.S.C. §§ 158(a)(3), 158(b)(2).18 In addition, Courts of Appeals have suggested that the obligations of union membership that may be required under union security clauses after seven days are limited to the normal financial obligations of membership.19 50 Finally, since the Denver Building Trades rule remains in effect, employees working for firms with whom a construction union has a primary dispute are protected against secondary picketing designed to force them off their current job. And as the Court of Appeals held in these cases, even where construction unions have negotiated secondary clauses that are sheltered by the proviso, they may not enforce them by picketing or other forms of concerted activity. See H.R.Conf.Rep.No.1147, 86th Cong., 1st Sess., 39 (1959), 1 Leg.Hist. 943; 105 Cong.Rec. 19772 (1959), 2 Leg.Hist. 1858 (postenactment memorandum of Sen. Goldwater). IV 51 Petitioner Woelke asks us to reverse the Court of Appeals' holding that unions do not violate § 8(b)(4)(A) when they picket to obtain a subcontracting clause sheltered by the construction industry proviso. However, the Court of Appeals was without jurisdiction to consider that question. The issue was not raised during the proceedings before the Board, either by the General Counsel or by Woelke. Thus, judicial review is barred by § 10(e) of the Act, 29 U.S.C. § 160(e), which provides that "[n]o objection that has not been urged before the Board . . . shall be considered by the court, unless the failure or neglect to urge such objection shall be excused because of extraordinary circumstances." See De ¢s666¢s Edison Co. v. NLRB, 440 U.S. 301, 311-312, n.10, 99 S.Ct. 1123, 1129-1130, n.10, 59 L.Ed.2d 333 (1979); Garment Workers v. Quality Mfg. Co., 420 U.S. 276, 281, n.3, 95 S.Ct. 972, 975, n.3, 43 L.Ed.2d 189 (1975); NLRB v. Ochoa Fertilizer Corp., 368 U.S. 318, 322, 82 S.Ct. 344, 347, 7 L.Ed.2d 312 (1961). 52 The § 10(e) bar applies even though the Board held that the picketing was not banned by § 8(b)(4)(A). See Carpenters Local No. 944, 239 N.L.R.B., at 251. Woelke could have objected to the Board's decision in a petition for reconsideration or rehearing. The failure to do so prevents consideration of the question by the courts. See Garment Workers v. Quality Mfg. Co., supra, at 281, n.3, 95 S.Ct., at 975, n.3. 53 Because the Court of Appeals lacks jurisdiction to review objections that were not urged before the Board, we do not reach the question whether the picketing was lawful. Instead, we vacate that portion of the Court of Appeals' judgment that relates to this issue, and remand with instructions to dismiss. V 54 We hold that the construction industry proviso to § 8(e) of the National Labor Relations Act ordinarily shelters union signatory subcontracting clauses that are sought or negotiated in the context of a collective-bargaining relationship, even when not limited in application to particular jobsites at which both union and nonunion workers are employed. This interpretation of the proviso is supported by its plain language, as well as the legislative history. Thus, we affirm the decision below, insofar as it holds that the clauses at issue here were sheltered by the proviso. We further hold that the Court of Appeals was without jurisdiction to decide whether a union violates § 8(b)(4)(A) when it pickets to obtain a lawful subcontracting clause. We vacate that portion of the judgment below, and remand for further proceedings consistent with this opinion. 55 It is so ordered. 1 The proposed clause provides in full: "The Contractor agrees that neither he nor any of his subcontractors on the jobsite will subcontract any work to be done at the site of construction, alteration, painting or repair of a building, structure or other work (including quarries, rock, san[d] and gravel plants, asphalt plants, ready-mix concrete plants, established on or adjacent to the jobsite to process or supply materials for the convenience of the Contractor for jobsite use) except to a person, firm or corporation, party to an appropriate, current labor agreement with the appropriate Union, or subordinate body signatory to this Agreement." 1 App. 86. The expiring contract contained a union signatory subcontracting clause that was similar in effect. Id., at 28. 2 Section 8(b)(4)(A) prohibits coercing "any employer or selfemployed person to join any labor or employer organization or to enter into any agreement which is prohibited by section 8(e)." 3 The clause provides in full: "Employers shall not contract any work covered by this Agreement to be done at the site of the construction, alteration, painting, or repair of a building, structure, or other work to any person, firm or company who does not have an existing labor agreement with the Union covering such work." 2 App. 9-10. 4 The Board reasoned that the use of self-help measures would violate § 8(b)(4)(B) of the Act, 29 U.S.C. § 158(b)(4)(B). That section makes it an unfair labor practice for a union to force any person to "cease doing business with any other person." 5 Petitioner Oregon AGC was technically a respondent in the Board proceeding. However, it supported the position taken by Pacific Northwest. 6 The only other Court of Appeals to confront this question reached the same conclusion. See Donald Schriver, Inc. v. NLRB, 204 U.S.App.D.C. 4, 25, 635 F.2d 859, 880 (1980), cert. denied, 451 U.S. 976, 101 S.Ct. 2058, 68 L.Ed.2d 357 (1981), petition for rehearing pending, No. 80-1257. 7 None of the petitioners sought review of the Court of Appeals' decision that economic pressure may not be used to enforce subcontracting agreements. 8 In Connell, the Court was confronted with a novel and apparently fool-proof organizational tactic: "stranger" picketing aimed at pressuring employers with whom the union had no collective-bargaining relationship, and whose employees it had no interest in representing, into signing union signatory subcontracting agreements. Because there was no recognitional objective to the picketing, it did not violate § 8(b)(7), 29 U.S.C. § 158(b)(7). And because the subcontracting clause appeared to be protected by the construction industry proviso, the picketing was arguably not prohibited by § 8(b)(4)(A), 29 U.S.C. § 158(b)(4)(A), which bans picketing to secure agreements made unlawful by § 8(e). The Court concluded, however, that the protection of the proviso "extends only to agreements in the context of collective-bargaining relationships." Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S., at 633, 95 S.Ct., at 1840. In these cases, we decide a question left unresolved in Connell : the extent to which the proviso shelters agreements sought or obtained within the context of a collective-bargaining relationship. 9 Since the proviso was added to § 8(e) at the Senate conferees' insistence, and since Senator Kennedy was chairman of the Senate conferees, his explanation of the clause is entitled to substantial weight. 10 The employer and union representatives who testified before the House Labor Committee, as well as Senator Curtis, opposed the use of subcontracting clauses, arguing that they forced nonunion contractors out of business. House Hearings 2366; Senate Hearings 752. Significantly, despite this testimony, Congress decided to exclude the construction industry from the scope of § 8(e). 11 In Operating Engineers Local Union No. 3 v. NLRB, the Court of Appeals approved the Board's conclusion that a contractor did not violate the Act by complying with a subcontracting clause under which it was not permitted to subcontract engineering work to a firm that did not observe the terms of the union's contract. The employer and union representatives who testified before the House Labor Committee referred to the case. House Hearings 2364, 2367. The text of the Court of Appeals decision was introduced into the record of the House Hearings. Id., at 801, 803-807. 12 Petitioners suggest that the Lunden study did not adequately distinguish between broad union signatory clauses like those at issue here and union standards clauses—clauses that permit general contractors to subcontract to nonunion subcontractors, so long as the subcontractors are willing to comply with the standards set forth in the union's contract. However, the Board's General Counsel, who conducted his own study of subcontracting restrictions in the construction industry, and who did distinguish between various types of restrictions, found union signatory clauses very similar to those at issue here in roughly 12% of the contracts studied. NLRB General Counsel's Memorandum, Dec. 15, 1976, 93 Lab.Rel.Rep. 390, 404, reprinted in 1976 Lab.Rel. Yearbook 295, 309. Broad subcontracting clauses have been part of the pattern of collective bargaining in southern California, where the Woelke & Romero case arose, since 1941. See Pierson, Building-Trades Bargaining Plan in Southern California, 70 Monthly Lab.Rev. 14, 17 (1950); see also 1 App. 20-22. The Pierson study also noted that similar bargaining arrangements existed in other localities, including the Portland, Ore., area, where the Pacific Northwest case arose. Pierson, supra, at 15. See also Aaron, The Labor-Management Reporting and Disclosure Act of 1959, 73 Harv.L.Rev. 1086, 1119 (1960) (construction industry proviso "approves the general practice of the building trades to secure from a contractor a promise that he will not subcontract work on the jobsite to a nonunion subcontractor"). 13 105 Cong.Rec. 17881, 2 Leg.Hist. 1425 (remarks of Sen. Morse); 105 Cong.Rec. 15541 (1959), 2 Leg.Hist. 1577 (memorandum by Reps. Thompson and Udall); 105 Cong.Rec. 15551-15552 (1959), 2 Leg.Hist. 1588 (memorandum by Rep. Elliott); 105 Cong.Rec. 15852 (1959), 2 Leg.Hist. 1684 (remarks by Rep. Goodell). 14 It is important to recognize, however, that reducing jobsite friction is a legitimate purpose. The clauses at issue here serve this goal by ensuring that members of the respondent unions need not work alongside nonunion employees. 15 It may be true, as petitioners emphasize, that the use of union signatory subcontracting clauses will give a particular union a monopoly position in a labor market. However, the Board has previously suggested that in most labor markets, apart from some minor overlaps, there is only one union representing a particular craft. See Carpenters Local 15, 240 N.L.R.B. 252, 261 (1979). In addition, as Congress recognized, see S.Rep.No.187, 86th Cong., 1st Sess., 28 (1959), U.S.Code Cong. & Admin.News 1959, p. 2318, a majority of the workers skilled in a particular craft will belong to that union. 16 Many of these protections would not have been available to limit the top-down organizing effect of the clauses at issue in Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S. 616, 95 S.Ct. 1830, 44 L.Ed.2d 418 (1975). 17 As we note below, see infra, at 2082-2083, we do not reach the question whether a union may use economic pressure to obtain a subcontracting agreement. We also do not reach the question whether a union may picket to obtain a prehire agreement. 18 See Radio Officers v. NLRB, 347 U.S. 17, 40-42, 74 S.Ct. 323, 335-336, 98 L.Ed. 455 (1954); Teamsters v. NLRB, 365 U.S. 667, 673-677, 81 S.Ct. 835, 838-840, 6 L.Ed.2d 11 (1961). 19 See NLRB v. Hershey Foods Corp., 513 F.2d 1083, 1085-1087 (CA9 1975); Local 1104, Communications Workers of America v. NLRB, 520 F.2d 411, 417-420 (CA2 1975), cert. denied, 423 U.S. 1051, 96 S.Ct. 778, 46 L.Ed.2d 639 (1976).

67

456 U.S. 667 102 S.Ct. 2083 72 L.Ed.2d 416 OREGON, Petitionerv.Bruce Alan KENNEDY. No. 80-1991. Argued March 29, 1982. Decided May 24, 1982. Syllabus During respondent's trial for theft in an Oregon state court, the State's expert witness testified as to the value and identity of the property involved. On cross-examination, he acknowledged that he had once filed an unrelated criminal complaint against respondent, but explained that no action had been taken on his complaint. On redirect examination, the court sustained a series of objections to the prosecutor's questions seeking to establish the reasons why the witness had filed a complaint against respondent. After eliciting from the witness that he had never done business with respondent, the prosecutor asked: "Is that because he is a crook?" The trial court then granted respondent's motion for a mistrial. On retrial, the court rejected respondent's contention that the Double Jeopardy Clause of the Fifth Amendment, as made applicable to the States by the Fourteenth Amendment, barred further prosecution, finding that "it was not the intention of the prosecutor in this case to cause a mistrial." Respondent was convicted, but the Oregon Court of Appeals reversed, sustaining the double jeopardy claim because the prosecutorial misconduct that had occasioned the mistrial, even if not intended to cause a mistrial, amounted to "overreaching." Held: 1. There is no merit to respondent's contentions that the Court of Appeals' decision was based upon an adequate and independent state ground, or that in the alternative the case should be remanded in order that the court may clarify the grounds upon which its judgment rested. A fair reading of the opinion below shows that the court rested its decision solely on federal law. Pp. 670-671. 2. Where a defendant in a criminal trial successfully moves for a mistrial, he may invoke the bar of double jeopardy in a second effort to try him only if the conduct giving rise to the successful motion for a mistrial was prosecutorial or judicial conduct intended to provoke the defendant into moving for a mistrial. A more general test of "overreaching" is rejected because it offers virtually no standards for its application and because such a rule may not aid defendants as a class. By contrast, a standard that examines the prosecutor's intent is a manageable standard to apply. Since the courts below both agreed that the prosecutor did not intend her conduct to provoke respondent into moving for a mistrial, that is the end of the matter for purposes of the Double Jeopardy Clause. Pp. 671-679. 49 Or.App. 415, 619 P.2d 948, reversed and remanded. David B. Frohnmayer, Atty. Gen., Salem, Or., for petitioner. Samuel A. Alito, Jr., Newark, N. J., for the United States as amicus curiae, by special leave of Court. Donald C. Walker, Portland, Or., for respondent. Justice REHNQUIST delivered the opinion of the Court. 1 The Oregon Court of Appeals decided that the Double Jeopardy Clause of the Fifth Amendment to the United States Constitution barred respondent's retrial after his first trial ended in a mistrial granted on his own motion. 49 Or.App. 415, 619 P.2d 948 (1980), cert. granted, 454 U.S. 891, 102 S.Ct. 385, 70 L.Ed.2d 205 (1981). The Court of Appeals concluded that retrial was barred because the prosecutorial misconduct that occasioned the mistrial in the first instance amounted to "overreaching." Because that court took an overly expansive view of the application of the Double Jeopardy Clause following a mistrial resulting from the defendant's own motion, we reverse its judgment. 2 * Respondent was charged with the theft of an oriental rug. During his first trial, the State called an expert witness on the subject of Middle Eastern rugs to testify as to the value and the identity of the rug in question. On cross-examination, respondent's attorney apparently attempted to establish bias on the part of the expert witness by asking him whether he had filed a criminal complaint against respondent. The witness eventually acknowledged this fact, but explained that no action had been taken on his complaint. On redirect examination, the prosecutor sought to elicit the reasons why the witness had filed a complaint against respondent, but the trial court sustained a series of objections to this line of inquiry.1 The following colloquy then ensued: 3 "Prosecutor: Have you ever done business with the Kennedys? 4 "Witness: No, I have not. 5 "Prosecutor: Is that because he is a crook?" 6 The trial court then granted respondent's motion for a mistrial. 7 When the State later sought to retry respondent, he moved to dismiss the charges because of double jeopardy. After a hearing at which the prosecutor testified, the trial court2 found as a fact that "it was not the intention of the prosecutor in this case to cause a mistrial." 49 Or.App., at 418, 619 P.2d, at 949. On the basis of this finding, the trial court held that double jeopardy principles did not bar retrial, and respondent was then tried and convicted. 8 Respondent then successfully appealed to the Oregon Court of Appeals, which sustained his double jeopardy claim. That court set out what it considered to be the governing principles in this kind of case: 9 "The general rule is said to be that the double jeopardy clause does not bar reprosecution, '. . . where circumstances develop not attributable to prosecutorial or judicial overreaching, . . . even if defendant's motion is necessitated by a prosecutorial error.' United States v. Jorn, 400 U.S. 470, 485 [91 S.Ct. 547, 557, 27 L.Ed.2d 543] (197[1] ). However, retrial is barred where the error that prompted the mistrial is intended to provoke a mistrial or is 'motivated by bad faith or undertaken to harass or prejudice' the defendant. United States v. Dinitz, 424 U.S. 600, 611 [96 S.Ct. 1075, 1081, 47 L.Ed.2d 267] (1976). Accord, State v. Rathbun, 37 Or.App. 259, 586 P.2d 1136 (1978), reversed on other grounds, 287 Or. 421, [600] P.2d [392] [329] (1979)." Id., at 417-418, 619 P.2d, at 949. 10 The Court of Appeals accepted the trial court's finding that it was not the intent of the prosecutor to cause a mistrial. Nevertheless, the court held that retrial was barred because the prosecutor's conduct in this case constituted what it viewed as "overreaching." Although the prosecutor intended to rehabilitate the witness, the Court of Appeals expressed the view that the question was in fact "a direct personal attack on the general character of the defendant." Id., at 418, 619 P.2d, at 949. This personal attack left respondent with a "Hobson's choice—either to accept a necessarily prejudiced jury, or to move for a mistrial and face the process of being retried at a later time." Id., at 418, 619 P.2d, at 950. 11 Before turning to the merits of the double jeopardy claim, we are met with the respondent's contention that the Court of Appeals' decision is based upon an adequate and independent state ground. Respondent contends in the alternative that the basis for the decision below is sufficiently uncertain that we ought to remand this case in order that the Court of Appeals may clarify the grounds upon which its judgment rested. See Delaware v. Prouse, 440 U.S. 648, 652, 99 S.Ct. 1391, 1395, 59 L.Ed.2d 660 (1979); California v. Krivda, 409 U.S. 33, 35, 93 S.Ct. 32, 33, 34 L.Ed.2d 45 (1972). 12 We reject both of these contentions. A fair reading of the opinion below convinces us that the Court of Appeals rested its decision solely on federal law. With one exception, the cases it cited in outlining the "general rule" that guided its decision are decisions of this Court. The Court of Appeals' citation to State v. Rathbun, 37 Or.App. 259, 586 P.2d 1136 (1978), rev'd, 287 Or. 421, 600 P.2d 392 (1979), was clearly to its own decision in that case, rather than the decision of the Oregon Supreme Court. Although the Supreme Court's decision in Rathbun was based on state statutory and constitutional grounds, the Court of Appeals' decision in Rathbun clearly rested on federal grounds, a fact which was so recognized by the Oregon Supreme Court. Id., at 430-431, 600 P.2d, at 396-397. Even if the case admitted of more doubt as to whether federal and state grounds for decision were intermixed, the fact that the state court relied to the extent it did on federal grounds requires us to reach the merits. Zacchini v. Scripps-Howard Broadcasting Co., 433 U.S. 562, 568, 97 S.Ct. 2849, 2853, 53 L.Ed.2d 965 (1977). II 13 The Double Jeopardy Clause of the Fifth Amendment3 protects a criminal defendant from repeated prosecutions for the same offense. United States v. Dinitz, 424 U.S. 600, 606, 96 S.Ct. 1075, 1079, 47 L.Ed.2d 267 (1976). As a part of this protection against multiple prosecutions, the Double Jeopardy Clause affords a criminal defendant a "valued right to have his trial completed by a particular tribunal." Wade v. Hunter, 336 U.S. 684, 689, 69 S.Ct. 834, 837, 93 L.Ed. 974 (1949). The Double Jeopardy Clause, however, does not offer a guarantee to the defendant that the State will vindicate its societal interest in the enforcement of the criminal laws in one proceeding. United States v. Jorn, 400 U.S. 470, 483-484, 91 S.Ct. 547, 556, 27 L.Ed.2d 543 (1971) (plurality opinion); Wade v. Hunter, 336 U.S., at 689, 69 S.Ct., at 837. If the law were otherwise, "the purpose of law to protect society from those guilty of crimes frequently would be frustrated by denying courts power to put the defendant to trial again." Ibid. 14 Where the trial is terminated over the objection of the defendant, the classical test for lifting the double jeopardy bar to a second trial is the "manifest necessity" standard first enunciated in Justice Story's opinion for the Court in United States v. Perez, 9 Wheat. 579, 580, 6 L.Ed. 165 (1824). Perez dealt with the most common form of "manifest necessity": a mistrial declared by the judge following the jury's declaration that it was unable to reach a verdict. While other situations have been recognized by our cases as meeting the "manifest necessity" standard, the hung jury remains the prototypical example. See, e.g., Arizona v. Washington, 434 U.S. 497, 509, 98 S.Ct. 824, 832, 54 L.Ed.2d 717 (1978); Illinois v. Somerville, 410 U.S. 458, 463, 93 S.Ct. 1066, 1070, 35 L.Ed.2d 425 (1973). The "manifest necessity" standard provides sufficient protection to the defendant's interests in having his case finally decided by the jury first selected while at the same time maintaining "the public's interest in fair trials designed to end in just judgments." Wade v. Hunter, supra, 336 U.S. at 689, 69 S.Ct., at 837. 15 But in the case of a mistrial declared at the behest of the defendant, quite different principles come into play. Here the defendant himself has elected to terminate the proceedings against him, and the "manifest necessity" standard has no place in the application of the Double Jeopardy Clause. United States v. Dinitz, supra, 424 U.S. at 607-610, 96 S.Ct., at 1079-81. Indeed, inUnited States v. Tateo, 377 U.S. 463, 467, 84 S.Ct. 1587, 1589, 12 L.Ed.2d 448 (1964), the Court stated: "If Tateo had requested a mistrial on the basis of the judge's comments, there would be no doubt that if he had been successful, the Government would not have been barred from retrying him" (emphasis in original). 16 Our cases, however, have indicated that even where the defendant moves for a mistrial, there is a narrow exception to the rule that the Double Jeopardy Clause is no bar to retrial. See, e.g., United States v. DiFrancesco, 449 U.S. 117, 130, 101 S.Ct. 426, 433, 66 L.Ed.2d 328 (1980); United States v. Dinitz, supra, 424 U.S. at 611, 96 S.Ct., at 1081; United States v. Jorn, supra, 400 U.S. at 485, 91 S.Ct., at 557; United States v. Tateo, supra, 377 U.S. at 468, n. 3, 84 S.Ct., at 1590 n. 3. The circumstances under which respondent's first trial was terminated require us to delineate the bounds of that exception more fully than we have in previous cases. 17 Since one of the principal threads making up the protection embodied in the Double Jeopardy Clause is the right of the defendant to have his trial completed before the first jury empaneled to try him, it may be wondered as a matter of original inquiry why the defendant's election to terminate the first trial by his own motion should not be deemed a renunciation of that right for all purposes. We have recognized, however, that there would be great difficulty in applying such a rule where the prosecutor's actions giving rise to the motion for mistrial were done "in order to goad the [defendant] into requesting a mistrial." United States v. Dinitz, supra, 424 U.S. at 611, 96 S.Ct., at 1081.4 In such a case, the defendant's valued right to complete his trial before the first jury would be a hollow shell if the inevitable motion for mistrial were held to prevent a later invocation of the bar of double jeopardy in all circumstances. But the precise phrasing of the circumstances which will allow a defendant to interpose the defense of double jeopardy to a second prosecution where the first has terminated on his own motion for a mistrial have been stated with less than crystal clarity in our cases which deal with this area of the law. In United States v. Dinitz, 424 U.S., at 611, 96 S.Ct., at 1081, we said: 18 "The Double Jeopardy Clause does protect a defendant against governmental actions intended to provoke mistrial requests and thereby to subject defendants to the substantial burdens imposed by multiple prosecutions." 19 This language would seem to follow the rule of United States v. Tateo, supra, 377 U.S. at 468, n. 3, 84 S.Ct., at 1590 n. 3, in limiting the exception to cases of governmental actions intended to provoke mistrial requests. But immediately following the quoted language we went on to say: 20 "[The Double Jeopardy Clause] bars retrials where 'bad-faith conduct by judge or prosecutor,' threatens the '[h]arassment of an accused by successive prosecutions or declaration of a mistrial so as to afford the prosecution a more favorable opportunity to convict' the defendant." United States v. Dinitz, 424 U.S., at 611, 96 S.Ct., at 1081 (citation omitted). 21 The language just quoted would seem to broaden the test from one of intent to provoke a motion for a mistrial to a more generalized standard of "bad faith conduct" or "harassment" on the part of the judge or prosecutor. It was upon this language that the Oregon Court of Appeals apparently relied in concluding that the prosecutor's colloquy with the expert witness in this case amount to "overreaching." 22 The difficulty with the more general standards which would permit a broader exception than one merely based on intent is that they offer virtually no standards for their application. Every act on the part of a rational prosecutor during a trial is designed to "prejudice" the defendant by placing before the judge or jury evidence leading to a finding of his guilt. Given the complexity of the rules of evidence, it will be a rare trial of any complexity in which some proffered evidence by the prosecutor or by the defendant's attorney will not be found objectionable by the trial court. Most such objections are undoubtedly curable by simply refusing to allow the proffered evidence to be admitted, or in the case of a particular line of inquiry taken by counsel with a witness, by an admonition to desist from a particular line of inquiry. 23 More serious infractions on the part of the prosecutor may provoke a motion for mistrial on the part of the defendant, and may in the view of the trial court warrant the granting of such a motion. The "overreaching" standard applied by the court below and urged today by Justice STEVENS, however, would add another classification of prosecutorial error, one requiring dismissal of the indictment, but without supplying any standard by which to assess that error.5 24 By contrast, a standard that examines the intent of the prosecutor, though certainly not entirely free from practical difficulties, is a manageable standard to apply. It merely calls for the court to make a finding of fact. Inferring the existence or nonexistence of intent from objective facts and circumstances is a familiar process in our criminal justice system. When it is remembered that resolution of double jeopardy questions by state trial courts are reviewable not only within the state court system, but in the federal court system on habeas corpus as well, the desirability of an easily applied principle is apparent. 25 Prosecutorial conduct that might be viewed as harassment or overreaching, even if sufficient to justify a mistrial on defendant's motion, therefore, does not bar retrial absent intent on the part of the prosecutor to subvert the protections afforded by the Double Jeopardy Clause. A defendant's motion for a mistrial constitutes "a deliberate election on his part to forgo his valued right to have his guilt or innocence determined before the first trier of fact." United States v. Scott, 437 U.S. 82, 93, 98 S.Ct. 2187, 2195, 57 L.Ed.2d 65 (1978). Where prosecutorial error even of a degree sufficient to warrant a mistrial has occurred, "[t]he important consideration, for purposes of the Double Jeopardy Clause, is that the defendant retain primary control over the course to be followed in the event of such error." United States v. Dinitz, supra, 424 U.S., at 609, 96 S.Ct., at 1080. Only where the governmental conduct in question is intended to "goad" the defendant into moving for a mistrial may a defendant raise the bar of double jeopardy to a second trial after having succeeded in aborting the first on his own motion. 26 Were we to embrace the broad and somewhat amorphous standard adopted by the Oregon Court of Appeals, we are not sure that criminal defendants as a class would be aided. Knowing that the granting of the defendant's motion for mistrial would all but inevitably bring with it an attempt to bar a second trial on grounds of double jeopardy, the judge presiding over the first trial might well be more loath to grant a defendant's motion for mistrial.6 If a mistrial were in fact warranted under the applicable law, of course, the defendant could in many instances successfully appeal a judgment of conviction on the same grounds that he urged a mistrial, and the Double Jeopardy Clause would present no bar to retrial.7 But some of the advantages secured to him by the Double Jeopardy Clause—the freedom from extended anxiety, and the necessity to confront the government's case only once—would be to a large extent lost in the process of trial to verdict, reversal on appeal, and subsequent retrial. See United States v. Dinitz, 424 U.S., at 608, 96 S.Ct., at 1080. 27 In adopting the position we now do, we recognize that language taken from our earlier opinions may well suggest a broader rule.8 The Court of Appeals in this case, for example, may have derived its "overreaching" standard from the following language in the plurality opinion in United States v. Jorn, 400 U.S., at 485, 91 S.Ct., at 557: 28 "Thus, where circumstances develop not attributable to prosecutorial or judicial overreaching, a motion by defendant for mistrial is ordinarily assumed to remove any barrier to reprosecution, even if the defendant's motion is necessitated by prosecutorial or judicial error." 29 A footnote attached to this sentence explains, however, that "where a defendant's mistrial motion is necessitated by judicial or prosecutorial impropriety designed to avoid an acquittal, reprosecution might well be barred." Id., at 485, n. 12, 91 S.Ct., at 557 n. 12. There are likewise statements in United States v. Dinitz, supra, 424 U.S. at 611, 96 S.Ct., at 1081, based largely on the plurality opinion in Jorn to the same effect. 30 Because of the confusion which these varying statements of the standard in question have occasioned in other courts, we deem it best to acknowledge the confusion and its justifiability in the light of these statements from previous decisions. We do not by this opinion lay down a flat rule that where a defendant in a criminal trial successfully moves for a mistrial, he may not thereafter invoke the bar of double jeopardy against a second trial. But we do hold that the circumstances under which such a defendant may invoke the bar of double jeopardy in a second effort to try him are limited to those cases in which the conduct giving rise to the successful motion for a mistrial was intended to provoke the defendant into moving for a mistrial. 31 Since the Oregon trial court found, and the Oregon Court of Appeals accepted, that the prosecutorial conduct culminating in the termination of the first trial in this case was not so intended by the prosecutor, that is the end of the matter for purposes of the Double Jeopardy Clause of the Fifth Amendment to the United States Constitution. The judgment of the Oregon Court of Appeals is reversed, and the cause is remanded for further proceedings not inconsistent with this opinion. 32 It is so ordered. 33 Justice POWELL, concurring. 34 I join the Court's opinion holding that the intention of a prosecutor determines whether his conduct, viewed by the defendant and the court as justifying a mistrial, bars a retrial of the defendant under the Double Jeopardy Clause. Because "subjective" intent often may be unknowable, I emphasize that a court—in considering a double jeopardy motion—should rely primarily upon the objective facts and circumstances of the particular case. See ante, at 675. 35 In the present case the mistrial arose from the prosecutor's conduct in pursuing a line of redirect examination of a key witness. The Oregon Court of Appeals identified a single question as constituting "overreaching" so serious as to bar a retrial. Yet, there are few vigorously contested lawsuits—whether criminal or civil—in which improper questions are not asked. Our system is adversarial and vigorous advocacy is encouraged. 36 Nevertheless, this would have been a close case for me if there had been substantial factual evidence of intent beyond the question itself. Here, however, other relevant facts and circumstances strongly support the view that prosecutorial intent to cause a mistrial was absent. First, there was no sequence of overreaching prior to the single prejudicial question. See ante, at 669-670, and n. 1. Moreover, it is evident from a colloquy between counsel and the court, out of the presence of the jury, that the prosecutor not only resisted, but also was surprised by, the defendant's motion for a mistrial. See App. 24-29. Finally, at the hearing on respondent's double jeopardy motion, the prosecutor testified—and the trial found as a fact and the appellate court agreed—that there was no " 'intention . . . to cause a mistrial.' " Ante, at 669. 37 In view of these circumstances, the Double Jeopardy Clause provides no bar to retrial. 38 Justice BRENNAN, with whom Justice MARSHALL joins, concurring in the judgment. 39 I concur in the judgment and join in the opinion of Justice STEVENS. However, it should be noted that nothing in the holding of the Court today prevents the state courts, on remand, from concluding that respondent's retrial would violate the provision of the Oregon Constitution that prohibits double jeopardy, Ore.Const., Art. I, § 12, as that provision has been interpreted by the state courts, State v. Rathbun, 287 Or. 421, 600 P.2d 392 (1979). See South Dakota v. Opperman, 428 U.S. 364, 396, 96 S.Ct. 3092, 3110, 49 L.Ed.2d 1000 (1976) (MARSHALL, J., dissenting), on remand, State v. Opperman, 247 N.W.2d 673 (S.D.1976) (original State Supreme Court judgment adhered to as a matter of state constitutional law); Oregon v. Hass, 420 U.S. 714, 726, 95 S.Ct. 1215, 1223, 43 L.Ed.2d 570 (1975) (MARSHALL, J., dissenting). 40 Justice STEVENS, with whom Justice BRENNAN, Justice MARSHALL, and Justice BLACKMUN join, concurring in the judgment. 41 Unless the Oregon Court of Appeals based its decision on Oregon law,1 this is a case in which the state court should have applied the general rule that a defendant's motion for a mistrial removes any double jeopardy bar to retrial. The prosecutor's mistake was not the kind of overreaching or harassment identified in our precedents as an exception to the general rule. Instead of explaining why that conclusion is required by settled law, this Court gratuitously lops off a portion of the previously recognized exception. This exercise in lawmaking is objectionable because it is wholly unnecessary and because it compromises an important protection provided by the Double Jeopardy Clause. 42 * The Double Jeopardy Clause represents a constitutional policy of finality for the defendant's benefit in criminal proceedings.2 If the defendant is acquitted by the jury, or if he is convicted and the conviction is upheld on appeal, he may not be prosecuted again for the same offense.3 The defendant's interest in finality is not confined to final judgments; he also has a protected interest in having his guilt or innocence decided in one proceeding.4 That interest must be balanced against society's interest in affording the prosecutor one full and fair opportunity to present his evidence to the jury.5 Our decisions in the mistrial setting accordingly have accommodated the defendant's double jeopardy interests6 with legitimate prosecutorial interests.7 43 The accommodation is reflected in two general rules that govern the permissibility of reprosecution after a mistrial. Which general rule applies turns on whether the defendant has retained control over the course to be followed once error has substantially tainted the initial proceeding.8 When a mistrial is declared over the defendant's objection, the general rule is that retrial is barred.9 An exception to this general rule exists for cases in which the mistrial was justified by "manifest necessity."10 The other general rule is that the defendant's motion for, or consent to, a mistrial removes any double jeopardy bar to reprosecution.11 There is an exception to this rule for cases in which the prosecutor12 intended to provoke a mistrial or otherwise engaged in "overreaching" or "harassment."13 The prosecutor has the burden of proving the former exception for manifest necessity, and the defendant has the burden of proving the latter exception for overreaching. 44 As an initial matter, it is useful to explain why the defendant's retention of control over the course to be followed once serious prosecutorial error has occurred represents a reasonable accommodation of double jeopardy and prosecutorial interests. A defendant cannot be guaranteed both that there will be only one proceeding and that it will be free of error.14 When unfair prejudice is injected into the proceeding by the prosecutor, the defendant may choose to continue the proceeding despite the taint and, if convicted, seek a reversal on appeal.15 Or he may choose to abort the tainted proceeding and begin anew.16 While it is true that prosecutorial error leaves the defendant with a "Hobson's choice,"17 it is also true that the prosecutor suffers substantial costs no matter how the defendant exercises this choice. If the defendant consents to a mistrial, the prosecutor must go to the time, trouble, and expense of starting all over with the criminal prosecution. If the defendant chooses to continue the proceeding and preserve his objection for appeal, the prosecutor must continue to completion a proceeding in which a conviction may not be sustainable.18 45 The rationale for the exception to the general rule permitting retrial after a mistrial declared with the defendant's consent is illustrated by the situation in which the prosecutor commits prejudicial error with the intent to provoke a mistrial.19 In this situation the defendant's choice to continue the tainted proceeding or to abort the proceeding and begin anew is inadequate to protect his double jeopardy interests. For, absent a bar to reprosecution, the defendant would simply play into the prosecutor's hands by moving for a mistrial. The defendant's other option—to continue the tainted proceeding—would be no option at all if, as we might expect given the prosecutor's intent, the prosecutorial error has virtually guaranteed conviction. There is no room in the balance of competing interests for this type of manipulation of the mistrial device. Or to put it another way, whereas we tolerate some incidental infringement upon a defendant's double jeopardy interests for the sake of society's interest in obtaining a verdict of guilt or innocence, when the prosecutor seeks to obtain an advantage by intentionally subverting double jeopardy interests, the balance invariably tips in favor of a bar to reprosecution.20 46 Today the Court once again recognizes that the exception properly encompasses the situation in which the prosecutor commits prejudicial error with the intent to provoke a mistrial. But the Court reaches out to limit the exception to that one situation,21 rejecting the previous recognition that prosecutorial overreaching or harassment is also within the exception.22 47 Even if I agreed that the balance of competing interests tipped in favor of a bar to reprosecution only in the situation in which the prosecutor intended to provoke a mistrial, I would not subscribe to a standard that conditioned such a bar on the determination that the prosecutor harbored such intent when he committed prejudicial error. It is almost inconceivable23 that a defendant could prove that the prosecutor's deliberate misconduct was motivated by an intent to provoke a mistrial instead of an intent simply to prejudice the defendant.24 The defendant must shoulder a strong burden to establish a bar to reprosecution when he has consented to the mistrial, but the Court's subjective intent standard would eviscerate the exception.25 48 A broader objection to the Court's limitation of the exception is that the rationale for the exception extends beyond the situation in which the prosecutor intends to provoke a mistrial. There are other situations in which the defendant's double jeopardy interests outweigh society's interest in obtaining a judgment on the merits even though the defendant has moved for a mistrial. For example, a prosecutor may be interested in putting the defendant through the embarrassment, expense, and ordeal of criminal proceedings even if he cannot obtain a conviction.26 In such a case, with the purpose of harassing the defendant the prosecutor may commit repeated prejudicial errors and be indifferent between a mistrial or mistrials and an unsustainable conviction or convictions. Another example is when the prosecutor seeks to inject enough unfair prejudice into the trial to ensure a conviction but not so much as to cause a reversal of that conviction.27 This kind of overreaching would not be covered by the Court's standard because, by hypothesis, the prosecutor's intent is to obtain a conviction, not to provoke a mistrial. Yet the defendant's choice—to continue the tainted proceeding or to abort it and begin anew—can be just as "hollow"28 in this situation as when the prosecutor intends to provoke a mistrial. 49 To invoke the exception for overreaching, a court need not divine the exact motivation for the prosecutorial error. It is sufficient that the court is persuaded that egregious prosecutorial misconduct has rendered unmeaningful the defendant's choice to continue or to abort the proceeding. It is unnecessary and unwise to attempt to identify all the factors that might inform the court's judgment, but several considerations follow from the rationale for recognizing the exception. First, because the exception is justified by the intolerance of intentional manipulation of the defendant's double jeopardy interests, a finding of deliberate misconduct normally would be a prerequisite to a reprosecution bar.29 Second, because the defendant's option to abort the proceeding after prosecutorial misconduct would retain real meaning for the defendant in any case in which the trial was going badly for him,30 normally a required finding would be that the prosecutorial error virtually eliminated, or at least substantially reduced, the probability of acquittal in a proceeding that was going badly for the government.31 It should be apparent from these observations that only in a rare and compelling case will a mistrial declared at the request of the defendant or with his consent bar a retrial. 50 No one case, of course, is a proper vehicle for identifying the limits of the exception. The Court repeatedly has shunned inflexible standards in applying the comparable "manifest necessity" exception to the general rule that a defendant is entitled to go to final judgment before the initial tribunal.32 The value of the overreaching standard, like "[t]he value of the [manifest necessity standard,] thus lies in [its] capacity for informed application under widely different circumstances without injury to defendants or to the public interest." Wade v. Hunter, 336 U.S. 684, 691, 69 S.Ct. 834, 838, 93 L.Ed. 974. The inexactitude of the standard used to protect defendants in the exceptional case surely should not concern the Court any more than the equally ill-defined formula used to protect prosecutors in the exceptional case. The scarcity of cases in which the exception has been invoked33 counsels against pre-empting the judgment reflected in our decisions that an exception for overreaching or harassment should remain available for the rare case in which it may be needed.34 We should simply decide this case on its facts, as we did in United States v. Dinitz and Lee v. United States,35 and thereby continue to give meaning to the "abstract formula"36 in the context of actual cases. II 51 The petitioner,37 and the state court that denied the respondent's motion to dismiss,38 have correctly pointed out that it is unnecessary to cut back on the recognized exception, or even to disavow the most liberal construction given it by the federal courts, to conclude that the exception has not been established on the facts of this case. The isolated prosecutorial error occurred early in the trial, too early to determine whether the case was going badly for the prosecution. If anyone was being harassed at that time, it was the prosecutor, who was frustrated by improper defense objections in her attempt to rehabilitate her witness. The gist of the comment that the respondent was a "crook" could fairly have been elicited from the witness, since defense counsel injected the respondent's past alleged improprieties into the trial by questioning the witness about his bias towards the defendant. The comment therefore could not have injected the kind of prejudice that would render unmeaningful the defendant's option to proceed with the trial. 52 Because the present case quite clearly does not come within the recognized exception, I join the Court's judgment. I cannot, however, join the Court's opinion because it totally fails to justify its disavowal of the Court's precedents. 1 The Court of Appeals later explained that respondent's "objections were not well taken, and the judge's rulings were probably wrong." 49 Or.App. 415, 417, 619 P.2d 948, 949 (1980). 2 These proceedings were not conducted by the same trial judge who presided over respondent's initial trial. 3 This Court held in Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969), that this Clause was made applicable to the States through the Due Process Clause of the Fourteenth Amendment. 4 Cf. United States v. Tateo, 377 U.S. 463, 468, n. 3, 84 S.Ct. 1587, 1590, n. 3, 12 L.Ed.2d 448 (1964) ("If there were any intimation in a case that prosecutorial or judicial impropriety justifying a mistrial resulted from a fear that the jury was likely to acquit the accused, different considerations would, of course, obtain"). 5 If the Court were to hold, as would Justice STEVENS, that such a determination requires an assessment of the facts and circumstances but without explaining how such an assessment ought to proceed, the Court would offer little guidance to the federal and state courts that must apply our decisions. Justice STEVENS disagrees with the decision below because his reaction to a cold record is different from that of the Oregon Court of Appeals. The Court of Appeals found "overreaching"; Justice STEVENS finds none. Neither articulates a basis for reaching their respective conclusions which can be applied to other factual situations. We are loath to adopt such an essentially standardless rule. 6 This Court has consistently held that the Double Jeopardy Clause imposes no limitation upon the power of the government to retry a defendant who has succeeded in persuading a court to set his conviction aside, unless the conviction has been reversed because of the insufficiency of the evidence. See, e.g., United States v. DiFrancesco, 449 U.S. 117, 130-131, 101 S.Ct. 426, 433, 66 L.Ed.2d 328 (1980). 7 Justice STEVENS' opinion concurring in the judgment criticizes the suggestion that the broader rule he espouses would make it less likely the judges would grant a motion for mistrial than if the narrower rule prevailed. Post, at 687-688, n. 22. Justice STEVENS' criticism of our conclusion appears to be based on the erroneous assumption that the courts in such a situation would be applying the narrow rule, rather than the broad rule. Tested by the correct assumption that the courts would be applying the all-encompassing standard denominated "overreaching," which he espouses, post, at 689-690, we do not find his criticisms persuasive. If appellate courts and trial courts alike in this branch of the law of double jeopardy were applying a rule of "black letter law" to a predetermined set of facts, it might be true that an appellate court would inevitably reach the conclusion that reprosecution should be barred in a number of cases where the trial court had actually denied the motion for mistrial. But there are two reasons why such a hypothesis is inapplicable here. First, the rule espoused by Justice STEVENS is anything but a rule of "black letter law." We are admonished that "[i]t is unnecessary and unwise to attempt to identify all the factors that might inform the court's judgment." Ibid. Second, appellate courts have traditionally given weight to a trial court's assessment as to the necessity for a mistrial in deciding questions of double jeopardy. As this Court said in Gori v. United States, 367 U.S. 364, 368, 81 S.Ct. 1523, 1526, 6 L.Ed.2d 901 (1961): "Where, for reasons deemed compelling by the trial judge, who is best situated intelligently to make such a decision, the ends of substantial justice cannot be obtained without discontinuing the trial, a mistrial may be declared without the defendant's consent and even over his objection, and he may be retried consistently with the Fifth Amendment." It seems entirely reasonable to expect, therefore, that appellate judges will continue to defer to the judgment of trial judges who are "on the scene" in this area, and that they will not inexorably reach the same conclusion on a cold record at the appellate stage that they might if any one of them had been sitting as a trial judge. And a trial judge trying to faithfully apply the amorphous standard enunciated by Justice STEVENS could surely be forgiven if in cases he regarded as extremely close he resolved the doubt in favor of continuing a trial to its conclusion rather than aborting it. 8 Justice STEVENS states that we "gratuitously lo[p] off a portion of the previously recognized exception." Post, at 681. This charge is simply not borne out by even a moderately careful reading of our cases on the point. The footnote in United States v. Tateo, 377 U.S., at 468, n. 3, 84 S.Ct., at 1590 n. 3, quoted in n. 4, supra, states the exception in terms of prosecutorial misconduct motivated by "fear that the jury was likely to acquit the accused." The plurality opinion in United States v. Jorn, 400 U.S. 470, 485, 91 S.Ct. 547, 557, 27 L.Ed.2d 543 (1971), quoted in the text, states the test in the broader terms of "prosecutorial or judicial overreaching"; the Court's opinion in United States v. Dinitz, 424 U.S. 600, 96 S.Ct. 1075, 47 L.Ed.2d 267 (1976), speaks at one point in terms of "governmental actions intended to provoke mistrial requests," 424 U.S., at 611, 96 S.Ct., at 1081, and at another point on the same page of " 'bad-faith conduct by judge or prosecutor' " which "threatens the '[h]arassment of an accused by successive prosecutions. . . .' " Only last Term, in United States v. DiFrancesco, supra, we said that "reprosecution of a defendant who has successfully moved for a mistrial is not barred, so long as the Government did not deliberately seek to provoke the mistrial request." Id., at 130, 101 S.Ct., at 433. Thus, it is quite inaccurate to suggest that our previous cases have single-mindedly adhered to one rule in cases such as this, and that we are now "lopping off" a part of that rule. However this case is decided, we are faced with a choice between two differing lines of authority in our own recent precedents; for the reasons stated in the text, supra, at 674-677, we think that the better arguments favor the rule which we adopt. But Justice STEVENS, no less than we, chooses one of two differing rules; the state of our case law indicates that the justification for the choice must be based upon principle, and not authority. 1 Although I am willing to accept the Court's reading of the Oregon Court of Appeals' opinion as having been based on federal law, I find the question somewhat more difficult than does the Court because the Oregon Supreme Court declined to review the case without explaining its reasons. Since the Oregon Supreme Court seems to have interpreted the state constitutional protection against double jeopardy to be broader than the federal provision, see State v. Rathbun, 287 Or. 421, 600 P.2d 392 (1979), it is entirely possible that that court's refusal to review the Court of Appeals' decision was predicated on its view that the decision was sound as a matter of state law regardless of whether it was compelled by federal precedents. 2 United States v. Jorn, 400 U.S. 470, 479, 91 S.Ct. 547, 554, 27 L.Ed.2d 543 (plurality opinion). 3 United States v. Ball, 163 U.S. 662, 16 S.Ct. 1192, 41 L.Ed. 300. 4 Arizona v. Washington, 434 U.S. 497, 503, 98 S.Ct. 824, 829, 54 L.Ed.2d 717. 5 Id., at 505, 98 S.Ct. at 830. The Court in Wade v. Hunter, 336 U.S. 684, 69 S.Ct. 834, 93 L.Ed. 974, explained: "The double-jeopardy provision of the Fifth Amendment . . . does not mean that every time a defendant is put to trial before a competent tribunal he is entitled to go free if the trial fails to end in a final judgment. Such a rule would create an insuperable obstacle to the administration of justice in many cases in which there is no semblance of the type of oppressive practices at which the double-jeopardy prohibition is aimed. There may be unforeseeable circumstances that arise during a trial making its completion impossible, such as the failure of a jury to agree on a verdict. In such event the purpose of law to protect society from those guilty of crimes frequently would be frustrated by denying courts power to put the defendant to trial again. And there have been instances where a trial judge has discovered facts during a trial which indicated that one or more members of a jury might be biased against the Government or the defendant. It is settled that the duty of the judge in this event is to discharge the jury and direct a retrial. What has been said is enough to show that a defendant's valued right to have his trial completed by a particular tribunal must in some instances be subordinated to the public's interest in fair trials designed to end in just judgments." Id., at 688-689, 69 S.Ct. at 837 (footnote omitted). 6 "The reasons why [the defendant's 'valued right to have his trial completed by a particular tribunal'] merits constitutional protection are worthy of repetition. Even if the first trial is not completed, a second prosecution may be grossly unfair. It increases the financial and emotional burden on the accused, prolongs the period in which he is stigmatized by an unresolved accusation of wrongdoing, and may even enhance the risk than an innocent defendant may be convicted. The danger of such unfairness to the defendant exists whenever a trial is aborted before it is completed. Consequently, as a general rule, the prosecutor is entitled to one, and only one, opportunity to require an accused to stand trial." Arizona v. Washington, supra, 434 U.S. at 503-505, 98 S.Ct., at 829-30 (footnotes omitted). 7 Society's interest, of course, is not simply to convict the guilty. Rather, its interest is "in fair trials designed to end in just judgments." Wade v. Hunter, supra, at 689, 69 S.Ct., at 837. 8 See United States v. Dinitz, 424 U.S. 600, 609, 96 S.Ct. 1075, 1080, 47 L.Ed.2d 267. 9 Arizona v. Washington, 434 U.S., at 505, 98 S.Ct., at 830. 10 Ibid. 11 United States v. Jorn, 400 U.S., at 485, 91 S.Ct., at 557. 12 The exception also encompasses comparable judicial misconduct. Because we are confronted with prosecutorial error, this opinion will address only that context. 13 Ibid. The Court has never invoked the exception to bar reprosecution after a mistrial. In only two cases has the Court actually been confronted with a claim that the exception applied. In United States v. Dinitz, supra, the trial court had granted the defendant's motion for a mistrial after expelling defense counsel for repeated misconduct. In holding that retrial was not barred by the Double Jeopardy Clause, this Court articulated the exception and the reasons why it was not established on the facts of that case: "The Double Jeopardy Clause does protect a defendant against governmental actions intended to provoke mistrial requests and thereby to subject defendants to the substantial burdens imposed by multiple prosecu- tions. It bars retrials where 'bad-faith conduct by judge or prosecutor,' United States v. Jorn, supra, at 485 [91 S.Ct., at 557], threatens the '[h]arassment of an accused by successive prosecutions or declaration of a mistrial so as to afford the prosecution a more favorable opportunity to convict' the defendant. Downum v. United States, 372 U.S., [734] at 736 [83 S.Ct. 1033 at 1034, 10 L.Ed.2d 100]. See Gori v. United States, 367 U.S., [364] at 369 [81 S.Ct. 1523 at 1526, 6 L.Ed.2d 901]; United States v. Jorn, supra, [400 U.S.] at 489 [91 S.Ct. at 559] (STEWART, J., dissenting); cf. Wade v. Hunter, 336 U.S., at 692 [69 S.Ct., at 838]. "But here the trial judge's banishment of Wagner from the proceedings was not done in bad faith in order to goad the respondent into requesting a mistrial or to prejudice his prospects for an acquittal. As the Court of Appeals noted, Wagner 'was guilty of improper conduct' during his opening statement which 'may have justified disciplinary action,' [United States v. Dinitz ] 492 F.2d, [53] at 60-61. Even accepting the appellate court's conclusion that the trial judge overreacted in expelling Wagner from the courtroom, ibid., the court did not suggest, the respondent has not contended, and the record does not show that the judge's action was motivated by bad faith or undertaken to harass or prejudice the respondent." 424 U.S., at 611, 96 S.Ct., at 1081 (footnote omitted). The exception was also unsuccessfully claimed in Lee v. United States, 432 U.S. 23, 97 S.Ct. 2141, 53 L.Ed.2d 80. In Lee the defendant had moved to dismiss a defective information prior to the attachment of jeopardy. The trial court tentatively denied the motion, but then granted it at the close of the evidence. Treating the motion to dismiss as a motion for a mistrial, this Court quoted extensively from Dinitz for the statement of the exception and then explained why the exception had not been established: "It follows under Dinitz that there was no double jeopardy barrier to petitioner's retrial unless the judicial or prosecutorial error that prompted petitioner's motion was 'intended to provoke' the motion or was otherwise 'motivated by bad faith or undertaken to harass or prejudice' petitioner. Supra, at 33, 97 S.Ct., at 2147. Here, two underlying errors are alleged: the prosecutor's failure to draft the information properly and the court's denial of the motion to dismiss prior to the attachment of jeopardy. Neither error—even assuming the court's action could be so characterized—was the product of the kind of overreaching outlined in Dinitz. The drafting error was at most an act of negligence, as prejudicial to the Government as to the defendant. And the court's failure to postpone the taking of evidence until it could give full consideration to the defendant's motion, far from evidencing bad faith, was entirely reasonable in light of the last-minute timing of the motion and the failure of counsel to request a continuance or otherwise im- press upon the court the importance to petitioner of not being placed in jeopardy on a defective charge." 432 U.S., at 33-34, 97 S.Ct., at 2147 (footnote omitted). For other descriptions of the overreaching or harassment exception, see, e.g., Arizona v. Washington, supra, 434 U.S. at 508, 98 S.Ct., at 831 ("using the superior resources of the State to harass or to achieve a tactical advantage over the accused") (footnote omitted); Illinois v. Somerville, 410 U.S. 458, 464, 93 S.Ct. 1066, 1070, 35 L.Ed.2d 425 (error "that would lend itself to prosecutorial manipulation"); United States v. Jorn, 400 U.S., at 485, 91 S.Ct., at 557 ("prosecutorial . . . overreaching"); id., at 485, n. 12, 91 S.Ct., at 557 n. 12 ("prosecutorial impropriety designed to avoid an acquittal"); United States v. Tateo, 377 U.S. 463, 468, n. 3, 84 S.Ct. 1587, 1590 n. 11, 12 L.Ed.2d 448 ("prosecutorial . . . impropriety . . . result[ing] from a fear that the jury was likely to acquit the accused"). 14 United States v. Jorn, 400 U.S., at 484, 91 S.Ct., at 556. 15 Id., at 484, n. 11, 91 S.Ct., at 557 n. 11. See also United States v. Tateo, supra, 377 U.S. at 474, 84 S.Ct., at 1593 (Goldberg, J., dissenting) ("Many juries acquit defendants after trials in which reversible error has been committed, and many experienced trial lawyers will forego a motion for a mistrial in favor of having his case decided by the jury"). 16 "[I]t is evident that when judicial or prosecutorial error seriously prejudices a defendant, he may have little interest in completing the trial and obtaining a verdict from the first jury. The defendant may reasonably conclude that a continuation of the tainted proceeding would result in a conviction followed by a lengthy appeal and, if a reversal is secured, by a second prosecution. In such circumstances, a defendant's mistrial request has objectives not unlike the interests served by the Double Jeopardy Clause—the avoidance of the anxiety, expense, and delay occasioned by multiple prosecutions." United States v. Dinitz, 424 U.S., at 608, 96 S.Ct., at 1080. 17 Id., at 609, 96 S.Ct., at 1081. 18 See Westen & Drubel, Toward a General Theory of Double Jeopardy, 1978 S.Ct.Rev. 81, 101-102. 19 The prosecutor might wish to provoke a mistrial "in order to shop for a more favorable trier of fact, or to correct deficiencies in [his] case, or to obtain an unwarranted preview of the defendant's evidence." Id., at 94. 20 Cf. Brock v. North Carolina, 344 U.S. 424, 429, 73 S.Ct. 349, 351, 97 L.Ed. 456 (Frankfurter, J., concurring). 21 Compare ante, at 675-676 ("Prosecutorial conduct that might be viewed as harassment or overreaching, even if sufficient to justify a mistrial on defendant's motion, therefore, does not bar retrial absent intent on the part of the prosecutor to subvert the protections afforded by the Double Jeopardy Clause"), with ante, at 676 ("Only where the governmental conduct in is intended to 'goad' the defendant into moving for a mistrial may a defendant raise the bar of double jeopardy to a second trial after having succeeded in aborting the first on his own motion"), and ante, at 679 ("But we do hold that the circumstances under which such a defendant may invoke the bar of double jeopardy in a second effort to try him are limited to those cases in which the conduct giving rise to the successful motion for a mistrial was intended to provoke the defendant into moving for a mistrial"). 22 The Court offers two reasons for cutting back on the exception. First, the Court states that "[t]he difficulty with the more general standards which would permit a broader exception than one merely based on intent is that they offer virtually no standards for their application." Ante, at 674. As I indicate in the text, however, some generality in the formula is a virtue and, in any event, meaningful and principled standards can be developed on a case-by-case basis that will not inhibit legitimate prosecution practices. See infra, at 689-692. Moreover, the general standards could hardly be more difficult to apply than the Court's subjective intent standard. On this point, it is noteworthy that Justice REHNQUIST recently cited the exception for "prosecutorial overreaching or misconduct" to illustrate that double jeopardy analysis rests on "balancing and fairness" rather than " 'bright line' distinction[s]." Finch v. United States, 433 U.S. 676, 680, 97 S.Ct. 2909, 2911, 53 L.Ed.2d 1048 (dissenting opinion). Second, the Court is "not sure that criminal defendants as a class would be aided" by a broader exception. Ante, at 676. If a mistrial will more frequently constitute a bar to reprosecution, the Court supposes that trial judges will tend to refuse the defendant's mistrial motion and permit the error to be corrected on appeal of the conviction, in which event there would be no bar to reprosecution. This reasoning is premised on the assumption that an appellate court that concluded not only that the defendant's mistrial motion should have been granted but also that the prosecutor intended to provoke a mistrial would not be obligated to bar reprosecution as well as reverse the conviction. The assumption is "irrational." Commonwealth v. Potter, 478 Pa. 251, 282, 386 A.2d 918, 933 (1978) (Roberts, J.); see id., at 259-260, 386 A.2d, at 921-922 (Pomeroy, J.); Westen & Drubel, supra, at 103, 106, n. 130; see generally Note, Double Jeopardy: An Illusory Remedy for Governmental Overreaching at Trial, 29 Buffalo L.Rev. 759, 773-776 (1980). 23 For an example of the kind of case that the Court's limited exception would cover, see Commonwealth v. Warfield, 424 Pa. 555, 227 A.2d 177 (1967). 24 "As an initial matter, I question the validity of the lower court's assumption that the Government in such cases tailors its misconduct to achieve one improper result as opposed to another. It is far more likely that in cases such as this, where the prosecution is concerned that the trial may result in an acquittal, that the Government engages in misconduct with the general purpose of prejudicing the defendant. In this case, for example, the Government stood to benefit from Dixon's misconduct, regardless of whether it resulted in a guilty verdict or a mistrial. Moreover, even if such subtle differences in motivation do exist, I suspect that a defendant seeking to prevent a retrial will seldom be able to prove the Government's actual motivation." Green v. United States, 451 U.S. 929, 931, n. 2, 101 S.Ct. 2005, 2006 n. 2, 68 L.Ed.2d 316 (MARSHALL, J., dissenting from denial of certiorari). 25 Moreover, a standard that requires a prosecutor to take the stand to explain his trial strategy and his train of thought prior to making a serious error is of questionable wisdom. 26 See, e.g., Shaw v. Garrison, 328 F.Supp. 390 (ED La.1971), aff'd, 467 F.2d 113 (CA5 1972), cert. denied, 409 U.S. 1024, 93 S.Ct. 467, 34 L.Ed.2d 317. 27 The defendant's successful argument for a bar to reprosecution in United States v. Kessler, 530 F.2d 1246, 1253 (CA5 1976), was that "otherwise 'a prosecutor would have the option of first trying his case with inadmissible, prejudicial, and irrelevant evidence—that is, committing known error—in hopes of "getting away" with it, with the ability to retry the case properly if the first trial is aborted by a mistrial.' " 28 Ante, at 673. 29 Deliberate misconduct generally must be inferred from the objective evidence. The more egregious the prosecutorial error, and the harsher its impact on the defendant, the more readily the inference could be drawn. 30 Justice Harlan aptly described the defendant's interest as "being able, once and for all, to conclude his confrontation with society through the verdict of a tribunal he might believe to be favorably disposed to his fate." United States v. Jorn, 400 U.S., at 486, 91 S.Ct., at 558. There is a corresponding societal interest against the historical and abhorrent practice of terminating trials whenever it appeared that the government's evidence was insufficient to convict. See Arizona v. Washington, 434 U.S., at 507-508, 98 S.Ct., at 831. 31 In a case in which the prosecutor's intent is primarily to harass the defendant, and only secondarily to obtain a conviction, this consideration would, of course, carry much less weight. The Double Jeopardy Clause protects a defendant not only from "declaration of a mistrial so as to afford the prosecution a more favorable opportunity to convict" but also from "[h]arassment of an accused by successive prosecutions." Downum v. United States, 372 U.S. 734, 736, 83 S.Ct. 1033, 1034, 10 L.Ed.2d 100. 32 See especially the Court's opinion in Illinois v. Somerville, 410 U.S. 458, 93 S.Ct. 1066, 35 L.Ed.2d 425. Justice MARSHALL criticized the majority in that case for "abandon[ing] th[e] tradition [of elaboration of rules which give increasing guidance as case after case is decided] and [adopting] a new balancing test whose elements are stated on such a high level of abstraction as to give judges virtually no guidance at all in deciding subsequent cases." Id., at 483, 93 S.Ct., at 1080 (dissenting opinion). 33 The petitioner and the United States as amicus curiae cite only a few cases in which the exception has been invoked to bar reprosecution. One commentator discovered only two cases in which a Federal Court of Appeals barred reprosecution. Note, Double Jeopardy: An Illusory Remedy for Governmental Overreaching at Trial, 29 Buffalo L.Rev. 759, 760, n. 16 (1980). 34 "We should not be so unmindful, even when constitutional questions are involved, of the principle of stare decisis, by whose circumspect observance the wisdom of this Court as an institution transcending the moment can alone be brought to bear on the difficult problems that confront us. . . . Furthermore, we are not here called upon to weigh considerations generated by changing concepts as to minimum standards of fairness, which interpretation of the Due Process Clause inevitably requires. Instead, the defense of double jeopardy is involved, whose contours are the product of history. In this situation the passage of time is not enough, and the conviction borne to the mind of the rightness of an overturning decision must surely be of a highly compelling quality to justify overruling a well-established precedent when we are presented with no considerations fairly deemed to have been wanting to those who preceded us." Green v. United States, 355 U.S. 184, 215, 78 S.Ct. 221, 238, 2 L.Ed.2d 199 (Frankfurter, J., dissenting). 35 See n. 13, supra. 36 Wade v. Hunter, 336 U.S., at 691, 69 S.Ct., at 837. 37 "The Oregon Court of Appeals' holding that retrial of this case was barred on double jeopardy grounds is erroneous by any standard of prosecutorial overreaching which the lower courts of this country had previously derived from the decisions of this Court." Brief for Petitioner 43. 38 "I have reviewed the transcript and the wording, as put, and I would agree that the question was improper as put. I do not find, however, that it constitutes bad faith or was intentional impropriety. The question of whether or not it constitutes overreaching is one of those gray areas where we have to determine what 'overreaching' means, and in looking to the case which the defense has cited, United States v. Kessler, prosecutorial overreaching is there defined as being such as must have been a result of gross negligence or intentional misconduct which prejudiced the defendant so that he cannot receive a fair trial, and I wouldn't find that the overreaching or the erroneous conduct in this matter reaches that degree of aggravation. I don't think it amounted to gross negligence or intentional misconduct." App. 53.

01

456 U.S. 844 102 S.Ct. 2182 72 L.Ed.2d 606 INWOOD LABORATORIES, INC., et al.,v.IVES LABORATORIES, INC. DARBY DRUG CO., INC., et al. v. IVES LABORATORIES, INC. Nos. 80-2182, 81-11. Argued Feb. 22, 1982. Decided June 1, 1982. Syllabus Respondent manufactured and marketed the patented prescription drug cyclandelate to wholesalers, retail pharmacists, and hospitals in colored capsules under the registered trademark CYCLOSPASMOL. After respondent's patent expired, several generic drug manufacturers, including petitioner manufacturers, began marketing the drug, intentionally copying the appearance of the CYCLOSPASMOL capsules. Respondent then brought an action against petitioner manufacturers and wholesalers in Federal District Court under, inter alia, § 32 of the Trademark Act of 1946, alleging that some pharmacists had dispensed generic drugs mislabeled as CYCLOSPASMOL and that petitioners' use of look-alike capsules and catalog entries comparing prices and revealing the colors of generic capsules contributed to the pharmacists' mislabeling. Respondent sought injunctive relief and damages. The District Court entered judgment for petitioners, finding that although the pharmacists had violated § 32, respondent had not made the necessary factual showing that petitioners had intentionally induced the pharmacists to mislabel generic drugs or continued to supply cyclandelate to pharmacists who the petitioners knew or should have known were mislabeling generic drugs. The Court of Appeals reversed, rejecting the District Court's findings and holding that the District Court failed to give sufficient weight to the evidence respondent offered to show a pattern of illegal substitution and mislabeling. After completing its own review of the evidence, the Court of Appeals further held that the evidence was 'clearly sufficient to establish a § 32 violation.' Held: The Court of Appeals erred in setting aside the District Court's findings of fact. Pp. 853-858. (a) In reviewing such findings, the Court of Appeals was bound by the 'clearly erroneous' standard of Federal Rule of Civil Procedure 52(a). P.855 (b) By rejecting the findings simply because it would have given more weight to evidence of mislabeling than did the trial court, the Court of Appeals clearly erred. Determining the weight and credibility of the evidence is the special province of the trier of fact. Because the District Court's findings concerning the significance of the instances of mislabeling were not clearly erroneous, they should not have been disturbed. Pp.855-856 (c) Moreover, each of the conclusions that the Court of Appeals made in holding that the evidence established a § 32 violation was contrary to the District Court's findings. An appellate court cannot substitute its interpretation of the evidence for that of the trial court simply because the reviewing court 'might give the facts another construction, resolve the ambiguities differently, and find a more sinister cast to actions which the District Court apparently deemed innocent.' United States v. Real Estate Boards, 339 U.S. 485, 495, 70 S.Ct. 711, 717, 94 L.Ed. 1007. Pp. 856-858. 638 F.2d 538, reversed and remanded. Milton A. Bass, New York City, for petitioners. Jerrold J. Ganzfried, New York City, for the United States as amicus curiae, by special leave to Court. Marie V. Driscoll, New York City, for respondent. Justice O'CONNOR delivered the opinion of the Court. 1 This action requires us to consider the circumstances under which a manufacturer of a generic drug, designed to duplicate the appearance of a similar drug marketed by a competitor under a registered trademark, can be held vicariously liable for infringement of that trademark by pharmacists who dispense the generic drug. 2 * In 1955, respondent Ives Laboratories, Inc. (Ives), received a patent on the drug cyclandelate, a vasodilator used in long-term therapy for peripheral and cerebral vascular diseases. Until its patent expired in 1972, Ives retained the exclusive right to make and sell the drug, which it did under the registered trademark CYCLOSPASMOL.1 Ives marketed the drug, a white powder, to wholesalers, retail pharmacists, and hospitals in colored gelatin capsules. Ives arbitrarily selected a blue capsule, imprinted with 'Ives 4124,' for its 200 mg dosage and a combination blue-red capsule, imprinted with 'Ives 4148,' for its 400 mg dosage. 3 After Ives' patent expired, several generic drug manufacturers, including petitioners Premo Pharmaceutical Laboratories, Inc., Inwood Laboratories, Inc., and MD Pharmaceutical Co., Inc. (collectively the generic manufacturers), began marketing cyclandelate.2 They intentionally copied the appearance of the CYCLOSPASMOL capsules, selling cyclandelate in 200 mg and 400 mg capsules in colors identical to those selected by Ives.3 4 The marketing methods used by Ives reflect normal industry practice. Because cyclandelate can be obtained only by prescription, Ives does not direct its advertising to the ultimate consumer. Instead, Ives' representatives pay personal visits to physicians, to whom they distribute product literature and 'starter samples.' Ives initially directed these efforts toward convincing physicians that CYCLOSPASMOL is superior to other vasodilators. Now that its patent has expired and generic manufacturers have entered the market, Ives concentrates on convincing physicians to indicate on prescriptions that a generic drug cannot be substituted for CYCLOSPASMOL.4 5 The generic manufacturers also follow a normal industry practice by promoting their products primarily by distribution of catalogs to wholesalers, hospitals, and retail pharmacies, rather than by contacting physicians directly. The catalogs truthfully describe generic cyclandelate as 'equivalent' or 'comparable' to CYCLOSPASMOL.5 In addition, some of the catalogs include price comparisons of the generic drug and CYCLOSPASMOL and some refer to the color of the generic capsules. The generic products reach wholesalers, hospitals, and pharmacists in bulk containers which correctly indicate the manufacturer of the product contained therein. 6 A pharmacist, regardless of whether he is dispensing CYCLOSPASMOL or a generic drug, removes the capsules from the container in which he receives them and dispenses them to the consumer in the pharmacist's own bottle with his own label attached. Hence, the final consumer sees no identifying marks other than those on the capsules themselves. II A. 7 Ives instituted this action in the United States District Court for the Eastern District of New York under §§ 32 and 43(a) of the Trademark Act of 1946 (Lanham Act), 60 Stat. 427, as amended, 15 U.S.C. § 1051 et seq., and under New York's unfair competition law, N.Y.Gen.Bus. Law § 368-d (McKinney 1968).6 8 Ives' claim under § 32, 60 Stat. 437, as amended, 15 U.S.C. § 1114,7 derived from its allegation that some pharmacists had dispensed generic drugs mislabeled as CYCLOSPASMOL.8 Ives contended that the generic manufacturers' use of look-alike capsules and of catalog entries comparing prices and revealing the colors of the generic capsules induced pharmacists illegally to substitute a generic drug for CYCLOSPASMOL and to mislabel the substitute drug CYCLOSPASMOL. Although Ives did not allege that the petitioners themselves applied the Ives trademark to the drug products they produced and distributed, it did allege that the petitioners contributed to the infringing activities of pharmacists who mislabeled generic cyclandelate. 9 Ives' claim under § 43(a), 60 Stat. 441, 15 U.S.C. § 1125(a),9 alleged that the petitioners falsely designated the origin of their products by copying the capsule colors used by Ives and by promoting the generic products as equivalent to CYCLOSPASMOL. In support of its claim, Ives argued that the colors of its capsules were not functional10 and that they had developed a secondary meaning for the consumers.11 10 Contending that pharmacists would continue to mislabel generic drugs as CYCLOSPASMOL so long as imitative products were available, Ives asked that the court enjoin the petitioners from marketing cyclandelate capsules in the same colors and form as Ives uses for CYCLOSPASMOL. In addition, Ives sought damages pursuant to § 35 of the Lanham Act, 60 Stat. 439, as amended, 15 U.S.C. § 1117. B 11 The District Court denied Ives' request for an order preliminarily enjoining the petitioners from selling generic drugs identical in appearance to those produced by Ives. 455 F.Supp. 939(1978). Referring to the claim based upon § 32, the District Court stated that, while the 'knowing and deliberate instigation' by the petitioners of mislabeling by pharmacists would justify holding the petitioners as well as the pharmacists liable for trademark infringement, Ives had made no showing sufficient to justify preliminary relief. Id., at 945. Ives had not established that the petitioners conspired with the pharmacists or suggested that they disregard physicians' prescriptions. 12 The Court of Appeals for the Second Circuit affirmed. 601 F.2d 631 (1979). To assist the District Court in the upcoming trial on the merits, the appellate court defined the elements of a claim based upon § 32 in some detail. Relying primarily upon Coca-Cola Co. v. Snow Crest Beverages, Inc., 64 F.Supp. 980 (Mass.1946), aff'd, 162 F.2d 280 (CA1), cert. denied, 332 U.S. 809, 68 S.Ct. 110, 92 L.Ed. 386 (1947), the court stated that the petitioners would be liable under § 32 either if they suggested, even by implication, that retailers fill bottles with generic cyclandelate and label the bottle with Ives' trademark or if the petitioners continued to sell cyclandelate to retailers whom they knew or had reason to know were engaging in infringing practices. 601 F.2d, at 636. C 13 After a bench trial on remand, the District Court entered judgment for the petitioners. 488 F.Supp. 394 (1980). Applying the test approved by the Court of Appeals to the claim based upon § 32, the District Court found that the petitioners had not suggested, even by implication, that pharmacists should dispense generic drugs incorrectly identified as CYCLOSPASMOL.12 14 In reaching that conclusion, the court first looked for direct evidence that the petitioners intentionally induced trademark infringement. Since the petitioners' representatives do not make personal visits to physicians and pharmacists, the petitioners were not in a position directly to suggest improper drug substitutions. Cf. William R. Warner & Co. v. Eli Lilly & Co., 265 U.S. 526, 530-531, 44 S.Ct. 615, 617, 68 L.Ed. 1161 (1924); Smith, Kline & French Laboratories v. Clark & Clark, 157 F.2d 725, 731 (CA3), cert. denied, 329 U.S. 796, 67 S.Ct. 482, 91 L.Ed. 681 (1946). Therefore, the court concluded, improper suggestions, if any, must have come from catalogs and promotional materials. The court determined, however, that those materials could not 'fairly be read' to suggest trademark infringement. 488 F.Supp., at 397. 15 The trial court next considered evidence of actual instances of mislabeling by pharmacists, since frequent improper substitutions of a generic drug for CYCLOSPASMOL could provide circumstantial evidence that the petitioners, merely by making available imitative drugs in conjunction with comparative price advertising, implicitly had suggested that pharmacists substitute improperly. After reviewing the evidence of incidents of mislabeling, the District Court concluded that such incidents occurred too infrequently to justify the inference that the petitioners' catalogs and use of imitative colors had 'impliedly invited' druggists to mislabel. Ibid. Moreover, to the extent mislabeling had occurred, the court found it resulted from pharmacists' misunderstanding of the requirements of the New York Drug Substitution Law, rather than from deliberate attempts to pass off generic cyclandelate as CYCLOSPASMOL. Ibid. 16 The District Court also found that Ives failed to establish its claim based upon § 43(a). In reaching its conclusion, the court found that the blue and blue-red colors were functional to patients as well as to doctors and hospitals: many elderly patients associate color with therapeutic effect; some patients commingle medications in a container and rely on color to differentiate one from another; colors are of some, if limited, help in identifying drugs in emergency situations; and use of the same color for brand name drugs and their generic equivalents helps avoid confusion on the part of those responsible for dispensing drugs. Id., at 398-399. In addition, because Ives had failed to show that the colors indicated the drug's origin, the court found that the colors had not acquired a secondary meaning. Id., at 399. 17 Without expressly stating that the District Court's findings were clearly erroneous, and for reasons which we discuss below, the Court of Appeals concluded that the petitioners violated § 32. 638 F.2d 538 (1981). The Court of Appeals did not reach Ives' other claims. We granted certiorari, 454 U.S. 891, 102 S.Ct. 386, 70 L.Ed.2d 205 (1981), and now reverse the judgment of the Court of Appeals. III A. 18 As the lower courts correctly discerned, liability for trademark infringement can extend beyond those who actually mislabel goods with the mark of another. Even if a manufacturer does not directly control others in the chain of distribution, it can be held responsible for their infringing activities under certain circumstances. Thus, if a manufacturer or distributor intentionally induces another to infringe a trademark, or if it continues to supply its product to one whom it knows or has reason to know is engaging in trademark infringement, the manufacturer or distributor is contributorially responsible for any harm done as a result of the deceit.13 See William R. Warner & Co. v. Eli Lilly & Co., supra; Coca-Cola Co. v. Snow Crest Beverages, Inc., supra. 19 It is undisputed that those pharmacists who mislabeled generic drugs with Ives' registered trademark violated § 32.14 However, whether these petitioners were liable for the pharmacists' infringing acts depended upon whether, in fact, the petitioners intentionally induced the pharmacists to mislabel generic drugs or, in fact, continued to supply cyclandelate to pharmacists whom the petitioners knew were mislabeling generic drugs. The District Court concluded that Ives made neither of those factual showings. B 20 In reviewing the factual findings of the District Court, the Court of Appeals was bound by the 'clearly erroneous' standard of Rule 52(a), Federal Rules of Civil Procedure. Pullman-Standard v. Swint, 456 U.S. 273, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982). That Rule recognizes and rests upon the unique opportunity afforded the trial court judge to evaluate the credibility of witnesses and to weigh the evidence. Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123, 89 S.Ct. 1562, 1576, 23 L.Ed.2d 129 (1969). Because of the deference due the trial judge, unless an appellate court is left with the 'definite and firm conviction that a mistake has been committed,' United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948), it must accept the trial court's findings.15 IV 21 In reversing the District Court's judgment, the Court of Appeals initially held that the trial court failed to give sufficient weight to the evidence Ives offered to show a 'pattern of illegal substitution and mislabeling in New York. . . .'16 638 F.2d, at 543. By rejecting the District Court's findings simply because it would have given more weight to evidence of mislabeling than did the trial court, the Court of Appeals clearly erred. Determining the weight and credibility of the evidence is the special province of the trier of fact. Because the trial court's findings concerning the significance of the instances of mislabeling were not clearly erroneous, they should not have been disturbed. 22 Next, after completing its own review of the evidence, the Court of Appeals concluded that the evidence was 'clearly sufficient to establish a § 32 violation.' Ibid. In reaching its conclusion, the Court of Appeals was influenced by several factors. First, it thought the petitioners reasonably could have anticipated misconduct by a substantial number of the pharmacists who were provided imitative, lower priced products which, if substituted for the higher priced brand name without passing on savings to consumers, could provide an economic advantage to the pharmacists. Ibid.17 Second, it disagreed with the trial court's finding that the mislabeling which did occur reflected confusion about state law requirements. Id., at 544.18 Third, it concluded that illegal substitution and mislabeling in New York are neitherde minimis nor inadvertent. Ibid.19 Finally, the Court of Appeals indicated it was further influenced by the fact that the petitioners did not offer 'any persuasive evidence of a legitimate reason unrelated to CYCLOSPASMOL' for producing an imitative product. Ibid.20 23 Each of those conclusions is contrary to the findings of the District Court. An appellate court cannot substitute its interpretation of the evidence for that of the trial court simply because the reviewing court 'might give the facts another construction, resolve the ambiguities differently, and find a more sinister cast to actions which the District Court apparently deemed innocent.' United States v. Real Estate Boards, 339 U.S. 485, 495, 70 S.Ct. 711, 717, 94 L.Ed. 1007 (1950). V 24 The Court of Appeals erred in setting aside findings of fact that were not clearly erroneous. Accordingly, the judgment of the Court of Appeals that the petitioners violated § 32 of the Lanham Act is reversed. 25 Although the District Court also dismissed Ives' claims alleging that the petitioners violated § 43(a) of the Lanham Act and the state unfair competition law, the Court of Appeals did not address those claims. Because § 43(a) prohibits a broader range of practices than does § 32, as may the state unfair competition law, the District Court's decision dismissing Ives' claims based upon those statutes must be independently reviewed. Therefore, we remand to the Court of Appeals for further proceedings consistent with this opinion. 26 Reversed and remanded. 27 Justice WHITE, with whom Justice MARSHALL joins, concurring in the result. 28 We granted certiorari in these cases in order to review the legal standard employed by the Second Circuit in finding that a generic drug manufacturer is vicariously liable for trademark infringement committed by pharmacists who dispense the generic drug. The Court implicitly endorses the legal standard purportedly employed by the Court of Appeals, ante, at 853-854, but finds that the court erred in setting aside factual findings that were not clearly erroneous. The question whether the Court of Appeals had misapplied the clearly-erroneous rule, however, was not presented in the petitions for certiorari. This was conceded at oral argument.1 Tr. of Oral Arg. 69. Our Rule 21.1(a) states that '[o]nly the questions set forth in the petition or fairly included therein will be considered by the Court.' The majority suggests no reason for ignoring our own Rule. Furthermore, if the issue presented in the petitions for certiorari had been whether the clearly-erroneous standard, although properly invoked, was erroneously applied, it is doubtful in my mind that this fact-bound issue would have warranted certiorari. I nevertheless concur in reversal because I believe that the Court of Appeals has watered down to an impermissible extent the standard for finding a violation of § 32 of the Lanham Act, 15 U.S.C. § 1114. 29 In its first opinion in this litigation, the Court of Appeals indicated that a 'manufacturer or wholesaler would be liable under § 32 if he suggested, even if only by implication, that a retailer fill a bottle with the generic capsules and apply Ives' mark to the label, or continued to sell capsules containing the generic drug which facilitated this to a druggist whom he knew or had reason to know was engaging in the practices just described.' 601 F.2d 631, 636 (1979) (Ives II). The District Court applied this test but concluded that no violation of § 32 had been shown. On appeal after trial, a majority of the Second Circuit found defendants liable for contributory infringement by revising and expanding the doctrine of contributory trademark infringement. 638 F.2d 538 (1981) (Ives IV): 30 'By using capsules of identical color, size, and shape, together with a catalog describing their appearance and listing comparable prices of CYCLOSPASMOL and generic cyclandelate, appellees could reasonably anticipate that their generic drug product would by a substantial number of druggists be substituted illegally . . . This amounted to a suggestion, at least by implication, that the druggists take advantage of the opportunity to engage in such misconduct.' Id., at 543 (emphasis added). 31 Ives II required a showing that petitioners intended illegal substitution or knowingly continued to supply pharmacists palming off generic cyclandelate as CYCLOSPASMOL; Ives IV was satisfied merely by the failure to 'reasonably anticipate' that illegal substitution by some pharmacists was likely. In my view, this is an erroneous construction of the statutory law governing trademark protection. 32 William R. Warner & Co. v. Eli Lilly & Co., 265 U.S. 526, 44 S.Ct. 615, 68 L.Ed. 1161 (1924), made clear that a finding of contributory infringement requires proof of either an intent to induce illegal substitution or continued sales to particular customers whom the manufacturer knows or should know are engaged in improper palming off. In that case, it was shown that the manufacturer's salesmen actively induced, either in direct terms or by insinuation, the filling of requests for Coco-Quinine with Quin-Coco. 'The wrong was in designedly enabling the dealers to palm off the preparation as that of the respondent.'2 Id., at 530, 44 S.Ct., at 617. Coca-Cola Co. v. Snow Crest Beverages, Inc., 64 F.Supp. 980, 989 (Mass.1946), aff'd, 162 F.2d 280 (CA1), cert. denied, 332 U.S. 809, 68 S.Ct. 110, 92 L.Ed. 386 (1947), the case upon which the Court of Appeals relied in Ives II, stands for this very proposition. There was no contributory infringement in Snow Crest's manufacture of a product identical in appearance to that of Coca-Cola. Judge Wyzanski observed that 33 'any man of common sense knows that in any line of business . . . there are some unscrupulous persons, who, when it is to their financial advantage to do so, will palm off on customers a different product from that ordered by the customer.' 64 F.Supp., at 988-989. 34 These cases reflect the general consensus. 2 J. McCarthy, Trademarks and Unfair Competition § 252 (1973): ('[T]he supplier's duty does not go so far as to require him to refuse to sell to dealers who merely might pass off its goods'). The mere fact that a generic drug company can anticipate that some illegal substitution will occur to some unspecified extent, and by some unknown pharmacists, should not by itself be a predicate for contributory liability. I thus am inclined to believe that the Court silently acquiesces in a significant change in the test for contributory infringement. 35 Diluting the requirement for establishing a prima facie case of contributory trademark infringement is particularly unjustified in the generic drugs field. Preventing the use of generic drugs of the same color to which customers had become accustomed in their prior use of the brand name product interferes with the important state policy, expressed in New York and 47 other States, of promoting the substitution of generic formulations. See Warner, Consumer Protection and Prescription Drugs: The Generic Drug Substitution Laws, 67 Ky.L.J. 384 (1978-1979). 36 The Court of Appeals concluded that there was no 'persuasive evidence of a legitimate reason' for petitioners to use imitative colors. The District Court, however, had expressly found that for purposes of § 43(a), the capsule colors were functional. With respect to functionality, I fully agree with the Court that the Court of Appeals erred in setting aside factual findings without finding that they were clearly erroneous. The District Court found that capsule color was functional in several respects: patient anxiety and confusion were likely if accustomed medicine were dispensed in a different color; capsule colors assist patients in identifying the correct pill to take; standard colors help physicians identify the drug involved in case of overdose.3 Clearly, the Court of Appeals could not reject these findings merely because it viewed the evidence as less persuasive than did the District Court. Rule 52(a) imposes a stricter standard. 37 Finally, although the Court states that a 'finding of functionality may also be relevant to an action involving § 32,' it does not explicate the relationship of functionality in a § 32 case. It is my view that a finding of functionality offers a complete affirmative defense to a contributory in fringement claim predicated solely on the reproduction of a functional attribute of the product. A functional characteristic is 'an important ingredient in the commercial success of the product,' 601 F.2d, at 643, and, after expiration of a patent, it is no more the property of the originator than the product itself. It makes no more sense to base contributory infringement upon the copying of functional colors than on the petitioners' decision to use the same formulation of the drug, or even to market the generic substitute in the first place. To be sure, the very existence of generic drugs 'facilitates' illegal substitution. But Ives no longer has a patent for cyclandelate, 'and the defendants have a right to reproduce it as nearly as they can.' Saxlehner v. Wagner, 216 U.S. 375, 380, 30 S.Ct. 298, 54 L.Ed. 525 (1910) (Holmes, J.). Reproduction of a functional attribute is legitimate competitive activity. 38 I am also mindful that functionality is a defense to a suit under § 43(a) of the Lanham Act alleging damages from a competitor's 'false designation of origin' on a good.4 The use of a product or package design that is so similar to that of another producer that it is likely to confuse purchasers as to the product's source may constitute 'false designation of origin' within the meaning of the Act.5 As the Court of Appeals noted in Ives II, § 43(a) 'goes beyond § 32 in making certain types of unfair competition federal statutory torts,' 601 F.2d, at 641. Section 43(a) offers the direct protection of Ives' interest in this case, and it is not surprising that the alleged § 43(a) violation was the primary claim in this litigation, as it has been in other cases of this genre. It would be anomalous for the imitation of a functional feature to constitute contributory infringement for purposes of § 32, while the same activity is not a 'false designation of origin' under § 43(a).6 39 I would reverse the decision of the Court of Appeals and remand for review of the District Court's findings consistent with the principles stated above. 40 Justice REHNQUIST, concurring in the result. 41 I agree that the judgment of the Court of Appeals should be reversed. That court set aside factual findings of the District Court without having found them to be clearly erroneous as required by Rule 52(a) of the Federal Rules of Civil Procedure. I disagree, however, with the Court's determining for itself that the findings of the District Court were not clearly erroneous. I think in the usual case this is a question best decided by the courts of appeals, who have a good deal more experience with the application of this principle than we do, and I see no reason to make an exception in this case. 42 I also assume, correctly I hope, that the Court's discussion of appellate review of trial court findings in bench trials, ante, at 855, is limited to cases in which the appellate court has not found the trial court findings to be 'clearly erroneous.' United States v. United States Gypsum Co., 333 U.S. 364, 68 S.Ct. 525, 92 L.Ed. 746 (1948), upon which the Court relies, establishes the authority of a reviewing court to make its own findings, contrary to those of the trial court, where it has determined the latter to be 'clearly erroneous.' 43 I agree with the Court that these cases should be remanded to the Court of Appeals to review the District Court's dismissal of respondent's claims under § 43(a) of the Lanham Act and its state-law claims. 1 Under the Trademark Act of 1946 (Lanham Act), 60 Stat. 427, as amended, 15 U.S.C. § 1051 et seq., the term 'trade-mark' includes 'any word, name, symbol, or device or any combination thereof adopted and used by a manufacturer or merchant to identify his goods and distinguish them from those manufactured or sold by others.' 15 U.S.C. § 1127. A 'registered mark' is one registered in the United States Patent and Trademark Office under the terms of the Lanham Act 'or under the Act of March 3, 1881, or the Act of February 20, 1905, or the Act of March 19, 1920.' Ibid. 2 The generic manufacturers purchase cyclandelate and empty capsules and assemble the product for sale to wholesalers and hospitals. The petitioner wholesalers, Darby Drug Co., Inc., Rugby Laboratories, Inc., and Sherry Pharmaceutical Co., Inc., in turn, sell to other wholesalers, physicians, and pharmacies. 3 Initially, the generic manufacturers did not place any identifying mark on their capsules. After Ives initiated this action, Premo imprinted 'Premo' on its capsules and Inwood imprinted 'Inwood 258.' 4 Since the early 1970's, most States have enacted laws allowing pharmacists to substitute generic drugs for brand name drugs under certain conditions. See generally Note, Consumer Protection and Prescription Drugs: The Generic Drug Substitution Laws, 67 Ky.L.J. 384 (1978-1979). The New York statutes involved in this action are typical of these generic substitution laws. New York law requires that prescription forms contain two lines, one of which a prescribing physician must sign. N.Y.Educ. Law § 6810 (McKinney Supp. 1981-1982). If the physician signs over the words 'substitution permissible,' substitution is mandatory if a substitute generic drug is on an approved list, N.Y. Educ. Law § 6816-a (McKinney Supp.1981-1982); N.Y.Pub. Health Law § 206.1(o) (McKinney Supp.1981-1982), and permissible if another generic drug is available. Unless the physician directs otherwise, the pharmacist must indicate the name of the generic manufacturer and the strength of the drug dispensed on the label. N.Y.Educ. Law § 6816-a(1)(c). In addition, the prescription form must specifically state that, unless the physician signs above the line 'dispense as written,' the prescription will be filled generically. § 6810(6)(a). If a pharmacist mislabels a drug or improperly substitutes, he is guilty of a misdemeanor and subject to a fine, §§ 6811, 6815, 6816, and to revocation of his license. § 6808. 5 Ives conceded that CYCLOSPASMOL and the petitioners' generic equivalents are bioequivalent and have the same bioavailability. See 455 F.Supp. 939, 942 (EDNY 1978), and 488 F.Supp. 394, 396 (EDNY 1980). Bioavailability is an absolute term which measures both the rate and the amount of a drug which reaches the general circulation from a defined dosage. Drugs are 'bioequivalent' if, when administered in equal amounts to the same individual, they reach general circulation at the same relative rate and to the same relative extent. Remington's Pharmaceutical Sciences 1368 (15th ed. 1975). 6 The state law claim was not discussed in the decision under review, and no further reference will be made to it here. 7 Section 32 of the Lanham Act, 60 Stat. 437, as amended, 15 U.S.C. § 1114, provides in part: '(1) Any person who shall, without the consent of the registrant—— '(a) use in commerce any reproduction, counterfeit, copy, or colorable imitation of a registered mark in connection with the sale, offering for sale, distribution, or advertising of any goods or services on or in connection with which such use is likely to cause confusion, or to cause mistake, or to deceive; or '(b) reproduce, counterfeit, copy, or colorably imitate a registered mark and apply such reproduction, counterfeit, copy, or colorable imitation to labels, signs, prints, packages, wrappers, receptacles or advertisements intended to be used in commerce upon or in connection with the sale, offering for sale, distribution, or advertising of goods or services on or in connection with which such use is likely to cause confusion, or to cause mistake, or to deceive, 'shall be liable in a civil action by the registrant for the remedies hereinafter provided. Under subsection (b) of this section, the registrant shall not be entitled to recover profits or damages unless the acts have been committed with knowledge that such imitation is intended to be used to cause confusion, or to cause mistake or to deceive.' 8 The claim involved two types of infringements. The first was 'direct' infringement, in which druggists allegedly filled CYCLOSPASMOL prescriptions marked 'dispense as written' with a generic drug and mislabeled the product as CYCLOSPASMOL. The second, 'intermediate' infringement, occurred when pharmacists, although authorized by the prescriptions to substitute, allegedly mislabeled a generic drug as CYCLOSPASMOL. The one retail pharmacy originally named as a defendant consented to entry of a decree enjoining it from repeating such actions. 455 F.Supp., at 942. 9 Section 43(a) of the Lanham Act, 60 Stat. 441, 15 U.S.C. § 1125(a), provides: '(a) Any person who shall affix, apply, or annex, or use in connection with any goods or services, or any container or containers for goods, a false designation of origin, or any false description or representation, including words or other symbols tending falsely to describe or represent the same, and shall cause such goods or services to enter into commerce, and any person who shall with knowledge of the falsity of such designation of origin or description or representation cause or procure the same to be transported or used in commerce or deliver the same to any carrier to be transported or used, shall be liable to a civil action by any person doing business in the locality falsely indicated as that of origin or in the region in which said locality is situated, or by any person who believes that he is or is likely to be damaged by the use of any such false description or representation.' 10 In general terms, a product feature is functional if it is essential to the use or purpose of the article or if it affects the cost or quality of the article. See Sears, Roebuck & Co. v. Stiffel Co., 376 U.S. 225, 232, 84 S.Ct. 784, 789, 11 L.Ed.2d 661 (1964); Kellogg Co. v. National Biscuit Co., 305 U.S. 111, 122, 59 S.Ct. 109, 115, 83 L.Ed. 73 (1938). 11 To establish secondary meaning, a manufacturer must show that, in the minds of the public, the primary significance of a product feature or term is to identify the source of the product rather than the product itself. See Kellogg Co. v. National Biscuit Co., supra, at 118, 59 S.Ct. at 113. 12 The District Court also found that the petitioners did not continue to provide drugs to retailers whom they knew or should have known were engaging in trademark infringement. 488 F.Supp., at 397. The Court of Appeals did not discuss that finding, and we do not address it. 13 Justice WHITE, in his opinion concurring in the result, voices his concern that we may have 'silently acquiesce[d] in a significant change in the test for contributory infringement.' Post, at 861. His concern derives from his perception that the Court of Appeals abandoned the standard enunciated by Judge Friendly in its first opinion, a standard which both we and Justice WHITE approve, post, at 859-860. The Court of Appeals, however, expressly premised its second opinion on 'the governing legal principles . . . set forth in Judge Friendly's opinion upon the earlier appeal, 601 F.2d 631 (2d Cir. 1979),' and explicitly claimed to have rendered its second decision by '[a]pplying those principles. . . .' 638 F.2d 538, 542 (1981). Justice WHITE's concern is based on a comment by the Court of Appeals that the generic manufacturers 'could reasonably anticipate' illegal substitution of their drugs. Id., at 543. If the Court of Appeals had relied upon that statement to define the controlling legal standard, the court indeed would have applied a 'watered down' and incorrect standard. As we read the Court of Appeals' opinion, however, that statement was intended merely to buttress the court's conclusion that the legal test for contributory infringement, as earlier defined, had been met. See infra, at 1289-1290. 14 Such blatant trademark infringement inhibits competition and subverts both goals of the Lanham Act. By applying a trademark to goods produced by one other than the trademark's owner, the infringer deprives the owner of the goodwill which he spent energy, time, and money to obtain. See S.Rep. No. 1333, 79th Cong., 2d Sess., 3 (1946). At the same time, the infringer deprives consumers of their ability to distinguish among the goods of competing manufacturers. See H.R.Rep. No. 944, 76th Cong., 1st Sess., 3 (1939.) 15 Of course, if the trial court bases its findings upon a mistaken impression of applicable legal principles, the reviewing court is not bound by the clearly erroneous standard. United States v. Singer Manufacturing Co., 374 U.S. 174, 194, n. 9, 83 S.Ct. 1773, 1784, n. 9, 10 L.Ed.2d 823 (1963). However, in this instance the District Court applied correct legal principles when it adopted the precise test developed by the Court of Appeals. Compare 601 F.2d 631, 636 (1979), with 488 F.Supp., at 397. 16 As the opinions from the lower courts reveal, more than one inference can be drawn from the evidence presented. Prior to trial, test shoppers hired by Ives gave CYCLOSPASMOL prescriptions on which the 'substitution permissible' line was signed to 83 New York pharmacists. Forty-eight of the pharmacists dispensed CYCLOSPASMOL; the rest dispensed a generic drug. Ten of the thirty-five pharmacists who dispensed a generic drug included the word CYCLOSPASMOL on the label, although 5 of those 10 also included some form of the word 'generic.' Nine of the ten told the consumer of the substitution. Only 1 of the 10 charged the brand name price for the generic drug. 488 F.Supp., at 397. The District Court concluded that that evidence did not justify the inference that the petitioners' catalogs invite pharmacists to mislabel. Ibid. The Court of Appeals, emphasizing that 10 of the 35 druggists who dispensed a generic drug mislabeled it as CYCLOSPASMOL, found a pattern of substitution and mislabeling. 638 F.2d, at 543. The dissenting judge on the appellate panel, emphasizing that only 1 of 83 pharmacists attempted an illegal substitution and reaped a profit made possible by the color imitation, concluded the facts supported the District Court's finding that mislabeling resulted from confusion about the substitution laws rather than from profit considerations. Id., at 546. On the basis of the record before us, the inferences drawn by the District Court are not, as a matter of law, unreasonable. 17 The Court of Appeals cited no evidence to support its conclusion, which apparently rests upon the assumption that a pharmacist who has been provided an imitative generic drug will be unable to resist the temptation to profit from illegal activity. We find no support in the record for such a far-reaching conclusion. Moreover, the assumption is inconsistent with the District Court's finding that only a 'few instances,' rather than a substantial number, of mislabelings occurred. 488 F.Supp., at 397. 18 The Court of Appeals characterized the District Court's finding as resting on 'a short and casual exchange with a witness . . ..' 638 F.2d, at 544. The District Court, however, stated that its conclusion that pharmacists did not understand the drug substitution law rested upon the fact that, in numerous instances, a pharmacist told a consumer that state law prohibited filling prescriptions with generic products, even though the consumer had presented a prescription allowing generic substitution. 488 F.Supp., at 397-398. 19 In reaching that conclusion, the Court of Appeals took judicial notice of the fact that, in May 1980, six indictments were handed down in New York City charging pharmacists with substituting cyclandelate for CYCLOSPASMOL. We note that the evidence of which the Court of Appeals took judicial notice not only involved no convictions but also reflected knowledge that was not available when the District Court rendered its decision. Moreover, even if the District Court failed to consider relevant evidence, which would have been an error of law, the Court of Appeals, rather than make its own factual determination, should have remanded for further proceedings to allow the trial court to consider the evidence. See Pullman-Standard v. Swint, 456 U.S., at 291-292, 102 S.Ct., at 1791. 20 The Court of Appeals reached that conclusion despite the District Court's express finding that, for purposes of § 43(a), the capsule colors were functional. See supra, at 853. As the dissent below noted, the Court of Appeals' majority either disregarded the District Court's finding of functionality, see 638 F.2d, at 545, n. 1 (Mulligan, J., dissenting), or implicitly rejected that finding as not 'persuasive.' See id., at 547. While the precise basis for the Court of Appeals' ruling on this issue is unclear, it is clear that the Court of Appeals erred. The appellate court was not entitled simply to disregard the District Court's finding of functionality. While the doctrine of functionality is most directly related to the question of whether a defendant has violated § 43(a) of the Lanham Act, see generally Note, The Problem of Functional Features: Trade Dress Infringement Under Section 43(a) of the Lanham Act, 82 Colum.L.Rev. 77 (1982), a finding of functionality may also be relevant to an action involving § 32. By establishing to the District Court's satisfaction that uniform capsule colors served a functional purpose, the petitioners offered a legitimate reason for producing an imitative product. Nor was the Court of Appeals entitled simply to dismiss the District Court's finding of functionality as not 'persuasive.' If the District Court erred as a matter of law, the Court of Appeals should have identified the District Court's legal error. If the Court of Appeals disagreed with the District Court's factual findings, it should not have dismissed them without finding them clearly erroneous. 1 The third question in petitioner Darby Drug Co.'s petition embraced the claim that the Court of Appeals had failed to observe Rule 52(a) in overturning the District Judge's finding of functionality. As discussed below, I agree with the Court's invocation of Rule 52 with respect to this aspect of the decision below. 2 Although Warner and other cases were decided before § 32 was enacted, the purpose of the Lanham Act was to codify and unify the common law of unfair competition and trademark protection. S.Rep.No. 1333, 79th Cong., 2d Sess. (1946). There is no suggestion that Congress intended to depart from Warner and other contemporary precedents. 3 'The reality is that for every link in the distributive chain (from producer to ultimate consumer) the color and shape of drugs dispensed by prescription do perform a function. For each of them, color or shape may be a convenient shorthand code by which to identify the drug and its milligram dosage so that mistakes can be avoided in the interests of pharmaceutical precaution and patient safety. For the patient-user, of course, the constancy of color and shape may be as psychologically reassuring and therefore as medically beneficial as the drug itself; in addition, they also serve to identify the drug for his ingestion. . . .' '[I]f the generic producer is constrained by § 43(a), trademark law, or the law of unfair competition to adopt a substantially different color . . . the therapeutic value of his generic drug might be seriously impaired and confusion at the pharmacist level could be compounded beyond redemption.' 3 R. Callmann, Unfair Competition, Trademarks and Monopolies § 82.1(m), pp. 217, 213 (Supp.1981). 4 See, e.g., International Order of Job's Daughters v. Lindeburg & Co., 633 F.2d 912, 917 (CA 9 1980), cert. denied, 452 U.S. 941, 101 S.Ct. 3086, 69 L.Ed.2d 956 (1981); Keebler Co. v. Rovira Biscuit Corp., 624 F.2d 366, 378 (CA 1 1980). See generally Note, The Problem of Functional Features: Trade Dress Infringement Under Section 43(a) of the Lanham Act, 82 Colum.L.Rev. 77, 81 (1982) ('Over the past three years the rule that functionality of a copied feature bars relief in section 43(a) claims for trade dress infringement or product imitation has become the plurality view'). 5 See, e.g., Truck Equip. Serv. Co. v. Fruehauf Corp., 536 F.2d 1210 (CA 8), cert. denied, 429 U.S. 861, 97 S.Ct. 164, 50 L.Ed.2d 139 (1976); Warner Bros., Inc. v. Gay Toys, Inc., 658 F.2d 76 (CA 2 1981). See also Note, 82 Colum.L.Rev., supra, at 78-80. 6 This is not to suggest that the copying of a functional feature protects a defendant from § 32 liability predicated on active inducement of trademark infringement or protects a defendant who has also reproduced nonfunctional features.

78

456 U.S. 728 102 S.Ct. 2118 72 L.Ed.2d 520 ARMY AND AIR FORCE EXCHANGE SERVICE, Petitionerv.Arthur Edward SHEEHAN. No. 80-1437. Argued Feb. 23, 1982. Decided June 1, 1982. Syllabus Respondent, while employed in a data processing position with petitioner Army and Air Force Exchange Service (AAFES), was selected for participation in the AAFES Executive Management Program (EMP). A regulation provided that EMP status could be withdrawn for conduct off the job reflecting discredit upon the AAFES. Respondent was discharged from the AAFES after pleading guilty to misdemeanor charges of violating state drug laws off the base. His administrative appeal was denied. While that appeal was pending before the Judge Advocate General of the Air Force, respondent filed suit against the AAFES in Federal District Court, alleging that his rights to due process and to a free and impartial appeal pursuant to AAFES regulations were infringed, and seeking reinstatement and damages, including backpay. The District Court dismissed the complaint for want of subject-matter jurisdiction. The Court of Appeals reversed, concluding that the Tucker Act, which gives the federal courts jurisdiction over certain suits against the United States founded upon express or implied contracts, provided a basis for jurisdiction over respondent's claims for monetary relief. The court held that, whether or not respondent's employment was initiated by appointment or contract, the AAFES regulations governing separation procedures created an implied-in-fact contract that the AAFES would adhere to those regulations while respondent continued in AAFES employment, and that respondent's allegation that his dismissal violated those regulations was equivalent to an allegation of breach of an implied-in-fact contract. Held : The Tucker Act did not confer jurisdiction over respondent's claim for money damages. Pp. 733-741. (a) Nothing in the record or relevant regulations indicates that respondent was employed pursuant to an express contract. Rather, the evidence shows that he was appointed to his positions. With respect to employment in the data processing position, regulations prohibited the AAFES from negotiating a contract with him, and his selection to the EMP clearly was pursuant to an appointment. There is no reason to remand for an evidentiary hearing on the nature of respondent's employment status. Pp. 735-738. (b) The Court of Appeals erred in implying a contract based solely on the AAFES personnel regulations and in premising Tucker Act jurisdiction on those regulations, which do not specifically authorize awards of money damages. United States v. Testan, 424 U.S. 392, 96 S.Ct. 948, 47 L.Ed.2d 114, is controlling. Moreover, Congress' intent to prohibit Back Pay Act claims by AAFES employees, as opposed to federal employees generally, would be subverted if an AAFES employee could sue under the Tucker Act whenever he asserted a violation of the AAFES regulations governing termination. In fact, the Court of Appeals' reasoning would extend Tucker Act jurisdiction to reach any complaint filed by a federal employee alleging the violation of a personnel statute or regulation. Pp. 738-741. 619 F.2d 1132, reversed. Samuel A. Alito, Jr., Newark, N. J., for petitioner. Ira E. Tobolowsky, Dallas, Tex., for respondent. Justice BLACKMUN delivered the opinion of the Court. 1 The issue presented by this case is whether the federal courts have jurisdiction over a civil action for monetary damages brought by a former military exchange employee who contests the validity of his discharge. The employee claims that federal jurisdiction exists under the Tucker Act, 28 U.S.C. § 1346(a)(2) (1976 ed., Supp. IV). 2 * A. 3 In 1962, respondent, Arthur Edward Sheehan, was selected for a data processing position with petitioner Army and Air Force Exchange Service (AAFES or Service).1 Five years later, respondent was designated by the AAFES commander for participation in the Service's Executive Management Program (EMP); this program is "intended to fulfill the continuing requirement of AAFES for highly qualified and dedicated executive employees who will be readily available to meet the worldwide executive personnel requirements of AAFES." Army Regulation (AR) 60-21/Air Force Regulation (AFR) 147-15, ch. 5, § II, ¶ 5-6 (1 Aug. 1979).2 Employees in the program enjoy special retention, insurance, and retirement benefits. On the other hand, those employees are subject to certain obligations, a principal one being that EMP personnel must accept transfer to any AAFES facility in this country or abroad. ¶ 5-9(a)(2). EMP status may be withdrawn for, among other things, "conduct off the job reflecting discredit upon AAFES." ¶ 5-9(c). Pursuant to the regulations governing the EMP, respondent was required to "acknowledg[e] in writing that he underst[ood] and accept[ed] the conditions of the EMP as prescribed by the Commander, AAFES." ¶ 5-7(b). 4 In 1975, while respondent was serving as a shopping center manager at Fort Jackson, S. C., he was arrested off the base for possession of controlled substances. Pursuant to a plea bargain, respondent pleaded guilty to four misdemeanor counts of violating state drug laws. He was sentenced to 18 months' probation and a $1,000 fine was imposed. 5 On March 16, 1976, respondent received advance written notice of separation from the Service for cause. Referring specifically to respondent's conviction, the notice stated that the reason for the separation was "conduct off the job which reflects discredit on the AAFES and which is of such a nature that your retention in any capacity is incompatible with the best interests of AAFES." App. 11. James J. Stapleton, the AAFES General Manager for the Piedmont Area Exchange, signed the notice, but, because of respondent's participation in the EMP, prior approval had been obtained from Major General C. W. Hospelhorn, Commander, AAFES. Following an investigation, Stapleton issued a final notice of separation for cause, effective April 19, 1976. Id., at 17. This notice advised respondent that he was to be dismissed "in view of the entire weight of evidence which resulted in your plea of guilty." Ibid. 6 Respondent, in accord with authorized AAFES procedures, filed an administrative appeal. The hearing examiner determined that the Service had acted in compliance with applicable laws and regulations, but concluded that respondent's conduct off the job did not reflect discredit on the AAFES and that his retention in some capacity was not incompatible with the interests of the Service. The examiner therefore recommended that respondent's appeal be granted and that he be reinstated with backpay to his former grade but transferred to an assignment in another region. General Hospelhorn, however, acting as the appellate authority, disagreed, and denied respondent's appeal. 7 In 1978, respondent, by a letter from counsel addressed to the new AAFES Commander, Major General Bobby W. Presley, requested reconsideration. Id., at 40. Respondent asserted that his separation was contrary to AAFES rules and regulations and that he had been denied due process of law. General Presley reopened the case and referred it to Lieutenant General Charles E. Buckingham, Chairman of the Board of Directors of AAFES. At General Buckingham's request, the administrative record was reviewed by the Judge Advocate General of the Air Force. He concluded that the record evidence supported the charge that respondent's conduct reflected discredit upon the AAFES and that his retention was inconsistent with the Service's best interests. The Judge Advocate General, however, agreed with respondent that General Hospelhorn was disqualified from acting as the appellate authority; he felt that it was appropriate for General Buckingham to act in that capacity, and he recommended that respondent's appeal be denied. General Buckingham followed that advice and denied respondent's appeal. B 8 While the matter was pending before the Judge Advocate General, respondent filed suit against the AAFES in the United States District Court for the Northern District of Texas. The first count of respondent's complaint alleged that his rights to due process and to a free and impartial appeal pursuant to AAFES regulations were infringed when General Hospelhorn acted as both the separation authority and the appellate authority. In the second count, respondent claimed that the denial of his appeal was arbitrary and capricious, an abuse of discretion, unsupported by substantial evidence and unwarranted by the facts, and in violation of statutory and constitutional provisions. Respondent sought reinstatement and damages, including backpay. 9 The District Court, without opinion, dismissed the complaint for want of subject-matter jurisdiction. App. to Pet. for Cert. 17a. 10 The United States Court of Appeals for the Fifth Circuit reversed. It concluded that the Tucker Act, 28 U.S.C. § 1346(a)(2), which gives the federal courts jurisdiction over certain suits against the United States founded upon express or implied contracts, provided a basis for jurisdiction over respondent's claims for monetary relief. 619 F.2d 1132 (1980). Whether respondent's employment was initiated by appointment or by contract, the court held, the AAFES regulations providing for separation for cause only under certain conditions and guaranteeing an administrative appeal "manifest[ed] the understanding of the parties concerning discharge procedures while Sheehan continued in AAFES employment." Id., at 1138 (emphasis in original). Accordingly, the court considered those regulations to be "part of a collateral implied-in-fact contract between Sheehan and the AAFES that the AAFES would adhere to the regulations in its dealings with him." Ibid. In the court's view, the understanding of the parties was reinforced by the well-established legal principle that a federal agency must comply with its own regulations. The court concluded that respondent's allegation that his dismissal violated applicable regulations was "equivalent to an allegation of breach of an implied-in-fact contract," ibid., and that the District Court therefore had erred in ruling that it had no jurisdiction to award respondent monetary relief.3 11 Because this ruling appeared to be in conflict with our precedents, we granted certiorari. 454 U.S. 813, 102 S.Ct. 88, 70 L.Ed.2d 81 (1981). II 12 The AAFES, like other military exchanges, is an " 'ar[m] of the government deemed by it essential for the performance of governmental functions . . . and partake[s] of whatever immunities it may have under the constitution and federal statutes.' " United States v. Mississippi Tax Comm'n, 421 U.S. 599, 606, 95 S.Ct. 1872, 1877, 44 L.Ed.2d 404 (1975), quoting, with approval, language of the District Court in the same case, 378 F.Supp. 558, 562-563 (SD Miss.1974). As a result, the federal courts may entertain actions against the Service only if Congress has consented to suit; "a waiver of the traditional sovereign immunity 'cannot be implied but must be unequivocally expressed.' " United States v. Testan, 424 U.S. 392, 399, 96 S.Ct. 948, 953, 47 L.Ed.2d 114 (1976), quoting United States v. King, 395 U.S. 1, 4, 89 S.Ct. 1501, 1502, 23 L.Ed.2d 52 (1969). 13 The Tucker Act effects one such explicit waiver when it provides in pertinent part: 14 "The district courts shall have original jurisdiction, concurrent with the Court of Claims of: 15 * * * * * 16 ". . . Any other civil action or claim against the United States, not exceeding $10,000 in amount, founded either upon the Constitution, or any Act of Congress, or any regulation of an executive department, or upon any express or implied contract with the United States, or for liquidated or unliquidated damages in cases not sounding in tort. . . . For the purpose of this paragraph, an express or implied contract with the Army and Air Force Exchange Service . . . shall be considered an express or implied contract with the United States."4 28 U.S.C. § 1346(a)(2) (1976 ed., Supp. IV) (emphasis added).5 17 Respondent does not assert Tucker Act jurisdiction on the basis of the Constitution or a specific statute or regulation. He claims only that the Tucker Act affords him a remedy because of an "express or implied contract with the United States" agreed to by the parties. Specifically, respondent urges that he became an AAFES employee, or at least entered the EMP, by virtue of an employment contract, not by appointment, and that the AAFES regulations governing dismissal of employees created an implied contract. We must reject both contentions. A. 18 In determining whether respondent's employment was the result of appointment or contract, we look to United States v. Hopkins, 427 U.S. 123, 96 S.Ct. 2508, 49 L.Ed.2d 361 (1976), a wrongful-discharge action brought by an AAFES employee who alleged that his separation from the Service constituted a breach of an employment contract. The Court in its per curiam opinion inHopkins noted that Tucker Act jurisdiction may be premised on an employment contract, as well as on one for goods or other services, id., at 126, 96 S.Ct., at 2510, and that the AAFES regulations authorize the Service to enter into service contracts. Id., at 127-128, 96 S.Ct., at 2511. But the Court also observed that many AAFES employees are appointed to their positions, and it remanded the case for consideration of the question whether the plaintiff had been employed by contract or by appointment, a determination dependent upon "an analysis of the statutes and regulations previously described in light of whatever evidence is adduced on remand as to plaintiff's particular status in this case." Id., at 130, 96 S.Ct., at 2512. 19 Although respondent alleges that he was employed, both initially and upon entering the EMP, by express employment contracts, he points to nothing in the record or in the relevant AAFES regulations that substantiates that claim. In fact, his complaint supports the contrary view. The complaint observes that respondent was first "employed" by the AAFES in 1962, App. 3; the regulations pertaining to "employees" refer to Service personnel as "Federal employees of an instrumentality of the United States" who are appointed to their positions. AR 60-21/AFR 147-15, ch. 1, § I, ¶ 1-6(a); ch. 2, § I, &Par; 2-2, 2-3 (1 Aug. 1979). Moreover, if, as respondent alleges, he was "employed" in a data processing position, AAFES regulations prohibit the Service from negotiating a contract with him. See AR 60-20/AFR 147-14, ch. 3, § III, ¶ 3-26(d) (15 Nov. 1978). 20 Respondent's selection to the EMP plainly was pursuant to appointment. The regulations governing the EMP appear in the provision entitled "Exchange Service Personnel Policies," AR 60-21/AFR 147-15, ch. 5, § II, rather than in the regulation providing for service contracts, AR 60-20/AFR 147-14, ch. 3, §§ II, III.6 And, in language that connotes appointment rather than contract, the EMP regulations refer to one's "nomination, selection, and designation to EMP status," AR 60-21/AFR 147-15, ch. 5, § II, ¶ 5-8.7 Furthermore, respondent complains that he was separated from the EMP in violation of discharge procedures described in the regulation applicable to appointed employees, not to those who have contracted with the AAFES to provide services. App. 4-5, 7; see AR 60-21/AFR 147-15, ch. 3. 21 Despite these clear indications that respondent was appointed to his position, he maintains, citing United States v. Hopkins, supra, that he is entitled to an evidentiary hearing aimed at ascertaining the nature of his employment status. In Hopkins, however, the plaintiff's complaint alleged that he had been employed pursuant to contract. The Court of Claims did not examine this allegation because it erroneously assumed that AAFES employees could never be appointed. This Court held that the plaintiff's allegation was sufficient to withstand the Government's motion to dismiss for want of jurisdiction and remanded the case because "the question of whether plaintiff was employed by virtue of a contract or by appointment is not susceptible of determination at this time." 427 U.S., at 130, 96 S.Ct., at 2512. Resolution of the question, the Court noted, depended upon an analysis of the applicable statutes and regulations "in light of whatever evidence is adduced on remand as to plaintiff's particular status in this case." Ibid.8 22 Respondent's complaint, in contrast, does not claim that he was employed pursuant to a contract. In fact, it supports the Government's view that he was appointed. Even after the AAFES moved in the District Court to dismiss for want of jurisdiction on the ground that respondent had been "an appointed (non-contract) employee," Memorandum of Points and Authorities in Support of Defendant's Motion to Dismiss or in the Alternative for Summary Judgment 8, respondent did not seek to amend his complaint and did not allege any facts indicating the existence of an employment contract. See Memorandum of Authorities in Opposition to Defendants' (sic ) Motion to Dismiss or in the Alternative, for Summary Judgment 5-6; see also id., at 1-3 (referring to respondent's status as an "employee" and to violations of regulations governing AAFES employees).9 Moreover, as discussed above, all the evidence in the record is to the effect that respondent was appointed to his positions with the AAFES. Under these circumstances, we conclude that a remand on this question would serve no purpose and that respondent will not now be able to adduce evidence, which he has heretofore declined to present, that he was employed—either initially or upon entering the EMP—pursuant to an express employment contract. B 23 The Court of Appeals' decision rests on a different theory that, whether or not respondent was initially employed by virtue of a contract or by appointment, the AAFES regulations governing separation procedures created an implied-in-fact contract that the Service would adhere to those regulations while respondent continued in AAFES employment.10 This approach, however, is foreclosed by our prior decisions. 24 In United States v. Testan, 424 U.S. 392, 96 S.Ct. 948, 47 L.Ed.2d 114 (1976), the Court concluded, without dissent, that the Tucker Act did not confer jurisdiction over a complaint filed by civil service employees who claimed that they were entitled to reclassification at a higher grade. The Act, the Court observed, "is itself only a jurisdictional statute; it does not create any substantive right enforceable against the United States for money damages." Id., at 398, 96 S.Ct., at 953. Rather, a plaintiff's "asserted entitlement to money damages depends upon whether any federal statute 'can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained.' " Id., at 400, 96 S.Ct., at 954, quoting Eastport S.S. Corp. v. United States, 178 Ct.Cl. 599, 607, 372 F.2d 1002, 1009 (1967). The Court explicitly rejected the argument that "the violation of any statute or regulation relating to federal employment automatically creates a cause of action against the United States for money damages." 424 U.S., at 401, 96 S.Ct., at 954; see also United States v. Hopkins, 427 U.S., at 130, 96 S.Ct., at 2512. 25 As Testan makes clear, jurisdiction over respondent's complaint cannot be premised on the asserted violation of regulations that do not specifically authorize awards of money damages.11 Respondent cannot escape the force of Testan by relying on the Court's observation that the plaintiffs in that case did not "rest their claims upon a contract," 424 U.S., at 400, 96 S.Ct., at 954, and distinguishing this case on the ground that the regulations effected an implied contract. To accept this reasoning would be to undermine the Court's ruling in Testan that the Tucker Act provides a remedy only where damages claims against the United States have been authorized explicitly. Admittedly, the Testan plaintiffs did not assert the existence of an employment contract, but neither did respondent until very late in the litigation. And if employment statutes and regulations create an implied-in-fact contract, surely the Court would have so noted in Testan instead of directing that the complaint be dismissed. See id., at 408, 96 S.Ct., at 958. Moreover, the plaintiff in Hopkins did claim that he had been employed pursuant to a contract; the Court's remand for consideration of the plaintiff's status as an appointed or contract employee, despite a claim that his discharge contravened applicable regulations, clearly suggests that employment regulations do not automatically give rise to an implied-in-fact contract.12 26 In addition to mandating different results in Testan and Hopkins, the Court of Appeals' approach would "rende[r] superfluous" "many of the federal statutes—such as the Back Pay Act—that expressly provide money damages as a remedy against the United States in carefully limited circumstances." United States v. Testan, 424 U.S., at 404, 96 S.Ct., at 956. The Back Pay Act, which permits an employee to recover lost wages due to "an unjustified or unwarranted personnel action which has resulted in the withdrawal or reduction of all or part" of the compensation to which he was otherwise entitled, 5 U.S.C. § 5596(b)(1) (1976 ed., Supp. IV), expressly denies that cause of action to AAFES personnel. See 5 U.S.C. § 2105(c)(1) (1976 ed., Supp. IV). Congress' intent to prohibit a backpay claim by a Service employee would obviously be subverted if the employee could sue under the Tucker Act whenever he asserted a violation of the Service's regulations governing termination. And the impact of the Court of Appeals' decision would not be limited to such circumstances: as counsel for respondent appeared to concede at oral argument, the Court of Appeals' reasoning would extend Tucker Act jurisdiction to reach any complaint filed by a federal employee alleging the violation of a personnel statute or regulation. Tr. of Oral Arg. 20-21. 27 We therefore conclude that Testan is controlling, and we hold that the Court of Appeals erred in implying a contract based solely on the existence of AAFES personnel regulations and in premising Tucker Act jurisdiction on those regulations, which do not explicitly authorize damages awards. Because the court's judgment may not be sustained on the ground that respondent was hired pursuant to an express employment contract, we find that the Tucker Act did not confer jurisdiction over respondent's claims for monetary relief. 28 The judgment of the Court of Appeals is therefore reversed. 29 It is so ordered. 30 THE CHIEF JUSTICE concurs in the judgment. 1 AAFES is a nonappropriated fund instrumentality of the United States, that is, one that does not receive funds by congressional appropriation. See 10 U.S.C. §§ 4779(c) and 9779(c). AAFES is under the control of the Secretaries of the Army and Air Force and, like other military post exchanges, is intended "to provide convenient and reliable sources where soldiers can obtain their ordinary needs at the lowest possible prices." Standard Oil Co. v. Johnson, 316 U.S. 481, 484-485, 62 S.Ct. 1168, 1169, 86 L.Ed. 1611 (1942). 2 The regulations cited are those currently in effect. They differ in no material respect from the regulations that were outstanding and applicable while respondent was employed by the AAFES. 3 Reasoning that § 1346(a)(2) does not confer federal jurisdiction to award nonmonetary relief, the Court of Appeals looked to the general federal-question jurisdictional provision, 28 U.S.C. § 1331(a), to support its finding of jurisdiction over respondent's request for reinstatement. Although the court concluded that § 1331(a) does not constitute a waiver of sovereign immunity, it interpreted the 1976 amendment to § 10 of the Administrative Procedure Act, 5 U.S.C. § 702, as effecting a waiver for actions against federal agencies, where the agency conduct is otherwise subject to judicial review. 619 F.2d, at 1138-1140. Given its determination that the District Court could provide respondent both monetary and nonmonetary relief under alternative statutes, the Court of Appeals held, finally, that the District Court did not have jurisdiction over respondent's complaint pursuant to the mandamus statute, 28 U.S.C. § 1361. 619 F.2d, at 1140-1141. Neither side seeks review of those rulings here. 4 The last sentence of § 1346(a)(2) was added in 1970 by Pub.L.91-350, 84 Stat. 449, following this Court's decision some years before in Standard Oil Co. v. Johnson, 316 U.S. 481, 62 S.Ct. 1168, 86 L.Ed. 1611 (1942). Relying on the Court's observation in that case that the "Government assumes none of the financial obligations" of military post exchanges, id., at 485, 62 S.Ct., at 1170, the Court of Claims, in a series of decisions, had held that it could not entertain contract claims against nonappropriated fund instrumentalities. See United States v. Hopkins, 427 U.S. 123, 125, 96 S.Ct. 2508, 2510, 49 L.Ed.2d 361 (1976). In 1970, Congress sought to close this "loophole" by expressly affording contractors a Tucker Act remedy against such instrumentalities. See id., at 126, 96 S.Ct., at 2511; S.Rep.No.91-268, p. 2 (1969); H.R.Rep.No.91-933, p. 2 (1970), U.S.Code Cong. & Admin.News, p. 3477. 5 Section 1346(a)(2) gives the district courts concurrent jurisdiction with the Court of Claims over all civil actions or claims seeking damages of $10,000 or less. The Court of Claims has sole jurisdiction under the Tucker Act, however, for claims greater than $10,000. See 28 U.S.C. § 1491 (1976 ed., Supp.IV). Both jurisdictional provisions are otherwise identical. See Richardson v. Morris, 409 U.S. 464, 466, 93 S.Ct. 629, 631, 34 L.Ed.2d 647 (1973); United States v. Sherwood, 312 U.S. 584, 590-591, 61 S.Ct. 767, 771, 85 L.Ed. 1058 (1941). 6 The AAFES regulations define "service contract" as follows: "A contract whereby a contractor performs a service for AAFES off a military installation, such as laundry, drycleaning, photo processing, and repair service. This type contract may also include procurement of direct services such as janitorial and window cleaning service." AR 60-20/AFR 147-14, App. A, ¶ A-8(e) (15 Nov. 1978). 7 Respondent points to the portion of the EMP regulations providing that an EMP employee must have "acknowledged in writing that he understands and accepts the conditions of the EMP as prescribed by the Commander, AAFES." AR 60-21/AFR 147-15, ch. 5, § II, ¶ 5-7(b). An employee's acknowledgment and acceptance of the conditions of his employment, however, hardly demonstrate that he is employed pursuant to a contract; surely, an employer could require a nominee to acknowledge and accept the conditions of his appointment. 8 We are advised that Hopkins' suit was settled on the remand, and that no further inquiry was made into his employment status. See Brief for Petitioner 16, n. 9; Brief for Respondent 14; Tr. of Oral Arg. 28. 9 Respondent did seek to amend his complaint, however, following the Court of Appeals' decision that the AAFES discharge regulations created an implied-in-fact contract between the parties. The amended complaint alleges that a contract was executed when respondent signed an acknowledgment of the conditions of the EMP, includes a breach-of-contract count, and refers repeatedly to the "employment agreement." First Amended Complaint 2, 6-7. 10 Claims grounded on implied-in-fact contracts may be brought under the Tucker Act, but the Act does not confer jurisdiction with respect to contracts implied in law. See Hatzlachh Supply Co. v. United States, 444 U.S. 460, 465, n. 5, 100 S.Ct. 647, 650, n. 5, 62 L.Ed.2d 614 (1980). 11 Like Testan, this case does not involve a suit "for money improperly exacted or retained" or a claim based on a regulation that promises money. 424 U.S., at 401, 402, 96 S.Ct., at 954. This case is therefore distinguishable from those cited by respondent where contracts were inferred from regulations promising payment. See Griffin v. United States, 215 Ct.Cl. 710, 714-715 (1978); New York Airways, Inc. v. United States, 177 Ct.Cl. 800, 816-817, 369 F.2d 743, 751-752 (1966); Radium Mines, Inc. v. United States, 139 Ct.Cl. 144, 147-148, 153 F.Supp. 403, 405-406 (1957); Aycock-Lindsey Corp. v. United States, 171 F.2d 518, 521 (CA5 1948); Augusta Aviation, Inc. v. United States, 500 F.Supp. 785, 786-787 (SD Ga.1980), rev'd, 671 F.2d 445 (CA11 1982). Because respondent has not demonstrated that the parties entered into an express contract, this case is also different from those where regulations were considered an implied part of an express contract. See Bodek v. Department of Treasury, Bureau of Public Debt, 532 F.2d 277, 279, n. 7 (CA2), cert. denied, 429 U.S. 849, 97 S.Ct. 137, 50 L.Ed.2d 122 (1976); Wolak v. United States, 366 F.Supp. 1106, 1110 (Conn.1973); Spicer v. United States, 217 F.Supp. 44, 50 (Kan.1963), aff'd, 332 F.2d 750 (CA10 1964). 12 The Court's observation in Testan that the case was "not one concerning a wrongful discharge or a wrongful suspension," 424 U.S., at 402, 96 S.Ct., at 955, does not indicate, as respondent urges, that any termination or suspension suit may be brought under the Tucker Act. In Hopkins, the Court relied on Testan in disposing, summarily and adversely, of the contention that "plaintiff's discharge in violation of executive regulations constituted a claim enforceable under the Tucker Act." 427 U.S., at 130, 96 S.Ct., at 2512.

89

456 U.S. 717 102 S.Ct. 2112 72 L.Ed.2d 511 SUMMIT VALLEY INDUSTRIES, INC., Petitionerv.LOCAL 112, UNITED BROTHERHOOD OF CARPENTERS AND JOINERS OF AMERICA. No. 81-497. Argued April 28, 1982. Decided June 1, 1982. Syllabus As the result of a labor dispute between petitioner employer and respondent union, petitioner filed an unfair labor practice charge against the union, alleging that it had violated the secondary boycott and jurisdictional picketing prohibitions of § 8(b)(4) of the National Labor Relations Act (NLRA). The National Labor Relations Board (Board) found against the union, and the Board's order was judicially enforced. Petitioner filed this action in Federal District Court pursuant to § 303 of the Labor Management Relations Act (LMRA), seeking damages resulting from the union's illegal activity in an amount that included both business losses and attorney's fees incurred during the Board proceedings. Section 303(a) makes it unlawful for a union to engage in conduct defined as an unfair labor practice under § 8(b)(4) of the NLRA, and § 303(b) provides that whoever is injured in his business or property because of a violation of § 303(a) may sue in a federal district court "and shall recover the damages by him sustained and the cost of the suit." The District Court entered judgment for petitioner, awarding it an amount that represented its business losses, but concluded that petitioner was not entitled to recover attorney's fees as part of its damages. The Court of Appeals affirmed. Held : Attorney's fees incurred during Board proceedings are not a proper element of damages under § 303(b) of the LMRA. Pp. 721-727. (a) Neither the language nor the legislative history of § 303 supports petitioner's contention that § 303 provides statutory authorization for such attorney's fees for purposes of the American Rule that attorney's fees are not ordinarily recoverable in the absence of a statute or enforceable contract providing therefor. The legislative history instead shows that Congress did not intend to expand the ordinary meaning of the term "damages" in § 303(b) to include attorney's fees. Cf. Teamsters v. Morton, 377 U.S. 252, 84 S.Ct. 1253, 12 L.Ed.2d 280. Pp. 721-724. (b) Nor can allowance of attorney's fees incurred during Board proceedings be justified on the asserted ground that it would further Congress' intent to protect employers from the adverse effects of a union's illegal secondary activity. This interest is adequately protected by the award of compensatory damages for the business losses resulting from the union's prohibited conduct. Even assuming that attorney's fees are necessary to "fully" compensate the victimized employer, this justification alone is not sufficient to create an exception to the American Rule in the absence of express congressional authority. Cf. F. D. Rich Co. v. United States ex rel. Industrial Lumber Co., 417 U.S. 116, 128-129, 94 S.Ct. 2157, 2164-2165, 40 L.Ed.2d 703. To adopt petitioner's analysis would authorize the recovery of attorney's fees in every case where the plaintiff has prevailed against the defendant in prior litigation involving the same issues. Such a result is clearly barred by this Court's prior decisions and by the American Rule. Pp. 724-726. 9 Cir., 652 F.2d 65, affirmed. Donald C. Robinson, Butte, Mont., for petitioner. David S. Paull, Portland, Or., for respondent. Justice MARSHALL delivered the opinion of the Court. 1 We granted certiorari to decide whether § 303 of the Labor Management Relations Act (LMRA), 61 Stat. 158, as amended, 29 U.S.C. § 187, authorizes the recovery of attorney's fees incurred in prior proceedings before the National Labor Relations Board (Board). 454 U.S. 1079, 102 S.Ct. 632, 70 L.Ed.2d 613 (1981). The Courts of Appeals have divided on this issue.1 In this case, the Court of Appeals for the Ninth Circuit held that attorney's fees may not be recovered. We affirm. 2 * Petitioner Summit Valley Industries, Inc. (Summit Valley), manufactures prefabricated modular homes. These homes are completed at petitioner's plant and sold directly to home buyers. The buyer then independently obtains the services of a local contractor to install the completed home and to attach ancillary structures. In June 1972, Summit Valley opened a plant in the Butte, Mont., area. Rather than utilizing skilled union carpenters at this plant, petitioner hired unskilled workers. These workers were represented by Butte Teamsters Union, Local No. 2 (Teamsters), under a collective-bargaining agreement between the Teamsters and Summit Valley. 3 Upon learning that Summit Valley employed no skilled carpenters, respondent Local 112 of the United Brotherhood of Carpenters and Joiners of America (Union) filed unfair labor practice charges against Summit Valley. The Union claimed that Summit Valley had violated a valid work-preservation agreement between the Union and area contractor associations. Although the Union withdrew these charges when it realized that Summit Valley was not a signatory to the agreement, it ordered its members not to work on the installation of petitioner's homes. On at least two occasions, Summit Valley sent its own employees to complete the installation work that would otherwise have been done by the Union. As a result, respondent began picketing petitioner's plant. 4 Summit Valley filed an unfair labor practice charge against the Union, alleging that respondent's work stoppage and picketing violated the secondary boycott and jurisdictional picketing prohibitions of the National Labor Relations Act (NLRA). §§ 8(b)(4)(B) and (D) of the NLRA, 29 U.S.C. §§ 158(b)(4)(B) and (D). The Regional Director of the Board initiated a § 10(l) proceeding, 29 U.S.C. § 160(l), in the united states district court for the District of Montana. The District Court imposed a temporary restraining order pending the outcome of the proceeding before the Board. Henderson v. United Brotherhood of Carpenters and Joiners of America, Local 112, Civ. Nos. 2235 & 2239 (1972). The Union ceased picketing in November 1972, and its members resumed the installation of Summit Valley's homes. 5 Summit Valley then initiated a § 10(k) proceeding, 29 U.S.C. § 160(k), seeking resolution of its claim that the Union had engaged in illegal jurisdictional picketing, prohibited by § 8(b)(4)(D). After a 2-day hearing, the Board found against the Union, holding that Summit Valley had validly assigned the work in question to the Teamsters. Carpenters, Local 112 (Summit Valley Industries), 202 N.L.R.B. 974, 83 LRRM 1013 (1973). The Union agreed not to pressure Summit Valley to reassign work in violation of its collective-bargaining agreement with the Teamsters. However, the Union maintained that it had the right to truthfully advise the public that petitioner did not employ its members in the construction of modular homes. 6 After the § 10(k) proceeding, an Administrative Law Judge (ALJ) conducted 15 days of hearings on the unfair labor practice charges. The ALJ concluded that the Union had violated §§ 8(b)(4)(B) and (D), but that the Union had sought to enforce the work-preservation clause on the good-faith belief that its actions were lawful. The Board adopted the findings of the ALJ, and it ordered the Union to cease and desist from these unfair labor practices, and to fully comply with the Board's order obtained as a result of the § 10(k) proceeding. Carpenters, Local 112 (Summit Valley Industries), 217 N.L.R.B. 902, 89 LRRM 1799 (1975). The Court of Appeals for the Ninth Circuit enforced the Board's order. Chamber of Commerce of United States v. NLRB, 574 F.2d 457 (1978). 7 This action was filed pursuant to § 303 of the LMRA in the United States District Court for the District of Montana. Summit Valley sought damages resulting from the Union's illegal secondary and jurisdictional activity in the amount of $17,279.33: $3,675.00 in business losses, and $13,604.33 in attorney's fees incurred during the Board proceedings. The District Court found that the Board's decision that the Union had committed unfair labor practices collaterally estopped the Union from relitigating this issue in the § 303 action. Relying on Mead v. Retail Clerks International Assn., 523 F.2d 1371 (CA9 1975), the District Court concluded that Summit Valley was not entitled to recover attorney's fees as part of its damages. The court entered judgment for Summit Valley, awarding it an amount that represented its business losses. 475 F.Supp. 665 (1979). The Court of Appeals affirmed in an unpublished per curiam. Civ. No. 79-4663 (CA9 1981); 652 F.2d 65 (1981) (affirmance order). II 8 Under the American Rule it is well established that attorney's fees "are not ordinarily recoverable in the absence of a statute or enforceable contract providing therefor." Fleischmann Distilling Corp. v. Maier Brewing Co., 386 U.S. 714, 717, 87 S.Ct. 1404, 1407, 18 L.Ed.2d 475 (1967). This Court has endorsed certain exceptions to this rule where necessary to further the interests of justice. See, e.g., Vaughan v. Atkinson, 369 U.S. 527, 82 S.Ct. 997, 8 L.Ed.2d 88 (1962) (bad faith); Toledo Scale Co. v. Computing Scale Co., 261 U.S. 399, 43 S.Ct. 458, 67 L.Ed. 719 (1923) (willful disobedience of a court order); Central Railroad & Banking Co. v. Pettus, 113 U.S. 116, 5 S.Ct. 387, 28 L.Ed. 915 (1885) (common fund). In the absence of one of these equitable exceptions, however, the rule has been consistently followed for almost 200 years. See Alyeska Pipeline Co. v. Wilderness Society, 421 U.S. 240, 249-250, 95 S.Ct. 1612, 1617-1618, 44 L.Ed.2d 141 (1975); Arcambel v. Wiseman, 3 Dall. 306, 1 L.Ed. 613 (1796). Recognizing this consistent adherence to the American Rule, petitioner contends that § 303 provides express statutory authorization for the recovery of attorney's fees incurred in prior Board proceedings. We find this contention unsupported by either the language or the legislative history of § 303. 9 Section 303 authorizes a private damages action for an employer who has been injured by a union's unfair labor practice. Section 303(a), 29 U.S.C. § 187(a), makes it unlawful for a labor organization to engage in conduct defined as an unfair labor practice under § 8(b)(4) of the NLRA. As a remedy for this conduct § 303(b) provides that "[w]hoever shall be injured in his business or property by reason of any violation of subsection (a) of this section may sue therefor in any district court of the United States . . . and shall recover the damages by him sustained and the cost of the suit." 29 U.S.C. § 187(b). 10 Section 303 does not expressly provide for the recovery of attorney's fees, so we are not presented with a situation where Congress has made "specific and explicit provisions for the allowance of" such fees. Alyeska Pipeline Co. v. Wilderness Society, supra, at 260, and n. 33, 95 S.Ct., at 1623, and n. 33 (collecting statutes). Nonetheless, Summit Valley argues that § 303 does specifically authorize a district court to award attorney's fees incurred for the purpose of terminating the Union's illegal secondary activity by providing for the recovery of "damages" resulting from this activity. Because attorney's fees expended to compel " 'resumption of work in effect take the place of other compensable damages which would continue to be suffered if work were not resumed,' " petitioner asserts that such fees are part of the damages caused by the Union's illegal activity. Brief for Petitioner 13, quoting Associated General Contractors v. Construction and General Laborers Local 563, 612 F.2d 1060, 1064 (CA8 1979). 11 In assessing petitioner's interpretation of the word "damages" in § 303(b), we begin with the "fundamental canon of statutory construction . . . that, unless otherwise defined, words will be interpreted as taking their ordinary, contemporary, common meaning." Perrin v. United States, 444 U.S. 37, 42, 100 S.Ct. 311, 314, 62 L.Ed.2d 199 (1979); Burns v. Alcala, 420 U.S. 575, 580-581, 95 S.Ct. 1180, 1184-1185, 43 L.Ed.2d 469 (1975). Ordinarily a statutory right to "damages" does not include an implicit authorization to award attorney's fees. Indeed, the American Rule presumes that the word "damages" means damages exclusive of fees. See, e.g., Arcambel v. Wiseman, supra; Day v. Woodworth, 13 How. 363, 14 L.Ed. 181 (1852); cf. Teamsters v. Morton, 377 U.S. 252, 260, n. 15, 84 S.Ct. 1253, 1259, n. 15, 12 L.Ed.2d 280 (1964). Thus, petitioner's claim can succeed only if an examination of the relevant legislative history demonstrates that Congress intended to give a broader than normal scope to the term "damages." 12 Our review of the legislative history of § 303 reveals no such intention. To the contrary, the little discussion pertaining to the scope of an employer's recovery under § 303(b) indicates that Congress did not intend to expand the term "damages" to include attorney's fees. The following colloquy between Senator Taft and Senator Morse is particularly instructive. In response to Senator Morse's suggestion that § 303(b) would impose virtually unlimited liability on unions, Senator Taft stated: "Under the Sherman Act the same question of boycott damage is subject to a suit for damages and attorneys' fees. In this case we simply provide for the amount of the actual damages." 93 Cong.Rec. 4872-4873 (1947) (emphasis added). We find these remarks persuasive evidence that Congress did not intend attorney's fees which were expended to stop a union from engaging in illegal activity to be recovered as "damages" under § 303(b).2 13 In this respect, petitioner's claim is analogous to that presented in Teamsters v. Morton, supra. In Teamsters, this Court reviewed the history and policies underlying § 303 in order to determine whether that provision authorized an award of punitive damages. Relying on the same colloquy quoted above, we concluded that "recovery for an employer's business losses caused by a union's peaceful secondary activity proscribed by § 303 should be limited to actual, compensatory damages." Id., at 260, 84 S.Ct., at 1259. As we have often noted, one of the primary justifications for the American Rule is that "one should not be penalized for merely defending or prosecuting a lawsuit." Fleischmann Distilling Corp. v. Maier Brewing Co., 386 U.S., at 718, 87 S.Ct., at 1407 (emphasis added). See also, F. D. Rich Co. v. United States ex rel. Industrial Lumber Co., 417 U.S. 116, 129, 94 S.Ct. 2157, 2165, 40 L.Ed.2d 703 (1974); Farmer v. Arabian American Oil Co., 379 U.S. 227, 235, 85 S.Ct. 411, 416, 13 L.Ed.2d 248 (1964). The congressional intention discerned in Teamsters to limit recovery to actual, compensatory damages, counsels against reading the word "damages" to include attorney's fees incurred during prior Board proceedings. 14 Nevertheless Summit Valley contends that allowing it to recover reasonable attorney's fees incurred during Board proceedings furthers the statutory intent to protect employers from the adverse effects of a union's illegal secondary activity. We agree that Congress was concerned about protecting employers from injury arising out of this type of activity when it enacted § 303. See generally S.Rep.No.105, 80th Cong., 1st Sess., 54-55 (1947), 1 Legislative History of the LMRA 460-461 (1974) (Leg.Hist.); 93 Cong.Rec. 5060 (1947), 2 Leg.Hist. 1371 (remarks of Sen. Taft). However, we fail to see why this interest cannot be adequately protected by the award of compensatory damages for the business losses resulting from the Union's prohibited conduct. 15 Ultimately, petitioner's argument rests on the assumption that "Congress plainly intended Section 303 to be fully remedial and to restore to the victimized employer all . . . losses caused by the illegal activity." Brief for Petitioner 23 (emphasis added). Even assuming that attorney's fees are necessary to achieve full compensation, this justification alone is not sufficient to create an exception to the American Rule in the absence of express congressional authority. See F. D. Rich Co. v. United States ex rel. Industrial Lumber Co., supra, at 128-129, 94 S.Ct., at 2164-2165. In F.D. Rich, this Court rejected the argument that attorney's fees should be awarded under the Miller Act, 49 Stat. 793, as amended, 80 Stat. 1139, 40 U.S.C. § 270a et seq., because the Act provided for recovery of "sums justly due," 40 U.S.C. § 270b(a), and, unless fees were awarded, the legislative intent in favor of full compensation would be frustrated. 417 U.S., at 128, 94 S.Ct., at 2164. 16 In squarely rejecting this claim, we found it to be nothing more than a "restate[ment] of one of the oft-repeated criticisms of the American Rule." Ibid. Although this Court acknowledged that "there is some force to the argument that a party who must bear the cost of his attorneys' fees out of his recovery is not made whole," we concluded that the countervailing considerations which support the American Rule argue against placing exclusive reliance on the need to provide full compensation. Id., at 129, 94 S.Ct., at 2165. These considerations include the possible deterrent effect that fee shifting would have on poor litigants with meritorious claims, the time, expense, and difficulty of litigating the fee question, and the possibility that the principle of independent advocacy might be threatened by having "the earnings of the attorney flow from the pen of the judge before whom he argues." Ibid. These same considerations persuade us not to infer that Congress intended to authorize fee shifting in § 303 actions in order to fully compensate an employer for the "damages sustained by him" as a result of a union's illegal activity. 17 Furthermore, petitioner's analysis would authorize the recovery of attorney's fees in every case where the plaintiff has prevailed against the defendant in prior litigation involving the same issues. See Mead v. Retail Clerks International Assn., 523 F.2d, at 1380.3 Quite simply, anytime a plaintiff must resort to litigation to enjoin the defendant from engaging in illegal conduct, arguably the plaintiff has been injured in the amount of its attorney's fees spent to obtain the injunction. Under petitioner's analysis, each plaintiff could recover the costs of this prior litigation as part of its damages in any subsequent action involving the same conduct. Such a result, however, is clearly barred by our prior decisions. The American Rule precludes courts from awarding attorney's fees incurred during prior proceedings in the same case. See Fleischmann Distilling Corp. v. Maier Brewing Co., supra. Similarly, courts have uniformly concluded that "where an action based on the same wrongful act has been prosecuted by the plaintiff against the defendant to a successful issue, he can not in a subsequent action recover, as damages, his costs and expenses in the former action." Ritter v. Ritter, 381 Ill. 549, 555, 46 N.E.2d 41, 44 (1943). See also Flanders v. Tweed, 15 Wall. 450, 21 L.Ed. 203 (1873); Mead v. Retail Clerks International Assn., supra, at 1380-1381; Ritter v. Ritter, 381 Ill., at 557, 46 N.E.2d, at 45 (collecting cases). 18 This rule is universally followed in order to avoid the endless stream of litigation that might ensue if successful litigants could recover their attorney's fees in subsequent actions: "immediately upon the entry of judgment the plaintiff would start another action against the defendant for his attorney fees and expenses incurred in obtaining the preceding judgment." Id., at 555, 46 N.E.2d, at 44. Under petitioner's construction of § 303(b) the word "damages" would always encompass attorney's fees expended in prior litigation. In the absence of clear support for this construction in the language or the legislative history of § 303 we decline to adopt such a broad exception to the American Rule. III 19 Attorney's fees incurred during prior Board proceedings are not a proper element of damages under § 303(b) of the LMRA. Accordingly, the judgment of the Court of Appeals for the Ninth Circuit is 20 Affirmed. 1 The Fifth, Sixth, and Eighth Circuits have held that attorney's fees may be recovered. See Texas Distributors, Inc. v. Local Union No. 100, 598 F.2d 393 (CA5 1979); Local Union No. 984, International Brotherhood of Teamsters v. Humko Co., 287 F.2d 231 (CA6), cert. denied, 366 U.S. 962, 81 S.Ct. 1922, 6 L.Ed.2d 1254 (1961); Associated General Contractors of Minnesota v. Construction and General Laborers Local No. 563, 612 F.2d 1060 (CA8 1979). The First Circuit has expressed approval of this rule in dicta. See F. F. Instrument Corp. v. Union de Tronquistas de Puerto Rico, 558 F.2d 607, 611 (1977). The Ninth Circuit alone has reached a contrary conclusion. See Mead v. Retail Clerks International Assn., 523 F.2d 1371 (1975). 2 Of course, these remarks are not conclusive. It is possible that Senator Taft's remarks may have been confined to a rejection of the recovery of attorney's fees expended during the § 303 proceeding itself. See Mead v. Retail Clerks International Assn., 523 F.2d, at 1380. 3 The rationale of petitioner's position would seem to require that fees be awarded whenever a defendant's wrongful conduct requires a plaintiff to incur more litigation expenses, even if those expenses are incurred in the same action. Furthermore, because the District Court gave the Board's findings collateral-estoppel effect in the § 303 action, petitioner did not have to relitigate the question whether the Union had committed an unfair labor practice. In effect, Summit Valley is therefore asking for fees that it would have incurred in the § 303 action had it not first litigated the issues before the Board.

56

456 U.S. 798 102 S.Ct. 2157 72 L.Ed.2d 572 UNITED STATES, Petitionerv.Albert ROSS, Jr. No. 80-2209. Argued March 1, 1982. Decided June 1, 1982. Syllabus Acting on information from an informant that a described individual was selling narcotics kept in the trunk of a certain car parked at a specified location, District of Columbia police officers immediately drove to the location, found the car there, and a short while later stopped the car and arrested the driver (respondent), who matched the informant's description. One of the officers opened the car's trunk, found a closed brown paper bag, and after opening the bag, discovered glassine bags containing white powder (later determined to be heroin). The officer then drove the car to headquarters, where another warrantless search of the trunk revealed a zippered leather pouch containing cash. Respondent was subsequently convicted of possession of heroin with intent to distribute—the heroin and currency found in the searches having been introduced in evidence after respondent's pretrial motion to suppress the evidence had been denied. The Court of Appeals reversed, holding that while the officers had probable cause to stop and search respondent's car—including its trunk without a warrant, they should not have opened either the paper bag or the leather pouch found in the trunk without first obtaining a warrant. Held: Police officers who have legitimately stopped an automobile and who have probable cause to believe that contraband is concealed somewhere within it may conduct a warrantless search of the vehicle that is as thorough as a magistrate could authorize by warrant. Pp. 804-825. (a) The "automobile exception" to the Fourth Amendment's warrant requirement established in Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543, applies to searches of vehicles that are supported by probable cause to believe that the vehicle contains contraband. In this class of cases, a search is not unreasonable if based on objective facts that would justify the issuance of a warrant, even though a warrant has not actually been obtained. Pp. 804-809. (b) However, the rationale justifying the automobile exception does not apply so as to permit a warrantless search of any movable container that is believed to be carrying an illicit substance and that is found in a public place—even when the container is placed in a vehicle (not otherwise believed to be carrying contraband). United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538; Arkansas v. Sanders, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235. Pp. 809-814. (c) Where police officers have probable cause to search an entire vehicle, they may conduct a warrantless search of every part of the vehicle and its contents, including all containers and packages, that may conceal the object of the search. The scope of the search is not defined by the nature of the container in which the contraband is secreted. Rather, it is defined by the object of the search and the places in which there is probable cause to believe that it may be found. For example, probable cause to believe that undocumented aliens are being transported in a van will not justify a warrantless search of a suitcase. Pp. 817-824. (d) The doctrine of stare decisis does not preclude rejection here of the holding in Robbins v. California, 453 U.S. 420, 101 S.Ct. 2841, 69 L.Ed.2d 744, and some of the reasoning in Arkansas v. Sanders, supra. Pp. 824-825. 210 U.S.App.D.C. 342, 655 F.2d 1159, reversed and remanded. Andrew L. Frey, Washington, D. C., for petitioner. William J. Garber, Washington, D. C., for respondent. Justice STEVENS delivered the opinion of the Court. 1 In Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543, the Court held that a warrantless search of an automobile stopped by police officers who had probable cause to believe the vehicle contained contraband was not unreasonable within the meaning of the Fourth Amendment. The Court in Carroll did not explicitly address the scope of the search that is permissible. In this case, we consider the extent to which police officers—who have legitimately stopped an automobile and who have probable cause to believe that contraband is concealed somewhere within it may conduct a probing search of compartments and containers within the vehicle whose contents are not in plain view. We hold that they may conduct a search of the vehicle that is as thorough as a magistrate could authorize in a warrant "particularly describing the place to be searched."1 2 * In the evening of November 27, 1978, an informant who had previously proved to be reliable telephoned Detective Marcum of the District of Columbia Police Department and told him that an individual known as "Bandit" was selling narcotics kept in the trunk of a car parked at 439 Ridge Street. The informant stated that he had just observed "Bandit" complete a sale and that "Bandit" had told him that additional narcotics were in the trunk. The informant gave Marcum a detailed description of "Bandit" and stated that the car was a "purplish maroon" Chevrolet Malibu with District of Columbia license plates. 3 Accompanied by Detective Cassidy and Sergeant Gonzales, Marcum immediately drove to the area and found a maroon Malibu parked in front of 439 Ridge Street. A license check disclosed that the car was registered to Albert Ross; a computer check on Ross revealed that he fit the informant's description and used the alias "Bandit." In two passes through the neighborhood the officers did not observe anyone matching the informant's description. To avoid alerting persons on the street, they left the area. 4 The officers returned five minutes later and observed the maroon Malibu turning off Ridge Street onto Fourth Street. They pulled alongside the Malibu, noticed that the driver matched the informant's description, and stopped the car. Marcum and Cassidy told the driver—later identified as Albert Ross, the respondent in this action—to get out of the vehicle. While they searched Ross, Sergeant Gonzales discovered a bullet on the car's front seat. He searched the interior of the car and found a pistol in the glove compartment. Ross then was arrested and handcuffed. Detective Cassidy took Ross' keys and opened the trunk, where he found a closed brown paper bag. He opened the bag and discovered a number of glassine bags containing a white powder. Cassidy replaced the bag, closed the trunk, and drove the car to headquarters. 5 At the police station Cassidy thoroughly searched the car. In addition to the "lunch-type" brown paper bag, Cassidy found in the trunk a zippered red leather pouch. He unzipped the pouch and discovered $3,200 in cash. The police laboratory later determined that the powder in the paper bag was heroin. No warrant was obtained. 6 Ross was charged with possession of heroin with intent to distribute, in violation of 21 U.S.C. § 841(a). Prior to trial, he moved to suppress the heroin found in the paper bag and the currency found in the leather pouch. After an evidentiary hearing, the District Court denied the motion to suppress. The heroin and currency were introduced in evidence at trial and Ross was convicted. 7 A three-judge panel of the Court of Appeals reversed the conviction. It held that the police had probable cause to stop and search Ross' car and that, under Carroll v. United States, supra, and Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419, the officers lawfully could search the automobile including its trunk—without a warrant. The court considered separately, however, the warrantless search of the two containers found in the trunk. On the basis of Arkansas v. Sanders, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235, the court concluded that the constitutionality of a warrantless search of a container found in an automobile depends on whether the owner possesses a reasonable expectation of privacy in its contents. Applying that test, the court held that the warrantless search of the paper bag was valid but the search of the leather pouch was not. The court remanded for a new trial at which the items taken from the paper bag, but not those from the leather pouch, could be admitted.2 8 The entire Court of Appeals then voted to rehear the case en banc. A majority of the court rejected the panel's conclusion that a distinction of constitutional significance existed between the two containers found in respondent's trunk; it held that the police should not have opened either container without first obtaining a warrant. The court reasoned: 9 "No specific, well-delineated exception called to our attention permits the police to dispense with a warrant to open and search 'unworthy' containers. Moreover, we believe that a rule under which the validity of a warrantless search would turn on judgments about the durability of a container would impose an unreasonable and unmanageable burden on police and courts. For these reasons, and because the Fourth Amendment protects all persons, not just those with the resources or fastidiousness to place their effects in containers that decision-makers would rank in the luggage line, we hold that the Fourth Amendment warrant requirement forbids the warrantless opening of a closed, opaque paper bag to the same extent that it forbids the warrantless opening of a small unlocked suitcase or a zippered leather pouch." 210 U.S.App.D.C. 342, 344, 655 F.2d 1159, 1161 (1981) (footnote omitted). 10 The en banc Court of Appeals considered, and rejected, the argument that it was reasonable for the police to open both the paper bag and the leather pouch because they were entitled to conduct a warrantlesssearch of the entire vehicle in which the two containers were found. The majority concluded that this argument was foreclosed by Sanders. 11 Three dissenting judges interpreted Sanders differently.3 Other courts also have read the Sanders opinion in different ways.4 Moreover, disagreement concerning the proper interpretation of Sanders was at least partially responsible for the fact that Robbins v. California, 453 U.S. 420, 101 S.Ct. 2841, 69 L.Ed.2d 744, was decided last Term without a Court opinion. 12 There is, however, no dispute among judges about the importance of striving for clarification in this area of the law. For countless vehicles are stopped on highways and public streets every day, and our cases demonstrate that it is not uncommon for police officers to have probable cause to believe that contraband may be found in a stopped vehicle. In every such case a conflict is presented between the individual's constitutionally protected interest in privacy and the public interest in effective law enforcement. No single rule of law can resolve every conflict, but our conviction that clarification is feasible led us to grant the Government's petition for certiorari in this case and to invite the parties to address the question whether the decision in Robbins should be reconsidered. 454 U.S. 891, 102 S.Ct. 386, 70 L.Ed.2d 205. II 13 We begin with a review of the decision in Carroll itself. In the fall of 1921, federal prohibition agents obtained evidence that George Carroll and John Kiro were "bootleggers" who frequently traveled between Grand Rapids and Detroit in an Oldsmobile Roadster.5 On December 15, 1921, the agents unexpectedly encountered Carroll and Kiro driving west on that route in that car. The officers gave pursuit, stopped the roadster on the highway, and directed Carroll and Kiro to get out of the car. 14 No contraband was visible in the front seat of the Oldsmobile and the rear portion of the roadster was closed. One of the agents raised the rumble seat but found no liquor. He raised the seat cushion and again found nothing. The officer then struck at the "lazyback" of the seat and noticed that it was "harder than upholstery ordinarily is in those backs." 267 U.S., at 174, 45 S.Ct., at 292. He tore open the seat cushion and discovered 68 bottles of gin and whiskey concealed inside. No warrant had been obtained for the search. 15 Carroll and Kiro were convicted of transporting intoxicating liquor in violation of the National Prohibition Act. On review of those convictions, this Court ruled that the warrantless search of the roadster was reasonable within the meaning of the Fourth Amendment. In an extensive opinion written by Chief Justice Taft, the Court held: 16 "On reason and authority the true rule is that if the search and seizure without a warrant are made upon probable cause, that is, upon a belief, reasonably arising out of circumstances known to the seizing officer, that an automobile or other vehicle contains that which by law is subject to seizure and destruction, the search and seizure are valid. The Fourth Amendment is to be construed in the light of what was deemed an unreasonable search and seizure when it was adopted, and in a manner which will conserve public interests as well as the interests and rights of individual citizens." Id., at 149, 45 S.Ct., at 283. 17 The Court explained at length the basis for this rule. The Court noted that historically warrantless searches of vessels, wagons, and carriages—as opposed to fixed premises such as a home or other building—had been considered reasonable by Congress. After reviewing legislation enacted by Congress between 1789 and 1799,6 the Court stated: tween goods subject to forfeiture, when concealed in a dwelling house or similar place, and like goods in course of transportation and concealed in a movable vessel where they readily could be put out of reach of a search warrant." Id., at 151, 45 S.Ct., at 284. 18 The Court reviewed additional legislation passed by Congress7 and again noted that 19 "the guaranty of freedom from unreasonable searches and seizures by the Fourth Amendment has been construed, practically since the beginning of the Government, as recognizing a necessary difference between a search of a store, dwelling house or other structure in respect of which a proper official warrant readily may be obtained, and a search of a ship, motor boat, wagon or automobile, for contraband goods, where it is not practicable to secure a warrant because the vehicle can be quickly moved out of the locality or jurisdiction in which the warrant must be sought." Id., at 153, 45 S.Ct., at 285. 20 Thus, since its earliest days Congress had recognized the impracticability of securing a warrant in cases involving the transportation of contraband goods.8 It is this impracticability, viewed in historical perspective, that provided the basis for the Carroll decision. Given the nature of an automobile in transit, the Court recognized that an immediate intrusion is necessary if police officers are to secure the illicit substance. In this class of cases, the Court held that a warrantless search of an automobile is not unreasonable.9 21 In defining the nature of this "exception" to the general rule that "[i]n cases where the securing of a warrant is reasonably practicable, it must be used," id., at 156, 45 S.Ct., at 285, the Court in Carroll emphasized the importance of the requirement that officers have probable cause to believe that the vehicle contains contraband. 22 "Having thus established that contraband goods concealed and illegally transported in an automobile or other vehicle may be searched for without a warrant, we come now to consider under what circumstances such search may be made. It would be intolerable and unreasonable if a prohibition agent were authorized to stop every automobile on the chance of finding liquor and thus subject all persons lawfully using the highways to the inconvenience and indignity of such a search. Travellers may be so stopped in crossing an international boundary because of national self protection reasonably requiring one entering the country to identify himself as entitled to come in, and his belongings as effects which may be lawfully brought in. But those lawfully within the country, entitled to use the public highways, have a right to free passage without interruption or search unless there is known to a competent official authorized to search, probable cause for believing that their vehicles are carrying contraband or illegal merchandise." Id., at 153-154, 45 S.Ct., at 285. 23 Moreover, the probable-cause determination must be based on objective facts that could justify the issuance of a warrant by a magistrate and not merely on the subjective good faith of the police officers. " '[A]s we have seen, good faith is not enough to constitute probable cause. That faith must be grounded on facts within knowledge of the [officer], which in the judgment of the court would make his faith reasonable.' " Id., at 161-162, 45 S.Ct., at 288 (quoting Director General of Railroads v. Kastenbaum, 263 U.S. 25, 28, 44 S.Ct. 52, 53, 68 L.Ed. 146).10 24 In short, the exception to the warrant requirement established in Carroll —the scope of which we consider in this case—applies only to searches of vehicles that are supported by probable cause.11 In this class of cases, a search is not unreasonable if based on facts that would justify the issuance of a warrant, even though a warrant has not actually been obtained.12 III 25 The rationale justifying a warrantless search of an automobile that is believed to be transporting contraband arguably applies with equal force to any movable container that is believed to be carrying an illicit substance. That argument, however, was squarely rejected in United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538. 26 Chadwick involved the warrantless search of a 200-pound footlocker secured with two padlocks. Federal railroad officials in San Diego became suspicious when they noticed that a brown footlocker loaded onto a train bound for Boston was unusually heavy and leaking talcum powder, a substance often used to mask the odor of marihuana. Narcotics agents met the train in Boston and a trained police dog signaled the presence of a controlled substance inside the footlocker. The agents did not seize the footlocker, however, at this time; they waited until respondent Chadwick arrived and the footlocker was placed in the trunk of Chadwick's automobile. Before the engine was started, the officers arrested Chadwick and his two companions. The agents then removed the footlocker to a secured place, opened it without a warrant, and discovered a large quantity of marihuana. 27 In a subsequent criminal proceeding, Chadwick claimed that the warrantless search of the footlocker violated the Fourth Amendment. In the District Court, the Government argued that as soon as the footlocker was placed in the automobile a warrantless search was permissible under Carroll. The District Court rejected that argument,13 and the Government did not pursue it on appeal.14 Rather, the Government contended in this Court that the warrant requirement of the Fourth Amendment applied only to searches of homes and other "core" areas of privacy. The Court unanimously rejected that contention.15 Writing for the Court, THE CHIEF JUSTICE stated: 28 "[I]f there is little evidence that the Framers intended the Warrant Clause to operate outside the home, there is no evidence at all that they intended to exclude from protection of the Clause all searches occurring outside the home. The absence of a contemporary outcry against warrantless searches in public places was because, aside from searches incident to arrest, such warrantless searches were not a large issue in colonial America. Thus, silence in the historical record tells us little about the Framers' attitude toward application of the Warrant Clause to the search of respondents' footlocker. What we do know is that the Framers were men who focused on the wrongs of that day but who intended the Fourth Amendment to safeguard fundamental values which would far outlast the specific abuses which gave it birth." 433 U.S., at 8-9, 97 S.Ct., at 2481-2482 (footnote omitted). 29 The Court in Chadwick specifically rejected the argument that the warrantless search was "reasonable" because a footlocker has some of the mobile characteristics that support warrantless searches of automobiles. The Court recognized that "a person's expectations of privacy in personal luggage are substantially greater than in an automobile," id., at 13, 97 S.Ct., at 2484, and noted that the practical problems associated with the temporary detention of a piece of luggage during the period of time necessary to obtain a warrant are significantly less than those associated with the detention of an automobile. Id., at 13, n. 7, 97 S.Ct., at 2484, n.7. In ruling that the warrantless search of the footlocker was unjustified, the Court reaffirmed the general principle that closed packages and containers may not be searched without a warrant. Cf. Ex parte Jackson, 96 U.S. 727, 24 L.Ed. 877; United States v. Van Leeuwen, 397 U.S. 249, 90 S.Ct. 1029, 25 L.Ed.2d 282. In sum, the Court in Chadwick declined to extend the rationale of the "automobile exception" to permit a warrantless search of any movable container found in a public place.16 30 The facts in Arkansas v. Sanders, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235, were similar to those in Chadwick. In Sanders, a Little Rock police officer received information from a reliable informant that Sanders would arrive at the local airport on a specified flight that afternoon carrying a green suitcase containing marihuana. The officer went to the airport. Sanders arrived on schedule and retrieved a green suitcase from the airline baggage service. Sanders gave the suitcase to a waiting companion, who placed it in the trunk of a taxi. Sanders and his companion drove off in the cab; police officers followed and stopped the taxi several blocks from the airport. The officers opened the trunk, seized the suitcase, and searched it on the scene without a warrant. As predicted, the suitcase contained marihuana. 31 The Arkansas Supreme Court ruled that the warrantless search of the suitcase was impermissible under the Fourth Amendment, and this Court affirmed. As in Chadwick, the mere fact that the suitcase had been placed in the trunk of the vehicle did not render the automobile exception of Carroll applicable; the police had probable cause to seize the suitcase before it was placed in the trunk of the cab and did not have probable cause to search the taxi itself.17 Since the suitcase had been placed in the trunk, no danger existed that its contents could have been secreted elsewhere in the vehicle.18 As THE CHIEF JUSTICE noted in his opinion concurring in the judgment: 32 The Court in Sanders did not, however, rest its decision solely on the authority of Chadwick. In rejecting the State's argument that the warrantless search of the suitcase was justified on the ground that it had been taken from an automobile lawfully stopped on the street, the Court broadly suggested that a warrantless search of a container found in an automobile could never be sustained as part of a warrantless search of the automobile itself.19 The Court did not suggest that it mattered whether probable cause existed to search the entire vehicle. It is clear, however, that in neither Chadwick nor Sanders did the police have probable cause to search the vehicle or anything within it except the footlocker in the former case and the green suitcase in the latter. 33 Robbins v. California, 453 U.S. 420, 101 S.Ct. 2841, 69 L.Ed.2d 744, however, was a case in which suspicion was not directed at a specific container. In that case the Court for the first time was forced to consider whether police officers who are entitled to conduct a warrantless search of an automobile stopped on a public roadway may open a container found within the vehicle. In the early morning of January 5, 1975, police officers stopped Robbins' station wagon because he was driving erratically. Robbins got out of the car, but later returned to obtain the vehicle's registration papers. When he opened the car door, the officers smelled marihuana smoke. One of the officers searched Robbins and discovered a vial of liquid; in a search of the interior of the car the officer found marihuana. The police officers then opened the tailgate of the station wagon and raised the cover of a recessed luggage compartment. In the compartment they found two packages wrapped in green opaque plastic. The police unwrapped the packages and discovered a large amount of marihuana in each. 34 Robbins was charged with various drug offenses and moved to suppress the contents of the plastic packages. The California Court of Appeal held that "[s]earch of the automobile was proper when the officers learned that appellant was smoking marijuana when they stopped him"20 and that the warrantless search of the packages was justified because "the contents of the packages could have been inferred from their outward appearance, so that appellant could not have held a reasonable expectation of privacy with respect to the contents." People v. Robbins, 103 Cal.App.3d 34, 40, 162 Cal.Rptr. 780, 783 (1980). 35 This Court reversed. Writing for a plurality, Justice Stewart rejected the argument that the outward appearance of the packages precluded Robbins from having a reasonable expectation of privacy in their contents. He also squarely rejected the argument that there is a constitutional distinction between searches of luggage and searches of "less worthy" containers. Justice Stewart reasoned that all containers are equally protected by the Fourth Amendment unless their contents are in plain view. The plurality concluded that the warrantless search was impermissible because Chadwick and Sanders had established that "a closed piece of luggage found in a lawfully searched car is constitutionally protected to the same extent as are closed pieces of luggage found anywhere else." 453 U.S., at 425, 101 S.Ct., at 2845. 36 In an opinion concurring in the judgment, Justice Powell, the author of the Court's opinion in Sanders, stated that "[t]he plurality's approach strains the rationales of our prior cases and imposes substantial burdens on law enforcement without vindicating any significant values of privacy." 453 U.S., at 429, 101 S.Ct., at 2847.21 He noted that possibly "the controlling question should be the scope of the automobile exception to the warrant requirement," id., at 435, 101 S.Ct., at 2850, and explained that under that view 37 "when the police have probable cause to search an automobile, rather than only to search a particular container that fortuitously is located in it, the exigencies that allow the police to search the entire automobile without a warrant support the warrantless search of every container found therein. See post, at 451, and n. 13 [101 S.Ct., at 2859, and n.13] (STEVENS, J., dissenting). This analysis is entirely consistent with the holdings in Chadwick and Sanders, neither of which is an 'automobile case,' because the police there had probable cause to search the double-locked footlocker and the suitcase respectively before either came near an automobile." Ibid. 38 The parties in Robbins had not pressed that argument, however, and Justice POWELL concluded that institutional constraints made it inappropriate to reexamine basic doctrine without full adversary presentation. He concurred in the judgment, since it was supported—although not compelled—by the Court's opinion in Sanders, and stated that a future case might present a better opportunity for thorough consideration of the basic principles in this troubled area. 39 That case has arrived. Unlike Chadwick and Sanders, in this case police officers had probable cause to search respondent's entire vehicle.22 Unlike Robbins, in this case the parties have squarely addressed the question whether, in the course of a legitimate warrantless search of an automobile, police are entitled to open containers found within the vehicle. We now address that question. Its answer is determined by the scope of the search that is authorized by the exception to the warrant requirement set forth inCarroll. IV 40 In Carroll itself, the whiskey that the prohibition agents seized was not in plain view. It was discovered only after an officer opened the rumble seat and tore open the upholstery of the lazyback. The Court did not find the scope of the search unreasonable. Having stopped Carroll and Kiro on a public road and subjected them to the indignity of a vehicle search—which the Court found to be a reasonable intrusion on their privacy because it was based on probable cause that their vehicle was transporting contraband—prohibition agents were entitled to tear open a portion of the roadster itself. The scope of the search was no greater than a magistrate could have authorized by issuing a warrant based on the probable cause that justified the search. Since such a warrant could have authorized the agents to open the rear portion of the roadster and to rip the upholstery in their search for concealed whiskey, the search was constitutionally permissible. 41 In Chambers v. Maroney the police found weapons and stolen property "concealed in a compartment under the dashboard." 399 U.S., at 44, 90 S.Ct., at 1977. No suggestion was made that the scope of the search was impermissible. It would be illogical to assume that the outcome of Chambers —or the outcome of Carroll itself—would have been different if the police had found the secreted contraband enclosed within a secondary container and had opened that container without a warrant. If it was reasonable for prohibition agents to rip open the upholstery in Carroll, it certainly would have been reasonable for them to look into a burlap sack stashed inside; if it was reasonable to open the concealed compartment in Chambers, it would have been equally reasonable to open a paper bag crumpled within it. A contrary rule could produce absurd results inconsistent with the decision in Carroll itself. 42 In its application of Carroll, this Court in fact has sustained warrantless searches of containers found during a lawful search of an automobile. In Husty v. United States, 282 U.S. 694, 51 S.Ct. 240, 75 L.Ed. 629, the Court upheld a warrantless seizure of whiskey found during a search of an automobile, some of which was discovered in "whiskey bags" that could have contained other goods.23 In Scher v. United States, 305 U.S. 251, 59 S.Ct. 174, 83 L.Ed. 151, federal officers seized and searched packages of unstamped liquor found in the trunk of an automobile searched without a warrant. As described by a police officer who participated in the search: "I turned the handle and opened the trunk and found the trunk completely filled with packages wrapped in brown paper, and tied with twine; I think somewhere around thirty packages, each one containing six bottles."24 In these cases it was not contended that police officers needed a warrant to open the whiskey bags or to unwrap the brown paper packages. These decisions nevertheless "have much weight, as they show that this point neither occurred to the bar or the bench." Bank of the United States v. Deveaux, 5 Cranch 61, 88, 3 L.Ed. 38 (Marshall, C. J.). The fact that no such argument was even made illuminates the profession's understanding of the scope of the search permitted under Carroll. Indeed, prior to the decisions in Chadwick and Sanders, courts routinely had held that containers and packages found during a legitimate warrantless search of an automobile also could be searched without a warrant.25 43 As we have stated, the decision in Carroll was based on the Court's appraisal of practical considerations viewed in the perspective of history. It is therefore significant that the practical consequences of the Carroll decision would be largely nullified if the permissible scope of a warrantless search of an automobile did not include containers and packages found inside the vehicle. Contraband goods rarely are strewn across the trunk or floor of a car; since by their very nature such goods must be withheld from public view, they rarely can be placed in an automobile unless they are enclosed within some form of container.26 The Court in Carroll held that "contraband goods concealed and illegally transported in an automobile or other vehicle may be searched for without a warrant." 267 U.S., at 153, 45 S.Ct., at 285 (emphasis added). As we noted in Henry v. United States, 361 U.S. 98, 104, 80 S.Ct. 168, 172, 4 L.Ed.2d 134, the decision in Carroll "merely relaxed the requirements for a warrant on grounds of practicability." It neither broadened nor limited the scope of a lawful search based on probable cause. 44 A lawful search of fixed premises generally extends to the entire area in which the object of the search may be found and is not limited by the possibility that separate acts of entry or opening may be required to complete the search.27 Thus, a warrant that authorizes an officer to search a home for illegal weapons also provides authority to open closets, chests, drawers, and containers in which the weapon might be found. A warrant to open a footlocker to search for marihuana would also authorize the opening of packages found inside. A warrant to search a vehicle would support a search of every part of the vehicle that might contain the object of the search. When a legitimate search is under way, and when its purpose and its limits have been precisely defined, nice distinctions between closets, drawers, and containers, in the case of a home, or between glove compartments, upholstered seats, trunks, and wrapped packages, in the case of a vehicle, must give way to the interest in the prompt and efficient completion of the task at hand.28 45 This rule applies equally to all containers, as indeed we believe it must. One point on which the Court was in virtually unanimous agreement in Robbins was that a constitutional distinction between "worthy" and "unworthy" containers would be improper.29 Even though such a distinction perhaps could evolve in a series of cases in which paper bags, locked trunks, lunch buckets, and orange crates were placed on one side of the line or the other,30 the central purpose of the Fourth Amendment forecloses such a distinction. For just as the most frail cottage in the kingdom is absolutely entitled to the same guarantees of privacy as the most majestic mansion,31 so also may a traveler who carries a toothbrush and a few articles of clothing in a paper bag or knotted scarf claim an equal right to conceal his possessions from official inspection as the sophisticated executive with the locked attache case. 46 As Justice Stewart stated in Robbins, the Fourth Amendment provides protection to the owner of every container that conceals its contents from plain view. 453 U.S., at 427, 101 S.Ct., at 2846 (plurality opinion). But the protection afforded by the Amendment varies in different settings. The luggage carried by a traveler entering the country may be searched at random by a customs officer; the luggage may be searched no matter how great the traveler's desire to conceal the contents may be. A container carried at the time of arrest often may be searched without a warrant and even without any specific suspicion concerning its contents. A container that may conceal the object of a search authorized by a warrant may be opened immediately; the individual's interest in privacy must give way to the magistrate's official determination of probable cause. 47 In the same manner, an individual's expectation of privacy in a vehicle and its contents may not survive if probable cause is given to believe that the vehicle is transporting contraband. Certainly the privacy interests in a car's trunk or glove compartment may be no less than those in a movable container. An individual undoubtedly has a significant interest that the upholstery of his automobile will not be ripped or a hidden compartment within it opened. These interests must yield to the authority of a search, however, which—in light of Carroll —does not itself require the prior approval of a magistrate. The scope of a warrantless search based on probable cause is no narrower—and no broader—than the scope of a search authorized by a warrant supported by probable cause. Only the prior approval of the magistrate is waived; the search otherwise is as the magistrate could authorize.32 48 The scope of a warrantless search of an automobile thus is not defined by the nature of the container in which the contraband is secreted. Rather, it is defined by the object of the search and the places in which there is probable cause to believe that it may be found. Just as probable cause to believe that a stolen lawnmower may be found in a garage will not support a warrant to search an upstairs bedroom, probable cause to believe that undocumented aliens are being transported in a van will not justify a warrantless search of a suitcase. Probable cause to believe that a container placed in the trunk of a taxi contains contraband or evidence does not justify a search of the entire cab. V 49 Our decision today is inconsistent with the disposition in Robbins v. California and with the portion of the opinion in Arkansas v. Sanders on which the plurality in Robbins relied. Nevertheless, the doctrine of stare decisis does not preclude this action. Although we have rejected some of the reasoning in Sanders, we adhere to our holding in that case; although we reject the precise holding in Robbins, there was no Court opinion supporting a single rationale for its judgment, and the reasoning we adopt today was not presented by the parties in that case. Moreover, it is clear that no legitimate reliance interest can be frustrated by our decision today.33 Of greatest importance, we are convinced that the rule we apply in this case is faithful to the interpretation of the Fourth Amendment that the Court has followed with substantial consistency throughout our history. 50 We reaffirm the basic rule of Fourth Amendment jurisprudence stated by Justice Stewart for a unanimous Court in Mincey v. Arizona, 437 U.S. 385, 390, 98 S.Ct. 2408, 2412, 57 L.Ed.2d 290: "The Fourth Amendment proscribes all unreasonable searches and seizures, and it is a cardinal principle that 'searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment—subject only to a few specifically established and well-delineated exceptions.' Katz v. United States, 389 U.S. 347, 357 [88 S.Ct. 507, 514, 19 L.Ed.2d 576] (footnotes omitted)." 51 The exception recognized in Carroll is unquestionably one that is "specifically established and well delineated." We hold that the scope of the warrantless search authorized by that exception is no broader and no narrower than a magistrate could legitimately authorize by warrant. If probable cause justifies the search of a lawfully stopped vehicle, it justifies the search of every part of the vehicle and its contents that may conceal the object of the search. 52 The judgment of the Court of Appeals is reversed. The case is remanded for further proceedings consistent with this opinion. 53 It is so ordered. 54 Justice BLACKMUN, concurring. 55 My dissents in prior cases have indicated my continuing dissatisfaction and discomfort with the Court's vacillation in what is rightly described as "this troubled area." Ante, at 817. See United States v. Chadwick, 433 U.S. 1, 17, 97 S.Ct. 2476, 2486, 53 L.Ed.2d 538 (1977); Arkansas v. Sanders, 442 U.S. 753, 768, 99 S.Ct. 2586, 2595, 61 L.Ed.2d 235 (1979); Robbins v. California, 453 U.S. 420, 436, 101 S.Ct. 2841, 2851, 69 L.Ed.2d 744 (1981). 56 I adhere to the views expressed in those dissents. It is important, however, not only for the Court as an institution, but also for law enforcement officials and defendants, that the applicable legal rules be clearly established. Justice STEVENS' opinion for the Court now accomplishes much in this respect, and it should clarify a good bit of the confusion that has existed. In order to have an authoritative ruling, I join the Court's opinion and judgment. 57 Justice POWELL, concurring. 58 In my opinion in Robbins v. California, 453 U.S. 420, 429, 101 S.Ct. 2841, 2847, 69 L.Ed.2d 744 (1981), concurring in the judgment, I stated that the judgment was justified, though not compelled, by the Court's opinion in Arkansas v. Sander, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235 (1979). I did not agree, however, with the "bright line" rule articulated by the plurality opinion. Rather, I repeated the view I long have held that one's "reasonable expectation of privacy" is a particularly relevant factor in determining the validity of a warrantless search. I have recognized that, with respect to automobiles in general, this expectation can be only a limited one. See Arkansas v. Sanders, supra, at 761, 99 S.Ct., at 2591; Almeida-Sanchez v. United States, 413 U.S. 266, 279, 93 S.Ct. 2535, 2542, 37 L.Ed.2d 596 (1973) (POWELL, J., concurring). I continue to think that in many situations one's reasonable expectation of privacy may be a decisive factor in a search case. 59 It became evident last Term, however, from the five opinions written in Robbins—in none of which THE CHIEF JUSTICE joined—that it is essential to have a Court opinion in automobile search cases that provides "specific guidance to police and courts in this recurring situation." Robbins v. California, supra, at 435, 101 S.Ct., at 2850 (POWELL, J., concurring in judgment). The Court's opinion today, written by Justice STEVENS and now joined by THE CHIEF JUSTICE and four other Justices, will afford this needed guidance. It is fair also to say that, given Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543 (1925), and Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970), the Court's decision does not depart substantially from Fourth Amendment doctrine in automobile cases. Moreover, in enunciating a readily understood and applied rule, today's decision is consistent with the similar step taken last Term in New York v. Belton, 453 U.S. 454, 101 S.Ct. 2860, 69 L.Ed.2d 768 (1981). 60 I join the Court's opinion. 61 Justice WHITE, dissenting. 62 I would not overrule Robbins v. California, 453 U.S. 420, 101 S.Ct. 2841, 69 L.Ed.2d 744 (1981). For the reasons stated by Justice Stewart in that case, I would affirm the judgment of the Court of Appeals. I also agree with much of Justice MARSHALL's dissent in this case. 63 Justice MARSHALL, with whom Justice BRENNAN joins, dissenting. 64 The majority today not only repeals all realistic limits on warrantless automobile searches, it repeals the Fourth Amendment warrant requirement itself. By equating a police officer's estimation of probable cause with a magistrate's, the Court utterly disregards the value of a neutral and detached magistrate. For as we recently, and unanimously, reaffirmed: 65 "The warrant traditionally has represented an independent assurance that a search and arrest will not proceed without probable cause to believe that a crime has been committed and that the person or place named in the warrant is involved in the crime. Thus, an issuing magistrate must meet two tests. He must be neutral and detached, and he must be capable of determining whether probable cause exists for the requested arrest or search. This Court long has insisted that inferences of probable cause be drawn by 'a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime.' " Shadwick v. City of Tampa, 407 U.S. 345, 350, 92 S.Ct. 2119, 2122, 32 L.Ed.2d 783 (1972), quoting Johnson v. United States, 333 U.S. 10, 14, 68 S.Ct. 367, 369, 92 L.Ed. 436 (1948). 66 A police officer on the beat hardly satisfies these standards. In adopting today's new rule, the majority opinion shows contempt for these Fourth Amendment values, ignores this Court's precedents, is internally inconsistent, and produces anomalous and unjust consequences. I therefore dissent. 67 * According to the majority, whenever police have probable cause to believe that contraband may be found within an automobile that they have stopped on the highway,1 they may search not only the automobile but also any container found inside it, without obtaining a warrant. The scope of the search, we are told, is as broad as a magistrate could authorize in a warrant to search the automobile. The majority makes little attempt to justify this rule in terms of recognized Fourth Amendment values. The Court simply ignores the critical function that a magistrate serves. And although the Court purports to rely on the mobility of an automobile and the impracticability of obtaining a warrant, it never explains why these concerns permit the warrantless search of a container, which can easily be seized and immobilized while police are obtaining a warrant. 68 The new rule adopted by the Court today is completely incompatible with established Fourth Amendment principles, and takes a first step toward an unprecedented "probable cause" exception to the warrant requirement. In my view, under accepted standards, the warrantless search of the containers in this case clearly violates the Fourth Amendment. 69 * "[I]t is a cardinal principle that 'searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment subject only to a few specifically established and well-delineated exceptions.' " Mincey v. Arizona, 437 U.S. 385, 390, 98 S.Ct. 2408, 2412, 57 L.Ed.2d 290 (1978), quoting Katz v. United States, 389 U.S. 347, 357, 88 S.Ct. 507, 514, 19 L.Ed.2d 576 (1967). The warrant requirement is crucial to protecting Fourth Amendment rights because of the importance of having the probable-cause determination made in the first instance by a neutral and detached magistrate. Time and again, we have emphasized that the warrant requirement provides a number of protections that a post hoc judicial evaluation of a policeman's probable cause does not. 70 The requirement of prior review by a detached and neutral magistrate limits the concentration of power held by executive officers over the individual, and prevents some overbroad or unjustified searches from occurring at all. See United States v. United States District Court, 407 U.S. 297, 317, 92 S.Ct. 2125, 2136, 32 L.Ed.2d 752 (1972); Abel v. United States, 362 U.S. 217, 252, 80 S.Ct. 683, 703, 4 L.Ed.2d 668 (1960) (BRENNAN, J., joined by Warren, C.J., and Black and Douglas, JJ., dissenting). Prior review may also "prevent hindsight from coloring the evaluation of the reasonableness of a search or seizure." United States v. Martinez-Fuerte, 428 U.S. 543, 565, 96 S.Ct. 3074, 3086, 49 L.Ed.2d 1116 (1976); see also Beck v. Ohio, 379 U.S. 89, 96, 85 S.Ct. 223, 228, 13 L.Ed.2d 142 (1964). Furthermore, even if a magistrate would have authorized the search that the police conducted, the interposition of a magistrate's neutral judgment reassures the public that the orderly process of law has been respected: 71 "The point of the Fourth Amendment, which often is not grasped by zealous officers, is not that it denies law enforcement the support of the usual inferences which reasonable men draw from evidence. Its protection consists in requiring that those inferences be drawn by a neutral and detached magistrate instead of being judged by the officer engaged in the often competitive enterprise of ferreting out crime." Johnson v. United States, supra, at 13-14, 68 S.Ct., at 368-369. 72 See also Marshall v. Barlow's, Inc., 436 U.S. 307, 323, 98 S.Ct. 1816, 1825, 56 L.Ed.2d 305 (1978); United States v. United States District Court, supra, at 321, 92 S.Ct., at 2138. The safeguards embodied in the warrant requirement apply as forcefully to automobile searches as to any others. 73 Our cases do recognize a narrow exception to the warrant requirement for certain automobile searches. Throughout our decisions, two major considerations have been advanced to justify the automobile exception to the warrant requirement. We have upheld only those searches that are actually justified by those considerations. 74 First, these searches have been justified on the basis of the exigency of the mobility of the automobile. See, e.g., Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970); Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543 (1925). This "mobility" rationale is something of a misnomer, cf. Cady v. Dombrowski, 413 U.S. 433, 442-443, 93 S.Ct. 2523, 2528-2529, 37 L.Ed.2d 706 (1973), since the police ordinarily can remove the car's occupants and secure the vehicle on the spot. However, the inherent mobility of the vehicle often creates situations in which the police's only alternative to an immediate search may be to release the automobile from their possession.2 This alternative creates an unacceptably high risk of losing the contents of the vehicle, and is a principal basis for the Court's automobile exception to the warrant requirement. See Chambers, supra, at 51, n. 9, 90 S.Ct., at 1981, n. 9. 75 In many cases, however, the police will, prior to searching the car, have cause to arrest the occupants and bring them to the station for booking. In this situation, the police can ordinarily seize the automobile and bring it to the station. Because the vehicle is now in the exclusive control of the authorities, any subsequent search cannot be justified by the mobility of the car. Rather, an immediate warrantless search of the vehicle is permitted because of the second major justification for the automobile exception: the diminished expectation of privacy in an automobile. 76 Because an automobile presents much of its contents in open view to police officers who legitimately stop it on a public way, is used for travel, and is subject to significant government regulation, this Court has determined that the intrusion of a warrantless search of an automobile is constitutionally less significant than a warrantless search of more private areas. See Arkansas v. Sanders, 442 U.S. 753, 761, 99 S.Ct. 2586, 2591, 61 L.Ed.2d 235 (1979) (collecting cases). This justification has been invoked for warrantless automobile searches in circumstances where the exigency of mobility was clearly not present. See, e.g., South Dakota v. Opperman, 428 U.S. 364, 367-368, 96 S.Ct. 3092, 3096, 49 L.Ed.2d 1000 (1976); Cady v. Dombrowski, supra, at 441-442, 93 S.Ct., at 2528. By focusing on the defendant's reasonable expectation of privacy, this Court has refused to require a warrant in situations where the process of obtaining such a warrant would be more intrusive than the actual search itself. Cf. Katz v. United States, supra. A defendant may consider the seizure of the car a greater intrusion than an immediate search. See Chambers, supra, at 51-52, 90 S.Ct., at 1981. Therefore, even where police can bring both the defendant and the automobile to the station safely and can house the car while they seek a warrant, the police are permitted to decide whether instead to conduct an immediate search of the car. In effect, the warrantless search is permissible because a warrant requirement would not provide significant protection of the defendant's Fourth Amendment interests. B 77 The majority's rule is flatly inconsistent with these established Fourth Amendment principles concerning the scope of the automobile exception and the importance of the warrant requirement. Historically, the automobile exception has been limited to those situations where its application is compelled by the justifications described above. Today, the majority makes no attempt to base its decision on these justifications. This failure is not surprising, since the traditional rationales for the automobile exception plainly do not support extending it to the search of a container found inside a vehicle. 78 The practical mobility problem—deciding what to do with both the car and the occupants if an immediate search is not conducted is simply not present in the case of movable containers, which can easily be seized and brought to the magistrate. See Sanders, 442 U.S., at 762-766, and nn. 10, 14, 99 S.Ct., at 2592-2594, and nn. 10, 14. The lesser-expectation-of-privacy rationale also has little force. A container, as opposed to the car itself, does not reflect diminished privacy interests. See id., at 762, 764-765, 99 S.Ct., at 2592, 2593. Moreover, the practical corollary that this Court has recognized—that depriving occupants of the use of a car may be a greater intrusion than an immediate search—is of doubtful relevance here, since the owner of a container will rarely suffer significant inconvenience by being deprived of its use while a warrant is being obtained. 79 Ultimately, the majority, unable to rely on the justifications underlying the automobile exception, simply creates a new "probable cause" exception to the warrant requirement for automobiles. We have soundly rejected attempts to create such an exception in the past, see Coolidge v. New Hampshire, 403 U.S. 443, 91 S.Ct. 2022, 29 L.Ed.2d 564 (1971), and we should do so again today. 80 In purported reliance on Carroll v. United States, supra, the Court defines the permissible scope of a search by reference to the scope of a probable-cause search that a magistrate could authorize. Under Carroll, however, the mobility of an automobile is what is critical to the legality of a warrantless search. Of course, Carroll properly confined the search to the probable-cause limits that would also limit a magistrate, but it did not suggest that the search could be as broad as a magistrate could authorize upon a warrant. A magistrate could authorize a search encompassing containers, even though the mobility rationale does not justify such a broad search. Indeed, the Court's reasoning might have justified the search of the entire car in Coolidge despite the fact that the car was not "mobile" at all. Thus, in blithely suggesting that Carroll "neither broadened nor limited the scope of a lawful search based on probable cause," ante, at 820, the majority assumes what has never been the law: that the scope of the automobile-mobility exception to the warrant requirement is as broad as the scope of a "lawful" probable-cause search of an automobile, i.e., one authorized by a magistrate. 81 The majority's sleight-of-hand ignores the obvious differences between the function served by a magistrate in making a determination of probable cause and the function of the automobile exception. It is irrelevant to a magistrate's function whether the items subject to search are mobile, may be in danger of destruction, or are impractical to store, or whether an immediate search would be less intrusive than a seizure without a warrant. A magistrate's only concern is whether there is probable cause to search them. Where suspicion has focused not on a particular item but only on a vehicle, home, or office, the magistrate might reasonably authorize a search of closed containers at the location as well. But an officer on the beat who searches an automobile without a warrant is not entitled to conduct a broader search than the exigency obviating the warrant justifies. After all, what justifies the warrantless search is not probable cause alone, but probable cause coupled with the mobility of the automobile. Because the scope of a warrantless search should depend on the scope of the justification for dispensing with a warrant, the entire premise of the majority's opinion fails to support its conclusion. 82 The majority's rule masks the startling assumption that a policeman's determination of probable cause is the functional equivalent of the determination of a neutral and detached magistrate. This assumption ignores a major premise of the warrant requirement—the importance of having a neutral and detached magistrate determine whether probable cause exists. See supra, at 828-829. The majority's explanation that the scope of the warrantless automobile search will be "limited" to what a magistrate could authorize is thus inconsistent with our cases, which firmly establish that an on-spot determination of probable cause is never the same as a decision by a neutral and detached magistrate. C 83 Our recent decisions in United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538 (1977), Arkansas v. Sanders, supra, and Robbins v. California, 453 U.S. 420, 101 S.Ct. 2841, 69 L.Ed.2d 744 (1981), clearly affirm that movable containers are different from automobiles for Fourth Amendment purposes. In Chadwick, the Court drew a constitutional distinction between luggage and automobiles in terms of substantial differences in expectations of privacy. 433 U.S., at 12, 97 S.Ct., at 2484. Moreover, the Court held that the mobility of such containers does not justify dispensing with a warrant, since federal agents had seized the luggage and safely transferred it to their custody under their exclusive control. Sanders explicitly held that "the warrant requirement of the Fourth Amendment applies to personal luggage taken from an automobile to the same degree it applies to such luggage in other locations." 442 U.S., at 766, 99 S.Ct., at 2594. And Robbins reaffirmed the Sanders rationale as applied to wrapped packages found in the unlocked luggage compartment of a vehicle. 453 U.S., at 425, 101 S.Ct., at 2845.3 84 In light of these considerations, I conclude that any movable container found within an automobile deserves precisely the same degree of Fourth Amendment warrant protection that it would deserve if found at a location outside the automobile. See Sanders, 442 U.S., at 763-765, and n. 13, 99 S.Ct., at 2592, 2593, and n. 13; Chadwick, supra, at 17, n. 1, 97 S.Ct., at 2486, n. 1 (BRENNAN, J., concurring). Chadwick, as the majority notes, "reaffirmed the general principle that closed packages and containers may not be searched without a warrant." Ante, at 812. Although there is no need to describe the exact contours of that protection in this dissenting opinion, it is clear enough that closed, opaque containers—regardless of whether they are "worthy" or are always used to store personal items—are ordinarily fully protected. Cf. Sanders, supra, at 764, n. 13, 99 S.Ct., at 2593, n. 13.4 85 Here, because respondent Ross had placed the evidence in question in a closed paper bag, the container could be seized, but not searched, without a warrant. No practical exigencies required the warrantless searches on the street or at the station: Ross had been arrested and was in custody when both searches occurred, and the police succeeded in transporting the bag to the station without inadvertently spilling its contents.5 II 86 In announcing its new rule, the Court purports to rely on earlier automobile search cases, especially Carroll v. United States. The Court's approach, however, far from being "faithful to the interpretation of the Fourth Amendment that the Court has followed with substantial consistency throughout our history," ante, at 824, is plainly contrary to the letter and the spirit of our prior automobile search cases. Moreover, the new rule produces anomalous and unacceptable consequences. 87 * The majority's argument that its decision is supported by our decisions in Carroll and Chambers is misplaced. The Court in Carroll upheld a warrantless search of an automobile for contraband on the basis of the impracticability of securing a warrant in cases involving the transportation of contraband goods. The Court did not, however, suggest that obtaining a warrant for the search of an automobile is always impracticable.6 "In cases where the securing of a warrant is reasonably practicable, it must be used. . . . In cases where seizure is impossible except without warrant, the seizing officer acts unlawfully and at his peril unless he can show the court probable cause." 267 U.S., at 156, 45 S.Ct., at 286 (emphasis added).7 As this Court reaffirmed in Chambers, 399 U.S., at 50, 90 S.Ct., at 1980, "[n]either Carroll, supra, nor other cases in this Court require or suggest that in every conceivable circumstance the search of an auto even with probable cause may be made without the extra protection for privacy that a warrant affords." 88 Notwithstanding the reasoning of these cases, the majority argues that Carroll and Chambers support its decisions because integral compartments of a car are functionally equivalent to containers found within a car, and because the practical advantages to the police of the Carroll doctrine "would be largely nullified if the permissible scope of a warrantless search of an automobile did not include containers and packages found inside the vehicle." Ante, at 820. Neither of these arguments is persuasive. First, the Court's argument that allowing warrantless searches of certain integral compartments of the car in Carroll and Chambers, while protecting movable containers within the car, would be "illogical" and "absurd," ante, at 818, ignores the reason why this Court has allowed warrantless searches of automobile compartments. Surely an integral compartment within a car is just as mobile, and presents the same practical problems of safekeeping, as the car itself. This cannot be said of movable containers located within the car. The fact that there may be a high expectation of privacy in both containers and compartments is irrelevant, since the privacy rationale is not, and cannot be, the justification for the warrantless search of compartments. 89 The Court's second argument, which focuses on the practical advantages to police of the Carroll doctrine, fares no better. The practical considerations which concerned the Carroll Court involved the difficulty of immobilizing a vehicle while a warrant must be obtained. The Court had no occasion to address whether containers present the same practical difficulties as the car itself or integral compartments of the car. They do not. See supra, at 832. Carroll hardly suggested, as the Court implies, ante, at 820, that a warrantless search is justified simply because it assists police in obtaining more evidence. 90 Although it can find no support for its rule in this Court's precedents or in the traditional justifications for the automobile exception, the majority offers another justification. In a footnote, the majority suggests that "practical considerations" militate against securing containers found during an automobile search and taking them to the magistrate. Ante, at 821, n. 28. The Court confidently remarks: "Prohibiting police from opening immediately a container in which the object of the search is most likely to be found and instead forcing them first to comb the entire vehicle would actually exacerbate the intrusion on privacy interests. Moreover, until the container itself was opened the police could never be certain that the contraband was not secreted in a yet undiscovered portion of the vehicle." Ibid. The vehicle would have to be seized while a warrant was obtained, a requirement inconsistent withCarroll and Chambers. Ante, at 821, n. 28. 91 This explanation is unpersuasive. As this Court explained in Sanders and as the majority today implicitly concedes, the burden to police departments of seizing a package or personal luggage simply does not compare to the burden of seizing and safeguarding automobiles. Sanders, 442 U.S., at 765, n. 14, 99 S.Ct., at 2593, n. 14; ante, at 811, and n. 16. Other aspects of the Court's explanation are also implausible. The search will not always require a "combing" of the entire vehicle, since police may be looking for a particular item and may discover it promptly. If, instead, they are looking more generally for evidence of a crime, the immediate opening of the container will not protect the defendant's privacy; whether or not it contains contraband, the police will continue to search for new evidence. Finally, the defendant, not the police, should be afforded the choice whether he prefers the immediate opening of his suitcase or other container to the delay incident to seeking a warrant. Cf. Sanders, supra, at 764, n. 12, 99 S.Ct., at 2593, n. 12. The more presumption, if a presumption is to replace the defendant's consent, is surely that the immediate search of a closed container will be a greater invasion of the defendant's privacy interests than a mere temporary seizure of the container.8 B 92 Finally, the majority's new rule is theoretically unsound and will create anomalous and unwarranted results. These consequences are readily apparent from the Court's attempt to reconcile its new rule with the holdings of Chadwick and Sanders.9 The Court suggests that probable cause to search only a container does not justify a warrantless search of an automobile in which it is placed, absent reason to believe that the contents could be secreted elsewhere in the vehicle. This, the majority asserts, is an indication that the new rule is carefully limited to its justification, and is not inconsistent with Chadwick and Sanders. But why is such a container more private, less difficult for police to seize and store, or in any other relevant respect more properly subject to the warrant requirement, than a container that police discover in a probable-cause search of an entire automobile?10 This rule plainly has peculiar and unworkable consequences: the Government "must show that the investigating officer knew enough but not too much, that he had sufficient knowledge to establish probable cause but insufficient knowledge to know exactly where the contraband was located." 210 U.S.App.D.C. 342, 384, 655 F.2d 1159, 1201 (1981) (en banc) (Wilkey, J., dissenting). 93 Alternatively, the majority may be suggesting that Chadwick and Sanders may be explained because the connection of the container to the vehicle was incidental in these two cases. That is, because police had pre-existing probable cause to seize and search the containers, they were not entitled to wait until the item was placed in a vehicle to take advantage of the automobile exception. Cf. Coolidge v. New Hampshire, 403 U.S. 443, 91 S.Ct. 2022, 29 L.Ed.2d 564 (1971); 2 W. LaFave, Search and Seizure 519-525 (1978). I wholeheartedly agree that police cannot employ a pretext to escape Fourth Amendment prohibitions and cannot rely on an exigency that they could easily have avoided. This interpretation, however, might well be an exception that swallows up the majority's rule. In neither Chadwick nor Sanders did the Court suggest that the delay of the police was a pretext for taking advantage of the automobile exception. For all that appears, the Government may have had legitimate reasons for not searching as soon as they had probable cause. In any event, asking police to rely on such an uncertain line in distinguishing between legitimate and illegitimate searches for containers in automobiles hardly indicates that the majority's approach has brought clarification to this area of the law. Ante, at 804; see Robbins, 453 U.S., at 435, 101 S.Ct., at 2850 (POWELL, J., concurring in judgment).11 III 94 The Court today ignores the clear distinction that Chadwick established between movable containers and automobiles. It also rejects all of the relevant reasoning of Sanders12 and offers a substitute rationale that appears inconsistent with the result. See supra, at 832. Sanders is therefore effectively overruled. And the Court unambiguously overrules "the disposition" of Robbins, ante, at 824, though it gingerly avoids stating that it is overruling the case itself. 95 The only convincing explanation I discern for the majority's broad rule is expediency: it assists police in conducting automobile searches, ensuring that the private containers into which criminal suspects often place goods will no longer be a Fourth Amendment shield. See ante, at 820. "When a legitimate search is under way," the Court instructs us, "nice distinctions between . . . glove compartments, upholstered seats, trunks, and wrapped packages . . . must give way to the interest in the prompt and efficient completion of the task at hand." Ante, at 821. No "nice distinctions" are necessary, however, to comprehend the well-recognized differences between movable containers (which, even after today's decision, would be subject to the warrant requirement if located outside an automobile), and the automobile itself, together with its integral parts. Nor can I pass by the majority's glib assertion that the "prompt and efficient completion of the task at hand" is paramount to the Fourth Amendment interests of our citizens. I had thought it well established that "the mere fact that law enforcement may be made more efficient can never by itself justify disregard of the Fourth Amendment." Mincey v. Arizona, 437 U.S., at 393, 98 S.Ct., at 2413.13 96 This case will have profound implications for the privacy of citizens traveling in automobiles, as the Court well understands. "For countless vehicles are stopped on highways and public streets every day and our cases demonstrate that it is not uncommon for police officers to have probable cause to believe that contraband may be found in a stopped vehicle." Ante, at 803-804. A closed paper bag, a toolbox, a knapsack, a suitcase, and an attache case can alike be searched without the protection of the judgment of a neutral magistrate, based only on the rarely disturbed decision of a police officer that he has probable cause to search for contraband in the vehicle.14 The Court derives satisfaction from the fact that its rule does not exalt the rights of the wealthy over the rights of the poor. Ante, at 822. A rule so broad that all citizens lose vital Fourth Amendment protection is no cause for celebration. 97 I dissent. 1 "The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized." U.S.Const., Amdt. 4. 2 The court rejected the Government's argument that the warrantless search of the leather pouch was justified as incident to respondent's arrest. App. to Pet. for Cert. 137a. The Government has not challenged this holding. 3 Judge Tamm, the author of the original panel opinion, reiterated the view that Sanders prohibited the warrantless search of the leather pouch but not the search of the paper bag. Judge Robb agreed that this result was compelled by Sanders, although he stated that in his opinion "the right to search an automobile should include the right to open any container found within the automobile, just as the right to search a lawfully arrested prisoner carries with it the right to examine the contents of his wallet and any envelope found in his pocket, and the right to search a room includes authority to open and search all the drawers and containers found within the room." 210 U.S.App.D.C., at 363, 655 F.2d, at 1180. Judge MacKinnon concurred with Judge Tamm that Sanders did not prohibit the warrantless search of the paper bag. Concerning the leather pouch, he agreed with Judge Wilkey, who dissented on the ground that Sanders should not be applied retroactively. 4 Many courts have held that Sanders requires that a warrant be obtained only for personal luggage and other "luggage-type" containers. See, e.g., United States v. Brown, 635 F.2d 1207 (CA6 1980); United States v. Jimenez, 626 F.2d 39 (CA7 1980). One court has held that Sanders does not apply if the police have probable cause to search an entire vehicle and not merely an isolated container within it. Cf. State v. Bible, 389 So.2d 42 (La.1980), vacated and remanded, 453 U.S. 918, 101 S.Ct. 3153, 69 L.Ed.2d 1001; State v. Hernandez, 408 So.2d 911 (La.1981); see also 210 U.S.App.D.C., at 363, 655 F.2d, at 1180 (Robb, J., dissenting). 5 On September 29, 1921, Carroll and Kiro met the agents in Grand Rapids and agreed to sell them three cases of whiskey. The sale was not consummated, however, possibly because Carroll learned the agents' true identity. In October, the agents discovered Carroll and Kiro driving the Oldsmobile Roadster on the road to Detroit, which was known as an active center for the introduction of illegal liquor into this country. The agents followed the roadster as far as East Lansing, but there abandoned the chase. 6 The legislation authorized customs officials to search any ship or vessel without a warrant if they had probable cause to believe that it concealed goods subject to duty. The same legislation required a warrant for searches of dwelling places. 267 U.S., at 150-151, 45 S.Ct., at 284. "Thus contemporaneously with the adoption of the Fourth Amendment we find in the first Congress, and in the following Second and Fourth Congresses, a difference made as to the necessity for a search warrant be- 7 In particular, the Court noted an 1815 statute that permitted customs officers not only to board and search vessels without a warrant "but also to stop, search and examine any vehicle, beast or person on which or whom they should suspect there was merchandise which was subject to duty or had been introduced into the United States in any manner contrary to law." Id., at 151, 45 S.Ct., at 284. 8 In light of this established history, individuals always had been on notice that movable vessels may be stopped and searched on facts giving rise to probable cause that the vehicle contains contraband, without the protection afforded by a magistrate's prior evaluation of those facts. 9 Subsequent cases make clear that the decision in Carroll was not based on the fact that the only course available to the police was an immediate search. As Justice Harlan later recognized, although a failure to seize a moving automobile believed to contain contraband might deprive officers of the illicit goods, once a vehicle itself has been stopped the exigency does not necessarily justify a warrantless search. Chambers v. Maroney, 399 U.S. 42, 62-64, 90 S.Ct. 1975, 1986-1987, 26 L.Ed.2d 419 (opinion of Harlan, J.). The Court in Chambers, however—with only Justice Harlan dissenting—refused to adopt a rule that would permit a warrantless seizure but prohibit a warrantless search. The Court held that if police officers have probable cause to justify a warrantless seizure of an automobile on a public roadway, they may conduct an immediate search of the contents of that vehicle. "For constitutional purposes, we see no difference between on the one hand seizing and holding a car before presenting the probable cause issue to a magistrate and on the other hand carrying out an immediate search without a warrant. Given probable cause to search, either course is reasonable under the Fourth Amendment." Id., at 52, 90 S.Ct., at 1981. The Court also has held that if an immediate search on the street is permissible without a warrant, a search soon thereafter at the police station is permissible if the vehicle is impounded. Chambers, supra; Texas v. White, 423 U.S. 67, 96 S.Ct. 304, 46 L.Ed.2d 209. These decisions are based on the practicalities of the situations presented and a realistic appraisal of the relatively minor protection that a contrary rule would provide for privacy interests. Given the scope of the initial intrusion caused by a seizure of an automobile—which often could leave the occupants stranded on the highway—the Court rejected an inflexible rule that would force police officers in every case either to post guard at the vehicle while a warrant is obtained or to tow the vehicle itself to the station. Similarly, if an immediate search on the scene could be conducted, but not one at the station if the vehicle is impounded, police often simply would search the vehicle on the street—at no advantage to the occupants, yet possibly at certain cost to the police. The rules as applied in particular cases may appear unsatisfactory. They reflect, however, a reasoned application of the more general rule that if an individual gives the police probable cause to believe a vehicle is transporting contraband, he loses the right to proceed on his way without official interference. 10 After reviewing the relevant authorities at some length, the Court concluded that the probable-cause requirement was satisfied in the case before it. The Court held that "the facts and circumstances within [the officers'] knowledge and of which they had reasonably trustworthy information were sufficient in themselves to warrant a man of reasonable caution in the belief that intoxicating liquor was being transported in the automobile which they stopped and searched." 267 U.S., at 162, 45 S.Ct., at 288. Cf. Brinegar v. United States, 338 U.S. 160, 176-177, 69 S.Ct. 1302, 1311, 93 L.Ed. 1879; Henry v. United States, 361 U.S. 98, 102, 80 S.Ct. 168, 171, 4 L.Ed.2d 134. 11 See Husty v. United States, 282 U.S. 694, 51 S.Ct. 240, 75 L.Ed. 629; Scher v. United States, 305 U.S. 251, 59 S.Ct. 174, 83 L.Ed. 151; Brinegar v. United States, supra; Henry v. United States, supra; Dyke v. Taylor Implement Mfg. Co., 391 U.S. 216, 88 S.Ct. 1472, 20 L.Ed.2d 538; Chambers v. Maroney, supra; Texas v. White, supra; Colorado v. Bannister, 449 U.S. 1, 101 S.Ct. 42, 66 L.Ed.2d 1. Warrantless searches of automobiles have been upheld in a variety of factual contexts quite different from that presented in Carroll. Cf. Cooper v. California, 386 U.S. 58, 87 S.Ct. 788, 17 L.Ed.2d 730; Cady v. Dombrowski, 413 U.S. 433, 93 S.Ct. 2523, 37 L.Ed.2d 706; South Dakota v. Opperman, 428 U.S. 364, 96 S.Ct. 3092, 49 L.Ed.2d 1000. Many of these searches do not require a showing of probable cause that the vehicle contains contraband. We are not called upon to—and do not—consider in this case the scope of the warrantless search that is permitted in those cases. 12 As the Court in Carroll concluded: "We here find the line of distinction between legal and illegal seizures of liquor in transport in vehicles. It is certainly a reasonable distinction. It gives the owner of an automobile or other vehicle seized under Section 26, in absence of probable cause, a right to have restored to him the automobile, it protects him under the Weeks [Weeks v. United States, 232 U.S. 383, 34 S.Ct. 341, 58 L.Ed. 652] and Amos [Amos v. United States, 255 U.S. 313, 41 S.Ct. 266, 65 L.Ed. 654] cases from use of the liquor as evidence against him, and it subjects the officer making the seizures to damages. On the other hand, in a case showing probable cause, the Government and its officials are given the opportunity which they should have, to make the investigation necessary to trace reasonably suspected contraband goods and to seize them." 267 U.S., at 156, 45 S.Ct., at 286. 13 The District Court noted: "In this case, there was no nexus between the search and the automobile, merely a coincidence. The challenged search in this case was one of a footlocker, not an automobile. The search took place not in an automobile, but in [the federal building]. The only connection that the automobile had to this search was that, prior to its seizure, the footlocker was placed on the floor of an automobile's open trunk." United States v. Chadwick, 393 F.Supp. 763, 772 (Mass.1975). 14 This Court specifically noted: "The Government does not contend that the footlocker's brief contact with Chadwick's car makes this an automobile search, but it is argued that the rationale of our automobile search cases demonstrates the reasonableness of permitting warrantless searches of luggage; the Government views such luggage as analogous to motor vehicles for Fourth Amendment purposes." 433 U.S., at 11-12, 97 S.Ct., at 2483-2484. 15 See id., at 17, 97 S.Ct., at 2486 (BLACKMUN, J., dissenting). 16 The Court concluded that there is a significant difference between the seizure of a sealed package and a subsequent search of its contents; the search of the container in that case was "a far greater intrusion into Fourth Amendment values than the impoundment of the footlocker." Id., 433 U.S., at 14, n. 8, 97 S.Ct., at 2485, n. 8. A temporary seizure of a package or piece of luggage often may be accomplished without as significant an intrusion upon the individual—and without as great a burden on the police—as in the case of the seizure of an automobile. See n. 9, supra. 17 The Arkansas Supreme Court carefully reviewed the facts of the case and concluded: "The information supplied to the police by the confidential informant is adequate to support the State's claim that the police had probable cause to believe that appellant's green suitcase contained a controlled substance when the police confiscated the suitcase and opened it." Sanders v. State, 262 Ark. 595, 599, 559 S.W.2d 704, 706 (1977). The court also noted: "The evidence in this case supports the conclusion that the relationship between the suitcase and the taxicab is coincidental." Id., at 600, n. 2, 559 S.W.2d, at 706, n. 2. 18 Moreover, none of the practical difficulties associated with the detention of a vehicle on a public highway that made the immediate search in Carroll reasonable could justify an immediate search of the suitcase, since the officers had no interest in detaining the taxi or its driver. "Because the police officers had probable cause to believe that respondent's green suitcase contained marihuana before it was placed in the trunk of the taxicab, their duty to obtain a search warrant before opening it is clear under United States v. Chadwick, 433 U.S. 1 [97 S.Ct. 2476, 53 L.Ed.2d 538] (1977). . . . ". . . Here, as in Chadwick, it was the luggage being transported by respondent at the time of the arrest, not the automobile in which it was being carried, that was the suspected locus of the contraband. The relationship between the automobile and the contraband was purely coincidental, as in Chadwick. The fact that the suitcase was resting in the trunk of the automobile at the time of respondent's arrest does not turn this into an 'automobile' exception case. The Court need say no more." 442 U.S., at 766-767, 99 S.Ct., at 2594. 19 The Court stated that "the extent to which the Fourth Amendment applies to containers and other parcels depends not at all upon whether they are seized from an automobile." 442 U.S., at 764, n. 13, 99 S.Ct., at 2593, n. 13. This general rule was limited only by the observation that "[n]ot all containers and packages found by police during the course of a search will deserve the full protection of the Fourth Amendment. Thus, some containers (for example a kit of burglar tools or a gun case) by their very nature cannot support any reasonable expectation of privacy because their contents can be inferred from their outward appearance. Similarly, in some cases the contents of a package will be open to 'plain view,' thereby obviating the need for a warrant." Ibid. 20 People v. Robbins, 103 Cal.App.3d 34, 39, 162 Cal.Rptr. 780, 782 (1980). 21 "While the plurality's blanket warrant requirement does not even purport to protect any privacy interest, it would impose substantial new burdens on law enforcement. Confronted with a cigar box or a Dixie cup in the course of a probable-cause search of an automobile for narcotics, the conscientious policeman would be required to take the object to a magistrate, fill out the appropriate forms, await the decision, and finally obtain the warrant. Suspects or vehicles normally will be detained while the warrant is sought. This process may take hours, removing the officer from his normal police duties. Expenditure of such time and effort, drawn from the public's limited resources for detecting or preventing crimes, is justified when it protects an individual's reasonable privacy interests. In my view, the plurality's requirement cannot be so justified. The aggregate burden of procuring warrants whenever an officer has probable cause to search the most trivial container may be heavy and will not be compensated by the advancement of important Fourth Amendment values." 453 U.S., at 433-434, 101 S.Ct., at 2849-2850 (POWELL, J., concurring in judgment). The substantial burdens on law enforcement identified by Justice POWELL would, of course, not be affected by the character of the container found during an automobile search. No comparable practical problems arise when the official suspicion is confined to a particular piece of luggage, as in Chadwick and Sanders. Cf. n. 19, supra. 22 The en banc Court of Appeals stated that "[b]ased on the tip the police received, Ross's car was properly stopped and searched, and the pouch and bag were properly seized." 210 U.S.App.D.C., at 361, 655 F.2d, at 1168 (footnote omitted). The court explained: "[W]e believe it clear that the police had ample and reasonable cause to stop Ross and to search his car. The informer had supplied accurate information on prior occasions, and he was an eyewitness to sales of narcotics by Ross. He said he had just seen Ross take narcotics from the trunk of his car in making a sale and heard him say he possessed additional narcotics." Id., at 361, n. 22, 655 F.2d, at 1168, n. 22. The court further noted: "In this case, the informant told the police that Ross had narcotics in the trunk of his car. No specific container was identified." Id., at 359, 655 F.2d, at 1166. 23 At the suppression hearing, defense counsel asked the police officer who had conducted the search: "Isn't it possible to put other goods in a bag that has the resemblance of a whiskey bag?" The officer responded: "I suppose it is. I did not think of that at that time. I knew it was whiskey, I was sure it was." App., O.T.1930, No. 477, p. 27. 24 App., O.T.1938, No. 49, p. 33. The brief of then Solicitor General Robert Jackson noted that the items searched "were wrapped in very heavy brown wrapping paper with at least two wrappings and with a heavy cord around them cross-wise so that they could readily be lifted." Brief for United States, O.T.1938, No. 49, p. 6. 25 See, e.g., United States v. Soriano, 497 F.2d 147, 149-150 (CA5 1974) (en banc); United States v. Vento, 533 F.2d 838, 867, n. 101 (CA3 1976); United States v. Tramunti, 513 F.2d 1087, 1104 (CA2 1975); United States v. Issod, 508 F.2d 990, 993 (CA7 1974); United States v. Evans, 481 F.2d 990, 994 (CA9 1973); United States v. Bowman, 487 F.2d 1229 (CA10 1973). Many courts continued to apply this rule following the decision in Chadwick. Cf. United States v. Milhollan, 599 F.2d 518, 526-527 (CA3 1979); United States v. Gaultney, 581 F.2d 1137, 1144-1145 (CA5 1978); United States v. Finnegan, 568 F.2d 637, 640-641 (CA9 1977). In ruling that police could search luggage and other containers found during a legitimate warrantless search of an automobile, courts often assumed that the "automobile exception" of Carroll applied whenever a container in an automobile was believed to contain contraband. That view, of course, has since been qualified by Chadwick and Sanders. 26 It is noteworthy that the early legislation on which the Court relied in Carroll concerned the enforcement of laws imposing duties on imported merchandise. See nn. 6 and 7, supra. Presumably such merchandise was shipped then in containers of various kinds, just as it is today. Since Congress had authorized warrantless searches of vessels and beasts for imported merchandise, it is inconceivable that it intended a customs officer to obtain a warrant for every package discovered during the search; certainly Congress intended customs officers to open shipping containers when necessary and not merely to examine the exterior of cartons or boxes in which smuggled goods might be concealed. During virtually the entire history of our country whether contraband was transported in a horse-drawn carriage, a 1921 roadster, or a modern automobile—it has been assumed that a lawful search of a vehicle would include a search of any container that might conceal the object of the search. 27 In describing the permissible scope of a search of a home pursuant to a warrant, Professor LaFave notes: "Places within the described premises are not excluded merely because some additional act of entry or opening may be required. 'In countless cases in which warrants described only the land and the buildings, a search of desks, cabinets, closets and similar items has been permitted.' " 2 W. LaFave, Search and Seizure 152 (1978) (quoting Massey v. Commonwealth, 305 S.W.2d 755, 756 (Ky.1957)). 28 The practical considerations that justify a warrantless search of an automobile continue to apply until the entire search of the automobile and its contents has been completed. Arguably, the entire vehicle itself (including its upholstery) could be searched without a warrant, with all wrapped articles and containers found during that search then taken to a magistrate. But prohibiting police from opening immediately a container in which the object of the search is most likely to be found and instead forcing them first to comb the entire vehicle would actually exacerbate the intrusion on privacy interests. Moreover, until the container itself was opened the police could never be certain that the contraband was not secreted in a yet undiscovered portion of the vehicle; thus in every case in which a container was found, the vehicle would need to be secured while a warrant was obtained. Such a requirement would be directly inconsistent with the rationale supporting the decisions in Carroll and Chambers. Cf. nn. 19 and 22, supra. 29 Cf. 453 U.S., at 426-427, 101 S.Ct., at 2845-2846 (plurality opinion); id., at 436, 101 S.Ct., at 2851 (BLACKMUN, J., dissenting); id., at 443, 101 S.Ct., at 2854 (REHNQUIST, J., dissenting); id., at 447, 101 S.Ct., at 2856 (STEVENS, J., dissenting). 30 If the distinction is based on the proposition that the Fourth Amendment protects only those containers that objectively manifest an individual's reasonable expectation of privacy, however, the propriety of a warrantless search necessarily would turn on much more than the fabric of the container. A paper bag stapled shut and marked "private" might be found to manifest a reasonable expectation of privacy, as could a cardboard box stacked on top of two pieces of heavy luggage. The propriety of the warrantless search seemingly would turn on an objective appraisal of all the surrounding circumstances. 31 " 'The poorest man may in his cottage bid defiance to all the forces of the Crown. It may be frail; its roof may shake; the wind may blow through it; the storm may enter; the rain may enter; but the King of England cannot enter—all his force dares not cross the threshold of the ruined tenement!' " Miller v. United States, 357 U.S. 301, 307, 78 S.Ct. 1190, 1194, 2 L.Ed.2d 1332 (quoting remarks attributed to William Pitt); cf. Payton v. New York, 445 U.S. 573, 601, n. 54, 100 S.Ct. 1371, 1388, n. 54, 63 L.Ed.2d 639. 32 In choosing to search without a warrant on their own assessment of probable cause, police officers of course lose the protection that a warrant would provide to them in an action for damages brought by an individual claiming that the search was unconstitutional. Cf. Monroe v. Pape, 365 U.S. 167, 81 S.Ct. 473, 5 L.Ed.2d 492. Although an officer may establish that he acted in good faith in conducting the search by other evidence, a warrant issued by a magistrate normally suffices to establish it. 33 Any interest in maintaining the status quo that might be asserted by persons who may have structured their business of distributing narcotics or other illicit substances on the basis of judicial precedents clearly would not be legitimate. 1 The Court confines its holding today to automobiles stopped on the highway which police have probable cause to believe contain contraband. I do not understand the Court to address the applicability of the automobile exception rule announced today to parked cars. Cf. Coolidge v. New Hampshire, 403 U.S. 443, 91 S.Ct. 2022, 29 L.Ed.2d 564 (1971). 2 The fact that the police are able initially to remove the occupants from the car does not remove the justification for an immediate search. If police could not conduct an immediate search of a stopped automobile, they would often be left with the difficult task of deciding what to do with the occupants while a warrant is obtained. In the case of a parked automobile, by contrast, if the automobile is unoccupied, this problem is not presented. See, e.g., Coolidge v. New Hampshire, supra. 3 The plurality stated: "[Chadwick and Sanders ] made clear, if it was not clear before, that a closed piece of luggage found in a lawfully searched car is constitutionally protected to the same extent as are closed pieces of luggage found anywhere else." Robbins v. California, 453 U.S., at 425, 101 S.Ct., at 2845. 4 This rule may present some line-drawing problems, but no greater than those presented when a movable container is in the arms of a citizen walking down the street. There is no justification for relying on marginal difficulties of definition to reject a warrant requirement in one situation but not the other. 5 The Government argues that less secure containers such as paper bags can easily spill their contents; thus, no privacy interest of the defendant is protected if police are required to seize the container and bring it to the station. Whatever the force of this argument in other contexts, here police succeeded in reclosing the bag after the initial search and transporting it to the station without incident. 6 The Court in Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543 (1925), seems to have assumed that the police could not arrest the occupants of the automobile, since the offense was a misdemeanor and was not deemed to have been committed in the officers' presence. See 2 W. LaFave, Search and Seizure 511 (1978). Accordingly, police were faced with an exigency often not encountered today in searches of stopped automobiles: in order to seize the car pending the securing of a warrant, they would have to leave the occupants stranded. 7 In Carroll, of course, no movable container was searched. Although in other early cases containers may in fact have been searched, see ante, at 818-819, the parties did not litigate in this Court the question whether containers deserve separate protection. The Court's suggestion that the absence of such an argument "illuminates the profession's understanding of the scope of the search permitted under Carroll," ante, at 819, is an unusual approach to constitutional interpretation. I would hesitate to rely upon the "profession's understanding" of the Fourteenth Amendment or of Plessy v. Ferguson, 163 U.S. 537, 16 S.Ct. 1138, 41 L.Ed. 256 (1896), in the early part of this century as justification for not granting Negroes constitutional protection. See Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954). Moreover, for a number of reasons, including the broad scope of the permitted search incident to arrest prior to Chimel v. California, 395 U.S. 752, 89 S.Ct. 2034, 23 L.Ed.2d 685 (1969), and the uncertain meaning of a "search" prior to Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967), the profession formerly advanced different arguments against automobile searches than it advances today. 8 Seizures of automobiles can be distinguished because of the greater interest of defendants in continuing possession of their means of transportation; in the case of automobiles, a seizure is more likely to be a greater intrusion than an immediate search. See Chambers v. Maroney, 399 U.S. 42, 51-52, 90 S.Ct. 1975, 1981, 26 L.Ed.2d 419 (1970). 9 Both cases would appear to fall within the majority's new rule. In United States v. Chadwick, 433 U.S. 1, 97 S.Ct. 2476, 53 L.Ed.2d 538 (1977), federal agents had probable cause to search a footlocker. Although the footlocker had been placed in the trunk of a car and the occupants were about to depart, the Court refused to rely on the automobile exception to uphold the search. (It is true that the United States did not argue in this Court that the search was justified pursuant to that exception, but the theory was hardly so novel that this Court could not have responsibly relied upon it.) In Arkansas v. Sanders, 442 U.S. 753, 99 S.Ct. 2586, 61 L.Ed.2d 235 (1979), too, the suitcase was mobile and police had probable cause to search it; it was carried in an automobile for several blocks before the automobile was stopped and the suitcase was seized and searched. Again, however, this Court invalidated the search. 10 In a footnote, the Court appears to suggest a more pragmatic rationale for distinguishing Chadwick and Sanders—that no practical problems comparable to those engendered by a general search of a vehicle would arise if the official suspicion is confined to a particular piece of luggage. Ante, at 816, n. 21. This suggestion is illogical. A general search might disclose only a single item worth searching; conversely, pre-existing suspicion might attach to a number of items later placed in a car. Surely the protection of the warrant requirement cannot depend on a numerical count of the items subject to search. 11 Unless one of these alternative explanations is adopted, the Court's attempt to distinguish the holdings in Chadwick and Sanders is not only unpersuasive but appears to contradict the Court's own theory. The Court suggests that in each case, the connection of the container to the vehicle was simply coincidental, and notes that the police did not have probable cause to search the entire vehicle. But the police assuredly did have probable cause to search the vehicle for the container. The Court states that the scope of the permitted warrantless search is determined only by what a magistrate could authorize. Ante, at 823. Once police found that container, according to the Court's own rule, they should have been entitled to search at least the container without a warrant. There was probable cause to search and the car was mobile in each case. 12 The Court suggests that it rejects "some of the reasoning in Sanders." Ante, at 824. But the Court in Sanders unambiguously stated: "[W]e hold that the warrant requirement of the Fourth Amendment applies to personal luggage taken from an automobile to the same degree it applies to such luggage in other locations." 442 U.S., at 766, 99 S.Ct., at 2594. The Court today instead adopts the reasoning of the opinion of THE CHIEF JUSTICE, joined by Justice STEVENS, who refused to join the majority opinion because of the breadth of its rationale. Ibid. 13 Of course, efficiency and promptness can never be substituted for due process and adherence to the Constitution. Is not a dictatorship the most "efficient" form of government? 14 The Court purports to restrict its rule to areas that the police have probable cause to search, as "defined by the object of the search and the places in which there is probable cause to believe that it may be found." Ante, at 824. I agree, of course, that the probable-cause component of the automobile exception must be strictly construed. I fear, however, that the restriction that the Court emphasizes may have little practical value. See 210 U.S.App.D.C. 342, 351, n. 21, 655 F.2d 1159, 1168, n. 21 (1981) (en banc). If police open a container within a car and find contraband, they may acquire probable cause to believe that other portions of the car, and other containers within it, will contain contraband. In practice, the Court's rule may amount to a wholesale authorization for police to search any car from top to bottom when they have suspicion, whether localized or general, that it contains contraband.

01

456 U.S. 865 102 S.Ct. 2290 72 L.Ed.2d 623 UNITED STATESv.LOUISIANA No. 9 Supreme Court of the United States June 1, 1982 1 On bill of complaint (Louisiana Boundary Case). ORDER 2 June 1, 1982. Pursuant to this Court's Final Decree herein entered on June 22, 1981, 452 U.S. 726 [101 S.Ct. 2605, 69 L.Ed.2d 368], the United States and the State of Louisiana filed their Final Accountings, and, subsequently, each party filed Objections to the Accounting of the other. By an Order entered on January 11, 1982, 454 U.S. 1135 [102 S.Ct. 990, 71 L.Ed.2d 287], the Court referred the said Objections to its Special Master. The parties have now agreed upon these matters, have submitted to the Special Master a proposed Order, and the Special Master, with the concurrence of both parties, has recommended its entry by the Court. 3 Accordingly, IT IS ORDERED, ADJUDGED, AND DECREED: 4 1. The Final Report of the Special Master is received and ordered filed. 5 2. The objections to accountings previously filed are sustained to the extent recommended in the Report of the Special Master. 6 3. All accountings required by the Court's Decree of June 22, 1981, have been made and, as supplemented by the ruling on the objections thereto, are now approved. 7 4. The United States is directed forthwith to pay over to the State of Louisiana the outstanding sum of $3,251,609.76. 8 5. After the payment directed by paragraph 4 above, neither party shall be accountable to the other for any further payment in respect of the matters in controversy in these proceedings between the United States and the State of Louisiana. 9 6. Upon receipt by the State of Louisiana of the payment directed by paragraph 4, above, the Interim Agreement of October 12, 1956, shall be deemed terminated for all purposes and all sums remaining in the impounded fund account established pursuant to that Agreement are unconditionally released to the United States. 10 7. The account submitted by the Special Master is approved and the balance owing to him shall be paid in equal shares by the United States and the State of Louisiana. 11 8. The Special Master is discharged in this case insofar as the proceedings involve the controversy between the United States and the State of Louisiana. 12 Justice MARSHALL took no part in the consideration or decision of this order.

910

456 U.S. 742 102 S.Ct. 2126 72 L.Ed.2d 532 FEDERAL ENERGY REGULATORY COMMISSION, et al., Appellants,v.MISSISSIPPI et al. No. 80-1749. Argued Jan. 19, 1982. Decided June 1, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1131, 103 S.Ct. 15. Syllabus The Public Utility Regulatory Policies Act of 1978 (PURPA) was enacted as part of a legislative package designed to combat the nationwide energy crisis. To further this effort, Titles I and III of PURPA direct state utility regulatory commissions and nonregulated utilities to "consider" the adoption and implementation of specific "rate design" and regulatory standards, and require state commissions to follow certain notice and comment procedures when acting on proposed federal standards. Section 210 of PURPA's Title II seeks to encourage the development of cogeneration and small power facilities, and directs the Federal Energy Regulatory Commission (FERC), in consultation with state regulatory authorities, to promulgate rules to carry out this goal. Section 210 then requires the state authorities, after notice and hearing, to implement such rules, and authorizes the FERC to exempt cogeneration and small power facilities from certain state and federal regulations. The State of Mississippi and the Mississippi Public Service Commission (appellees) brought an action in Federal District Court against the FERC and the Secretary of Energy (appellants), seeking a declaratory judgment that Titles I and III and § 210 are unconstitutional because they exceed congressional power under the Commerce Clause and constitute an invasion of state sovereignty in violation of the Tenth Amendment. The District Court so held and pronounced the challenged provisions void. Held : 1. The challenged provisions are within Congress' power under the Commerce Clause. Pp. 753-758. (a) To assert that PURPA is facially unconstitutional because it does not regulate "commerce," or because it does not have "a substantial effect" on such activity, disregards the specific congressional finding in § 2 of PURPA that the regulated activities do have an immediate effect on interstate commerce. Pp. 754-755. (b) The legislative history amply supports the congressional conclusion that limited federal regulation of retail sales of electricity and natural gas, and of the relationships between cogenerators and electric utilities, was essential to protect interstate commerce and the Nation's economy. Pp. 756-758. 2. The challenged provisions do not trench on state sovereignty in violation of the Tenth Amendment. Pp. 758-771. (a) Insofar as § 210 authorizes the FERC to exempt qualified power facilities from state laws and regulations, it does nothing more than pre-empt conflicting state enactments in the traditional way. Because of the substantial interstate effect of such activity, Congress may pre-empt the States completely in the regulation of retail sales by electric and gas utilities and of transactions between such utilities and cogenerators. With respect to § 210's requirement that state authorities implement FERC's rules, the statute and its implementing regulations simply require state commissions to settle disputes arising under the statute, the very type of adjudicatory activity customarily engaged in by the Mississippi Public Service Commission. Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967. Pp.759-761. (b) The "mandatory consideration" provisions of Titles I and III do not involve the compelled exercise of Mississippi's sovereign powers or set a mandatory agenda to be considered in all events by state legislative or administrative decisionmakers, but simply establish requirements for continued state activity in an otherwise pre-emptible field. Cf. Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 101 S.Ct. 2389, 69 L.Ed.2d 1. Pp. 761-770. (c) Similarly, the procedural requirements of Titles I and III do not compel the exercise of a State's sovereign power or purport to set standards to be followed in all areas of the state commission's endeavors. If Congress may require a state administrative body to consider proposed federal regulations as a condition to its continued involvement in a pre-emptible field, it may require the use of certain procedural minima during that body's deliberations on the subject. Pp.770-771 Reversed. Sol. Gen. Rex E. Lee, Washington, D. C., for appellants. Alex A. Alston, Jr., Jackson, Miss., for appellees. [Amicus Curiae Information from page 744-745 intentionally omitted] Justice BLACKMUN delivered the opinion of the Court. 1 In this case, appellees successfully challenged the constitutionality of Titles I and III, and of § 210 of Title II, of the Public Utility Regulatory Policies Act of 1978, Pub.L. 95-617, 92 Stat. 3117 (PURPA or Act). We conclude that appellees' challenge lacks merit and we reverse the judgment below. 2 * On November 9, 1978, President Carter signed PURPA into law.1 The Act was part of a package of legislation,2 approved the same day, designed to combat the nationwide energy crisis. 3 At the time, it was said that the generation of electricity consumed more than 25% of all energy resources used in the United States. S.Rep.No.95-442, p. 7 (1977), U.S.Code Cong. & Admin.News 1978, p. 7659. Approximately one-third of the electricity in this country was generated through use of oil and natural gas, and electricity generation was one of the fastest growing segments of the Nation's economy. S.Rep.No.95-361, p. 32 (1977). In part because of their reliance on oil and gas, electricity utilities were plagued with increasing costs and decreasing efficiency in the use of their generating capacities; each of these factors had an adverse effect on rates to consumers and on the economy as a whole. S.Rep.No.95-442, at 9. Congress accordingly determined that conservation by electricity utilities of oil and natural gas was essential to the success of any effort to lessen the country's dependence on foreign oil, to avoid a repetition of the shortage of natural gas that had been experienced in 1977, and to control consumer costs. A. Titles I and III 4 PURPA's Titles I and III, which relate to regulatory policies for electricity and gas utilities, respectively, are administered (with minor exceptions) by the Secretary of Energy. These provisions are designed to encourage the adoption of certain retail regulatory practices. The Titles share three goals: (1) to encourage "conservation of energy supplied by . . . utilities"; (2) to encourage "the optimization of the efficiency of use of facilities and resources" by utilities; and (3) to encourage "equitable rates to . . . consumers." §§ 101 and 301, 92 Stat. 3120 and 3149, 16 U.S.C. § 2611 (1976 ed., Supp. IV), 15 U.S.C. § 3201 (1976 ed., Supp. IV).3 To achieve these goals, Titles I and III direct state utility regulatory commissions and nonregulated utilities to "consider" the adoption and implementation of specific "rate design" and regulatory standards. 5 Section 111(d) of the Act, 16 U.S.C. § 2621(d), requires each state regulatory authority and nonregulated utility to consider the use of six different approaches to structuring rates: (1) promulgation, for each class of electricity consumers, of rates that, "to the maximum extent practicable," would "reflect the costs of . . . service to such class"; (2) elimination of declining block rates;4 (3) adoption of time-of-day rates;5 (4) promulgation of seasonal rates;6 (5) adoption of interruptible rates;7 and (6) use of load management techniques.8 The Act directed each state authority and nonregulated utility to consider these factors not later than two years after PURPA's enactment, that is, by November 8, 1980, and provided that the authority or utility by November 8, 1981, was to have made a decision whether to adopt the standards. § 2622(b). The statute does not provide penalties for failure to meet these deadlines; the state authority or nonregulated utility is merely directed to consider the standards at the first rate proceeding initiated by the authority after November 9, 1980. § 2622(c). 6 Section 113 of PURPA, 16 U.S.C. § 2623, requires each state regulatory authority and nonregulated utility to consider the adoption of a second set of standards relating to the terms and conditions of electricity service: (1) prohibition of master-metering in new buildings;9 (2) restrictions on the use of automatic adjustment clauses;10 (3) disclosure to consumers of information regarding rate schedules; (4) promulgation of procedural requirements relating to termination of service; and (5) prohibition of the recovery of advertising costs from consumers. Similarly, § 303, 15 U.S.C. § 3203, requires consideration of the last two standards—procedures for termination of service and the nonrecovery of advertising costs—for natural gas utilities. A decision as to the standards contained in §§ 113 and 303 was to have been made by November 1980, although, again, no penalty was provided by the statute for failure to meet the deadline. 7 Finally, § 114 of the Act, 16 U.S.C. § 2624, directs each state authority and nonregulated utility to consider promulgation of "lifeline rates"—that is, lower rates for service that meets the essential needs of residential consumers—if such rates have not been adopted by November 1980. 8 Titles I and III also prescribe certain procedures to be followed by the state regulatory authority and the nonregulated utility when considering the proposed standards. Each standard is to be examined at a public hearing after notice, and a written statement of reasons must be made available to the public if the standards are not adopted. 16 U.S.C. §§ 2621(b) and (c)(2), and §§ 2623(a) and (c); 15 U.S.C. §§ 3203(a) and (c). "Any person" may bring an action in state court to enforce the obligation to hold a hearing and make determinations on the PURPA standards. 16 U.S.C. § 2633(c)(1); 15 U.S.C. § 3207(b)(1). 9 The Secretary of Energy, any affected utility, and any consumer served by an affected utility is given the right to intervene and participate in any rate-related proceeding considering the Title I standards. 16 U.S.C. § 2631(a). Under Title III, the Secretary alone has the right to intervene. 15 U.S.C. § 3205. Any person (including the Secretary) who intervenes or otherwise participates in the proceeding may obtain review in state court of any administrative determination concerning the Title I standards, 16 U.S.C. § 2633(c)(1), and the Secretary has the right to participate as an amicus in any Title III judicial review proceeding initiated by another. 15 U.S.C. § 3207(b)(2). The right to intervene is enforceable against the state regulatory authority by an action in federal court. 16 U.S.C. § 2633(b); 15 U.S.C. § 3207(a)(2). 10 Titles I and III also set forth certain reporting requirements. Within one year of PURPA's enactment, and annually thereafter for 10 years, each state regulatory authority and nonregulated utility is to report to the Secretary "respecting its consideration of the standards established." 16 U.S.C. § 2626(a); 15 U.S.C. § 3209(a). The Secretary, in turn, is to submit a summary and analysis of these reports to Congress. 16 U.S.C. § 2626(b); 15 U.S.C. § 3209(b). Electricity utilities also are required to collect information concerning their service costs. 16 U.S.C. § 2643. This information is to be filed periodically with appellant Federal Energy Regulatory Commission (FERC) and with appropriate state regulatory agencies, and is to be made available to the public. Title III requires the Secretary, in consultation with FERC, state regulatory authorities, gas utilities, and gas consumers, to submit a report to Congress on gas utility rate design. 15 U.S.C. § 3206. 11 Despite the extent and detail of the federal proposals, however, no state authority or nonregulated utility is required to adopt or implement the specified rate design or regulatory standards. Thus, 16 U.S.C. §§ 2621(a) and 2623(a) and 15 U.S.C. § 3203(a) all provide: "Nothing in this subsection prohibits any State regulatory authority or nonregulated . . . utility from making any determination that it is not appropriate to implement [or adopt] any such standard, pursuant to its authority under otherwise applicable State law." Similarly, 16 U.S.C. § 2627(b) and 15 U.S.C. § 3208 make it clear that any state regulatory authority or nonregulated utility may adopt regulations or rates that are "different from any standard established by this [subchapter or] chapter." B Section 210 12 Section 210 of PURPA's Title II, 92 Stat. 3144, 16 U.S.C. § 824a-3, seeks to encourage the development of cogeneration and small power production facilities.11 Congress believed that increased use of these sources of energy would reduce the demand for traditional fossil fuels. But it also felt that two problems impeded the development of nontraditional generating facilities: (1) traditional electricity utilities were reluctant to purchase power from, and to sell power to, the nontraditional facilities,12 and (2) the regulation of these alternative energy sources by state and federal utility authorities imposed financial burdens upon the nontraditional facilities and thus discouraged their development.13 13 In order to overcome the first of these perceived problems, § 210(a) directs FERC, in consultation with state regulatory authorities, to promulgate "such rules as it determines necessary to encourage cogeneration and small power production," including rules requiring utilities to offer to sell electricity to, and purchase electricity from, qualifying cogeneration and small power production facilities. Section 210(f), 16 U.S.C. § 824a-3(f), requires each state regulatory authority and nonregulated utility to implement FERC's rules. And § 210(h), 16 U.S.C. § 824a-3(h), authorizes FERC to enforce this requirement in federal court against any state authority or nonregulated utility; if FERC fails to act after request, any qualifying utility may bring suit. 14 To solve the second problem perceived by Congress, § 210(e), 16 U.S.C. § 824a-3(e), directs FERC to prescribe rules exempting the favored cogeneration and small power facilities from certain state and federal laws governing electricity utilities. 15 Pursuant to this statutory authorization, FERC has adopted regulations relating to purchases and sales of electricity to and from cogeneration and small power facilities. See 18 CFR pt. 292 (1980); 45 Fed.Reg. 12214-12237 (1980). These afford state regulatory authorities and nonregulated utilities latitude in determining the manner in which the regulations are to be implemented. Thus, a state commission may comply with the statutory requirements by issuing regulations, by resolving disputes on a case-by-case basis, or by taking any other action reasonably designed to give effect to FERC's rules.14 II 16 In April 1979, the State of Mississippi and the Mississippi Public Service Commission, appellees here, filed this action in the United States District Court for the Southern District of Mississippi against FERC and the Secretary of Energy, seeking a declaratory judgment that PURPA's Titles I and III and § 210 are unconstitutional. App. 3.15 Appellees maintained that PURPA was beyond the scope of congressional power under the Commerce Clause and that it constituted an invasion of state sovereignty in violation of the Tenth Amendment.16 17 Following cross-motions for summary judgment, the District Court, in an unreported opinion, held that in enacting PURPA Congress had exceeded its powers under the Commerce Clause. App. to Juris. Statement 1a. The court observed that the Mississippi Public Service Commission by state statute possessed the "power and authority to regulate and control intrastate activities and policies of all utilities operating within the sovereign state of Mississippi." Id., at 2a. Relying on Carter v. Carter Coal Co., 298 U.S. 238, 56 S.Ct. 855, 80 L.Ed. 1160 (1936), the court stated: "There is literally nothing in the Commerce Clause of the Constitution which authorizes or justifies the federal government in taking over the regulation and control of public utilities. These public utilities were actually unknown at the writing of the Constitution." App. to Juris. Statement 4a. Indeed, in the court's view, the legislation "does not even attempt to regulate commerce among the several states but it is a clear usurpation of power and authority which the United States simply does not have under the Commerce Clause of the Constitution." Id., at 7a. 18 Relying on National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), the court also concluded that PURPA trenches on state sovereignty.17 It therefore pronounced the statutory provisions void because "they constitute a direct intrusion on integral and traditional functions of the State of Mississippi." App. to Juris. Statement 8a-9a. For reasons it did not explain, the court also relied on the guarantee of a republican form of government, U.S.Const., Art. IV, § 4, and on the Supremacy Clause, Art. VI, cl. 2. App. to Juris. Statement 2a, n. 1, and 9a. 19 FERC and the Secretary of Energy appealed directly to this Court pursuant to 28 U.S.C. § 1252. See Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S. 264, 274, n. 15, 101 S.Ct. 2352, 2359, n. 15, 69 L.Ed.2d 1 (1981). We noted probable jurisdiction. 452 U.S. 936, 101 S.Ct. 3077, 69 L.Ed.2d 950 (1981). III The Commerce Clause 20 We readily conclude that the District Court's analysis and the appellees' arguments are without merit so far as they concern the Commerce Clause. To say that nothing in the Commerce Clause justifies federal regulation of even the intrastate operations of public utilities misapprehends the proper role of the courts in assessing the validity of federal legislation promulgated under one of Congress' plenary powers. The applicable standard was reiterated just last Term in Hodel v. Indiana, 452 U.S. 314, 101 S.Ct. 2376, 69 L.Ed.2d 40 (1981): "It is established beyond peradventure that 'legislative Acts adjusting the burdens and benefits of economic life come to the Court with a presumption of constitutionality . . . .' Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15 [96 S.Ct. 2882, 2892, 49 L.Ed.2d 752] (1976). . . . A court may invalidate legislation enacted under the Commerce Clause only if it is clear that there is no rational basis for a congressional finding that the regulated activity affects interstate commerce, or that there is no reasonable connection between the regulatory means selected and the asserted ends." Id., at 323-324, 101 S.Ct., at 2382.18 21 Despite these expansive observations by this Court, appellees assert that PURPA is facially unconstitutional because it does not regulate "commerce"; instead, it is said, the Act directs the nonconsenting State to regulate in accordance with federal procedures. This, appellees continue, is beyond Congress' power: "In exercising the authority conferred by this clause of the Constitution, Congress is powerless to regulate anything which is not commerce, as it is powerless to do anything about commerce which is not regulation." Carter v. Carter Coal Co., 298 U.S., at 297, 56 S.Ct., at 866. The "governance of commerce" by the State is to be distinguished from commerce itself, for regulation of the former is said to be outside the plenary power of Congress.19 22 It is further argued that the proper test is not whether the regulated activity merely "affects" interstate commerce but, instead, whether it has "a substantial effect" on such commerce, citing Justice REHNQUIST's opinion concurring in the judgment in the Hodel cases, 452 U.S., at 311-312, 101 S.Ct., at 2391. PURPA, appellees maintain, does not meet this standard. 23 The difficulty with these arguments is that they disregard entirely the specific congressional finding, in § 2 of the Act, 16 U.S.C. § 2601, that the regulated activities have an immediate effect on interstate commerce. Congress there determined that "the protection of the public health, safety, and welfare, the preservation of national security, and the proper exercise of congressional authority under the Constitution to regulate interstate commerce require," among other things, a program for increased conservation of electric energy, increased efficiency in the use of facilities and resources by electricity utilities, and equitable retail rates for electricity consumers, as well as a program to improve the wholesale distribution of electric energy, and a program for the conservation of natural gas while ensuring that rates to gas consumers are equitable. 16 U.S.C. § 2601. The findings, thus, are clear and specific. 24 The Court heretofore has indicated that federal regulation of intrastate power transmission may be proper because of the interstate nature of the generation and supply of electric power. FPC v. Florida Power & Light Co., 404 U.S. 453, 92 S.Ct. 637, 30 L.Ed.2d 600 (1972). Our inquiry, then, is whether the congressional findings have a rational basis. Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 277, 101 S.Ct., at 2360; Hodel v. Indiana, 452 U.S., at 323-324, 101 S.Ct., at 2382. 25 The legislative history provides a simple answer: there is ample support for Congress' conclusions. The hearings were extensive. Committees in both Houses of Congress noted the magnitude of the Nation's energy problems and the need to alleviate those problems by promoting energy conservation and more efficient use of energy resources. See S.Rep.No.95-442, at 7-10; H.R.Rep.No.95-543, vol. I, pp. 5-10 (1977); H.R.Rep.No.95-496, pt. 4, pp. 3-7, 125-130 (1977).20 Congress was aware that domestic oil production had lagged behind demand and that the Nation had become increasingly dependent on foreign oil. Id., at 3. The House Committee observed: "Reliance upon imported oil to meet the bulk of U. S. oil demands could seriously jeopardize the stability of the Nation's economy and could undermine the independence of the United States." Ibid. See H.R.Rep.No.95-543, vol. I, at 5-6. Indeed, the Nation had recently experienced severe shortages in its supplies of natural gas. Id., at 7. The House and Senate Committees both noted that the electricity industry consumed more than 25% of the total energy resources used in this country while supplying only 12% of the user demand for energy. S.Rep.No.95-442, at 7-8; H.R.Rep.No.95-496, pt. 4, at 125. In recent years, the electricity utility industry had been beset by numerous problems, id., at 129, which resulted in higher bills for the consuming public, a result exacerbated by the rate structures employed by most utilities. S.Rep.No.95-442, at 26. Congress naturally concluded that the energy problem was nationwide in scope,21 and that these developments demonstrated the need to establish federal standards regarding retail sales of electricity, as well as federal attempts to encourage conservation and more efficient use of scarce energy resources. Seeid., at 24-32; H.R.Rep.No.95-496, pt. 4, at 131-133, 136-138, 170-171. 26 Congress also determined that the development of cogeneration and small power production facilities would conserve energy. The evidence before Congress showed the potential contribution of these sources of energy: it was estimated that if proper incentives were provided, industrial cogeneration alone could account for 7%-10% of the Nation's electrical generating capacity by 1987. S.Rep.No.95-442, at 21, 23. 27 We agree with appellants that it is difficult to conceive of a more basic element of interstate commerce than electric energy, a product used in virtually every home and every commercial or manufacturing facility. No State relies solely on its own resources in this respect. See FPC v. Florida Power & Light Co., supra. Indeed, the utilities involved in this very case, Mississippi Power & Light Company and Mississippi Power Company, sell their retail customers power that is generated in part beyond Mississippi's borders, and offer reciprocal services to utilities in other States. App. 93-94. The intrastate activities of these utilities, although regulated by the Mississippi Public Service Commission, bring them within the reach of Congress' power over interstate commerce. See FPC v. Florida Power & Light Co., 404 U.S., at 458, 92 S.Ct., at 641; New England Power Co. v. New Hampshire, 454 U.S. 331, 102 S.Ct. 1096, 71 L.Ed.2d 188 (1982).22 28 Even if appellees were correct in suggesting that PURPA will not significantly improve the Nation's energy situation, the congressional findings compel the conclusion that " 'the means chosen by [Congress are] reasonably adapted to the end permitted by the Constitution.' " Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 276, 101 S.Ct., at 2359, quoting Heart of Atlanta Motel, Inc. v. United States, 379 U.S. 241, 262, 85 S.Ct. 348, 360, 13 L.Ed.2d 258 (1964). It is not for us to say whether the means chosen by Congress represent the wisest choice. It is sufficient that Congress was not irrational in concluding that limited federal regulation of retail sales of electricity and natural gas, and of relationships between cogenerators and electric utilities, was essential to protect interstate commerce. That is enough to place the challenged portions of PURPA within Congress' power under the Commerce Clause.23 Because PURPA's provisions concern private nonregulated utilities as well as state commissions, the statute necessarily is valid at least insofar as it regulates private parties. See Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 286, 101 S.Ct., at 2365. IV The Tenth Amendment 29 Unlike the Commerce Clause question, the Tenth Amendment issue presented here is somewhat novel. This case obviously is related to National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), insofar as both concern principles of state sovereignty. But there is a significant difference as well. National League of Cities, like Fry v. United States, 421 U.S. 542, 95 S.Ct. 1792, 44 L.Ed.2d 363 (1975), presented a problem the Court often confrontss the extent to which state sovereignty shields the States from generally applicable federal regulations. In PURPA, in contrast, the Federal Government attempts to use state regulatory machinery to advance federal goals. To an extent, this presents an issue of first impression. 30 PURPA, for all its complexity, contains essentially three requirements: (1) § 210 has the States enforce standards promulgated by FERC; (2) Titles I and III direct the States to consider specified ratemaking standards; and (3) those Titles impose certain procedures on state commissions. We consider these three requirements in turn: 31 A. Section 210. On its face, this appears to be the most intrusive of PURPA's provisions. The question of its constitutionality, however, is the easiest to resolve. Insofar as § 210 authorizes FERC to exempt qualified power facilities from "State laws and regulations," it does nothing more than pre-empt conflicting state enactments in the traditional way. Clearly, Congress can pre-empt the States completely in the regulation of retail sales by electricity and gas utilities and in the regulation of transactions between such utilities and cogenerators. Cf. Southern Pacific Co. v. Arizona, 325 U.S. 761, 769, 65 S.Ct. 1515, 1520, 89 L.Ed. 1915 (1945). The propriety of this type of regulation—so long as it is a valid exercise of the commerce power—was made clear in National League of Cities, and was reaffirmed in Hodel v. Virginia Surface Mining & Recl. Assn.: the Federal Government may displace state regulation even though this serves to "curtail or prohibit the States' prerogatives to make legislative choices respecting subjects the States may consider important." 452 U.S., at 290, 101 S.Ct., at 2367. 32 Section 210's requirement that "each State regulatory authority shall, after notice and opportunity for public hearing, implement such rule (or revised rule) for each electric utility for which it has ratemaking authority," 16 U.S.C. § 824a-3(f)(1) (emphasis added), is more troublesome. The statute's substantive provisions require electricity utilities to purchase electricity from, and to sell it to, qualifying cogenerator and small power production facilities. § 824a-3(a). Yet FERC has declared that state commissions may implement this by, among other things, "an undertaking to resolve disputes between qualifying facilities and electric utilities arising under [PURPA]." 18 CFR § 292.401(a) (1980). In essence, then, the statute and the implementing regulations simply require the Mississippi authorities to adjudicate disputes arising under the statute. Dispute resolution of this kind is the very type of activity customarily engaged in by the Mississippi Public Service Commission. See, e.g., Miss.Code Ann. §§ 77-1-31, 77-3-5, 77-3-13(3), 77-3-21, 77-3-405 (1973). 33 Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947), is instructive and controlling on this point. There, the Emergency Price Control Act, 56 Stat. 34, as amended, created a treble-damages remedy, and gave jurisdiction over claims under the Act to state as well as federal courts. The courts of Rhode Island refused to entertain such claims, although they heard analogous state causes of action. This Court upheld the federal program. It observed that state courts have a unique role in enforcing the body of federal law, and that the Rhode Island courts had "jurisdiction adequate and appropriate under established local law to adjudicate this action." 330 U.S., at 394, 67 S.Ct., at 814. Thus the state courts were directed to heed the constitutional command that "the policy of the federal Act is the prevailing policy in every state," id., at 393, 67 S.Ct., at 814, " 'and should be respected accordingly in the courts of the State.' " Id., at 392, 67 S.Ct., at 813, quoting Mondou v. New York, N. H. & H. R. Co., 223 U.S. 1, 57, 32 S.Ct. 169, 178, 56 L.Ed. 327 (1912). 34 So it is here. The Mississippi Commission has jurisdiction to entertain claims analogous to those granted by PURPA, and it can satisfy § 210's requirements simply by opening its doors to claimants. That the Commission has administrative as well as judicial duties is of no significance.24 Any other conclusion would allow the States to disregard both the preeminent position held by federal law throughout the Nation, cf. Martin v. Hunter's Lessee, 1 Wheat. 304, 340-341, 4 L.Ed. 97 (1816), and the congressional determination that the federal rights granted by PURPA can appropriately be enforced through state adjudicatory machinery. Such an approach, Testa emphasized, "flies in the face of the fact that the States of the Union constitute a nation," and "disregards the purpose and effect of Article VI of the Constitution." 330 U.S., at 389, 67 S.Ct., at 812. 35 B. Mandatory Consideration of Standards. We acknowledge that "the authority to make . . . fundamental . . . decisions" is perhaps the quintessential attribute of sovereignty. See National League of Cities v. Usery, 426 U.S., at 851, 96 S.Ct., at 2474. Indeed, having the power to make decisions and to set policy is what gives the State its sovereign nature. See Bates v. State Bar of Arizona, 433 U.S. 350, 360, 97 S.Ct. 2691, 2697, 53 L.Ed.2d 810 (1977) (State Supreme Court speaks as sovereign because it is the "ultimate body wielding the State's power over the practice of law"). It would follow that the ability of a state legislative (or, as here, administrative) body—which makes decisions and sets policy for the State as a whole—to consider and promulgate regulations of its choosing must be central to a State's role in the federal system. Indeed, the 19th-century view, expressed in a well-known slavery case, was that Congress "has no power to impose on a State officer, as such, any duty whatever, and compel him to perform it." Kentucky v. Dennison, 24 How. 66, 107, 16 L.Ed. 717 (1861). 36 Recent cases, however, demonstrate that this rigid and isolated statement from Kentucky v. Dennison —which suggests that the States and the Federal Government in all circumstances must be viewed as co-equal sovereigns—is not representative of the law today.25 While this Court never has sanctioned explicitly a federal command to the States to promulgate and enforce laws and regulations, cf. EPA v. Brown, 431 U.S. 99, 97 S.Ct. 1635, 52 L.Ed.2d 166 (1977), there are instances where the Court has upheld federal statutory structures that in effect directed state decisionmakers to take or to refrain from taking certain actions. In Fry v. United States, 421 U.S. 542, 95 S.Ct. 1792, 44 L.Ed.2d 363 (1975), for example, state executives were held restricted, with respect to state employees, to the wage and salary limitations established by the Economic Stabilization Act of 1970. Washington v. Washington State Commercial Passenger Fishing Vessel Assn., 443 U.S. 658, 99 S.Ct. 3055, 61 L.Ed.2d 823 (1979), acknowledged a federal court's power to enforce a treaty by compelling a state agency to "prepare" certain rules "even if state law withholds from [it] the power to do so." Id., at 695, 99 S.Ct., at 3079.26 And certainly Testa v. Katt, supra, by declaring that "the policy of the federal Act is the prevailing policy in every state," 330 U.S., at 393, 67 S.Ct., at 814, reveals that the Federal Government has some power to enlist a branch of state government—there the judiciary—to further federal ends.27 In doing so, Testa clearly cut back on both the quoted language and the analysis of the Dennison case of the preceding century.28 37 Whatever all this may forebode for the future, or for the scope of federal authority in the event of a crisis of national proportions, it plainly is not necessary for the Court in this case to make a definitive choice between competing views of federal power to compel state regulatory activity. Titles I and III of PURPA require only consideration of federal standards. And if a State has no utilities commission, or simply stops regulating in the field, it need not even entertain the federal proposals. As we have noted, the commerce power permits Congress to pre-empt the States entirely in the regulation of private utilities. In a sense, then, this case is only one step beyond Hodel v. Virginia Surface Mining & Recl. Assn., supra. There, the Federal Government could have pre-empted all surface mining regulations; instead, it allowed the States to enter the field if they promulgated regulations consistent with federal standards. In the Court's view, this raised no Tenth Amendment problem: "We fail to see why the Surface Mining Act should become constitutionally suspect simply because Congress chose to allow the States a regulatory role." 452 U.S., at 290, 101 S.Ct., at 2367. "[T]here can be no suggestion that the Act commandeers the legislative processes of the States by directly compelling them to enact and enforce a regulatory program." Id., at 288, 101 S.Ct., at 2366. 38 Similarly here, Congress could have pre-empted the field, at least insofar as private rather than state activity is concerned; PURPA should not be invalid simply because, out of deference to state authority, Congress adopted a less intrusive scheme and allowed the States to continue regulating in the area on the condition that they consider the suggested federal standards.29 While the condition here is affirmative in nature—that is, it directs the States to entertain proposals—nothing in this Court's cases suggests that the nature of the condition makes it a constitutionally improper one. There is nothing in PURPA "directly compelling" the States to enact a legislative program. In short, because the two challenged Titles simply condition continued state involvement in a pre-emptible area on the consideration of federal proposals, they do not threaten the States' "separate and independent existence," Lane County v. Oregon, 7 Wall. 71, 76, 19 L.Ed. 101 (1869); Coyle v. Oklahoma, 221 U.S. 559, 580, 31 S.Ct. 688, 695, 55 L.Ed. 853 (1911), and do not impair the ability of the States "to function effectively in a federal system." Fry v. United States, 421 U.S., at 547, n.7, 95 S.Ct., at 1795, n.7; National League of Cities v. Usery, 426 U.S., at 852, 96 S.Ct., at 2474. To the contrary, they offer the States a vehicle for remaining active in an area of overriding concern. 39 We recognize, of course, that the choice put to the States that of either abandoning regulation of the field altogether or considering the federal standards—may be a difficult one. And that is particularly true when Congress, as is the case here, has failed to provide an alternative regulatory mechanism to police the area in the event of state default. Yet in other contexts the Court has recognized that valid federal enactments may have an effect on state policy—and may, indeed, be designed to induce state action in areas that otherwise would be beyond Congress' regulatory authority. Thus in Oklahoma v. CSC, 330 U.S. 127, 67 S.Ct. 544, 91 L.Ed. 794 (1947), the Court upheld Congress' power to attach conditions to grants-in-aid received by the States, although the condition under attack involved an activity that "the United States is not concerned with, and has no power to regulate." Id., at 143, 67 S.Ct., at 553. The Tenth Amendment, the Court declared, "has been consistently construed 'as not depriving the national government of authority to resort to all means for the exercise of a granted power which are appropriate and plainly adapted to the permitted end,' " ibid., quoting United States v. Darby, 312 U.S. 100, 124, 61 S.Ct. 451, 462, 85 L.Ed. 609 (1941)—the end there being the disbursement of federal funds. Thus it cannot be constitutionally determinative that the federal regulation is likely to move the States to act in a given way, or even to "coerc[e] the States" into assuming a regulatory role by affecting their "freedom to make decisions in areas of 'integral governmental functions.' " Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 289, 101 S.Ct., at 2366. 40 Equally as important, it has always been the law that state legislative and judicial decisionmakers must give preclusive effect to federal enactments concerning nongovernmental activity, no matter what the strength of the competing local interests. See Martin v. Hunter's Lessee, 1 Wheat., at 340-341. This requirement follows from the nature of governmental regulation of private activity. "[I]ndividual businesses necessarily [are] subject to the dual sovereignty of the government of the Nation and of the State in which they reside," National League of Cities v. Usery, 426 U.S., at 845, 96 S.Ct., at 2471; when regulations promulgated by the sovereigns conflict, federal law necessarily controls. This is true though Congress exercises its authority "in a manner that displaces the States' exercise of their police powers," Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 291, 101 S.Ct., at 2367, or in such a way as to "curtail or prohibit the States' prerogatives to make legislative choices respecting subjects the States may consider important," id., at 290, 101 S.Ct., at 2367 or, to put it still more plainly, in a manner that is "extraordinarily intrusive." Id., at 305, 101 S.Ct., at 2374 (POWELL, J., concurring). Thus it may be unlikely that the States will or easily can abandon regulation of public utilities to avoid PURPA's requirements. But this does not change the constitutional analysis: as in Hodel v. Virginia Surface Mining & Recl. Assn., "[t]he most that can be said is that the . . . Act establishes a program of cooperative federalism that allows the States, within limits established by federal minimum standards, to enact and administer their own regulatory programs, structured to meet their own particular needs." Id., at 289, 101 S.Ct., at 2366.30 41 To be sure, PURPA gives virtually any affected person the right to compel consideration of the statutory standards through judicial action. We fail to see, however, that this places any particularly onerous burden on the State. Mississippi by statute already grants "[a]ny interested person . . . the right to petition the [Public Service] [C]ommission for issuance, amendment or repeal of a rule or regulation," Miss.Code Ann. § 77-3-45 (1973) (emphasis added), and provides that "any party aggrieved by any final finding, order or judgment of the commission shall have the right, regardless of the amount involved, of appeal in chancery court." Miss.Code Ann. § 77-3-67(1) (Supp.1981) (emphasis added). Indeed, "[a]ny person whose rights may be directly affected by said appeal may appear and become a party. . . ." Ibid. And "[a]ppeals in accordance with law may be had to the supreme court of the State of Mississippi from any final judgment of the chancery court." Miss.Code Ann. § 77-3-71 (1973). 42 It is hardly clear on the statute's face, then, that PURPA's standing and appeal provisions grant any rights beyond those presently accorded by Mississippi law, and appellees point to no specific provision of the Act expanding on the State's existing, liberal approach to public participation in ratemaking.31 In this light, we again find the principle of Testa v. Katt, supra, controlling: the State is asked only to make its administrative tribunals available for the vindication of federal as well as state-created rights. PURPA, of course, establishes as federal policy the requirement that state commissions consider various ratemaking standards, and it gives individuals a right to enforce that policy; once it is established that the requirement is constitutionally supportable, "the obligation of states to enforce these federal laws is not lessened by reason of the form in which they are cast or the remedy which they provide." Testa v. Katt, 330 U.S., at 391, 67 S.Ct., at 813. See Second Employers' Liability Cases, 223 U.S. 1, 57, 32 S.Ct. 169, 178, 56 L.Ed. 327 (1912). 43 In short, Titles I and III do not involve the compelled exercise of Mississippi's sovereign powers. And, equally important, they do not set a mandatory agenda to be considered in all events by state legislative or administrative decisionmakers. As we read them, Titles I and III simply establish requirements for continued state activity in an otherwise pre-emptible field.32 Whatever the constitutional problems associated with more intrusive federal programs, the "mandatory consideration" provisions of Titles I and III must be validated under the principle of Hodel v. Virginia Surface Mining & Recl. Assn.33 44 C. The Procedural Requirements. Titles I and III also require state commissions to follow certain notice and comment procedures when acting on the proposed federal standards. In a way, these appear more intrusive than the "consideration" provisions; while the latter are essentially hortatory, the procedural provisions obviously are prescriptive. Appellants and amici Maryland et al. argue that the procedural requirements simply establish minimum due process standards, something Mississippi appears already to provide,34 and therefore may be upheld as an exercise of Congress' Fourteenth Amendment powers. We need not go that far, however, for we uphold the procedural requirements under the same analysis employed above in connection with the "consideration" provisions. If Congress can require a state administrative body to consider proposed regulations as a condition to its continued involvement in a pre-emptible field—and we hold today that it can there is nothing unconstitutional about Congress' requiring certain procedural minima as that body goes about undertaking its tasks. The procedural requirements obviously do not compel the exercise of the State's sovereign powers, and do not purport to set standards to be followed in all areas of the state commission's endeavors. 45 The judgment of the District Court is reversed. 46 It is so ordered. 47 Justice POWELL, concurring in part and dissenting in part. 48 The Public Utility Regulatory Policies Act of 1978, Pub.L. 95-617, 92 Stat. 3117 et seq. (PURPA), imposes unprecedented burdens on the States. As Justice O'CONNOR ably demonstrates, it intrusively requires them to make a place on their administrative agenda for consideration and potential adoption of federally proposed "standards." The statute does not simply ask States to consider quasi-legislative matters that Congress believes they would do well to adopt. It also prescribes administrative and judicial procedures that States must follow in deciding whether to adopt the proposed standards. At least to this extent, I think the PURPA violates the Tenth Amendment. 49 * Most, if not all, of the States have administrative bodies usually commissions—that regulate electric and gas public utility companies. As these utilities normally are given monopoly jurisdiction, they are extensively regulated both substantively and procedurally by state law. Until now, with limited exceptions, the Federal Government has not attempted to pre-empt this important state function, and certainly has not undertaken to prescribe the procedures by which state regulatory bodies make their decisions. The PURPA, for the first time, breaks with this longstanding deference to principles of federalism. 50 Now, regardless of established procedures before state administrative regulatory agencies and of state law with respect to judicial review, the PURPA forces federal procedures on state regulatory institutions. The PURPA prescribes rules directing that "the Secretary [of Energy], any affected electric utility, or any electric consumer of an affected electric utility may intervene and participate as a matter of right" in regulatory proceedings required by the PURPA respecting electrical rates.1 It directs that "[a]ny person (including the Secretary) may bring an action to enforce" the obligations with respect to electrical rate consideration that the PURPA lays upon state regulatory commissions.2 The statute provides that "[a]ny person (including the Secretary) may obtain [judicial] review of any determination" made by a state regulatory commission regarding the PURPA's electrical rate policies.3 The foregoing requirements by the PURPA intrude upon—in effect pre-empt—core areas of a State's administrative and judicial procedure. II 51 In sustaining these provisions of the Act, the Court reasons that Congress can condition the utility regulatory activities of States on any terms it pleases since, under the Commerce Clause, Congress has the power to pre-empt completely all such activities. Ante, at 765-766. Under this "threat of preemption" reasoning, Congress—one supposes—could reduce the States to federal provinces. But as National League of Cities v. Usery, 426 U.S. 833, 841, 96 S.Ct. 2465, 2469, 49 L.Ed.2d 245 (1976), stated, and indeed as the structure of the Court's opinion today makes plain, ante, at 753 and 758, the Commerce Clause and the Tenth Amendment embody distinct limitations on federal power. That Congress has satisfied the one demonstrates nothing as to whether Congress has satisfied the other.4 52 "The general rule, bottomed deeply in belief in the importance of state control of state judicial procedure, is that federal law takes the state courts as it finds them." Hart, The Relations Between State and Federal Law, 54 Colum.L.Rev. 489, 508 (1954). I believe the same principle must apply to other organs of state government. It may be true that the procedural provisions of the PURPA that prompt this dissent may not effect dramatic changes in the laws and procedures of some States. But I know of no other attempt by the Federal Government to supplant state-prescribed procedures that in part define the nature of their administrative agencies. If Congress may do this, presumably it has the power to pre-empt state-court rules of civil procedure and judicial review in classes of cases found to affect commerce. This would be the type of gradual encroachment hypothesized by Professor Tribe: "Of course, no one expects Congress to obliterate the states, at least in one fell swoop. If there is any danger, it lies in the tyranny of small decisions—in the prospect that Congress will nibble away at state sovereignty, bit by bit, until someday essentially nothing is left but a gutted shell."5 53 I limit this dissent to the provisions of the PURPA identified above. Despite the appeal—and indeed wisdom—of Justice O'CONNOR's evocation of the principles of federalism, I believe precedents of this Court support the constitutionality of the substantive provisions of this Act on this facial attack. See Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 101 S.Ct. 2352, 69 L.Ed.2d 1 (1981); Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947). Accordingly, to the extent the procedural provisions may be separable, I would affirm in part and reverse in part. 54 Justice O'CONNOR, with whom THE CHIEF JUSTICE and Justice REHNQUIST join, concurring in the judgment in part and dissenting in part. 55 I agree with the Court that the Commerce Clause supported Congress' enactment of the Public Utility Regulatory Policies Act of 1978, Pub.L. 95-617, 92 Stat. 3117 (PURPA). I disagree, however, with much of the Court's Tenth Amendment analysis. Titles I and III of PURPA conscript state utility commissions into the national bureaucratic army. This result is contrary to the principles of National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), antithetical to the values of federalism, and inconsistent with our constitutional history. Accordingly, I dissent from Parts IV-B and IV-C of the Court's opinion.1 56 * Titles I and III of PURPA require state regulatory agencies to decide whether to adopt a dozen federal standards governing gas and electric utilities.2 The statute describes, in some detail, the procedures state authorities must follow when evaluating these standards,3 but does not compel the States to adopt the suggested federal standards. 15 U.S.C. § 3203(a) (1976 ed., Supp. IV); 16 U.S.C. §§ 2621(a), 2623(a), 2627(b) (1976 ed., Supp. IV). The latter, deceptively generous feature of PURPA persuades the Court that the statute does not intrude impermissibly into state sovereign functions. The Court's conclusion, however, rests upon a fundamental misunderstanding of the role that state governments play in our federalist system. 57 State legislative and administrative bodies are not field offices of the national bureaucracy. Nor are they think tanks to which Congress may assign problems for extended study. Instead, each State is sovereign within its own domain, governing its citizens and providing for their general welfare. While the Constitution and federal statutes define the boundaries of that domain, they do not harness state power for national purposes. The Constitution contemplates "an indestructible Union, composed of indestructible States," a system in which both the State and National Governments retain a "separate and independent existence." Texas v. White, 7 Wall. 700, 725, 19 L.Ed. 227 (1869); Lane County v. Oregon, 7 Wall. 71, 76, 19 L.Ed. 101 (1869). 58 Adhering to these principles, the Court has recognized that the Tenth Amendment restrains congressional action that would impair "a state's ability to function as a state." Transportation Union v. Long Island R. Co., 455 U.S. 678, 686, 102 S.Ct. 1349, 1354, 71 L.Ed.2d 547 (1982); National League of Cities v. Usery, 426 U.S., at 842-852, 96 S.Ct., at 2470-2474; Fry v. United States, 421 U.S. 542, 547, n. 7, 95 S.Ct. 1792, 1795, n. 7, 44 L.Ed.2d 363 (1975). See also Lafayette v. Louisiana Power & Light Co., 435 U.S. 389, 423-424, 98 S.Ct. 1123, 1142, 55 L.Ed.2d 364 (1978) (BURGER, C.J., concurring in judgment). For example, in National League of Cities v. Usery, supra, the Court held that Congress could not prescribe the minimum wages and maximum hours of state employees engaged in "traditional governmental functions," id., 426 U.S., at 852, 96 S.Ct., at 2474, because the power to set those wages and hours is an "attribute of state sovereignty" that is " 'essential to [a] separate and independent existence.' " Id., at 845, 96 S.Ct., at 2471 (quoting Lane County v. Oregon, supra, 7 Wall., at 76). 59 Just last Term this Court identified three separate inquiries underlying the result in National League of Cities. A congressional enactment violates the Tenth Amendment, we observed, if it regulates the " 'States as States,' " addresses "matters that are indisputably 'attribute[s] of state sovereignty,' " and "directly impair[s] [the States'] ability to 'structure integral operations in areas of traditional governmental functions.' " Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 287-288, 101 S.Ct. 2352, 2365-2366, 69 L.Ed.2d 1 (1981) (quoting National League of Cities, supra, 426 U.S., at 854, 845, 852, 96 S.Ct., at 2475, 2471, 2474). See also Transportation Union, supra, at 684, 102 S.Ct., at 1353.4 60 Application of these principles to the present case reveals the Tenth Amendment defects in Titles I and III. Plainly those Titles regulate the "States as States." While the statute's ultimate aim may be the regulation of private utility companies, PURPA addresses its commands solely to the States. Instead of requesting private utility companies to adopt lifeline rates, declining block rates, or the other PURPA standards, Congress directed state agencies to appraise the appropriateness of those standards. It is difficult to argue that a statute structuring the regulatory agenda of a state agency is not a regulation of the "State." 61 I find it equally clear that Titles I and III address "attribute[s] of state sovereignty." Even the Court recognizes that "the power to make decisions and to set policy is what gives the State its sovereign nature." Ante, at 761. The power to make decisions and set policy, however, embraces more than the ultimate authority to enact laws; it also includes the power to decide which proposals are most worthy of consideration, the order in which they should be taken up, and the precise form in which they should be debated. PURPA intrudes upon all of these functions. It chooses 12 proposals, forcing their consideration even if the state agency deems other ideas more worthy of immediate attention. In addition, PURPA hinders the agency's ability to schedule consideration of the federal standards.5 Finally, PURPA specifies, with exacting detail, the content of the standards that will absorb the agency's time.6 62 If Congress routinely required the state legislatures to debate bills drafted by congressional committees, it could hardly be questioned that the practice would affect an attribute of state sovereignty. PURPA, which sets the agendas of agencies exercising delegated legislative power in a specific field, has a similarly intrusive effect. 63 Finally, PURPA directly impairs the States' ability to "structure integral operations in areas of traditional governmental functions." Utility regulation is a traditional function of state government,7 and the regulatory commission is the most integral part of that function. By taxing the limited resources of these commissions, and decreasing their ability to address local regulatory ills, PURPA directly impairs the power of state utility commissions to discharge their traditional functions efficiently and effectively.8 64 The Court sidesteps this analysis, suggesting that the States may escape PURPA simply by ceasing regulation of public utilities. Even the Court recognizes that this choice "may be a difficult one," ante, at 766, and that "it may be unlikely that the States will or easily can abandon regulation of public utilities to avoid PURPA's requirements." Ante, at 767. In fact, the Court's "choice" is an absurdity, for if its analysis is sound, the Constitution no longer limits federal regulation of state governments. Under the Court's analysis, for example, National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), would have been wrongly decided, because the States could have avoided the Fair Labor Standards Act by "choosing" to fire all employees subject to that Act and to close those branches of state government.9 Similarly, Congress could dictate the agendas and meeting places of state legislatures, because unwilling States would remain free to abolish their legislative bodies.10 I do not agree that this dismemberment of state government is the correct solution to a Tenth Amendment challenge. 65 The choice put to the States by the Surface Mining Control and Reclamation Act of 1977, 91 Stat. 447, 30 U.S.C. § 1201 et seq. (1976 ed., Supp.IV), the federal statute upheld in Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 101 S.Ct. 2352, 69 L.Ed.2d 1 (1981), and discussed by the Court, ante, at 764-765, 768, n. 30, is quite different from the decision PURPA mandates. The Surface Mining Act invites the States to submit proposed surface mining regulations to the Secretary of the Interior. 30 U.S.C. § 1253 (1976 ed., Supp.IV). If the Secretary approves a state regulatory program, then the State enforces that program. If a State chooses not to submit a program, the Secretary develops and implements a program for that State. § 1254. Even States in the latter category, however, may supplement the Secretary's program with consistent state laws.11 The Surface Mining Act does not force States to choose between performing tasks set by Congress and abandoning all mining or land use regulation. That statute is "a program of cooperative federalism," Hodel, supra, at 289, 101 S.Ct., at 2366, because it allows the States to choose either to work with Congress in pursuit of federal surface mining goals or to devote their legislative resources to other mining and land use problems. By contrast, there is nothing "cooperative" about a federal program that compels state agencies either to function as bureaucratic puppets of the Federal Government or to abandon regulation of an entire field traditionally reserved to state authority.12 Yet this is the "choice" the Court today forces upon the States. 66 The Court defends its novel decision to permit federal conscription of state legislative power by citing three cases upholding statutes that "in effect directed state decisionmakers to take or to refrain from taking certain actions." Ante, at 762. Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947), is the most suggestive of these decisions.13 In Testa, the Court held that state trial courts may not refuse to hear a federal claim if "th[e] same type of claim arising under [state] law would be enforced by that State's courts." Id., at 394, 67 S.Ct., at 814. A facile reading of Testa might suggest that state legislatures must also entertain congressionally sponsored business, as long as the federal duties are similar to existing state obligations. Application of Testa to legislative power, however, vastly expands the scope of that decision. Because trial courts of general jurisdiction do not choose the cases that they hear, the requirement that they evenhandedly adjudicate state and federal claims falling within their jurisdiction does not infringe any sovereign authority to set an agenda.14 As explained above, however, the power to choose subjects for legislation is a fundamental attribute of legislative power, and interference with this power unavoidably undermines state sovereignty. Accordingly, the existence of a congressional authority to "enlist . . . the [state] judiciary . . . to further federal ends," ante, at 762, does not imply an equivalent power to impress state legislative bodies into federal service. 67 The Court, finally, reasons that because Congress could have pre-empted the entire field of intrastate utility regulation, the Constitution should not forbid PURPA's "less intrusive scheme." Ante, at 765, and n.29.15 The Court's evaluation of intrusiveness, however, is simply irrelevant to the constitutional inquiry. The Constitution permits Congress to govern only through certain channels. If the Tenth Amendment principles articulated in National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), and Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 101 S.Ct. 2352, 69 L.Ed.2d 1 (1981), foreclose PURPA's approach, it is no answer to argue that Congress could have reached the same destination by a different route. This Court's task is to enforce constitutional limits on congressional power, not to decide whether alternative courses would better serve state and federal interests.16 68 I do not believe, moreover, that Titles I and III of PURPA are less intrusive than pre-emption.17 When Congress preempts a field, it precludes only state legislation that conflicts with the national approach. The States usually retain the power to complement congressional legislation, either by regulating details unsupervised by Congress or by imposing requirements that go beyond the national threshold.18 Most importantly, after Congress pre-empts a field, the States may simply devote their resources elsewhere. This country does not lack for problems demanding legislative attention. PURPA, however, drains the inventive energy of state governmental bodies by requiring them to weigh its detailed standards, enter written findings, and defend their determinations in state court. While engaged in these congressionally mandated tasks, state utility commissions are less able to pursue local proposals for conserving gas and electric power. The States might well prefer that Congress simply impose the standards described in PURPA; this, at least, would leave them free to exercise their power in other areas. 69 Federal pre-emption is less intrusive than PURPA's approach for a second reason. Local citizens hold their utility commissions accountable for the choices they make. Citizens, moreover, understand that legislative authority usually includes the power to decide which ideas to debate, as well as which policies to adopt. Congressional compulsion of state agencies, unlike pre-emption, blurs the lines of political accountability and leaves citizens feeling that their representatives are no longer responsive to local needs.19 70 The foregoing remarks suggest that, far from approving a minimally intrusive form of federal regulation, the Court's decision undermines the most valuable aspects of our federalism. Courts and commentators frequently have recognized that the 50 States serve as laboratories for the development of new social, economic, and political ideas.20 This state innovation is no judicial myth. When Wyoming became a State in 1890, it was the only State permitting women to vote.21 That novel idea did not bear national fruit for another 30 years.22 Wisconsin pioneered unemployment insurance,23 while Massachusetts initiated minimum wage laws for women and minors.24 After decades of academic debate, state experimentation finally provided an opportunity to observe no-fault automobile insurance in operation.25 Even in the field of environmental protection, an area subject to heavy federal regulation, the States have supplemented national standards with innovative and far-reaching statutes.26 Utility regulation itself is a field marked by valuable state invention.27 PURPA, which commands state agencies to spend their time evaluating federally proposed standards and defending their decisions to adopt or reject those standards, will retard this creative experimentation. 71 In addition to promoting experimentation, federalism enhances the opportunity of all citizens to participate in representative government. Alexis de Tocqueville understood well that participation in local government is a cornerstone of American democracy: 72 "It is incontestably true that the love and the habits of republican government in the United States were engendered in the townships and in the provincial assemblies. [I]t is this same republican spirit, it is these manners and customs of a free people, which are engendered and nurtured in the different States, to be afterwards applied to the country at large." 1 A. de Tocqueville, Democracy in America 181 (H. Reeve trans. 1961).28 73 Citizens, however, cannot learn the lessons of self-government if their local efforts are devoted to reviewing proposals formulated by a faraway national legislature. If we want to preserve the ability of citizens to learn democratic processes through participation in local government, citizens must retain the power to govern, not merely administer, their local problems. 74 Finally, our federal system provides a salutary check on governmental power. As Justice Harlan once explained, our ancestors "were suspicious of every form of all-powerful central authority." Harlan, supra n. 16, at 944. To curb this evil, they both allocated governmental power between state and national authorities, and divided the national power among three branches of government. Unless we zealously protect these distinctions, we risk upsetting the balance of power that buttresses our basic liberties. In analyzing this brake on governmental power, Justice Harlan noted that "[t]he diffusion of power between federal and state authority . . . takes on added significance as the size of the federal bureaucracy continues to grow." Ibid.29 Today, the Court disregards this warning and permits Congress to kidnap state utility commissions into the national regulatory family. Whatever the merits of our national energy legislation, I am not ready to surrender this state legislative power to the Federal Energy Regulatory Commission. II 75 As explained above, the Court's decision to uphold Titles I and III violates the principles of National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 49 L.Ed.2d 245 (1976), and threatens the values promoted by our federal system. The Court's result, moreover, is at odds with our constitutional history, which demonstrates that the Framers consciously rejected a system in which the National Legislature would employ state legislative power to achieve national ends. 76 The principal defect of the Articles of Confederation, 18th-century writers agreed, was that the new National Government lacked the power to compel individual action. Instead, the central government had to rely upon the cooperation of state legislatures to achieve national goals. Thus, Alexander Hamilton explained that "[t]he great and radical vice in the construction of the existing Confederation is in the principle of legislation for states or governments, in their corporate or collective capacities and as contradistinguished from the individuals of whom they consist." The Federalist No. 15, p. 93 (J. Cooke ed. 1961) (emphasis omitted). He pointed out, for example, that the National Government had "an indefinite discretion to make requisitions for men and money," but "no authority to raise either by regulations extending to the individual citizens of America." Ibid. 77 The Constitution cured this defect by permitting direct contact between the National Government and the individual citizen, a change repeatedly acknowledged by the delegates assembled in Philadelphia. George Mason, for example, declared: 78 "Under the existing Confederacy, Congress represent[s] the States not the people of the States: their acts operate on the States not on the individuals. The case will be changed in the new plan of Government." 1 The Records of the Federal Convention of 1787, p. 133 (M. Farrand ed. 1911) (hereinafter Farrand) (abbreviations expanded in this and subsequent quotations). 79 Hamilton subsequently explained to the people of New York that the Constitution marked the "difference between a league and a government," because it "extend[ed] the authority of the union to the persons of the citizens,—the only proper objects of government." The Federalist No. 15, supra, at 95. Similarly, Charles Pinckney told the South Carolina House of Representatives that "the necessity of having a government which should at once operate upon the people, and not upon the states, was conceived to be indispensable by every delegation present; . . . however they may have differed with respect to the quantum of power, no objection was made to the system itself." 4 Elliot's Debates on the Federal Convention 256 (2d ed.1863). 80 The speeches and writings of the Framers suggest why they adopted this means of strengthening the National Government. Mason, for example, told the Convention that because "punishment could not [in the nature of things be executed on] the States collectively," he advocated a National Government that would "directly operate on individuals." 1 Farrand 34. Hamilton predicted that a National Government forced to work through the States would "degenerate into a military despotism" because it would have to maintain a "large army, continually on foot" to enforce its will against the States. The Federalist No. 16, p. 101 (J. Cooke ed. 1961). See also id., at 102; The Federalist No. 15, supra, at 95-96. 81 Thus, the Framers concluded that government by one sovereign through the agency of a second cannot be satisfactory. At one extreme, as under the Articles of Confederation, such a system is simply ineffective. At the other, it requires a degree of military force incompatible with stable government and civil liberty.30 For this reason, the Framers concluded that "the execution of the laws of the national government . . . should not require the intervention of the State Legislatures," The Federalist No. 16, supra, at 103, and abandoned the Articles of Confederation in favor of direct national legislation. 82 At the same time that the members of the Constitutional Convention fashioned this principle, they rejected two proposals that would have given the National Legislature power to supervise directly state governments. The first proposal would have authorized Congress "to call forth the force of the Union against any member of the Union failing to fulfill its duty under the articles thereof." 1 Farrand 21. The delegates never even voted on this suggestion. James Madison moved to postpone it, stating that "the more he reflected on the use of force, the more he doubted the practicability, the justice and the efficacy of it when applied to people collectively and not individually." Id., at 54. Several other delegates echoed his concerns,31 and Madison ultimately reported that "[t]he practicability of making laws, with coercive sanctions, for the States as political bodies [has] been exploded on all hands." 2 id. at 9. 83 The second proposal received more favorable consideration. Virginia's Governor Randolph suggested that Congress should have the power "to negative all laws passed by the several States, contravening in the opinion of the National Legislature the articles of Union." 1 id., at 21. On May 31, 1787, the Committee of the Whole approved this proposal without debate. Id., at 61. A week later, Pinckney moved to extend the congressional negative to all state laws "which [Congress] should judge to be improper." Id., at 164. Numerous delegates criticized this attempt to give Congress unbounded control over state lawmaking. Hugh Williamson, for example, thought "the State Legislatures ought to possess independent powers in cases purely local," id., at 171, while Elbridge Gerry thought Pinckney's idea might "enslave the States." Id., at 165. After much debate, the Convention rejected Pinckney's suggestion. 84 Late in July, the delegates reversed their approval of even Randolph's more moderate congressional veto. Several delegates now concluded that the negative would be "terrible to the States," "unnecessary," and "improper." 2 id., at 27.32 Omission of the negative, however, left the new system without an effective means of adjusting conflicting state and national laws. To remedy this defect, the delegates adopted the Supremacy Clause, providing that the Federal Constitution, laws, and treaties are "the supreme Law of the Land" and that "the Judges in every State shall be bound thereby." Art. VI, cl. 2. Thus, the Framers substituted judicial review of state laws for congressional control of state legislatures. 85 While this history demonstrates the Framers' commitment to a strong central government, the means that they adopted to achieve that end are as instructive as the end itself.33 Under the Articles of Confederation, the National Legislature operated through the States. The Framers could have fortified the central government, while still maintaining the same system, if they had increased Congress' power to demand obedience from state legislatures. In time, this scheme might have relegated the States to mere departments of the National Government, a status the Court appears to endorse today. The Framers, however, eschewed this course, choosing instead to allow Congress to pass laws directly affecting individuals, and rejecting proposals that would have given Congress military or legislative power over state governments. In this way, the Framers established independent state and national sovereigns. The National Government received the power to enact its own laws and to enforce those laws over conflicting state legislation. The States retained the power to govern as sovereigns in fields that Congress cannot or will not pre-empt.34 This product of the Constitutional Convention, I believe, is fundamentally inconsistent with a system in which either Congress or a state legislature harnesses the legislative powers of the other sovereign.35 III 86 During his last Term of service on this Court, Justice Black eloquently explained that our notions of federalism subordinate neither national nor state interests: 87 "The concept does not mean blind deference to 'States' Rights' any more than it means centralization of control over every important issue in our National Government and its courts. The Framers rejected both these courses. What the concept does represent is a system in which there is sensitivity to the legitimate interests of both State and National Governments, and in which the National Government, anxious though it may be to vindicate and protect federal rights and federal interests, always endeavors to do so in ways that will not unduly interfere with the legitimate activities of the States." Younger v. Harris, 401 U.S. 37, 44, 91 S.Ct. 746, 750, 27 L.Ed.2d 669 (1971). 88 In this case, I firmly believe that a proper "sensitivity to the legitimate interests of both State and National Governments" requires invalidation of Titles I and III of PURPA insofar as they apply to state regulatory authorities. Accordingly, I respectfully dissent from the Court's decision to uphold those portions of the statute. 1 The Senate vote was taken on October 9, 1978. The Mississippi Senators voted against the bill. See 124 Cong.Rec. 34780. The House vote was taken on October 14, 1978. The five-member Mississippi delegation voted three "ayes" and two "nays." See id., at 38503. 2 In addition to PURPA, the package included the Energy Tax Act of 1978, Pub.L. 95-618, 92 Stat. 3174; the National Energy Conservation Policy Act, Pub.L. 95-619, 92 Stat. 3206; the Powerplant and Industrial Fuel Use Act of 1978, Pub.L. 95-620, 92 Stat. 3289; and the Natural Gas Policy Act of 1978, Pub.L. 95-621, 92 Stat. 3351. 3 For simplicity of citation, and to avoid repetition, unless otherwise noted herein, any reference to 15 or 16 U.S.C. relates to Supplement IV of the 1976 edition of the Code. 4 "Declining block rates" are a traditional and still common approach used by utilities in their charges for electricity. The highest unit rate is charged for basic electrical consumption, with a declining per-unit price for each block of additional consumption. See S.Rep.No. 95-442, pp. 26-27 (1977). 5 "Time-of-day rates" are designed to reduce "peak load," the term used to describe the greatest demand for a utility's electricity. Demand varies by hour and season, usually reaching a daily maximum in the afternoon and a seasonal maximum in midsummer or midwinter. A utility must have enough generating capacity to meet that demand; steps that reduce peak demand also reduce the required amount of generating capacity and the use of "peaking" generating equipment, which frequently is gas- or oil-fueled. Under time-of-day rates, utilities charge more for electricity consumed during peak load hours. See id., at 29. 6 "Seasonal rates" operate to reduce peak load by imposing higher rates during the seasons when demand is greatest. 7 "Interruptible rates" tend to reduce peak load by charging less for service which the utility can interrupt, or stop, during peak demand periods. 8 "Load management techniques" are methods used to reduce the demand for electricity at peak times. For example, a utility might employ remote-control devices that temporarily turn off appliances during periods when the demand is particularly great. 9 "Master-metering" is the use of one meter for several living units. Studies have shown that tenants of master-metered buildings use 35% more electricity, on the average, than tenants of buildings where each apartment has its own meter. See id., at 31. 10 An "automatic adjustment clause" provides that as a utility's fuel costs rise it may increase its rates without public hearing or review by the state regulatory authority. A clause of this kind provides the utility with no incentive to reduce its costs or to shift away from oil- or gas-fueled generating facilities, and therefore tends to discourage the efficient use of energy resources. 11 A "cogeneration facility" is one that produces both electric energy and steam or some other form of useful energy, such as heat. 16 U.S.C. § 796(18)(A). A "small power production facility" is one that has a production capacity of no more than 80 megawatts and uses biomass, waste, or renewable resources (such as wind, water, or solar energy) to produce electric power. § 796(17)(A). 12 See 123 Cong.Rec. 25848 (1977) (remarks of Sen. Percy); id., at 32403 (remarks of Sen. Durkin); id., at 32437 (remarks of Sen. Haskell); id., at 32419 (remarks of Sen. Hart); National Energy Act: Hearings on H.R. 6831 et al. before the Subcommittee on Energy and Power of the House Committee on Interstate and Foreign Commerce, 95th Cong., 1st Sess., 552-553 (1977). 13 See H.R.Conf.Rep.No. 95-1750, p. 98 (1978); H.R.Rep.No. 95-496, pt. 4, p. 157 (1977); 123 Cong.Rec. 32399 (1977) (remarks of Sen. Cranston); id., at 32660 (remarks of Sen. Percy). 14 Congress recognized that a State's compliance with the requirements of PURPA would involve the expenditure of funds. Accordingly, it authorized the Secretary of Energy to make grants to state regulatory authorities to assist them in carrying out the provisions of Titles I and III, including the reporting requirements, and the provisions of § 210. See 42 U.S.C. § 6807 (1976 ed., Supp. IV). For each of the fiscal years 1979 and 1980, Congress authorized for appropriation up to $40 million to help state regulatory authorities defray the costs of complying with PURPA. Pub.L. 95-617, § 142(1), 92 Stat. 3134, 42 U.S.C. § 6808(1) (1976 ed., Supp. IV). 15 Mississippi Power & Light Company was permitted to intervene in the action as a plaintiff and is also an appellee here. 16 "The powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people." U.S.Const., Amdt. 10. 17 "The sovereign state of Mississippi is not a robot, or lackey which may be shuttled back and forth to suit the whim and caprice of the federal government." App. to Juris. Statement 2a. 18 In the companion case decided the same day, this Court observed: "Judicial review in this area is influenced above all by the fact that the Commerce Clause is a grant of plenary authority to Congress . . .. This power is 'complete in itself, may be exercised to its utmost extent, and acknowledges no limitations, other than are prescribed in the constitution.' Gibbons v. Ogden, 9 Wheat. 1, 196 [6 L.Ed. 23] (1824). Moreover, this Court has made clear that the commerce power extends not only to 'the use of channels of interstate or foreign commerce' and to 'protection of the instrumentalities of interstate commerce . . . or persons or things in commerce,' but also to 'activities affecting commerce.' Perez v. United States, 402 U.S. 146, 150 [91 S.Ct. 1357, 1359, 28 L.Ed.2d 686] (1971). As we explained in Fry v. United States, 421 U.S. 542, 547 [95 S.Ct. 1792, 1795, 44 L.Ed.2d 363] (1975), '[e]ven activity that is purely intrastate in character may be regulated by Congress, where the activity, combined with like conduct by others similarly situated, affects commerce among the States or with foreign nations.' " Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S. 264, 276-277, 101 S.Ct. 2352, 2359-2360, 69 L.Ed.2d 1 (1981). 19 For this proposition, appellees rely on Brown v. EPA, 521 F.2d 827, 839 (CA9 1975), vacated and remanded, 431 U.S. 99, 97 S.Ct. 1635, 52 L.Ed.2d 166 (1977), and District of Columbia v. Train, 172 U.S.App.D.C. 311, 332, 521 F.2d 971, 992 (1975), vacated and remanded sub nom. EPA v. Brown, 431 U.S. 99, 97 S.Ct. 1635, 52 L.Ed.2d 166 (1977). 20 See also 124 Cong.Rec. 34558 (1978) (remarks of Sen. Jackson); id., at 34560 (remarks of Sen. Bumpers); id., at 34776 (remarks of Sen. Robert C. Byrd); id., at 38350 (remarks of Rep. Ashley); id., at 38370-38371 (remarks of Rep. Dingell); 123 Cong.Rec. 25894 (1977) (remarks of Rep. Ashley); id., at 25916-25917 (remarks of Rep. Ottinger); id., at 27063-27064 (remarks of Rep. Wolff). 21 See, e.g., id., at 32437-32438 (remarks of Sen. Brooke); id., at 32444 (remarks of Sen. Percy). 22 PURPA could be upheld even if some of its provisions were not directly related to the purpose of fostering interstate commerce: "A complex regulatory program . . . can survive a Commerce Clause challenge without a showing that every single facet of the program is independently and directly related to a valid congressional goal. It is enough that the challenged provisions are an integral part of the regulatory program and that the regulatory scheme when considered as a whole satisfies this test." Hodel v. Indiana, 452 U.S. 314, 329, n. 17, 101 S.Ct. 2376, 2385, n.17, 69 L.Ed.2d 40 (1981). 23 This is not to say the Congress can regulate in an area that is only tangentially related to interstate commerce. See Maryland v. Wirtz, 392 U.S. 183, 196-197, n. 27, 88 S.Ct. 2017, 2023-2024, n.27, 20 L.Ed.2d 1020 (1968). That obviously is not the case here. 24 In another context, the Court has noted that "the role of the modern federal hearing examiner or administrative law judge . . . is 'functionally comparable' to that of a judge." Butz v. Economou, 438 U.S. 478, 513, 98 S.Ct. 2894, 2914, 57 L.Ed.2d 895 (1978). 25 Justice O'CONNOR reviews the constitutional history at some length, ultimately deriving the proposition that the Framers intended to deny the Federal Government the authority to exercise "military or legislative power over state governments," instead "allow[ing] Congress to pass laws directly affecting individuals." Post, at 795. If Justice O'CONNOR means this rhetorical assertion to be taken literally, it is demonstrably incorrect. See, e.g., Transportation Union v. Long Island R. Co., 455 U.S. 678, 102 S.Ct. 1349, 71 L.Ed.2d 547 (1982); Fry v. United States, 421 U.S. 542, 95 S.Ct. 1792, 44 L.Ed.2d 363 (1975); Parden v. Terminal R. Co., 377 U.S. 184, 84 S.Ct. 1207, 12 L.Ed.2d 233 (1964); California v. Taylor, 353 U.S. 553, 77 S.Ct. 1037, 1 L.Ed.2d 1034 (1957); Case v. Bowles, 327 U.S. 92, 66 S.Ct. 438, 90 L.Ed. 552 (1946); United States v. California, 297 U.S. 175, 56 S.Ct. 421, 80 L.Ed. 567 (1936). 26 The Court did express doubt as to whether a state agency "may be ordered actually to promulgate regulations having effect as a matter of state law." 443 U.S., at 695, 99 S.Ct., at 3079. As we have noted, however, PURPA does not require promulgation of particular regulations. 27 Justice O'CONNOR's partial dissent finds each of these cases inapposite. Yet the purported distinctions are little more than exercises in the art of ipse dixit. Thus she suggests that Testa v. Katt provides no support for the imposition of federal responsibilities on state legislatures, because "the requirement that [state courts] evenhandedly adjudicate state and federal claims falling within their jurisdiction does not infringe any sovereign authority to set an agenda." Post, at 2150-2151. Yet the courts have always been recognized as a coequal part of the State's sovereign decision-making apparatus, see Bates v. State Bar of Arizona, 433 U.S. 350, 360, 97 S.Ct. 2691, 2697, 53 L.Ed.2d 810 (1977), and it seems evident that requiring state tribunals to entertain federal claims interferes, at least to a degree, with the State's sovereign prerogatives, see n.25, supra, as well as with the amount of time that state courts may devote to adjudicating state claims. Conversely, it is difficult to perceive any fundamental distinction between the state legislature's power to establish limits on the jurisdiction of state courts, and its prerogative to set ratemaking criteria for use by quasi-legislative utilities commissions. Justice O'CONNOR fails to explain, however, why this does not implicate her concern that "[w]hile engaged in . . . congressionally mandated tasks, state utility commissions are less able to pursue local proposals . . . ." Post, at 787. The partial dissent finds Fry v. United States inapposite because the wage freeze there at issue " 'displaced no state choices as to how governmental operations should be structured . . . . Instead, it merely required that the wage scales and employment relationships which the States themselves had chosen be maintained . . ..' " Post, at 784, n. 13, quoting National League of Cities v. Usery, 426 U.S. 833, 853, 96 S.Ct. 2465, 2474, 49 L.Ed.2d 245 (1976). It seems absurd to suggest, however, that a federal veto of the States' chosen method of structuring their employment relationships is less intrusive in any realistic sense than are PURPA's mandatory consideration provisions. Finally, Justice O'CONNOR would distinguish Fishing Vessel Assn. as involving only "[t]he power of a court to enjoin adjudicated violations of federal law." Post, at 784, n. 13. In doing so, however, the Court unambiguously held that federal law could impose an affirmative obligation upon state officials to prepare administrative regulations—a holding of obvious relevance to this case. 28 In Dennison, the Court concluded that the state courts entertained federal actions solely as a discretionary "matter of comity, which the several sovereignties extended to one another for their mutual benefit. It was not regarded by either party as an obligation imposed by the Constitution." 24 How., at 109. That analysis cannot survive Testa, which squarely held "that state courts do not bear the same relation to the United States that they do to foreign countries." 330 U.S., at 389, 67 S.Ct., at 812. And Testa, of course, placed the obligation of state officials to enforce federal law squarely in the Supremacy Clause. Our recent cases also demonstrate that the Federal Government, at least in certain circumstances, can structure the State's exercise of its sovereign powers. In National League of Cities v. Usery, supra, for example, the Court made clear that the State's regulation of its relationship with its employees is an "undoubted attribute of state sovereignty." 426 U.S., at 845, 96 S.Ct., at 2471. Yet, by holding "unimpaired" California v. Taylor, 353 U.S. 553, 77 S.Ct. 1037, 1 L.Ed.2d 1034 (1957), which upheld a federal labor regulation as applied to state railroad employees, 426 U.S., at 854, n.18, 96 S.Ct., at 2475, n.18, National League of Cities acknowledged that not all aspects of a State's sovereign authority are immune from federal control. This analysis was restated in Hodel v. Virginia Surface Mining & Recl. Assn., supra, which indicated that federal regulations are subject to Tenth Amendment attack only if they "regulat[e] the 'States as States,' " "address matters that are indisputably 'attributes of state sovereignty,' " and impair the States' "ability 'to structure integral operations in areas of traditional functions.' " 452 U.S., at 287-288, 101 S.Ct., at 2365-2366, quoting National League of Cities v. Usery, 426 U.S., at 854, 845, 852, 96 S.Ct., at 2475, 2471, 2474. And even when these requirements are met, "[t]here are situations in which the nature of the federal interest advanced may be such that it justifies state submission." Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 288, n. 29, 101 S.Ct., at 2366, n. 29. 29 It seems evident that Congress intended to defer to state prerogatives—and expertise—in declining to pre-empt the utilities field entirely. See, e.g., S.Rep.No. 95-442, pp. 9, 13-14 (1977); 124 Cong.Rec. 34558 (1978) (remarks of Sen. Jackson); id., at 34560 (remarks of Sen. Bumpers); id., at 34763 (remarks of Sen. Metzenbaum); id., at 34768 (remarks of Sen. Durkin); 123 Cong.Rec. 32430 (1977) (remarks of Sen. Johnston); id., at 32395 (remarks of Sen. Bartlett). Justice O'CONNOR's partial dissent's response to this is peculiar. On the one hand, she suggests that the States might prefer that Congress simply pre-empt the field, since that "would leave them free to exercise their power in other areas." Post, at 787. Yet Justice O'CONNOR elsewhere acknowledges the importance of utilities regulation to the States, post, at 781, and emphasizes that local experimentation and self-determination are essential aspects of the federal system. Post, at 787-791. PURPA, of course, permits the States to play a continued role in the utilities field, and gives full force to the States' ultimate policy choices. Certainly, it is a curious type of federalism that encourages Congress to pre-empt a field entirely, when its preference is to let the States retain the primary regulatory role. 30 Justice O'CONNOR's partial dissent suggests that our analysis is an "absurdity," post, at 781, and variously accuses us of "conscript[ing] state utility commissions into the national bureaucratic army," of transforming state legislative bodies into "field offices of the national bureaucracy," of approving the "dismemberment of state government," of making state agencies "bureaucratic puppets of the Federal Government," and—most colorfully—of permitting "Congress to kidnap state utility commissions." Post, at 775, 777, 782, 783,790.. While these rhetorical devices make for absorbing reading, they unfortunately are substituted for useful constitutional analysis. For while Justice O'CONNOR articulates a view of state sovereignty that is almost mystical, she entirely fails to address our central point. The partial dissent does not quarrel with the propositions that Congress may pre-empt the States in the regulation of private conduct, that Congress may condition the validity of State enactments in a pre-emptible area on their conformity with federal law, and that Congress may attempt to "coerce" the States into enacting nationally desirable legislation. Given this, the partial dissent fails to identify precisely what is "absurd" about a scheme that gives the States a choice between regulating in conformity with federal requirements, or abandoning regulation in a given field. Though the partial dissent finds Hodel v. Virginia Surface Mining & Recl. Assn. inapposite, in our view the parallel is striking: there, the States were directed to legislate consistently with congressional enactments, or not at all; here, the States are asked to regulate in conformity with federal requirements, or not at all. While it is true that PURPA conditions continued state regulatory activity on the performance of certain affirmative tasks, the partial dissent nowhere explains why—so long as the field is preemptible the nature of the condition is relevant. And while PURPA's requirements in practice may be more intrusive and more difficult for the States to avoid than was the legislation at issue in Hodel v. Virginia Surface Mining & Recl. Assn., Justice O'CONNOR herself acknowledges that an "evaluation of intrusiveness . . . is simply irrelevant to the constitutional inquiry." Post, at 785-786. Similarly, the difference between PURPA and the Surface Mining Control and Reclamation Act of 1977 identified by the partial dissent cannot be that only the former affects a "traditional function of state government," post, at 781, for regulation of land use is perhaps the quintessential state activity. In short, while the area of state action potentially foreclosed by PURPA may be broader than was the case in Hodel, the partial dissent has pointed to no constitutionally significant theoretical distinction between the two statutory schemes. 31 We believe that this seemingly precise parallel between state and federal procedures suffices to overcome Justice POWELL's objections to PURPA, at least where, as here, the statute is subjected to a facial attack. See also n.34, infra. 32 Justice O'CONNOR's partial dissent accuses us of undervaluing National League of Cities, and maintains that our analysis permits Congress to "dictate the agendas and meeting places of state legislatures." Post, at 782. These apocalyptic observations, while striking, are overstated and patently inaccurate. We hold only that Congress may impose conditions on the State's regulation of private conduct in a pre-emptible area. This does not foreclose a Tenth Amendment challenge to federal interference with the State's ability "to structure employer-employee relationships," 426 U.S., at 851, 96 S.Ct., at 2474, while providing "those governmental services which [its] citizens require," id., at 847, 96 S.Ct., at 2472, as was the case in National League of Cities. It does not suggest that the Federal Government may impose conditions on state activities in fields that are not pre-emptible, or that are solely of intrastate concern. And it does not purport to authorize the imposition of general affirmative obligations on the States. 33 As we note above, PURPA imposes certain reporting requirements on state commissions. But because these attach only if the State chooses to continue its regulatory efforts in the field, we find them supportable for the reasons addressed in connection with the other provisions of Titles I and III. Appellees nevertheless suggest that PURPA's requirements must fall because compliance will impose financial burdens on the States. We are unconvinced: in a Tenth Amendment challenge to congressional activity, "the determinative factor . . . [is] the nature of the federal action, not the ultimate economic impact on the States." Hodel v. Virginia Surface Mining & Recl. Assn., Inc., 452 U.S., at 292, n.33, 101 S.Ct., at 2368, n.33. In any event, Congress has taken steps to reduce or eliminate the economic burden of compliance. See n.14, supra. 34 Mississippi law provides for reasonable notice in the fixing of rates and conditions of service of utilities. Miss.Code Ann. § 77-3-33(2) (1973). It also requires the Public Service Commission to keep a "full and complete record" of all proceedings, § 77-3-63, and to "make and file its findings and order, and its opinion, if any," § 77-3-59. Indeed, the state statute requires that "[a]ll findings of the commission and the determination of every matter by it shall be in writing and placed upon its minutes." § 77-1-41. These "shall be deemed a public record, and shall at all seasonable times be subject to the inspection of the public." Ibid. Thus, the requirements that appellees characterize as an extraordinary burden on the State appear to accord few, if any, procedural rights not already established by Mississippi law. 1 16 U.S.C. § 2631(a) (1976 ed., Supp.IV). "[A]ny electric utility or electric consumer" may enforce its intervention and participation rights in federal court. § 2633(b)(2). See also § 2633(b)(1). The PURPA grants similar intervention and participation rights to the Secretary with respect to state natural gas utility rate proceedings. See 15 U.S.C. § 3205 (1976 ed., Supp.IV). These rights also are specified to be enforceable in federal court. See § 3207(a)(2). 2 16 U.S.C. § 2633(c)(1) (1976 ed., Supp.IV). A similar enforcement right is granted in the case of natural gas rate proceedings. 15 U.S.C. § 3207(b)(1) (1976 ed., Supp.IV). Under the PURPA's Title II, § 210, States must implement federal rules relating to the interconnection of electrical utilities with qualifying cogeneration and small power production facilities. 16 U.S.C. § 824a-3 (1976 ed., Supp.IV). The Federal Energy Regulatory Commission and (under certain conditions) "[a]ny electrical utility, qualifying cogenerator, or qualifying small power producer" may bring judicial actions against state regulatory commissions to require the implementation of the federal rules prescribed by the PURPA. §§ 824a-3(h)(2)(A) and (B). 3 16 U.S.C. § 2633(c)(1) (1976 ed., Supp.IV). The PURPA also makes available a right of judicial review in the same manner with respect to the interconnection of electrical utilities with cogeneration and small power production facilities. § 824a-3(g)(1). No similar right is available in the case of natural gas rate proceedings. See 15 U.S.C. § 3207(b)(2) (1976 ed., Supp.IV). As a separate matter, the PURPA specifies the procedural requirements for the state regulatory agencies' consideration and determination of the PURPA's federally proposed standards. See § 3203(c); 16 U.S.C. § 2621(b)(1) (1976 ed., Supp.IV). 4 The Court cites Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947), in support of the proposition that under some conditions the Federal Government may call upon state governmental institutions to decide matters of federal policy. But Testa recognized that, when doing so, Congress must respect the state institution's own decisionmaking structure and method. That opinion limited its holding to circumstances under which the state court has "jurisdiction adequate and appropriate under established local law to adjudicate this [federal] action." Id., at 394, 67 S.Ct., at 814 (emphasis added). The Testa Court then emphasized its meaning by citing Herb v. Pitcairn, 324 U.S. 117, 65 S.Ct. 459, 89 L.Ed. 789 (1945), where the Court stated that "[i]t would not be open to us" to insist on adjudication in a state court of a federal claim arising beyond the jurisdiction of the local court. Id., at 121, 65 S.Ct., at 461. See Note, Utilization of State Courts to Enforce Federal Penal and Criminal Statutes: Development in Judicial Federalism, 60 Harv.L.Rev. 966, 971 (1947) (nothing in Testa upsets "the traditional doctrine that Congress may not interfere with a state's sovereign right to determine and control the jurisdictional requirements of its own courts"). The Court also cites Washington v. Washington State Commercial Passenger Fishing Vessel Assn., 443 U.S. 658, 99 S.Ct. 3055, 61 L.Ed.2d 823 (1979), to support its holding. Ante, 762. The case stands for the unremarkable proposition that a district court, after adjudicating a contest under federal law between a State and Indian tribes over fishing rights, may order the losing State to abide by the court's decision. Nothing in our Fishing Vessel Assn. opinion authorized the federal court to amend the structure of a state political institution. 5 L. Tribe, American Constitutional Law 302 (1978). 1 I concur in the Court's decision to uphold Title II, § 210, of PURPA against appellees' facial attack. As the Court explains, part of that section permits the Federal Energy Regulatory Commission (FERC) to exempt cogeneration and small power production facilities from otherwise applicable state and federal laws. 16 U.S.C. § 824a-3(e) (1976 ed., Supp. IV). This exemption authority does not violate the Tenth Amendment, for it merely pre-empts state control of private conduct, rather than regulating the "States as States." See Hodel v. Virginia Surface Mining & Reclamation Assn., Inc., 452 U.S. 264, 287-293, 101 S.Ct. 2352, 2365-2369, 69 L.Ed.2d 1 (1981). Section 210's requirement that the States "implement" rules promulgated by the Secretary of Energy, 16 U.S.C. § 824a-3(f) (1976 ed., Supp. IV), is more disturbing. Appellants, however, have interpreted this statutory obligation to include "an undertaking to resolve disputes between qualifying facilities and electric utilities arising under [§ 210], or any other action reasonably designed to implement [that section]." 18 CFR § 292.401(a) (1981). It appears, therefore, that state regulatory authorities may satisfy § 210's implementation requirement simply by adjudicating private disputes arising under that section. As the Court points out, ante, at 760-761, the Mississippi Public Service Commission has jurisdiction over similar state disputes, and it is settled that a State may not exercise its judicial power in a manner that discriminates between analogous federal and state causes of action. See Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947). Under these circumstances, but without foreclosing the possibility that particular applications of § 210's implementation provision might uncover hidden constitutional defects, I would not sustain appellees' facial attack on the provision. Section 210 also authorizes FERC, electric utilities, cogenerators, and small power producers to "enforce" the above implementation provision against state utility commissions. 16 U.S.C. § 824a-3(h)(2) (1976 ed., Supp. IV). As applied, it is conceivable that this enforcement provision would raise troubling federalism issues. Once again, however, I decline to accept appellees' facial challenge to the provision, preferring to consider the constitutionality of this provision in the setting of a concrete controversy. 2 The statute imposes the same requirements upon nonregulated utilities. In this respect, it regulates purely private conduct and does not violate the Tenth Amendment. Throughout this dissent, I consider only the constitutionality of Titles I and III as applied to state regulatory authorities. I would allow the District Court, on remand, to decide whether the constitutionally defective aspects of Titles I and III are severable from the unobjectionable portions. 3 See ante, at 748-749. The Court overlooks several of PURPA's procedural mandates. For example, with respect to six of the standards, the state agency must publish a written determination, including findings, even if it decides to adopt the federal standard. 16 U.S.C. § 2621(b) (1976 ed., Supp. IV). In addition, PURPA guarantees certain rights to discover information, § 2631(b); requires the State to provide transcripts, at the cost of reproduction, to parties to ratemaking proceedings or other "regulatory proceeding[s] relating to [electric utility] rates or rate design," § 2632(c); and, under some circumstances, mandates compensation for reasonable attorney's fees, expert witness fees, and other costs to consumers who contribute substantially to the adoption of a Title I standard, §§ 2632(a), (b). These requirements, as well as the ones described by the Court, may impose special burdens on state administrative agencies. I do not weigh the constitutionality of these individual procedural requirements, however, because I would invalidate the entire regimen that Titles I and III impose on state regulatory authorities. 4 In both Hodel and United Transportation Union we further noted that, even when these three requirements are met, "the nature of the federal interest advanced may be such that it justifies state submission." Hodel, 452 U.S., at 288, n.29, 101 S.Ct., at 2366, n.29; Transportation Union, 455 U.S., at 684, n.9, 102 S.Ct., at 1353, n.9. Neither of those cases involved such an exception to National League of Cities, and the Court has not yet explored the circumstances that might justify such an exception. 5 As the Court recognizes, ante, at 748, PURPA permits "[a]ny person" to bring an action in state court to enforce the agency's obligation to consider the federal standards. 15 U.S.C. § 3207(b)(1) (1976 ed., Supp. IV); 16 U.S.C. § 2633(c)(1) (1976 ed., Supp. IV). The Secretary of Energy, moreover, may intervene in any ongoing ratemaking proceeding to require consideration of PURPA's standards. 15 U.S.C. § 3205(a) (1976 ed., Supp. IV); 16 U.S.C. §§ 2631(a), 2622(a) (1976 ed., Supp. IV). Title I grants affected utilities and consumers the same right of intervention. 16 U.S.C. § 2631(a) (1976 ed., Supp. IV). Because of these rights of intervention and enforcement, state agencies lack even the power to schedule their consideration of PURPA's standards. 6 For example, the proposed standards governing advertising provide that "[n]o electric [or gas] utility may recover from any person other than the shareholders (or other owners) of such utility any direct or indirect expenditure by such utility for promotional or political advertising as [further] defined in . . . this title." 16 U.S.C. § 2623(b)(5) (1976 ed., Supp. IV); 15 U.S.C. § 3203(b)(2) (1976 ed., Supp. IV). PURPA then defines the terms "advertising," "political advertising," and "promotional advertising": "(1) For purposes of this section and section 2623(b)(5) of this title— "(A) The term 'advertising' means the commercial use, by an electric utility, of any media, including newspaper, printed matter, radio, and television, in order to transmit a message to a substantial number of members of the public or to such utility's electric consumers. "(B) The term 'political advertising' means any advertising for the purpose of influencing public opinion with respect to legislative, administrative, or electoral matters, or with respect to any controversial issue of public importance. "(C) The term 'promotional advertising' means any advertising for the purpose of encouraging any person to select or use the service or additional service of an electric utility or the selection or installation of any appliance or equipment designed to use such utility's service. "(2) For purposes of this subsection and section 2623(b)(5) of this title, the terms 'political advertising' and 'promotional advertising' do not include— "(A) advertising which informs electric consumers how they can conserve energy or can reduce peak demand for electric energy, "(B) advertising required by law or regulation, including advertising required under part 1 of title II of the National Energy Conservation Policy Act . . ., "(C) advertising regarding service interruptions, safety measures, or emergency conditions, "(D) advertising concerning employment opportunities with such utility, "(E) advertising which promotes the use of energy efficient appliances, equipment or services, or "(F) any explanation or justification of existing or proposed rate schedules, or notifications of hearings thereon." 16 U.S.C. § 2625(h) (1976 ed., Supp. IV). See also 15 U.S.C. § 3204(b) (1976 ed., Supp. IV) (containing similar provisions for gas utilities). 7 The Court has not explored fully the extent of "traditional" state functions. Utility regulation, however, should fall within any definition of that term. See generally W. Jones, Cases and Materials on Regulated Industries 25-44 (2d ed. 1976) (tracing history of state regulation of utilities). 8 PURPA thus offends each of the criteria named in Hodel. I do not believe, moreover, that this is a case in which "the nature of the federal interest advanced may be such that it justifies state submission." See n.4, supra. Whatever the ultimate content of that standard, it must refer not only to the weight of the asserted federal interest, but also to the necessity of vindicating that interest in a manner that intrudes upon state sovereignty. In this case, the Government argues that PURPA furthers vital national interests in energy conservation. Although the congressional goal is a noble one, appellants have not shown that Congress needed to commandeer state utility commissions to achieve its aim. Consistent with the Tenth Amendment, Congress could have assigned PURPA's tasks to national officials. Alternatively, it could have requested state commissions to comply with Titles I and III and directed the Secretary to shoulder the burden of any State choosing not to comply. 9 The Court attempts to distinguish National League of Cities, suggesting that it involved "the State's ability 'to structure employer-employee relationships,' . . . while providing 'those governmental services which [its] citizens require.' " Ante, at 770, n.32 (quoting National League of Cities, 426 U.S., at 851, 847, 96 S.Ct., at 2474, 2472). This case, the Court declares, "hold[s] only that Congress may impose conditions on the State's regulation of private conduct in a pre-emptible area." Ante, at 769-770, n.32. The Court, however, does not explain why our National League of Cities opinion did not consider compliance with the Fair Labor Standards Act in fields such as "licensing of occupations and businesses, . . . preservation of environmental quality, . . . [and] protection of the public against fraud and sharp practice," App. in National League of Cities v. Usery, O.T.1975, No. 74-876, p. 16 (reprinting complaint), a "conditio[n] on the State's regulation of private conduct in a pre-emptible area." In that case, Congress had required the States to pay their employees specified amounts if they wished to continue regulating a variety of pre-emptible fields. Here, it has required the States to burden their officials with evaluation of a dozen legislative proposals if they wish to continue regulating private utilities. To me, the parallel is obvious, not "overstated." Ante, at 769, n.32. I am nevertheless confident that, as the Court itself stresses, ibid., today's decision is not intended to overrule National League of Cities. Instead, the novelty of PURPA's scheme, see ante, at 758-759, merely seems to have obscured the relevance of National League of Cities to this case. 10 But cf. Coyle v. Oklahoma, 221 U.S. 559, 565, 31 S.Ct. 688, 689, 55 L.Ed. 853 (1911) ("The power to locate its own seat of government and to determine when and how it shall be changed from one place to another . . . are essentially and peculiarly state powers. That one of the . . . States could now be shorn of such powers by an act of Congress would not be for a moment entertained"). 11 Title 30 U.S.C. § 1254(g) (1976 ed., Supp.IV) only pre-empts state laws "insofar as they interfere with the achievement of the purposes and the requirements of this chapter and the Federal program." Similarly, § 1255(a) provides that no state law or regulation "shall be superseded by any provision of this chapter or any regulation issued pursuant thereto, except insofar as such State law or regulation is inconsistent with the provisions of this chapter." Section 1255(b) explains that neither state laws that are more stringent than the federal standards nor state laws governing operations "for which no provision is contained in this chapter" are "inconsistent" with the congressional Act. 12 As one scholar has written: "A federal system implies a partnership, all members of which are effective players on the team and all of whom retain the capacity for independent action. It does not imply a system of collaboration in which one of the collaborators is annihilated by the other." L. White, The States and the Nation 3 (1953). 13 The other two decisions, Fry v. United States, 421 U.S. 542, 95 S.Ct. 1792, 44 L.Ed.2d 363 (1975), and Washington v. Washington State Commercial Passenger Fishing Vessel Assn., 443 U.S. 658, 99 S.Ct. 3055, 61 L.Ed.2d 823 (1979), are readily distinguishable. Fry upheld a temporary wage freeze as applied to state and local governmental employees. As we subsequently observed, this emergency restraint "displaced no state choices as to how governmental operations should be structured, nor did it force the States to remake such choices themselves. Instead, it merely required that the wage scales and employment relationships which the States themselves had chosen be maintained during [a] period of . . . emergency." National League of Cities v. Usery, supra, at 853, 96 S.Ct. 2465, 2474. In Washington State Fishing Vessel Assn., state agencies were defendants to a suit charging violations of federal treaties, and we upheld the lower court's power to enforce its judgment by ordering the defendants to comply with federal law. The power of a court to enjoin adjudicated violations of federal law, however, is far different from the power of Congress to demand state legislative action in the absence of any showing that the State has violated existing federal duties. See Hart, The Relations Between State and Federal Law, 54 Colum.L.Rev. 489, 515-516 (1954); Salmon, The Federalist Principle: The Interaction of the Commerce Clause and the Tenth Amendment in the Clean Air Act, 2 Colum.J.Envtl.L. 290, 334-337 (1976). 14 The Court suggests, ante, at 762-763, n.27, that the requirement that state courts adjudicate federal claims may, as a practical matter, undermine the capacity of those courts to decide state controversies. Whatever the force of that observation, it does not demonstrate Testa's relevance to this case. State legislative bodies possess at least one attribute of sovereignty, the power to set an agenda, that trial courts lack. This difference alone persuades me not to embrace the Court's expansion of Testa. 15 The Court's suggestion is somewhat disingenuous because Congress concluded that federal pre-emption of the matters governed by Titles I and III would be inappropriate. The administration's original proposal, as well as the version of PURPA approved by the House, would have preempted state law by establishing minimum federal ratemaking standards. See generally H.R.Conf.Rep.No. 95-1750, pp. 63-65 (1978); S.Conf.Rep.No. 95-1292, pp. 63-65 (1978). The Senate Committee on Energy and Natural Resources, however, rejected this approach because "the committee felt that setting minimum federal standards for utility rates, or mandating the use of certain costing methods for ratesetting, would be an unnecessary intrusion into an area which has traditionally been regulated by the States. It was apparent to the committee that many State utility commissions are currently involved in innovative ratemaking and are working toward the goal of conservation of energy through rate reform. At present, the State regulatory agencies rather than the Federal Government, possess the expertise to conduct the detailed costing and demand studies required to implement rate structure revision. Moreover, the committee recognized that rate structures must reflect the individual needs and local peculiarities of each utilities' service area. . . . Finally the committee felt that the potential uncertainty and delays accompanying Federal regulation threatened to have an adverse impact on the financial health of the utility industry which outweighed the projected savings in capital expenditures claimed by supporters of the administration's proposal." S.Rep. No. 95-442, p. 9 (1977), U.S.Code Cong. & Admin.News, p. 7906. See also 123 Cong.Rec. 32392-32393 (1977) (remarks of Sen. Johnston); id., at 32394 (remarks of Sen. Domenici). The Senate version of PURPA, accordingly, eschewed the pre-emption route. See H.R.Conf.Rep.No. 95-1750, supra, at 65-66; S.Conf.Rep. No. 95-1292, supra, at 65-66. While the Conferees produced a compromise bill, they too stopped short of preemption. Today's decision, therefore, permits Congress to set state legislative agendas in a field that Congress might have occupied but expressly found unsuited to pre-emption. 16 Justice Harlan once commented that times of "international unrest and domestic uncertainty" are "bound to produce temptations and pressures to depart from or temporize with traditional constitutional precepts or even to short-cut the processes of change which the Constitution establishes." Harlan, Thoughts at a Dedication: Keeping the Judicial Function in Balance, 49 A.B.A.J. 943 (1963). Justice Harlan then cautioned that it "[i]s . . . the special responsibility of lawyers, whether on or off the bench, to see to it that such things do not happen." Ibid. 17 In 1975, then Attorney General Edward H. Levi responded to a similar argument that the "greater" power of pre-emption includes the "lesser" power of demanding affirmative action from state governments. Attorney General Levi remarked that "it is an insidious point to say that there is more federalism by compelling a State instrumentality to work for the Federal Government." Hearings on S. 354 before the Senate Committee on Commerce, 94th Cong., 1st Sess., 503 (1975). In a similar vein, he warned against "lov[ing] the States to their demise." Id., at 507. 18 In rare instances, Congress so occupies a field that any state regulation is inconsistent with national goals. The Court, however, is reluctant to infer such expansive pre-emption "in the absence of persuasive reasons." Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142, 83 S.Ct. 1210, 1217, 10 L.Ed.2d 248 (1963). 19 See generally Stewart, Pyramids of Sacrifice? Problems of Federalism in Mandating State Implementation of National Environmental Policy, 86 Yale L.J. 1196, 1239-1247 (1977); Comment, Redefining the National League of Cities State Sovereignty Doctrine, 129 U.Pa.L.Rev. 1460, 1477-1478 (1981). Daniel Elazar, testifying before the Advisory Commission on Intergovernmental Relations in March 1980, commented upon this problem of garbled political responsibility. He suggested that national officials tend to force state governments to administer unpopular programs, thus transferring political liability for those programs to the States. Advisory Commission on Intergovernmental Relations, The Federal Role in the Federal System: The Dynamics of Growth, Hearings on the Federal Role 32 (Oct. 1980). As an example, he cited the President's attempt in 1979 to force state Governors to establish and enforce unpopular gas rationing mechanisms. Id., at 85 (formal statement of Professor Elazar). 20 See, e.g., Chandler v. Florida, 449 U.S. 560, 579, 101 S.Ct. 802, 812, 66 L.Ed.2d 740 (1981); Reeves, Inc. v. Stake, 447 U.S. 429, 441, 100 S.Ct. 2271, 2279, 65 L.Ed.2d 244 (1980); Whalen v. Roe, 429 U.S. 589, 597, and n.20, 97 S.Ct. 869, 875, and n.20, 51 L.Ed.2d 64 (1977); New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386, 76 L.Ed. 747 (1932) (Brandeis, J., dissenting); Hart, supra n.13, at 540, 542; A. Macmahon, The Problems of Federalism: A Survey, in Federalism: Mature and Emergent 3, 10-11 (A. Macmahon ed., 1955); N. Rockefeller, The Future of Federalism 8-9 (1962); Stewart, supra n.19, at 1210; White, supra n.12, at 46-47. 21 Wyoming's policy followed a practice it had adopted as a Territory. Compare Act of Jan. 21, 1891, ch. 100, § 4, 1890-1891 Wyo.Sess.Laws 394, with Act of Mar. 14, 1890, ch. 80, § 7, 1890 Wyo.Terr.Sess.Laws 158. See generally C. Beard & M. Beard, The Rise of American Civilization 563 (rev. ed. 1937). 22 The Nineteenth Amendment, ratified in 1920, prohibits abridgment of the right to vote "on account of sex." 23 See Act of Jan. 28, 1932, ch. 20, 1931-1932 Wis.Laws 57; Act of June 1, 1933, ch. 186, 1933 Wis.Laws 448; Act of June 2, 1933, ch. 194, 1933 Wis.Laws 491; W. Leuchtenburg, Franklin D. Roosevelt and the New Deal, 1932-1940, p. 130 (1963); Rockefeller, supra n.20, at 16. 24 See Act of June 4, 1912, ch. 706, 1912 Mass. Acts 780; R. Morris, Encyclopedia of American History 768 (bicentennial ed. 1976). 25 See C. Morris & C. Morris, Jr., Morris on Torts 244-245 (2d ed. 1980); Friendly, Federalism: A Foreword, 86 Yale L.J. 1019, 1034 (1977). 26 Florida, for example, has enacted particularly strict legislation against oil spills. Fla.Stat. §§ 376.011-.376.21 (1974 and Supp. 1982). This Court upheld that legislation in Askew v. American Waterways Operators, Inc., 411 U.S. 325, 93 S.Ct. 1590, 36 L.Ed.2d 280 (1973). 27 See FPC v. East Ohio Gas Co., 338 U.S. 464, 489, 70 S.Ct. 266, 279, 94 L.Ed. 268 (1950) (Jackson, J., dissenting) ("Long before the Federal Government could be stirred to regulate utilities, courageous states took the initiative and almost the whole body of utility practice has resulted from their experiences"). 28 See also I. Silone, The School for Dictators 119 (W. Weaver trans. 1963) ("A regime of freedom should receive its lifeblood from the self-government of local institutions. When democracy, driven by some of its baser tendencies, suppresses such autonomies, it is only devouring itself. If in the factory the master's word is law, if bureaucracy takes over the trade union, if the central government's representative runs the city and the province, . . . then you can no longer speak of democracy"). 29 See also Stewart, supra n.19, at 1241-1244 (discussing "political safeguards of federalism"); Rockefeller, supra n.20, at 10. 30 Henry M. Hart, Jr., agreed that the Framers were well aware "of the delicacy, and the difficulties of enforcement, of affirmative mandates from a federal government to the governments of the member states." Hart, supra n.13, at 515. Until the second half of this century, Congress apparently heeded this wisdom. "Federal law," Hart observed in 1954, "often says to the states, 'Don't do any of these things,' leaving outside the scope of its prohibition a wide range of alternative courses of action. But it is illuminating to observe how rarely it says, 'Do this thing,' leaving no choice but to go ahead and do it." Ibid. 31 Governor Randolph of Virginia, for example, opposed a similar proposal for national coercion on the grounds that it was "impracticable, expensive, [and] cruel to individuals." Instead, he advocated "resort . . . to a national Legislation over individuals." 1 Farrand 256 (emphasis deleted). Mason eloquently argued that "[t]he most jarring elements of nature; fire & water themselves are not more incompatible that [sic ] such a mixture of civil liberty and military execution." Id., at 339. 32 Thomas Jefferson disapproved of the congressional veto as soon as he heard of it. Writing to Madison from Paris, he declared: "The negative proposed to be given [the national legislators] on all the acts of the several Legislatures is now for the first time suggested to my mind. Prima facie I do not like it." C. Warren, The Making of the Constitution 168 (1937). Notably, Jefferson suggested that "an appeal from the State Judicatures to a Federal Court, in all cases where the Act of Confederation controuled the question, [would] be as effectual a remedy." Id., at 168-169. 33 Experience under the Articles of Confederation taught the Framers that multiple state legislatures, unchecked by any central power, "threat[en] danger not to the harmony only, but to the tranquillity of the Union." Id., at 166 (quoting Madison). My analysis of the Framers' intent does not detract from the proper role of federal power in a federalist system, but merely requires the exercise of that power in a manner that does not destroy state independence. 34 This Court quickly recognized that Congress' strength derives from its own enumerated powers, not from the ability to direct state legislatures. In McCulloch v. Maryland, 4 Wheat. 316, 4 L.Ed. 579 (1819), the historic decision affirming Congress' power to establish a national bank, Chief Justice Marshall declared: "No trace is to be found in the constitution of an intention to create a dependence of the government of the Union on those of the states, for the execution of the great powers assigned to it. Its means are adequate to its ends; and on those means alone was it expected to rely for the accomplishment of its ends." Id., at 424 (emphasis added). See also S. Davis, The Federal Principle 114 (1978) (after examining history of Constitutional Convention, "only the principle of duality articulated in a single constitutional system of two distinct governments, national and state, each acting in its own right, each acting directly on individuals, and each qualified master of a limited domain of action, stands out as the clearest fact"); Salmon, supra n.13, at 359 (discussing history of Constitutional Convention and concluding that substitution of Supremacy Clause for negative on state laws "evidenced the clear distinction in [the Framers'] minds between the supremacy of the nation, which they approved, and the power of the nation to control the functioning of the states, which they rejected"). 35 After the Convention, several thinkers suggested that the National Government might rely upon state officers to perform some of its tasks. Madison, for example, thought that Congress might rely upon state officials to collect national revenue. The Federalist No. 45, pp. 312-313 (J. Cooke ed. 1961). None of these suggestions, however, went so far as to propose congressional control of state legislative power. The suggestions, moreover, seemed to assume that the States would consent to national use of their officials. See also W. Anderson, The Nation and the States, Rivals or Partners? 86-87 (1957) (noting that First Congress rejected proposals to rely upon state officials to enforce federal law and suggesting that this decision to leave "the states free to work out, and to concentrate their attention and resources upon, their own functions" has become part of our constitutional understanding).

910

456 U.S. 694 102 S.Ct. 2099 72 L.Ed.2d 492 INSURANCE CORP. OF IRELAND, LTD. et al., Petitioners,v.COMPAGNIE des BAUXITES de GUINEE. No. 81-440. Argued March 23, 1982. Decided June 1, 1982. Syllabus Federal Rule of Civil Procedure 37(b)(2)(A) provides that a district court, as a sanction for failure to comply with discovery orders, may enter "[a]n order that the matters regarding which the [discovery] order was made or any other designated facts shall be taken to be established for the purposes of the action in accordance with the claim of the party obtaining the order." Asserting diversity jurisdiction, respondent, a Delaware corporation with its principal place of business in the Republic of Guinea, filed suit against various insurance companies in the United States District Court for the Western District of Pennsylvania to recover on a business interruption policy. When certain of the defendants (a group of foreign insurance companies, including petitioners) raised the defense of lack of personal jurisdiction, respondent attempted to use discovery in order to establish jurisdictional facts. After petitioners repeatedly failed to comply with the court's orders for production of the requested information, the court warned them that unless they complied by a specified date, it would assume, pursuant to Rule 37(b)(2)(A), that it had personal jurisdiction. When petitioners again failed to comply, the court imposed the sanction, and the Court of Appeals affirmed, concluding that imposition of the sanction fell within the trial court's discretion under Rule 37(b)(2)(A) and that the sanction did not violate petitioners' due process rights. Held: 1. Rule 37(b)(2)(A) may be applied to support a finding of personal jurisdiction without violating due process. Unlike subject-matter jurisdiction, which is an Art. III as well as a statutory requirement, the requirement that a court have personal jurisdiction flows from the Due Process Clause and protects an individual liberty interest. Because it protects an individual interest, it may be intentionally waived, or for various reasons a defendant may be estopped from raising the issue. Due process is violated by a rule establishing legal consequences of a failure to produce evidence only if the defendant's behavior will not support the presumption that "the refusal to produce evidence material to the administration of due process was but an admission of the want of merit in the asserted defense." Hammond Packing Co. v. Arkansas, 212 U.S. 322, 351, 29 S.Ct. 370, 380, 53 L.Ed. 530. A proper application of Rule 37(b)(2)(A) will, as a matter of law, support such a presumption. Pp. 701-707. 2. The District Court did not abuse its discretion in applying Rule 37(b)(2)(A) in this case. The record establishes that imposition of the sanction here satisfied the Rule's requirements that the sanction be both "just" and specifically related to the particular "claim" that was at issue in the discovery order. Pp. 707-709. 651 F.2d 877, affirmed. Edmund K. Trent, Pittsburgh, Pa., for petitioners. Cloyd R. Mellott, Pittsburgh, Pa., for respondent. Justice WHITE delivered the opinion of the Court. 1 Rule 37(b), Federal Rules of Civil Procedure, provides that a district court may impose sanctions for failure to comply with discovery orders. Included among the available sanctions is: 2 "An order that the matters regarding which the order was made or any other designated facts shall be taken to be established for the purposes of the action in accordance with the claim of the party obtaining the order." Rule 37(b)(2)(A). 3 The question presented by this case is whether this Rule is applicable to facts that form the basis for personal jurisdiction over a defendant. May a district court, as a sanction for failure to comply with a discovery order directed at establishing jurisdictional facts, proceed on the basis that personal jurisdiction over the recalcitrant party has been established? Petitioners urge that such an application of the Rule would violate due process: If a court does not have jurisdiction over a party, then it may not create that jurisdiction by judicial fiat.1 They contend also that until a court has jurisdiction over a party, that party need not comply with orders of the court; failure to comply, therefore, cannot provide the ground for a sanction. In our view, petitioners are attempting to create a logical conundrum out of a fairly straightforward matter. 4 * Respondent Compagnie des Bauxites de Guinee (CBG) is a Delaware corporation, 49% of which is owned by the Republic of Guinea and 51% is owned by Halco (Mining) Inc. CBG's principal place of business is in the Republic of Guinea, where it operates bauxite mines and processing facilities. Halco, which operates in Pennsylvania, has contracted to perform certain administrative services for CBG. These include the procurement of insurance. 5 In 1973, Halco instructed an insurance broker, Marsh & McLennan, to obtain $20 million worth of business interruption insurance to cover CBG's operations in Guinea. The first half of this coverage was provided by the Insurance Company of North America (INA). The second half, or what is referred to as the "excess" insurance, was provided by a group of 21 foreign insurance companies,2 14 of which are petitioners in this action (the excess insurers).3 6 Marsh & McLennan requested Bland Payne to obtain the excess insurance in the London insurance market. Pursuant to normal business practice 7 "[i]n late January and in February, 1974, Bland Payne presented to the excess insurer [petitioners] a placing slip in the amount of $10,000,000, in excess of the first $10,000,000. [Petitioners] initialed said placing slip, effective February 12, 1974, indicating the part of said $10,000,000 each was willing to insure."4 Finding 27 of the District Court, 2 App. 347a. 8 Once the offering was fully subscribed, Bland Payne issued a cover note indicating the amount of the coverage and specifying the percentage of the coverage that each excess insurer had agreed to insure. No separate policy was issued; the excess insurers adopted the INA policy "as far as applicable." 9 Sometime after February 12, CBG allegedly experienced mechanical problems in its Guinea operation, resulting in a business interruption loss in excess of $10 million. Contending that the loss was covered under its policies, CBG brought suit when the insurers refused to indemnify CBG for the loss. Whatever the mechanical problems experienced by CBG, they were perhaps minor compared to the legal difficulties encountered in the courts. 10 In December 1975, CBG filed a two-count suit in the Western District of Pennsylvania, asserting jurisdiction based on diversity of citizenship. The first count was against INA; the second against the excess insurers. INA did not challenge personal or subject-matter jurisdiction of the District Court. The answer of the excess insurers, however, raised a number of defenses, including lack of in personam jurisdiction. Subsequently, this alleged lack of personal jurisdiction became the basis of a motion for summary judgment filed by the excess insurers.5 The issue in this case requires an account of respondent's attempt to use discovery in order to demonstrate the court's personal jurisdiction over the excess insurers. 11 Respondent's first discovery request—asking for "[c]opies of all business interruption insurance policies issued by Defendant during the period from January 1, 1972 to December 31, 1975"—was served on each defendant in August 1976. In January 1977, the excess insurers objected, on grounds of burdensomeness, to producing such policies. Several months later, respondent filed a motion to compel petitioners to produce the requested documents. In June 1978, the court orally overruled petitioners' objections. This was followed by a second discovery request in which respondent narrowed the files it was seeking to policies which "were delivered in. . . Pennsylvania . . . or covered a risk located in . . . Pennsylvania." Petitioners now objected that these documents were not in their custody or control; rather, they were kept by the brokers in London. The court ordered petitioners to request the information from the brokers, limiting the request to policies covering the period from 1971 to date. That was in July 1978; petitioners were given 90 days to produce the information. On November 8, petitioners were given an additional 30 days to complete discovery. On November 24, petitioners filed an affidavit offering to make their records, allegedly some 4 million files, available at their offices in London for inspection by respondent. Respondent countered with a motion to compel production of the previously requested documents. On December 21, 1978, the court, noting that no conscientious effort had yet been made to produce the requested information and that no objection had been entered to the discovery order in July, gave petitioners 60 more days to produce the requested information. The District Judge also issued the following warning: 12 "[I]f you don't get it to him in 60 days, I am going to enter an order saying that because you failed to give the information as requested, that I am going to assume, under Rule of Civil Procedure 37(b), subsection 2(A), that there is jurisdiction." 1 App. 115a. 13 A few moments later he restated the warning as follows: "I will assume that jurisdiction is here with this court unless you produce statistics and other information in that regard that would indicate otherwise." Id., at 116a. 14 On April 19, 1979, the court, after concluding that the requested material had not been produced, imposed the threatened sanction, finding that "for the purpose of this litigation the Excess Insurers are subject to the in personam jurisdiction of this Court due to their business contacts with Pennsylvania." Id., at 201a. Independently of the sanction, the District Court found two other grounds for holding that it had personal jurisdiction over petitioners. First, on the record established, it found that petitioners had sufficient business contacts with Pennsylvania to fall within the Pennsylvania long-arm statute. Second, in adopting the terms of the INA contract with CBG—a Pennsylvania insurance contract—the excess insurers implicitly agreed to submit to the jurisdiction of the court.6 15 Except with respect to three excess insurers, the Court of Appeals for the Third Circuit affirmed the jurisdictional holding, relying entirely upon the validity of the sanction.7 Compagnie des Bauxites de Guinea v. Insurance Co. of North America, 651 F.2d 877 (1981). That court specifically found that the discovery orders of the District Court did not constitute an abuse of discretion and that imposition of the sanction fell within the limits of trial court discretion under Rule 37(b): 16 "The purpose and scope of the ordered discovery were directly related to the issue of jurisdiction and the rule 37 sanction was tailored to establish as admitted those jurisdictional facts that, because of the insurers' failure to comply with discovery orders, CBG was unable to adduce through discovery." 651 F.2d, at 885. 17 Furthermore, it held that the sanction did not violate petitioners' due process rights, because it was no broader than "reasonably necessary" under the circumstances. 18 Because the decision below directly conflicts with the decision of the Court of Appeals for the Fifth Circuit in Familia de Boom v. Arosa Mercantil, S.A., 629 F.2d 1134 (1980), we granted certiorari.8 454 U.S. 963, 102 S.Ct. 502, 70 L.Ed.2d 377 (1981). II 19 In McDonald v. Mabee, 243 U.S. 90, 37 S.Ct. 343, 61 L.Ed. 608 (1917), another case involving an alleged lack of personal jurisdiction, Justice Holmes wrote for the Court, "great caution should be used not to let fiction deny the fair play that can be secured only by a pretty close adhesion to fact." Id., at 91, 37 S.Ct., at 343-344. Petitioners' basic submission is that to apply Rule 37(b)(2) to jurisdictional facts is to allow fiction to get the better of fact and that it is impermissible to use a fiction to establish judicial power, where, as a matter of fact, it does not exist. In our view, this represents a fundamental misunderstanding of the nature of personal jurisdiction. 20 The validity of an order of a federal court depends upon that court's having jurisdiction over both the subject matter and the parties. Stoll v. Gottlieb, 305 U.S. 165, 171-172, 59 S.Ct. 134, 137-138, 83 L.Ed. 104 (1938); Thompson v. Whitman, 18 Wall. 457, 465, 21 L.Ed. 897 (1874). The concepts of subject-matter and personal jurisdiction, however, serve different purposes, and these different purposes affect the legal character of the two requirements. Petitioners fail to recognize the distinction between the two concepts—speaking instead in general terms of "jurisdiction"—although their argument's strength comes from conceiving of jurisdiction only as subject-matter jurisdiction. 21 Federal courts are courts of limited jurisdiction. The character of the controversies over which federal judicial authority may extend are delineated in Art. III, § 2, cl. 1. Jurisdiction of the lower federal courts is further limited to those subjects encompassed within a statutory grant of jurisdiction. Again, this reflects the constitutional source of federal judicial power: Apart from this Court, that power only exists "in such inferior Courts as the Congress may from time to time ordain and establish." Art. III, § 1. 22 Subject-matter jurisdiction, then, is an Art. III as well as a statutory requirement; it functions as a restriction on federal power, and contributes to the characterization of the federal sovereign. Certain legal consequences directly follow from this. For example, no action of the parties can confer subject-matter jurisdiction upon a federal court. Thus, the consent of the parties is irrelevant, California v. LaRue, 409 U.S. 109, 93 S.Ct. 390, 34 L.Ed.2d 342 (1972), principles of estoppel do not apply, American Fire & Casualty Co. v. Finn, 341 U.S. 6, 17-18, 71 S.Ct. 534, 541-542, 95 L.Ed. 702 (1951), and a party does not waive the requirement by failing to challenge jurisdiction early in the proceedings. Similarly, a court, including an appellate court, will raise lack of subject-matter jurisdiction on its own motion. "[T]he rule, springing from the nature and limits of the judicial power of the United States is inflexible and without exception, which requires this court, of its own motion, to deny its jurisdiction, and, in the exercise of its appellate power, that of all other courts of the United States, in all cases where such jurisdiction does not affirmatively appear in the record." Mansfield, C. & L. M. R. Co. v. Swan, 111 U.S. 379, 382, 4 S.Ct. 510, 511, 28 L.Ed. 462 (1884).9 23 None of this is true with respect to personal jurisdiction. The requirement that a court have personal jurisdiction flows not from Art. III, but from the Due Process Clause. The personal jurisdiction requirement recognizes and protects an individual liberty interest. It represents a restriction on judicial power not as a matter of sovereignty, but as a matter of individual liberty.10 Thus, the test for personal jurisdiction requires that "the maintenance of the suit . . . not offend 'traditional notions of fair play and substantial justice.' " International Shoe Co. v. Washington, 326 U.S. 310, 316, 66 S.Ct. 154, 158, 90 L.Ed. 95 (1945), quoting Milliken v. Meyer, 311 U.S. 457, 463, 61 S.Ct. 339, 342, 85 L.Ed. 278 (1940). 24 Because the requirement of personal jurisdiction represents first of all an individual right, it can, like other such rights, be waived. In McDonald v. Mabee, supra, the Court indicated that regardless of the power of the State to serve process, an individual may submit to the jurisdiction of the court by appearance. A variety of legal arrangements have been taken to represent express or implied consent to the personal jurisdiction of the court. In National Equipment Rental, Ltd. v. Szukhent, 375 U.S. 311, 316, 84 S.Ct. 411, 414, 11 L.Ed.2d 354 (1964), we stated that "parties to a contract may agree in advance to submit to the jurisdiction of a given court," and in Petrowski v. Hawkeye-Security Co., 350 U.S. 495, 76 S.Ct. 490, 100 L.Ed. 639 (1956), the Court upheld the personal jurisdiction of a District Court on the basis of a stipulation entered into by the defendant. In addition, lower federal courts have found such consent implicit in agreements to arbitrate. See Victory Transport Inc. v. Comisaria General de Abastecimientos y Transportes, 336 F.2d 354 (CA2 1964); 2 J. Moore & J. Lucas, Moore's Federal Practice ¶ 4.02[3], n. 22 (1982) and cases listed there. Furthermore, the Court has upheld state procedures which find constructive consent to the personal jurisdiction of the state court in the voluntary use of certain state procedures. See Adam v. Saenger, 303 U.S. 59, 67-68, 58 S.Ct. 454, 458, 82 L.Ed. 649 (1938) ("There is nothing in the Fourteenth Amendment to prevent a state from adopting a procedure by which a judgment in personam may be rendered in a cross-action against a plaintiff in its courts . . . It is the price which the state may exact as the condition of opening its courts to the plaintiff"); Chicago Life Ins. Co. v. Cherry, 244 U.S. 25, 29-30, 37 S.Ct. 492, 493, 61 L.Ed. 966 (1917) ("[W]hat acts of the defendant shall be deemed a submission to [a court's] power is a matter upon which States may differ"). Finally, unlike subject-matter jurisdiction, which even an appellate court may review sua sponte, under Rule 12(h), Federal Rules of Civil Procedure, "[a] defense of lack of jurisdiction over the person . . . is waived" if not timely raised in the answer or a responsive pleading. 25 In sum, the requirement of personal jurisdiction may be intentionally waived, or for various reasons a defendant may be estopped from raising the issue. These characteristics portray it for what it is—a legal right protecting the individual. The plaintiff's demonstration of certain historical facts may make clear to the court that it has personal jurisdiction over the defendant as a matter of law—i.e., certain factual showings will have legal consequences—but this is not the only way in which the personal jurisdiction of the court may arise. The actions of the defendant may amount to a legal submission to the jurisdiction of the court, whether voluntary or not. 26 The expression of legal rights is often subject to certain procedural rules: The failure to follow those rules may well result in a curtailment of the rights. Thus, the failure to enter a timely objection to personal jurisdiction constitutes, under Rule 12(h)(1), a waiver of the objection. A sanction under Rule 37(b)(2)(A) consisting of a finding of personal jurisdiction has precisely the same effect. As a general proposition, the Rule 37 sanction applied to a finding of personal jurisdiction creates no more of a due process problem than the Rule 12 waiver. Although "a court cannot conclude all persons interested by its mere assertion of its own power," Chicago Life Ins. Co. v. Cherry, supra, at 29, 37 S.Ct., at 493, not all rules that establish legal consequences to a party's own behavior are "mere assertions" of power. 27 Rule 37(b)(2)(A) itself embodies the standard established in Hammond Packing Co. v. Arkansas, 212 U.S. 322, 29 S.Ct. 370, 53 L.Ed. 530 (1909), for the due process limits on such rules.11 There the Court held that it did not violate due process for a state court to strike the answer and render a default judgment against a defendant who failed to comply with a pretrial discovery order. Such a rule was permissible as an expression of "the undoubted right of the lawmaking power to create a presumption of fact as to the bad faith and untruth of an answer begotten from the suppression or failure to produce the proof ordered . . . [T]he preservation of due process was secured by the presumption that the refusal to produce evidence material to the administration of due process was but an admission of the want of merit in the asserted defense." Id., at 350-351, 29 S.Ct., at 380. 28 The situation in Hammond was specifically distinguished from that in Hovey v. Elliott, 167 U.S. 409, 17 S.Ct. 841, 42 L.Ed. 215 (1897), in which the Court held that it did violate due process for a court to take similar action as "punishment" for failure to obey an order to pay into the registry of the court a certain sum of money. Due process is violated only if the behavior of the defendant will not support the Hammond Packing presumption. A proper application of Rule 37(b)(2) will, as a matter of law, support such a presumption. See Societe Internationale v. Rogers, 357 U.S. 197, 209-213, 78 S.Ct. 1087, 1094-1096, 2 L.Ed.2d 1255 (1958). If there is no abuse of discretion in the application of the Rule 37 sanction, as we find to be the case here (see Part III), then the sanction is nothing more than the invocation of a legal presumption, or what is the same thing, the finding of a constructive waiver. 29 Petitioners argue that a sanction consisting of a finding of personal jurisdiction differs from all other instances in which a sanction is imposed, including the default judgment in Hammond Packing, because a party need not obey the orders of a court until it is established that the court has personal jurisdiction over that party. If there is no obligation to obey a judicial order, a sanction cannot be applied for the failure to comply. Until the court has established personal jurisdiction, moreover, any assertion of judicial power over the party violates due process. 30 This argument again assumes that there is something unique about the requirement of personal jurisdiction, which prevents it from being established or waived like other rights. A defendant is always free to ignore the judicial proceedings, risk a default judgment, and then challenge that judgment on jurisdictional grounds in a collateral proceeding. See Baldwin v. Traveling Men's Assn., 283 U.S. 522, 525, 51 S.Ct. 517, 75 L.Ed. 1244 (1931). By submitting to the jurisdiction of the court for the limited purpose of challenging jurisdiction, the defendant agrees to abide by that court's determination on the issue of jurisdiction: That decision will be res judicata on that issue in any further proceedings. Id., at 524, 51 S.Ct., at 517; American Surety Co. v. Baldwin, 287 U.S. 156, 166, 53 S.Ct. 98, 101, 77 L.Ed. 234 (1932). As demonstrated above, the manner in which the court determines whether it has personal jurisdiction may include a variety of legal rules and presumptions, as well as straightforward factfinding. A particular rule may offend the due process standard of Hammond Packing, but the mere use of procedural rules does not in itself violate the defendant's due process rights. III 31 Even if Rule 37(b)(2) may be applied to support a finding of personal jurisdiction, the question remains as to whether it was properly applied under the circumstances of this case. Because the District Court's decision to invoke the sanction was accompanied by a detailed explanation of the reasons for that order and because that decision was upheld as a proper exercise of the District Court's discretion by the Court of Appeals, this issue need not detain us for long. What was said in National Hockey League v. Metropolitan Hockey Club, Inc., 427 U.S. 639, 642, 96 S.Ct. 2778, 2780, 49 L.Ed.2d 747 (1976), is fully applicable here: "The question, of course, is not whether this Court, or whether the Court of Appeals, would as an original matter have [applied the sanction]; it is whether the District Court abused its discretion in so doing" (citations omitted). For the reasons that follow, we hold that it did not. 32 Rule 37(b)(2) contains two standards—one general and one specific—that limit a district court's discretion. First, any sanction must be "just"; second, the sanction must be specifically related to the particular "claim" which was at issue in the order to provide discovery. While the latter requirement reflects the rule of Hammond Packing, supra, the former represents the general due process restrictions on the court's discretion. 33 In holding that the sanction in this case was "just," we rely specifically on the following. First, the initial discovery request was made in July 1977. Despite repeated orders from the court to provide the requested material, on December 21, 1978, the District Court was able to state that the petitioners "haven't even made any effort to get this information up to this point." 1 App. 112a. The court then warned petitioners of a possible sanction. Confronted with continued delay and an obvious disregard of its orders, the trial court's invoking of its powers under Rule 37 was clearly appropriate. Second, petitioners repeatedly agreed to comply with the discovery orders within specified time periods. In each instance, petitioners failed to comply with their agreements. Third, respondent's allegation that the court had personal jurisdiction over petitioners was not a frivolous claim, and its attempt to use discovery to substantiate this claim was not, therefore, itself a misuse of judicial process. The substantiality of the jurisdictional allegation is demonstrated by the fact that the District Court found, as an alternative ground for its jurisdiction, that petitioners had sufficient contacts with Pennsylvania to fall within the State's long-arm statute. Supra, at 699. Fourth, petitioners had ample warning that a continued failure to comply with the discovery orders would lead to the imposition of this sanction. Furthermore, the proposed sanction made it clear that, even if there was not compliance with the discovery order, this sanction would not be applied if petitioners were to "produce statistics and other information" that would indicate an absence of personal jurisdiction. 1 App. 116a. In effect, the District Court simply placed the burden of proof upon petitioners on the issue of personal jurisdiction.12 Petitioners failed to comply with the discovery order; they also failed to make any attempt to meet this burden of proof. This course of behavior, coupled with the ample warnings, demonstrates the "justice" of the trial court's order. 34 Neither can there be any doubt that this sanction satisfies the second requirement. CBG was seeking through discovery to respond to petitioners' contention that the District Court did not have personal jurisdiction. Having put the issue in question, petitioners did not have the option of blocking the reasonable attempt of CBG to meet its burden of proof. It surely did not have this option once the court had overruled petitioners' objections. Because of petitioners' failure to comply with the discovery orders, CBG was unable to establish the full extent of the contacts between petitioners and Pennsylvania, the critical issue in proving personal jurisdiction. Petitioners' failure to supply the requested information as to its contacts with Pennsylvania supports "the presumption that the refusal to produce evidence . . . was but an admission of the want of merit in the asserted defense." Hammond Packing, 212 U.S., at 351, 29 S.Ct., at 380. The sanction took as established the facts—contacts with Pennsylvania—that CBG was seeking to establish through discovery. That a particular legal consequence—personal jurisdiction of the court over the defendants—follows from this, does not in any way affect the appropriateness of the sanction. IV 35 Because the application of a legal presumption to the issue of personal jurisdiction does not in itself violate the Due Process Clause and because there was no abuse of the discretion granted a district court under Rule 37(b)(2), we affirm the judgment of the Court of Appeals. 36 So ordered. 37 Justice POWELL, concurring in the judgment. 38 The Court rests today's decision on a constitutional distinction between "subject matter" and "in personam" jurisdiction. Under this distinction, subject-matter jurisdiction defines an Art. III limitation on the power of federal courts. By contrast, the Court characterizes the limits on in personam jurisdiction solely in terms of waivable personal rights and notions of "fair play." Having done so, it determines that fundamental questions of judicial power do not arise in this case concerning the personal jurisdiction of a federal district court. 39 In my view the Court's broadly theoretical decision misapprehends the issues actually presented for decision. Federal courts are courts of limited jurisdiction. Their personal jurisdiction, no less than their subject-matter jurisdiction, is subject both to constitutional and to statutory definition. When the applicable limitations on federal jurisdiction are identified, it becomes apparent that the Court's theory could require a sweeping but largely unexplicated revision of jurisdictional doctrine. This revision could encompass not only the personal jurisdiction of federal courts but "sovereign" limitations on state jurisdiction as identified in World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 291-293, 100 S.Ct. 559, 564-565, 62 L.Ed.2d 490 (1980). Fair resolution of this case does not require the Court's broad holding. Accordingly, although I concur in the Court's judgment, I cannot join its opinion. 40 * This lawsuit began when the respondent Compagnie des Bauxites brought a contract action against the petitioner insurance companies in the United States District Court for the Western District of Pennsylvania. Alleging diversity jurisdiction, respondent averred that the District Court had personal jurisdiction of the petitioners, all foreign corporations, under the long-arm statute of the State of Pennsylvania. See Compagnie des Bauxites de Guinea v. Insurance Co. of North America, 651 F.2d 877, 880-881 (CA3 1981). Petitioners, however, denied that they were subject to the court's personal jurisdiction under that or any other statute. Viewing the question largely as one of fact, the court ordered discovery to resolve the dispute. 41 Meantime, while respondent unsuccessfully sought compliance with its discovery requests, petitioners brought a parallel action in England's High Court of Justice, Queens Bench Division. It was at this juncture that the current issues arose. Seeking to enjoin the English proceedings, respondent sought an injunction in the District Court. Petitioners protested that they were not subject to that court's personal jurisdiction and thus that they lay beyond its injunctive powers. But the District Court disagreed. As a jurisdictional prerequisite to its entry of the injunction, the court upheld its personal jurisdiction over petitioners.1 It characterized its finding of jurisdiction partly as a sanction for petitioners' noncompliance with its discovery orders under Federal Rule of Civil Procedure 37(b).2 42 Rule 37(b) is not, however, a jurisdictional provision. As recognized by the Court of Appeals, the governing jurisdictional statute remains the long-arm statute of the State of Pennsylvania. See 651 F.2d, at 881. In my view the Court fails to make clear the implications of this central fact: that the District Court in this case relied on state law to obtain personal jurisdiction. 43 As courts of limited jurisdiction, the federal district courts possess no warrant to create jurisdictional law of their own. Under the Rules of Decision Act, 28 U.S.C. § 1652, they must apply state law "except where the Constitution or treaties of the United States or Acts of Congress otherwise require or provide. . . ." See generally Erie R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). Thus, in the absence of a federal rule or statute establishing a federal basis for the assertion of personal jurisdiction, the personal jurisdiction of the district courts is determined in diversity cases by the law of the forum State. See, e.g., Intermeat, Inc. v. American Poultry Co., 575 F.2d 1017 (CA2 1978); Wilkerson v. Fortuna Corp., 554 F.2d 745 (CA5), cert. denied, 434 U.S. 939, 98 S.Ct. 430, 54 L.Ed.2d 299 (1977); Poyner v. Erma Werke Gmbh, 618 F.2d 1186, 1187 (CA6 1980); Lakeside Bridge & Steel Co. v. Mountain State Constr. Co., 597 F.2d 596 (CA7 1979), cert. denied, 445 U.S. 907, 100 S.Ct. 1087, 63 L.Ed.2d 325 (1980); Lakota Girl Scout Council, Inc. v. Havey Fundraising Management, Inc., 519 F.2d 634 (CA8 1975); Arrowsmith v. United Press International, 320 F.2d 219, 226 (CA2 1963); Forsythe v. Overmyer, 576 F.2d 779, 782 (CA9), cert. denied, 439 U.S. 864, 99 S.Ct. 188, 58 L.Ed.2d 174 (1978); Quarles v. Fuqua Industries, Inc., 504 F.2d 1358 (CA10 1974).3 44 As a result of the District Court's dependence on the law of Pennsylvania to establish personal jurisdiction—a dependence mandated by Congress under 28 U.S.C. § 1652—its jurisdiction in this case normally would be subject to the same due process limitations as a state court. See, e.g., Forsythe v. Overmyer, supra, at 782; Washington v. Norton Mfg., Inc., 588 F.2d 441, 445 (CA5 1979); Fisons Ltd. v. United States, 458 F.2d 1241, 1250 (CA7 1972).4 Thus, the question arises how today's decision is related to cases restricting the personal jurisdiction of the States. 45 Before today our decisions had established that "minimum contacts" represented a constitutional prerequisite to the exercise of in personam jurisdiction over an unconsenting defendant. See, e.g., World-Wide Volkswagen Corp. v. Wood son, 444 U.S., at 291-293, 100 S.Ct., at 564-565; Hanson v. Denckla, 357 U.S. 235, 251, 78 S.Ct. 1228, 1238, 2 L.Ed.2d 1283 (1958); International Shoe Co. v. Washington, 326 U.S. 310, 316, 66 S.Ct. 154, 158, 90 L.Ed. 95 (1945). In the absence of a showing of minimum contacts, a finding of personal jurisdiction over an unconsenting defendant, even as a sanction, therefore would appear to transgress previously established constitutional limitations. The cases cannot be reconciled by a simple distinction between the constitutional limits on state and federal courts. Because of the District Court's reliance on the Pennsylvania long-arm statute—the applicable jurisdictional provision under the Rules of Decisions Act—the relevant constitutional limits would not be those imposed directly on federal courts by the Due Process Clause of the Fifth Amendment, but those applicable to state jurisdictional law under the Fourteenth. 46 The Court's decision apparently must be understood as related to our state jurisdictional cases in one of two ways. Both involve legal theories that fail to justify the doctrine adopted by the Court in this case. A. 47 Under traditional principles, the due process question in this case is whether "minimum contacts" exist between petitioners and the forum State that would justify the State in exercising personal jurisdiction. See, e.g., World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S., at 291-293, 100 S.Ct., at 564-565; Shaffer v. Heitner, 433 U.S. 186, 216, 97 S.Ct. 2569, 2586, 53 L.Ed.2d 683 (1977); Hanson v. Denckla, supra, 357 U.S., at 251, 78 S.Ct., at 1238. By finding that the establishment of minimum contacts is not a prerequisite to the exercise of jurisdiction to impose sanctions under Federal Rule of Civil Procedure 37, the Court may be understood as finding that "minimum contacts" no longer are a constitutional requirement for the exercise by a state court of personal jurisdiction over an unconsenting defendant.5 Whenever the Court's notions of fairness are not offended, jurisdiction apparently may be upheld. 48 Before today, of course, our cases had linked minimum contacts and fair play as jointly defining the "sovereign" limits on state assertions of personal jurisdiction over unconsenting defendants. See World-Wide Volkswagen Corp. v. Woodson, supra, 444 U.S., at 292-293, 100 S.Ct., at 564-565; see Hanson v. Denckla, supra, 357 U.S., at 251, 78 S.Ct., at 1238. The Court appears to abandon the rationale of these cases in a footnote. See ante, at 702-703, n. 10. But it does not address the implications of its action. By eschewing reliance on the concept of minimum contacts as a "sovereign" limitation on the power of States—for, again, it is the State's long-arm statute that is invoked to obtain personal jurisdiction in the District Court—the Court today effects a potentially substantial change of law. For the first time it defines personal jurisdiction solely by reference to abstract notions of fair play. And, astonishingly to me, it does so in a case in which this rationale for decision was neither argued nor briefed by the parties. B 49 Alternatively, it is possible to read the Court opinion, not as affecting state jurisdiction, but simply as asserting that Rule 37 of the Federal Rules of Civil Procedure represents a congressionally approved basis for the exercise of personal jurisdiction by a federal district court. On this view Rule 37 vests the federal district courts with authority to take jurisdiction over persons not in compliance with discovery orders. This of course would be a more limited holding. Yet the Court does not cast its decision in these terms. And it provides no support for such an interpretation, either in the language or in the history of the Federal Rules. 50 In the absence of such support, I could not join the Court in embracing such a construction of the Rules of Civil Procedure.6 There is nothing in Rule 37 to suggest that it is intended to confer a grant of personal jurisdiction. Indeed, the clear language of Rule 82 seems to establish that Rule 37 should not be construed as a jurisdictional grant: "These rules shall not be construed to extend . . . the jurisdiction of the United States district courts or the venue of actions therein." Moreover, assuming that minimum contacts remain a constitutional predicate for the exercise of a State's in personam jurisdiction over an unconsenting defendant, constitutional questions would arise if Rule 37 were read to permit a plaintiff in a diversity action to subject a defendant to a "fishing expedition" in a foreign jurisdiction. A plaintiff is not entitled to discovery to establish essentially speculative allegations necessary to personal jurisdiction. Nor would the use of Rule 37 sanctions to enforce discovery orders constitute a mere abuse of discretion in such a case.7 For me at least, such a use of discovery would raise serious questions as to the constitutional as well as the statutory authority of a federal court—in a diversity case—to exercise personal jurisdiction absent some showing of minimum contacts between the unconsenting defendant and the forum State. II 51 In this case the facts alone—unaided by broad jurisdictional theories—more than amply demonstrate that the District Court possessed personal jurisdiction to impose sanctions under Rule 37 and otherwise to adjudicate this case. I would decide the case on this narrow basis. 52 As recognized both by the District Court and the Court of Appeals, the respondent adduced substantial support for its jurisdictional assertions. By affidavit and other evidence, it made a prima facie showing of "minimum contacts." See 651 F.2d, at 881-882, 886, and n. 9. In the view of the District Court, the evidence adduced actually was sufficient to sustain a finding of personal jurisdiction independently of the Rule 37 sanction. App. to Pet. for Cert. 51a, 53a.8 53 Where the plaintiff has made a prima facie showing of minimum contacts, I have little difficulty in holding that its showing was sufficient to warrant the District Court's entry of discovery orders. And where a defendant then fails to comply with those orders, I agree that the prima facie showing may be held adequate to sustain the court's finding that minimum contacts exist, either under Rule 37 or under a theory of "presumption" or "waiver." 54 Finding that the decision of the Court of Appeals should be affirmed on this ground, I concur in the judgment of the Court. 1 The petition with which we deal in this case was filed as a cross-petition in response to the petition for certiorari filed in No. 81-290, Compagnie des Bauxites de Guinee v. Insurance Corp. of Ireland, Ltd. We granted the cross-petition, limiting the grant to the question of the validity of the Rule 37(b)(2) sanction. 454 U.S. 963, 102 S.Ct. 502, 71 L.Ed.2d 377 (1981). We shall refer to the cross-petitioners as "petitioners" and to the cross-respondent as "respondent." 2 The District Court described these excess insurers as follows: "Of the 21 Excess Insurers, five are English companies representing English domestic interests but insuring risks throughout the world, particularly in Pennsylvania. Seven are English companies which represent non English parents, or affiliates. The United States, Japan and Israel are the nationalities of two each of the Excess Insurer Defendants. Switzerland and the Republic of Ireland are the nationalities of one each of the Excess Insurer Defendants. The remaining Excess Insurer Defendant is a Belgium Company which represents the United States parent." 1 App. 196a. 3 Four of the excess insurers did not contest personal jurisdiction in the District Court. Id., at 105a. The Court of Appeals directed the dismissal of the complaint with respect to three others. Compagnie des Bauxites de Guinea v. Insurance Co. of North America, 651 F.2d 877, 886 (1981). CBG challenges the latter action in its petition for certiorari in No. 81-290. 4 One of the excess insurers, L'Union Atlantique S.A. d'Assurances, does business in Brussels, and was sent a separate placing slip. 5 The motion for summary judgment was filed on May 20, 1977. In it, 17 of the excess insurers alleged a lack of in personam jurisdiction and all 21 excess insurers sought dismissal on the ground of forum non conveniens. The District Court denied the motion on April 19, 1979. 6 On March 22, 1979, the excess insurers instituted a suit against CBG in England, attacking the validity of the insurance contract. In its April 19 decision, the District Court found that "the commencement of the separate action in England [was] oppressive, unfair, and an act of bad faith under all of the circumstances." 1 App. 203a. It, therefore, enjoined the continuation of that suit. This aspect of the District Court decision was reversed by the Court of Appeals. Respondent seeks certiorari review of that decision (see n. 1, supra ). 7 It reversed as to three of the excess insurers on the grounds that they had complied with the discovery orders and that their contacts with Pennsylvania were not sufficient to justify exercise of the Pennsylvania long-arm statute. It also held that the District Court had abused its discretion in enjoining the action in England. Judge Gibbons dissented on the propriety of the sanction, arguing that the District Court had abused its discretion. He also expressed some doubt that a Rule 37 sanction could ever be used as the source of personal jurisdiction. 651 F.2d, at 892, n. 4. 8 In Familia de Boom, the Fifth Circuit held that a sanction under Rule 37(b)(2) is valid only if the court has personal jurisdiction over the party that has refused compliance with a court order. Personal jurisdiction must, it held, appear from the record independently of the sanction. The Courts of Appeals for the Fourth and Eighth Circuits, on the other hand, have agreed with the Third Circuit on the appropriateness of a sanction on the issue of personal jurisdiction. Lekkas v. Liberian M/V Caledonia, 443 F.2d 10, 11 (CA4 1971); English v. 21st Phoenix Corp., 590 F.2d 723 (CA8 1979). 9 A party that has had an opportunity to litigate the question of subject-matter jurisdiction may not, however, reopen that question in a collateral attack upon an adverse judgment. It has long been the rule that principles of res judicata apply to jurisdictional determinations—both subject matter and personal. See Chicot County Drainage Dist. v. Baxter State Bank, 308 U.S. 371, 60 S.Ct. 317, 84 L.Ed. 329 (1940); Stoll v. Gottlieb, 305 U.S. 165, 59 S.Ct. 134, 83 L.Ed. 104 (1938). 10 It is true that we have stated that the requirement of personal jurisdiction, as applied to state courts, reflects an element of federalism and the character of state sovereignty vis-a-vis other States. For example, in World-Wide Volkswagen Corp. v. Woodson, 444 U.S. 286, 291-292, 100 S.Ct. 559, 564, 62 L.Ed.2d 490 (1980), we stated: "[A] state court may exercise personal jurisdiction over a nonresident defendant only so long as there exist 'minimum contacts' between the defendant and the forum State. The concept of minimum contacts, in turn, can be seen to perform two related, but distinguishable, functions. It protects the defendant against the burdens of litigating in a distant or inconvenient forum. And it acts to ensure that the States, through their courts, do not reach out beyond the limits imposed on them by their status as coequal sovereigns in a federal system." (Citation omitted.) Contrary to the suggestion of Justice POWELL, post, at 713-714, our holding today does not alter the requirement that there be "minimum contacts" between the nonresident defendant and the forum State. Rather, our holding deals with how the facts needed to show those "minimum contacts" can be established when a defendant fails to comply with court-ordered discovery. The restriction on state sovereign power described in World-Wide Volkswagen Corp., however, must be seen as ultimately a function of the individual liberty interest preserved by the Due Process Clause. That Clause is the only source of the personal jurisdiction requirement and the Clause itself makes no mention of federalism concerns. Furthermore, if the federalism concept operated as an independent restriction on the sovereign power of the court, it would not be possible to waive the personal jurisdiction requirement: Individual actions cannot change the powers of sovereignty, although the individual can subject himself to powers from which he may otherwise be protected. 11 The Advisory Committee Notes to the Rule specifically stated that "the provisions of the rule find support in [Hammond Packing Co. v. Arkansas, 212 U.S. 322 [29 S.Ct. 370, 53 L.Ed. 530] (1909) ]." Final Report of Advisory Committee on Rules for Civil Procedure 25 (1937). See also Societe Internationale v. Rogers, 357 U.S. 197, 209, 78 S.Ct. 1087, 1094, 2 L.Ed.2d 1255 (1958). 12 Counsel for petitioners agreed to this characterization of the sanction at oral argument. Tr. of Oral Arg. 47-48. 1 A district court must have personal jurisdiction over a party before it can enjoin its actions. Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 111-112, 89 S.Ct. 1562, 1570, 23 L.Ed.2d 129 (1969). 2 The court also found that petitioners in fact had undertaken sufficient business activity in the State to bring them within the reach of the Pennsylvania long-arm statute. See App. to Pet. for Cert. 51a, 53a. 3 As Judge Friendly explained in the leading case of Arrowsmith v. United Press International, 320 F.2d, at 226: "State statutes determining what foreign corporations may be sued, for what, and by whom, are not mere whimsy; like most legislation they represent a balancing of various considerations for example, affording a forum for wrongs connected with the state and conveniencing resident plaintiffs, while avoiding the discouragement of activity within the state by foreign corporations. We see nothing in the concept of diversity jurisdiction that should lead us to read into the governing statutes a Congressional mandate, unexpressed by Congress itself, to disregard the balance thus struck by the states." 4 It is not contended that there is any federal basis for the exercise of personal jurisdiction by the District Court. 5 The Court refers to the respondent's prima facie showing of "minimum contacts" only as one factor indicating that the District Court did not abuse its discretion in entering a finding of personal jurisdiction as a sanction under Rule 37(b). See ante, at 708. Generally it views the requirement of personal jurisdiction as a right that may be "established or waived like other rights." Ante, at 706. 6 Jurisdiction over the person generally is dealt with by Rule 4, governing the methods of service through which personal jurisdiction may be obtained. Although Rule 4 deals expressly only with service of process, not with the underlying jurisdictional prerequisites, jurisdiction may not be obtained unless process is served in compliance with applicable law. See, e.g., Intermeat, Inc. v. American Poultry Co., 575 F.2d 1017 (CA2 1978); Washington v. Norton Mfg., Inc., 588 F.2d 441, 445 (CA5 1979); D. Currie, Federal Courts 858 (2d ed. 1975). For this reason Rule 4 frequently has been characterized as a jurisdictional provision. See, e.g., 374 U.S. 869 (1963) (statement of Black and Douglas, JJ., dissenting from adoption of amendments to the Federal Rules of Civil Procedure); Currie, supra, at 858; Foster, Long-Arm Jurisdiction in Federal Courts, 1969 Wis.L.Rev. 9, 11. As applicable here, Rule 4 relies expressly on state law. See Fed.Rules Civ.Proc. 4(d)(7) and (e). 7 Compare the Court's view. Ante, at 707. 8 The Court of Appeals deemed it unnecessary to review this alternative basis for the District Court's finding of jurisdiction. See 651 F.2d, at 886, and n. 9.

34

456 U.S. 867 102 S.Ct. 2291 72 L.Ed.2d 625 CALIFORNIAv.NEVADA No. 73 Supreme Court of the United States June 1, 1982 1 On bill of complaint. 2 June 1, 1982. The Report of the Special Master is received and ordered filed. DECREE 3 For the purpose of giving effect to the opinion of this Court announced on June 10, 1980, 447 U.S. 125, 100 S.Ct. 2064, 65 L.Ed.2d 1, and the stipulation of the parties entered into on February 5, 1982, and approved by the Special Master: 4 IT IS ORDERED, ADJUDGED, AND DECREED AS FOLLOWS: 5 1. That the location of the boundary between the States of California and Nevada is as hereafter specified: 6 Beginning at the Initial Point on the California-Nevada boundary as established under authority of the Act of June 10, 1872, 17 Stat. 358, by Allexey W. Von Schmidt in 1872 as an eight (8) inch square wood post set in a large mound of stone at the intersection to the 42nd parallel of north latitude along the southern boundary of the State of Oregon and the 120th degree of longitude west from Greenwich, each as reported by him; thence running southerly along said 1872 boundary as surveyed and marked by Allexey W. Von Schmidt to a cast iron monument designated as Von Schmidt Milepost 191 on said boundary; thence S 00x09'15".21 W along a line directed toward a standard National Geodetic Survey brass marker cemented in a granite formation and stamped "VS 120, 1981", located at the geographic position of: Latitude 38x53'51".59866 Longitude 120x00'20".93116, a distance of 80,191.63 feet to an intersection in Lake Tahoe with a line extending between two points, the first of which is a standard National Geodetic Survey brass marker cemented in granite and stamped "EAGLE ROCK, 1981," which has a geographic position of: Latitude 39x06'33".17268, Longitude 120x09'38".06504, and the second being a granite stone with copper bolt identified as No. 1 (Initial 1894), being the first monumented point on the California-Nevada Oblique 1893-1899 Boundary survey as determined and marked by the U. S. Coast and Geodetic Survey; thence from said intersection point in Lake Tahoe S 48x46'02".80 E, 21,568.48 feet to said Monument No. 1 (Initial 1894); thence proceeding southeasterly along the oblique California-Nevada boundary line as surveyed and marked by the U. S. Coast and Geodetic Survey under authority of the Act of August 5, 1892, 27 Stat. 357, during the period 1893-1899, to Point No. 1 described in the Interstate Compact Defining the Boundary Between the States of Arizona and California executed on March 12, 1963, being a point common to Point No. 1 described in the Interstate Compact Defining A Portion of the Arizona-Nevada Boundary On The Colorado River, executed February 6, 1960. Bearings, distances, and geographic positions used hereinbefore are based upon the 1927 North American Datum and are denoted on the attached map, Exhibit A, which is incorporated herein by reference. 7 2. That the intersection of the two lines described herein shall form the only angle point of the true California-Nevada state boundary within the waters of Lake Tahoe, which angle point shall be considered as satisfying the definition of the intersection of the 120th meridian west of Greenwich with the 39th parallel of north latitude as prescribed in the constitutions of California and Nevada. 8 3. That the necessary expenses incurred by the Special Master incident to this litigation and all other proper expenses incurred jointly by the parties shall be equally borne by the parties. 9 4. That, except as provided in paragraph 3 herein, each party shall bear its own expenses. 10 5. That any unexpended funds contributed by the parties to the Special Master for necessary expenses be returned to the parties in proportion to their contributions. EXHIBIT An 11

1011

457 U.S. 15 102 S.Ct. 2202 72 L.Ed.2d 639 JACKSON TRANSIT AUTHORITY, et al., Petitionersv. LOCAL DIVISION 1285, AMALGAMATED TRANSIT UNION, AFL-CIO-CLC. No. 81-411. Argued April 21, 1982. Decided June 7, 1982. Syllabus Section 13(c) of the Urban Mass Transportation Act of 1964 requires a state or local government to make arrangements to preserve transit workers' existing collective-bargaining rights before that government may receive federal financial assistance for the acquisition of a privately owned transit company. Petitioner city entered into a "§ 13(c) agreement" with respondent transit union in order to obtain federal funds to acquire a failing private bus company and convert it into petitioner Jackson Transit Authority. Thereafter, the Authority's unionized workers were covered by a series of collective-bargaining agreements. In 1975, however, the Authority notified the union that it no longer considered itself bound by the newest of the collective-bargaining agreements. The union subsequently filed suit in Federal District Court, seeking damages and injunctive relief and alleging that petitioners had breached the § 13(c) and collective-bargaining agreements. The District Court held that it lacked subject-matter jurisdiction because the complaint rested on contract rights that should be enforced only in a state court. The Court of Appeals reversed, holding that there was subject-matter jurisdiction because the claim arose under a federal law, specifically § 13(c), and that § 13(c) implicitly provided a federal private right of action. Held : Section 13(c) does not provide the union with federal causes of action for alleged breaches of the § 13(c) and collective-bargaining agreements. While § 13(c)'s language supplies no definitive answer, the legislative history is conclusive that Congress intended that such agreements be governed by state law applied in state courts. Congress designed § 13(c) as a means to accommodate state law to collective bargaining, not as a means to substitute a federal law of collective bargaining for state labor law. Pp. 20-29. 650 F.2d 1379, reversed and remanded. Joseph S. Kaufman, Baltimore, Md., for petitioners. Linda R. Hirshman, Chicago, Ill., for respondent. Justice BLACKMUN delivered the opinion of the Court. 1 Under § 13(c) of the Urban Mass Transportation Act of 1964 (Act or UMTA), 78 Stat. 307, as amended, 49 U.S.C. § 1609(c),1 a state or local government must make arrangements to preserve transit workers' existing collective-bargaining rights before that government may receive federal financial assistance for the acquisition of a privately owned transit company. This case presents the issue whether § 13(c) by itself permits a union to sue in federal court for alleged violations of an arrangement of this kind or of the collective-bargaining agreement between the union and the local government transit authority. 2 * A. 3 When the Act was under consideration in the Congress, that body was aware of the increasingly precarious financial condition of a number of private transportation companies across the country, and it feared that communities might be left without adequate mass transportation. See S.Rep.No. 82, 88th Cong., 1st Sess., 4-5, 19-20 (1963). The Act was designed in part to provide federal aid for local governments in acquiring failing private transit companies so that communities could continue to receive the benefits of mass transportation despite the collapse of the private operations. See §§ 2(b) and 3, as amended, 49 U.S.C. §§ 1601(b) and 1602. 4 At the same time, however, Congress was aware that public ownership might threaten existing collective-bargaining rights of unionized transit workers employed by private companies. If, for example, state law forbade collective bargaining by state and local government employees, the workers might lose their collective-bargaining rights when a private company was acquired by a local government. See Urban Mass Transportation—1963, Hearings on S.6 and S.917 before a Subcommittee of the Senate Committee on Banking and Currency, 88th Cong., 1st Sess., 318-323 (1963) (Senate Hearings) (statement of Andrew J. Biemiller, Director, Department of Legislation, AFL-CIO). To prevent federal funds from being used to destroy the collective-bargaining rights of organized workers, Congress included § 13(c) in the Act. See H.R.Rep.No.204, 88th Cong., 1st Sess., 15-16 (1963). 5 Section 13(c) requires, as a condition of federal assistance under the Act, that the Secretary of Labor certify that "fair and equitable arrangements" have been made "to protect the interests of employees affected by [the] assistance." The statute lists several protective steps that must be taken before a local government may receive federal aid; among these are the preservation of benefits under existing collective-bargaining agreements and the continuation of collective-bargaining rights. The protective arrangements must be specified in the contract granting federal aid.2 B 6 In 1966, petitioner city of Jackson, Tenn., applied for federal aid to convert a failing private bus company into a public entity, petitioner Jackson Transit Authority. See App. 12a-16a. In order to satisfy § 13(c), the Authority so created entered into a "§ 13(c) agreement" with respondent Local Division 1285, Amalgamated Transit Union, AFL-CIO-CLC, the union that represented the private company's employees. See 29 CFR pt. 215 (1981). Among other things, the § 13(c) agreement guaranteed the preservation of the transit workers' collective-bargaining rights. App. 16a-20a. The Secretary of Labor certified that the agreement was "fair and equitable." Its substance was made a part of the grant contract between the city and the United States, and the city received approximately $279,000 in federal aid. 7 Thereafter, until 1975, the Authority's unionized workers were covered by a series of collective-bargaining agreements. Six months after a new 3-year collective-bargaining agreement was signed in 1975, see id., at 31a, however, the Authority notified the union that it no longer considered itself bound by that contract. See id., at 45a.3 8 Ultimately, the union filed suit in the United States District Court for the Western District of Tennessee. It sought damages and injunctive relief, alleging that petitioners had breached the § 13(c) agreement and the collective-bargaining contract. App. 8a, 10a-11a.4 The District Court concluded that it lacked subject-matter jurisdiction to hear the suit because the complaint rested on contract rights that should be enforced only in a state court. 447 F.Supp. 88 (1977). 9 The United States Court of Appeals for the Sixth Circuit reversed. 650 F.2d 1379 (1981). Relying on Bell v. Hood, 327 U.S. 678, 66 S.Ct. 773, 90 L.Ed. 939 (1946), that court first determined that it had subject-matter jurisdiction under 28 U.S.C. § 1331, because the union's claim arose under the laws of the United States, specifically § 13(c). The court then held that § 13(c) implicitly provides a federal private right of action. Section 13(c) reflects national labor policy, the Court of Appeals reasoned, and the rights protected by the statute are thus federal rights. The court concluded that it was consistent with the congressional intent behind § 13(c) to permit enforcement of these federal rights in federal court. 10 Because of the importance of the interpretation of § 13(c) for local transit labor relations,5 we granted certiorari. 454 U.S. 1079, 102 S.Ct. 632, 70 L.Ed.2d 613 (1981). II 11 While the Court of Appeals treated this as a private right of action case, see, e.g., Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 102 S.Ct. 1825, 72 L.Ed.2d 182 (1982), it does not fit comfortably in that mold. Indeed, since § 13(c) contemplates protective arrangements between grant recipients and unions as well as subsequent collective-bargaining agreements between those parties, see H.R.Rep.No.204, 88th Cong., 1st Sess., 16 (1963), it is reasonable to conclude that Congress expected the § 13(c) agreement and the collective-bargaining agreement, like ordinary contracts, to be enforceable by private suit upon a breach. See Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 18-19, 100 S.Ct. 242, 246-47, 62 L.Ed.2d 146 (1979). The gist of the union's position is not that § 13(c) creates an implied right of action to sue for violations of the statute. Instead, the union argues that "[i]t was the intent of Congress that federal law would determine the binding effect of labor protective agreements under § 13(c) and of the collective-bargaining agreements reached pursuant to § 13(c) between unions and recipients of UMTA funds" so that those agreements "are enforceable in the federal courts." Brief for Respondent 24. 12 The issue, then, is not whether Congress intended the union to be able to bring contract actions for breaches of the two contracts, but whether Congress intended such contract actions to set forth federal, rather than state, claims. Admittedly, since the private right of action decisions address the related question whether Congress intended that a particular party be able to bring suit under a federal statute, those decisions may provide assistance in resolving this case. But the precise question before us is whether the union's contract actions are federal causes of action, not whether the union can bring suit at all to enforce its contracts. See Local Div. 732, Amalgamated Transit Union v. Metropolitan Atlanta Rapid Transit Authority, 667 F.2d 1327, 1329-1334 (CA11 1982).6 13 As the union points out, on several occasions the Court has determined that a plaintiff stated a federal claim when he sued to vindicate contractual rights set forth by federal statutes, despite the fact that the relevant statutes lacked express provisions creating federal causes of action. In Machinists v. Central Airlines, Inc., 372 U.S. 682, 83 S.Ct. 956, 10 L.Ed.2d 67 (1963), the Court held that a union had a federal cause of action to enforce an award of an airline adjustment board included in a collective-bargaining contract pursuant to § 204 of the Railway Labor Act, 45 U.S.C. § 184 (1958 ed.). Similarly, in Norfolk & Western R. Co. v. Nemitz, 404 U.S. 37, 92 S.Ct. 185, 30 L.Ed.2d 198 (1971), the Court ruled that a railroad's employees made out federal claims when they sought to enforce assurances made by the railroad to secure the Interstate Commerce Commission's approval of a consolidation under a provision of the Interstate Commerce Act, 49 U.S.C. § 5(2)(f) (1970 ed.). And recently, in an analogous private right of action decision, the Court permitted a federal suit for rescission of a contract declared void by § 215 of the Investment Advisers Act of 1940, 15 U.S.C. § 80b-15 (1976), although the statute itself made no express provision for private suits. Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S., at 18-19, 100 S.Ct., at 246-47. See also Mills v. Electric Auto-Lite Co., 396 U.S. 375, 388, 90 S.Ct. 616, 623, 24 L.Ed.2d 593 (1970) (recognizing federal right to rescind contracts rendered void by § 29(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78cc(b)); American Surety Co. v. Shulz, 237 U.S. 159, 35 S.Ct. 525, 59 L.Ed. 892 (1915) (finding federal-question jurisdiction to hear suit on supersedeas bond required by Rev.Stat. § 1007). 14 These decisions demonstrate that suits to enforce contracts contemplated by federal statutes may set forth federal claims and that private parties in appropriate cases may sue in federal court to enforce contractual rights created by federal statutes. But they do not dictate the result in this case. Whenever we determine the scope of rights and remedies under a federal statute, the critical factor is the congressional intent behind the particular provision at issue. See, e.g., Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. at 18, 100 S.Ct., at 246; Cannon v. University of Chicago, 441 U.S. 677, 688, 99 S.Ct. 1946, 1952, 60 L.Ed.2d 560 (1979); Machinists v. Central Airlines, Inc., 372 U.S., at 685-692; see also n. 9, 83 S.Ct., at 960-962 also n. 9, infra. Thus, if Congress intended that § 13(c) agreements and collective-bargaining agreements be "creations of federal law," Machinists v. Central Airlines, Inc., 372 U.S., at 692, 83 S.Ct., at 962, and that the rights and duties contained in those contracts be federal in nature, see id., at 695, 83 S.Ct., at 963, then the union's suit states federal claims. Otherwise, the union's complaint presents only state-law claims. See Miree v. DeKalb County, 433 U.S. 25, 97 S.Ct. 2490, 53 L.Ed.2d 557 (1977). III 15 We begin with the language of the statute itself. See, e.g., Universities Research Assn., Inc. v. Coutu, 450 U.S. 754, 771, 101 S.Ct. 1451, 1462, 67 L.Ed.2d 662 (1981). The bare language of § 13(c) is not conclusive. In some ways, the statute seems to make § 13(c) agreements and collective-bargaining contracts creatures of federal law. Section 13(c) demands "fair and equitable arrangements" as prerequisites for federal aid; it requires the approval of the Secretary of Labor for those arrangements; it specifies five different varieties of protective provisions that must be included among the § 13(c) arrangements; and it expressly incorporates the protective arrangements into the grant contract between the recipient and the Federal Government.7 See n. 2, supra. On the other hand, labor relations between local governments and their employees are the subject of a longstanding statutory exemption from the National Labor Relations Act. 29 U.S.C. § 152(2). Section 13(c) evinces no congressional intent to upset the decision in the National Labor Relations Act to permit state law to govern the relationships between local governmental entities and the unions representing their employees. See Cort v. Ash, 422 U.S. 66, 78, 95 S.Ct. 2080, 2087, 45 L.Ed.2d 26 (1975) (noting reluctance to permit suit in federal court when "the cause of action [is] one traditionally relegated to state law"). 16 While the statutory language supplies no definitive answer, the legislative history is conclusive. A consistent theme runs throughout the consideration of § 13(c): Congress intended that labor relations between transit workers and local governments would be controlled by state law. 17 In 1963, Secretary of Labor Wirtz presented the original version of § 13(c) to the relevant House and Senate Committees. Before both Committees, Members of Congress expressed concern about the effect of the statute on state laws. And Secretary Wirtz explained to both Committees that, while attempts would be made to accommodate state law to the preservation of collective-bargaining rights, state law would control local transit labor relations. The Secretary told the House Committee that "this proposal is submitted on this basis, . . . that the State laws must control." Urban Mass Transportation Act of 1963, Hearings on H.R. 3881 before the House Committee on Banking and Currency, 88th Cong., 1st Sess., 482 (1963) (House Hearings). A Committee member raised the issue again; the Secretary repeated that "State laws would be controlling in the situation," though he suggested that there "would be few, if any, situations" where state law and § 13(c) could not be reconciled. House Hearings, at 486. When similar concerns were expressed during his testimony before the Senate Committee, the Secretary reiterated: "I should like it quite clear that I think that there could be no superseding here of the State law." Senate Hearings, at 313. 18 The House and Senate Reports took the Secretary at his word. The House Report advised that § 13(c) would ensure protection of the interests of workers, but that "subject to the basic standards set forth in the bill, specific conditions for worker protection will normally be the product of local and negotiation." H.R.Rep.No.204, 88th Cong., 1st Sess., 16 (1963). The Senate Report was more direct: "In regard to the question as to whether these provisions would supersede State labor laws, the committee concurs in a statement made by the Secretary of Labor 'that there could be no superseding of State laws by a provision of this kind.' " S.Rep.No.82, 88th Cong., 1st Sess., 29 (1963). 19 During the debates, the role of state law under § 13(c) was discussed at length. Senators Goldwater and Tower suggested that § 13(c) would supplant state law with federal law. 109 Cong.Rec. 5416 (1963). Senator Williams, one of the bill's chief sponsors, replied: "The legislative history has to be corrected" because "we must have a record that will show that the bill does not preempt State law; it does not control or dominate with irrevocable authority local situations." Id., at 5417. The proposed statute, Senator Williams continued, would not "preempt or be a substitute for State law." Ibid. Senator Goldwater remained adamant that "we are attempting a major alteration in the Nation's labor laws." Id., at 5418. But Senator Sparkman, the Chairman of the Senate Committee, repeated the Secretary's assurance that § 13(c) "will not supersede or displace or override" state law. 109 Cong.Rec. 5418 (1963).8 20 The Senate returned to the issue during a colloquy between Senator Goldwater and Senator Morse. Senator Goldwater feared that the proposed statute would override state laws denying public employees the right to strike. Id., at 5673. Senator Morse assured Senator Goldwater otherwise that "the State law would supervene." Ibid. When Senator Goldwater inquired about state laws other than those concerning the right to strike, Senator Morse replied in the same vein: "The amendment does not supersede any State policy." Ibid. 21 In an important exchange, Senator Goldwater noted that local government employers were excluded from the coverage of the National Labor Relations Act, see 29 U.S.C. § 152(2), and asked whether § 13(c) would be inconsistent with that exclusion. 109 Cong.Rec. 5673-5674 (1963). Senator Morse responded that the language of the bill "make[s] it clear that the Taft-Hartley exemptions are not changed by the amendment." Id., at 5674. See also id., at 5422 (remarks of Sen. Javits) (state law could not be overridden "under any phase of the Taft-Hartley law"). Senator Morse underscored the purpose of the amendment: "I cannot emphasize the point more than I already have done in the legislative history in our debate. It deals with municipal and State problems, and not Federal problems." Id., at 5674. Finally, Senator Goldwater asked whether state law would control if there were no specific state law forbidding strikes by public employees. Senator Morse adhered to the same course: "In the absence of any local law, it would be for the State court to decide whether [the employees] had that right." Ibid. 22 A similar, but more abbreviated, interchange took place on the House floor. When some Congressmen questioned the effect of § 13(c) on state law, they were reassured by Congressman Multer that "[n]othing in this bill . . . will infringe upon local law, whether it be of a State or municipality." 110 Cong.Rec. 14980 (1964). And Congressman Rains repeated, "there is not one line in this bill that would vitiate in any way any State or local law." Ibid. 23 Thus, Congress made it absolutely clear that it did not intend to create a body of federal law applicable to labor relations between local governmental entities and transit workers.9 Section 13(c) would not supersede state law, it would leave intact the exclusion of local government employers from the National Labor Relations Act, and state courts would retain jurisdiction to determine the application of state policy to local government transit labor relations. Congress intended that § 13(c) would be an important tool to protect the collective-bargaining rights of transit workers, by ensuring that state law preserved their rights before federal aid could be used to convert private companies into public entities.10 See 109 Cong.Rec. 5673 (1963) (remarks of Sen. Morse) (if city proposed to reject collective bargaining, it would be ineligible for federal aid). But Congress designed § 13(c) as a means to accommodate state law to collective bargaining, not as a means to substitute a federal law of collective bargaining for state labor law.11 IV 24 Given this explicit legislative history, we cannot read § 13(c) to create federal causes of action for breaches of § 13(c) agreements and collective-bargaining contracts between UMTA aid recipients and transit unions.12 The legislative history indicates that Congress intended those contracts to be governed by state law applied in state courts.13 25 Accordingly, the judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. 26 It is so ordered. 27 Justice POWELL, with whom Justice O'CONNOR joins, concurring. 28 As the Court notes, this case "does not fit comfortably in [the] mold" of our implied right of action cases. Ante, at 20. Congress here provided for the making of contracts that it must have intended to be enforced. The Court thus identifies the question correctly as whether Congress intended those contracts to be enforced in federal court. Ante, at 21. This of course is precisely the question on which implied rights of action cases properly are decided. See, e.g., Middlesex County Sewerage Authority v. National Sea Clammers Assn., 453 U.S. 1, 13, 101 S.Ct. 2615, 2622-23, 69 L.Ed.2d 435 (1981); Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 639, 101 S.Ct. 2061, 2066, 68 L.Ed.2d 500 (1981). 29 There are other parallels between this case and those in the more familiar implied right of action "mold." Most significantly to me, both kinds of cases involve the same fundamental issues of congressional and judicial power. By enforcing contract rights not within the jurisdictional grant conferred by Congress, as much as by improperly "inferring" a right of action, "a court of limited jurisdiction necessarily extends its authority to embrace a dispute Congress has not assigned it to resolve. . . . This runs contrary to the established principle that '[t]he jurisdiction of the federal courts is carefully guarded against expansion by judicial interpretation. . .,' American Fire & Cas. Co. v. Finn, 341 U.S. 6, 17, 71 S.Ct. 534, 541, 95 L.Ed. 702 (1951), and conflicts with the authority of Congress under Art. III to set the limits of federal jurisdiction." Cannon v. University of Chicago, 441 U.S. 677, 746-747, 99 S.Ct. 1946, 1983, 60 L.Ed.2d 560 (1979) (POWELL, J., dissenting). 30 Because a federal court should exercise extreme caution before assuming jurisdiction not clearly conferred by Congress, we should not condone the implication of federal jurisdiction over contract claims in the absence of an unambiguous expression of congressional intent. As I do not view this position as inconsistent with the reasoning of the Court, I join its opinion. 1 Originally, § 13(c) was § 10(c). In 1966, the Act was amended, and the section received its present designation. Pub.L. 89-562, § 2(b)(1), 80 Stat. 716. Throughout this opinion, it is referred to as § 13(c). 2 Section 13(c) reads in full: "It shall be a condition of any assistance under section 3 of this Act that fair and equitable arrangements are made, as determined by the Secretary of Labor, to protect the interests of employees affected by such assistance. Such protective arrangements shall include, without being limited to, such provisions as may be necessary for (1) the preservation of rights, privileges, and benefits (including continuation of pension rights and benefits) under existing collective bargaining agreements or otherwise; (2) the continuation of collective bargaining rights; (3) the protection of individual employees against a worsening of their positions with respect to their employment; (4) assurances of employment to employees of acquired mass transportation systems and priority of reemployment of employees terminated or laid off; and (5) paid training or retraining programs. Such arrangements shall include provisions protecting individual employees against a worsening of their positions with respect to their employment which shall in no event provide benefits less than those established pursuant to section 5(2)(f) of the Act of February 4, 1887 (24 Stat. 379), as amended. The contract for the granting of any such assistance shall specify the terms and conditions of the protective arrangements." 3 According to the union's complaint, the Authority since 1966 had contracted with a private individual, T. O. Petty, for the management of the transportation system. App. 6a-7a. The union negotiated its 1975 contract with Petty; when he left as manager of the system, the Authority claimed it was not bound by the contract he had negotiated. Id., at 47a. The union alleged that petitioners had promised in the § 13(c) agreement to be bound by the contracts Petty had signed and that petitioners violated both the § 13(c) agreement and the 1975 collective-bargaining contract. App. 7a-8a. 4 Prior to filing suit, the union asked the Secretary of Labor and the Secretary of Transportation to find that petitioners had violated the § 13(c) agreement and to ensure that petitioners complied with the agreement. Both Secretaries refused. See 650 F.2d 1379, 1381 (CA6 1981). In addition to relief directed at petitioners, the union requested that the two Secretaries be ordered to take appropriate enforcement action against petitioners to compel compliance with the § 13(c) agreement. App. 11a. Both the District Court and the Court of Appeals refused the union's request. See 447 F.Supp. 88, 90-92 (1977); 650 F.2d, at 1387-1388. The union has not sought review of this ruling, and we express no opinion on that aspect of the litigation. 5 Several Courts of Appeals, in addition to the Sixth Circuit, have decided that § 13(c) authorizes federal suits for violations of § 13(c) agreements and collective-bargaining contracts between recipients of UMTA funds and transit unions. Division 587, Amalgamated Transit Union, AFL-CIO v. Municipality of Metropolitan Seattle, 663 F.2d 875 (CA9 1981); Local Div. 714, Amalgamated Transit Union, AFL-CIO v. Greater Portland Transit District, 589 F.2d 1 (CA1 1978); Local Div. 519, Amalgamated Transit Union, AFL-CIO v. LaCrosse Municipal Transit Utility, 585 F.2d 1340 (CA7 1978); Division 1287, Amalgamated Transit Union, AFL-CIO v. Kansas City Area Transportation Authority, 582 F.2d 444 (CA8 1978), cert. denied, 439 U.S. 1090, 99 S.Ct. 872, 59 L.Ed.2d 56 (1979). One Court of Appeals has reached the opposite conclusion. Local Div. 732, Amalgamated Transit Union v. Metropolitan Atlanta Rapid Transit Authority, 667 F.2d 1327 (CA11 1982). In a related decision, the First Circuit has concluded that the terms of § 13(c) agreements do not override conflicting provisions of state law. Local Div. 589, Amalgamated Transit Union, AFL-CIO v. Massachusetts, 666 F.2d 618 (1st Cir. 1981), cert. pending, No. 81-1817. 6 Thus, we agree with the Court of Appeals that, strictly speaking, the District Court had jurisdiction under 28 U.S.C. § 1331 to hear the union's suit. Under Bell v. Hood, 327 U.S. 678, 681, 66 S.Ct. 773, 775, 90 L.Ed. 939 (1946), jurisdiction exists if the complaint is "drawn so as to claim a right to recover under the Constitution and laws of the United States." The complaint alleged a violation of the § 13(c) agreement required by the UMTA and of the subsequent collective-bargaining agreement contemplated by the Act, and prayed for relief under federal law. We do not consider the union's asserted federal claims to be "wholly insubstantial and frivolous," 327 U.S., at 682-683, 66 S.Ct., at 776, so that the District Court lacked jurisdiction to entertain the union's suit. Thus, the District Court had jurisdiction for the purposes of determining whether the union stated a cause of action on which relief could be granted. Id., at 682, 66 S.Ct., at 776. See also Wheeldin v. Wheeler, 373 U.S. 647, 83 S.Ct. 1441, 10 L.Ed.2d 605 (1963). 7 The statute also provides that the protective "arrangements shall include provisions . . . which shall in no event provide benefits less than those established pursuant to section 5(2)(f). . . ." As we explain, see n. 9, infra, this portion of the statute strengthens the union's position, but we do not consider it at all determinative. 8 The union points to the fact that Congress rejected amendments that would have required the continuation of collective-bargaining rights only to the extent not inconsistent with state law. See 109 Cong.Rec. 5422 (1963); id., at 5582; id., at 5684; 110 Cong.Rec. 14980 (1964). But, as Senator Williams explained, those amendments were rejected not because Congress thought § 13(c) would supplant state labor law but because such an amendment was "clearly . . . unnecessary" to guarantee that § 13(c) would not "supersede or preempt or override State law." 109 Cong.Rec. 5421 (1963). Accord: 110 Cong.Rec. 14980 (1964) (remarks of Reps. Multer and Rains). Beyond the explanation given by Senator Williams and repeated on the House floor, the defeat of these amendments merely reflected a congressional intent that the Federal Government be able to seek changes in state law and ultimately to refuse financial assistance when state law prevented compliance with § 13(c). See 109 Cong.Rec. 5684 (1963) (remarks of Sen. Morse); id., at 5422 (remarks of Sen. Javits). 9 In light of the legislative history of § 13(c), we do not find Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 100 S.Ct. 242, 62 L.Ed.2d 146 (1979), or Machinists v. Central Airlines, Inc., 372 U.S. 682, 83 S.Ct. 956, 10 L.Ed.2d 67 (1963), to be controlling. Both cases turned on the language, purpose, and legislative history of the particular statute involved, see Transamerica, 444 U.S., at 18-19, 100 S.Ct., at 246-247; Machinists, 372 U.S., at 685-695, 83 S.Ct., at 960-963, and we read the congressional intent behind § 13(c) to be far different from the congressional purpose underlying the statutes at issue in those cases. Norfolk & Western R. Co. v. Nemitz, 404 U.S. 37, 92 S.Ct. 185, 30 L.Ed.2d 198 (1971), of course, is not to be overlooked. In that case, the Court decided that a railroad's employees stated federal claims when they alleged a breach of an agreement entered into by the railroad under § 5(2)(f) of the Interstate Commerce Act, 49 U.S.C. § 5(2)(f) (1970 ed.). Section 13(c) refers to § 5(2)(f) and provides that the protective arrangements shall not provide benefits less than those established by § 5(2)(f). See nn. 2 and 7, supra. If, when it passed § 13(c), Congress had expressed an awareness that § 5(2)(f) assurances could be enforced in federal court, or if there was reason to presume such awareness, the reference in § 13(c) to § 5(2)(f) would make this a different case. See generally Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 102 S.Ct. 1825, 72 L.Ed.2d 182 (1982). But the legislative history contains no such congressional recognition. Furthermore, Nemitz was decided several years after § 13(c) was enacted. Consequently, we find the specific legislative history of § 13(c), not the holding of Nemitz, to be determinative. 10 The union relies upon the fact that Congress strengthened the language of § 13(c) during the course of its passage. Congress wrote § 13(c) to require protective provisions "necessary for" the protection of collective-bargaining rights, rather than provisions "as are found to be appropriate for" the protection of those rights, as the Kennedy administration had recommended. See House Hearings, at 476; 110 Cong.Rec. 14976 (1964). In addition, § 13(c)(2) was amended at the request of Senator Morse to require the "continuation" of collective-bargaining rights rather than the mere "encouragement" of the continuation of those rights. See 109 Cong.Rec. 5627 (1963); id., at 5685. But these alterations in the bill demonstrate only that Congress demanded that the Secretary ensure that state law preserved collective-bargaining rights before he provided federal aid for acquisition of a private transit company. When viewed in conjunction with the discussion of state law in the legislative history, the modifications of the original bill do not prove that Congress intended that federal rather than state law would govern a contract between a UMTA aid recipient and a union representing its employees. 11 Senator Javits summarized: "[W]e have a balanced scheme. We do not override the [state] law; at the same time, we do not compel the Federal Government to go in where the law is adverse to the interest of labor and labor's own point of view, and perhaps also even give encouragement to exempt a situation of this kind where the State desires to get this type of Federal help." 109 Cong.Rec. 5422 (1963). There remains the possibility that Congress might have intended a federal court to hear the union's claims, but to apply state law. Such an anomalous result would be inconsistent with the emphasis in the legislative history that § 13(c) addresses "municipal and State problems, and not Federal problems," id., at 5674. Thus, unless there is an independent source of jurisdiction, such as diversity or pendent jurisdiction, the union must sue in state court. See Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S., at 19, n. 8, 100 S.Ct., at 247 n. 8; Shoshone Mining Co. v. Rutter, 177 U.S. 505, 506-507, 20 S.Ct. 726, 727, 44 L.Ed. 864 (1900). 12 Based on the legislative history, we could not permit the union to bring a federal suit if we characterized its complaint as alleging that petitioners violated § 13(c) itself by virtue of the alleged contractual breaches. See Brief for Respondent 33-34. Nor does 42 U.S.C. § 1983 (1976 ed., Supp.IV) permit the union to bring suit. As the Court held in Maine v. Thiboutot, 448 U.S. 1, 100 S.Ct. 2502, 65 L.Ed.2d 555 (1980), § 1983 encompasses deprivations of rights secured by all "laws" of the United States, of which § 13(c) is, of course, one. But because we have determined that Congress did not intend that breaches of § 13(c) agreements or collective-bargaining contracts would constitute deprivations of federal rights secured by § 13(c), the union has no cause of action under § 1983. See Middlesex County Sewerage Authority v. National Sea Clammers Assn., 453 U.S. 1, 19-21, 101 S.Ct. 2615, 2626, 69 L.Ed.2d 435 (1981). 13 There are other possible remedies for violations of § 13(c) agreements and collective-bargaining contracts. The union, of course, can pursue a contract action in state court. In addition, the Federal Government can respond by threatening to withhold additional financial assistance. See Local Div. 589, 666 F.2d, at 634-635. While we hold that the union cannot sue in federal court to enforce its contracts, we express no view on the entirely separate question whether the Federal Government could bring a federal suit against a UMTA funding recipient for violating the terms of its grant agreement with the Government. Such a suit would involve a different contract from the § 13(c) agreement and the collective-bargaining agreement at issue in this case. See generally Local Div. 732, 667 F.2d, at 1338-1339; Local Div. No. 714, 589 F.2d, at 13.

89

457 U.S. 52 102 S.Ct. 2223 72 L.Ed.2d 668 Kathy Sue JOHNSON, etc., et al.v.BOARD OF EDUCATION OF the CITY OF CHICAGO et al. No. 81-1097. June 7, 1982. PER CURIAM. 1 This case was commenced by petitioners challenging the voluntary adoption by the Board of Education of the city of Chicago of racial quotas on enrollment at two high schools. Petitioners alleged that the quotas, purportedly designed to arrest "white flight," were unlawful because they resulted in the denial of admission to those schools of some black applicants but no white applicants. The District Court upheld the plan, and the Court of Appeals affirmed. 604 F.2d 504 (CA7 1979). We granted certiorari, 448 U.S. 910, 100 S.Ct. 3055, 65 L.Ed.2d 1139 (1980), but then vacated the judgment and remanded the case "for further consideration in light of the subsequent development described in the suggestion of mootness filed by respondents." 449 U.S. 915, 101 S.Ct. 339, 66 L.Ed.2d 162 (1980). That development was the entry of a consent decree in a related case, United States v. Board of Education of Chicago, 88 F.Supp. 679 (ND Ill.), in which the Board of Education agreed to develop a systemwide integration plan, and the Board's announcement that it had abandoned use of the racial quotas at the two high schools. The Court of Appeals, 645 F.2d 75, remanded to the District Court to consider the suggestion of mootness. 645 F.2d 75 (1981). That court, finding that the Board had readopted the quotas, concluded without taking further evidence that the challenge was not moot. The Court of Appeals, 664 F.2d 1069, agreeing that the case was not moot and relying upon the doctrine of the law of the case, affirmed without reconsidering the constitutional challenge to the racial quotas in light of the subsequent development that the Board argued eliminated or reduced any discriminatory effects of the quotas. 664 F.2d 1069 (1981). Petitioners have now renewed their request for review. 2 We agree with the Court of Appeals that the case is not moot and that the subsequent development does not undermine that court's original decision upholding the racial quotas. However, since if we were to grant certiorari we would consider the constitutional challenge as an original matter, the subsequent development might well be relevant to that consideration. It was for that reason that we vacated the Court of Appeals' judgment for further consideration in light of the subsequent development. No additional evidence was taken and therefore neither the record nor the District Court or Court of Appeals opinions reflect the subsequent development. We therefore grant certiorari, vacate the judgment, and remand the case with the direction that the matter be consolidated with the ongoing proceeding in the District Court in United States v. Board of Education of Chicago, 80-C-5124, so that court may decide petitioners' challenge on the basis of a complete factual record. Because we have vacated the Court of Appeals' judgments in this case, the doctrine of the law of the case does not constrain either the District Court or, should an appeal subsequently be taken, the Court of Appeals. 3 It is so ordered. 4 Justice BRENNAN would grant the petition for a writ of certiorari and set the case for oral argument. 5 Justice WHITE took no part in the consideration or decision of this case. 6 Justice REHNQUIST, with whom Justice MARSHALL joins, dissenting. 7 Title 28 U.S.C. § 2106 provides that "[t]he Supreme Court . . . may affirm, modify, vacate, set aside or reverse any judgment, decree, or order of a court lawfully brought before it for review. . . ." Our practice over many years indicates that implicit in this grant of authority is a requirement that we specify our reasons for acting as we do. Here the Court departs from that implicit requirement. The ultimate disposition of the case is the vacation of the judgment of the Court of Appeals and a remand so that this case may be consolidated with another proceeding in the District Court for the Northern District of Illinois. A reading of the Court's per curiam suggests that the Court is vaguely dissatisfied with the opinion of the Court of Appeals which it purportedly reviews, but no substantive judgment is made as to whether that opinion was correct or incorrect in whole or in part. Nothing in the record before us suggests to me any reason why we should assume a function more properly exercised by the Court of Appeals or by the District Court, and order consolidation of this case with another pending action in the District Court. But even if I were disposed to agree as to the propriety of the disposition now made by the Court, I would hope that something in the nature of an opinion explaining the reasons for the action would accompany the disposition. Since the Court's per curiam makes no effort at such an explanation, I dissent.

89

457 U.S. 1 102 S.Ct. 2194 72 L.Ed.2d 628 Jesus Rivera RODRIGUEZ et al., Appellants,v.POPULAR DEMOCRATIC PARTY et al. No. 81-328. Argued March 22, 1982. Decided June 7, 1982. Syllabus A member of appellee Popular Democratic Party (hereafter appellee) who was elected in a 1980 general election to the Puerto Rico House of Representatives from District 31, died in 1981. The Governor of Puerto Rico subsequently called for a "by-election" open to all qualified voters in District 31—to fill the vacancy. Appellee then filed suit in the Superior Court of Puerto Rico, alleging that the Puerto Rico statutes under which the Governor purported to act authorized only candidates and electors affiliated with appellee to participate in the by-election. Appellants, qualified electors in District 31 who are not affiliated with appellee, intervened as defendants. The court entered judgment for appellee. The Puerto Rico Supreme Court modified the Superior Court's judgment, holding, inter alia, that the pertinent statute, as properly construed, requires a by-election only if the party of the legislator vacating the seat fails to designate a replacement within 60 days after the vacancy occurs, and that if the party selects a single candidate within such period, that candidate is declared "automatically elected to fill the vacancy." The court rejected appellants' contention that this procedure violated the Federal Constitution. While the case was pending before the Puerto Rico Supreme Court, appellee held a primary election in which only its members were permitted to participate and which resulted in the selection of a person who, pursuant to the Supreme Court's mandate, was sworn in as the new representative from District 31. Held: The Puerto Rico statute, as interpreted by the Puerto Rico Supreme Court to vest in a political party the initial authority to appoint an interim replacement for one of its members who vacates a position as a district senator or representative, does not violate the Federal Constitution. Pp. 5-14. (a) The voting rights of Puerto Rico citizens are constitutionally protected to the same extent as those of all other United States citizens. At the same time, Puerto Rico, like a state, is an autonomous political entity, "sovereign over matters not ruled by the Constitution," Calero-Toledo v. Pearson Yacht Leasing Co., 416 U.S. 663, 673, 94 S.Ct. 2080, 2086, 40 L.Ed.2d 452, and the methods by which its people and their representatives have chosen to structure the Commonwealth's electoral system are entitled to substantial deference. Pp. 7-8. (b) The right to vote, per se, is not a constitutionally protected right, and the Constitution does not compel a fixed method of choosing state or local officers or representatives. While a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction when a state or the Commonwealth of Puerto Rico has provided that its representatives be elected, the Puerto Rico statute at issue does not restrict access to the electoral process or afford unequal treatment to different classes of voters, candidates, or political parties. All qualified voters have an equal opportunity to select a district representative in the general election; and the interim appointment provision applies uniformly to all legislative vacancies, whenever they arise. Cf. Valenti v. Rockefeller, 393 U.S. 405, 89 S.Ct. 689, 21 L.Ed.2d 635. Moreover, the interim appointment system serves the legitimate purpose of ensuring that vacancies are filled promptly, without the necessity of the expense and inconvenience of a special election. Pp. 8-12. (c) Nor is Puerto Rico's appointment mechanism rendered constitutionally defective by virtue of the fact that the interim appointment power is given to the political party with which the previous incumbent was affiliated. The Puerto Rico Legislature could reasonably conclude that appointment by the previous incumbent's political party would more fairly reflect the will of the voters than appointment by the Governor or some other elected official, particularly where such official is a member of a different party. And in light of Puerto Rico's special interest in ensuring minority representation in its legislature, it was not unreasonable for the legislature, in establishing the appointment system for filling vacancies, to make provision for continuity of party representation. Pp. 12-13. (d) Appellants' rights of association and equal protection of the laws were not violated by their exclusion, because of their party affiliation, from appellee's special election held to select the interim representative. Puerto Rico law authorized appellee to designate the interim replacement, and it was entitled to adopt its own procedures for such selection. Appellee was not required to include participation by nonmembers. P. 14. ____ P.R.R. ____, affirmed. Philip Lacovara, Washington, D.C., for appellants. Abe Fortas, Washington, D.C., for appellees. Chief Justice BURGER delivered the opinion of the Court. 1 The question presented by this appeal is whether Puerto Rico may by statute vest in a political party the power to fill an interim vacancy in the Puerto Rico Legislature. The Supreme Court of Puerto Rico held that such a procedure did not violate the United States Constitution. We noted probable jurisdiction, 454 U.S. 938, 102 S.Ct. 472, 70 L.Ed.2d 246 (1981), and we affirm. 2 * In the November 4, 1980, Puerto Rico general election, Ramon Muniz, a member of appellee Popular Democratic Party, was elected to the Puerto Rico House of Representatives from District 31.1 Muniz died on January 28, 1981. The Governor of Puerto Rico, a member of the opposition New Progressive Party, subsequently called for a "by-election"—open to all qualified voters in District 31—to fill the vacancy caused by Muniz' death. The Governor purported to act pursuant to Articles 5.006 and 5.007 of the Electoral Law of Puerto Rico, P.R. Laws Ann., Tit. 16, §§ 3206, 3207 (Supp.1980).2 3 On March 3, 1981, the Popular Democratic Party instituted this action in the Superior Court of Puerto Rico, alleging that Articles 5.006 and 5.007 authorized only candidates and electors affiliated with the Party to participate in the by-election. Appellants, 10 qualified electors in District 31 who are not affiliated with the Popular Democratic Party, intervened as defendants. On March 20, 1981, the Superior Court entered judgment for the Popular Democratic Party; it ordered the Governor and General Administrator of Elections to limit participation in the by-election to Party members. App. to Juris. Statement 36a. 4 A divided Supreme Court of Puerto Rico modified the Superior Court's judgment. It interpreted Articles 5.006 and 5.007 to require a by-election only in the event that the party of the legislator vacating the seat fails to designate a replacement within 60 days after the vacancy occurred; if the party selects a single candidate within the 60-day period, that candidate is "automatically elected to fill the vacancy," rendering a by-election unnecessary. Popular Democratic Party v. Barcelo, ____ P.R.R. ____ (1981). The court held further that if the party presents more than one candidate during the 60-day period, a by-election must be conducted in which only party-affiliated candidates may run but in which all qualified electors may vote. In the event no candidate is presented within the 60-day period, candidates affiliated with any party, as well as independent candidates, are permitted to run in the by-election. Because of the delay already occasioned by the litigation, the court permitted appellee party only 30 days from the entry of judgment, May 8, 1981, to present a "slate" of candidates to the Commonwealth Election Commission. The court ordered that "[i]f said slate is limited to only one candidate, he shall be certified by the General Administrator of Elections as the person entitled to hold the vacant seat." Id., at ____.3 5 The court rejected appellants' contention that this procedure violated the United States Constitution. It noted that the Constitution does not expressly require a fixed method for filling vacancies in a state or commonwealth legislature. The court also held that Puerto Rico's party appointment system serves several "compelling interests," such as ensuring the stability and continuity of the "legislative balance" until the next general election; protecting the "electoral mandate" of the previous election; and reducing "inter-partisan political campaigns to once every four years." Id., at ____. II 6 Puerto Rico, in common with many of the States, has adopted means of filling interim vacancies in elective commonwealth offices without the necessity of a full-scale special election.4 If a vacancy occurs in the office of Governor, it is automatically filled by the Secretary of State, an officer appointed by the Governor. P.R.Const., Art. IV, § 7. Mayoral vacancies and vacancies in the municipal assemblies are filled by appointment upon the recommendation of the political party to which the incumbent belonged. P.R. Laws Ann., Tit. 21, §§ 1161, 1259 (1974). Similarly, the Commonwealth Constitution provides that vacancies in the posts of at-large senators and representatives, see n.13, infra, shall be filled "upon recommendation of the political party to which belonged the Senator or Representative causing the vacancy. . . ." Art. III, § 8. Article 5.006 of the Puerto Rico Electoral Law, as interpreted by the Supreme Court of Puerto Rico in this case, likewise confers on a political party the initial opportunity to appoint an interim replacement for one of its members who vacates a position as a district senator or representative. In each case, the appointee serves only until the next regularly scheduled election.5 7 Appellants' challenge to the procedure mandated by Article 5.006 is essentially two-pronged. Appellants first contend that qualified voters have a federal constitutional right to elect their representatives to the Puerto Rico Legislature, and that vacancies in legislative offices therefore must be filled by a special election open to all qualified electors, not by interim appointment of any kind. Alternatively, appellants maintain that even if legislative vacancies may be filled by an interim appointment of the Governor or some other elected official, Puerto Rico's party appointment mechanism impermissibly infringes upon their right of association under the First Amendment and denies them equal protection of the laws. A. 8 It is not disputed that the fundamental protections of the United States Constitution extend to the inhabitants of Puerto Rico. See Torres v. Puerto Rico, 442 U.S. 465, 469-470, 99 S.Ct. 2425, 2428-29, 61 L.Ed.2d 1 (1979). Cf. Dorr v. United States, 195 U.S. 138, 148, 24 S.Ct. 808, 812, 49 L.Ed. 128 (1904). In particular, we have held that Puerto Rico is subject to the constitutional guarantees of due process and equal protection of the laws. Examining Board v. Flores de Otero, 426 U.S. 572, 599-601, 96 S.Ct. 2264, 2279-80, 49 L.Ed.2d 65 (1976); Calero-Toledo v. Pearson Yacht Leasing Co., 416 U.S. 663, 94 S.Ct. 2080, 40 L.Ed.2d 452 (1974).6 We thus think it is clear that the voting rights of Puerto Rico citizens are constitutionally protected to the same extent as those of all other citizens of the United States. 9 At the same time, Puerto Rico, like a state, is an autonomous political entity, "sovereign over matters not ruled by the Constitution." Calero-Toledo, supra, at 673, 94 S.Ct., at 2086 (quoting Mora v. Mejias, 115 F.Supp. 610 (PR 1953)). See Cordova & Simonpietri Ins. Agency Inc. v. Chase Manhattan Bank, 649 F.2d 36, 39-42 (CA1 1981). The methods by which the people of Puerto Rico and their representatives have chosen to structure the Commonwealth's electoral system are entitled to substantial deference. Moreover, we should accord weight to the Puerto Rico Supreme Court's assessment of the justification and need for particular provisions to fill vacancies caused by the death, resignation, or removal of a member of the legislature. Bearing these considerations in mind, we turn to appellants' constitutional challenges. B 10 No provision of the Federal Constitution expressly mandates the procedures that a state or the Commonwealth of Puerto Rico must follow in filling vacancies in its own legislature. Cf. U.S.Const., Art. I, § 2; Amdt. 17, cl. 2.7 Appellants nevertheless maintain that qualified electors have an absolute constitutional right to vote for the members of a state or commonwealth legislature, even when a special election is required for this purpose.8 However, this Court has often noted that the Constitution "does not confer the right of suffrage upon any one," Minor v. Happersett, 21 Wall. 162, 178, 22 L.Ed. 627 (1875), and that "the right to vote, per se, is not a constitutionally protected right," San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 35, n. 78, 93 S.Ct. 1278, 1298, n. 78, 36 L.Ed.2d 16 (1973). See McPherson v. Blacker, 146 U.S. 1, 38-39, 13 S.Ct. 3, 11-12, 36 L.Ed. 869 (1892). Moreover, we have previously rejected claims that the Constitution compels a fixed method of choosing state or local officers or representatives. 11 For example, in Fortson v. Morris, 385 U.S. 231, 234, 87 S.Ct. 446, 448, 17 L.Ed.2d 330 (1966), Justice Black, speaking for the Court, stated: 12 "There is no provision of the United States Constitution or any of its amendments which either expressly or impliedly dictates the method a State must use to select its Governor." 13 In Fortson, the Court sustained a Georgia constitutional provision empowering the state legislature to elect a Governor from the two candidates receiving the highest number of votes cast in the general election, in the event neither received a majority. Similarly, in Sailors v. Board of Education, 387 U.S. 105, 87 S.Ct. 1549, 18 L.Ed.2d 650 (1967), the Court upheld a statute authorizing appointment rather than election of the members of a county school board.9 14 To be sure, when a state or the Commonwealth of Puerto Rico has provided that its representatives be elected, "a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction." Dunn v. Blumstein, 405 U.S. 330, 336, 92 S.Ct. 995, 1000, 31 L.Ed.2d 274 (1972). See Kramer v. Union Free School District, 395 U.S. 621, 626-629, 89 S.Ct. 1886, 1889-90, 23 L.Ed.2d 583 (1969); Gray v. Sanders, 372 U.S. 368, 379-380, 83 S.Ct. 801, 807-08, 9 L.Ed.2d 821 (1963). However, the Puerto Rico statute at issue here does not restrict access to the electoral process or afford unequal treatment to different classes of voters or political parties. All qualified voters have an equal opportunity to select a district representative in the general election; and the interim appointment provision applies uniformly to all legislative vacancies, whenever they arise.10 15 In Valenti v. Rockefeller, 393 U.S. 405, 89 S.Ct. 689, 21 L.Ed.2d 635 (1969), the Court sustained the authority of the Governor of New York to fill a vacancy in the United States Senate by appointment pending the next regularly scheduled congressional election—in that case, a period of over 29 months.11 Thus, although most Members of the United States Senate hold office by virtue of popular election, some Members, at any given time, may hold office by virtue of an interim appointment. The Court found nothing invidious or arbitrary in this distinction in Valenti, nor do we here. As the three-judge District Court observed in Valenti: 16 "In this case we are confronted with no fundamental imperfection in the functioning of democracy. No political party or portion of the state's citizens can claim it is permanently disadvantaged . . . or that it lacks effective means of securing legislative reform if the statute is regarded as unsatisfactory. We have, rather, only the unusual, temporary, and unfortunate combination of a tragic event and a reasonable statutory scheme." Valenti v. Rockefeller, 292 F.Supp. 851, 867 (SDNY 1968). 17 Valenti, of course, unlike this case, involved an interpretation of the Seventeenth Amendment, which explicitly outlines the procedures for filling vacancies in the United States Senate. See n.7, supra. However, the fact that the Seventeenth Amendment permits a state, if it chooses, to forgo a special election in favor of a temporary appointment to the United States Senate suggests that a state is not constitutionally prohibited from exercising similar latitude with regard to vacancies in its own legislature. We discern nothing in the Federal Constitution that imposes greater constraints on the Commonwealth of Puerto Rico. 18 The Commonwealth's choice to fill legislative vacancies by appointment rather than by a full-scale special election may have some effect on the right of its citizens to elect the members of the Puerto Rico Legislature; however, the effect is minimal, and like that in Valenti, it does not fall disproportionately on any discrete group of voters, candidates, or political parties. See n.10, supra. Moreover, the interim appointment system plainly serves the legitimate purpose of ensuring that vacancies are filled promptly, without the necessity of the expense and inconvenience of a special election. The Constitution does not preclude this practical and widely accepted means of addressing an infrequent problem. C 19 Puerto Rico's appointment mechanism is not rendered constitutionally defective by virtue of the fact that the interim appointment power is given to the political party with which the previous incumbent was affiliated. Appellants maintain that the power to make interim appointments must be vested in an elected official, such as the Governor of the Commonwealth, so that the appointments will have "the legitimacy of derivative voter approval and control." Reply Brief for Appellants 15. However, that such control may often be largely illusory is illustrated by this case, where the Governor and the incumbent belonged to different parties. The Puerto Rico Legislature could reasonably conclude that appointment by the previous incumbent's political party would more fairly reflect the will of the voters than appointment by the Governor or some other elected official.12 20 The Supreme Court of Puerto Rico held that party appointment was a legitimate mechanism serving to protect the mandate of the preceding election and to preserve the "legislative balance" until the next general election is held. Such protection is particularly important in light of Puerto Rico's special interest in ensuring minority representation in its legislature.13 See Garcia v. Barcelo, 671 F.2d 1, 6-7 (CA1 1982). It was thus not unreasonable for the Puerto Rico Legislature, in establishing an appointment system for filling legislative vacancies, to make provision for continuity of party representation. Cf. Kaelin v. Warden, 334 F.Supp. 602, 607-608 (ED Pa. 1971) (three-judge District Court).14 Absent some clear constitutional limitation, Puerto Rico is free to structure its political system to meet its "special concerns and political circumstances," Garcia, supra, at 7. 21 Finally, appellants argue that their rights of association and equal protection of the laws were violated by their exclusion, based solely upon their party affiliation, from the Party-sponsored election held to select Muniz' successor, see n.3, supra. Cf. Branti v. Finkel, 445 U.S. 507, 100 S.Ct. 1287, 63 L.Ed.2d 574 (1980). However, appellants' argument misconceives the nature of the election held in this case. Puerto Rico law authorized the Popular Democratic Party to designate an interim replacement to fill Muniz' seat. The Party was entitled to adopt its own procedures to select this replacement; it was not required to include nonmembers in what can be analogized to a party primary election. Cf. Democratic Party of U.S. v. Wisconsin, 450 U.S. 107, 101 S.Ct. 1010, 67 L.Ed.2d 82 (1981); Cousins v. Wigoda, 419 U.S. 477, 95 S.Ct. 541, 42 L.Ed.2d 595 (1975). Appellants' exclusion from this election did not violate their rights of association, nor did it deprive them of equal protection of the laws. III 22 We hold that the mechanism adopted by the Puerto Rico Legislature for filling legislative vacancies is not foreclosed by the Federal Constitution. Accordingly, the judgment of the Supreme Court of Puerto Rico is 23 Affirmed. 1 The Puerto Rico Legislative Assembly consists of two chambers—the House of Representatives, with 51 members, and the Senate, with 27 members. P.R.Const., Art. III, §§ 1 and 2. A single general election is held in Puerto Rico every four years for all elective officials. Art. VI, § 4; P.R.Laws Ann., Tit. 16, §§ 3201, 3205 (Supp.1980). 2 Article 5.006 provides, in pertinent part: "When a vacancy occurs in the office of a senator or representative elected as an independent candidate for a district, or when a vacancy occurs in the office of a senator or representative for a district, nominated by a party before the fifteen (15) months immediately preceding the date of the following general election, the Governor, with the advice of the [Commonwealth Election] Commission shall, within the thirty (30) days following the date on which the vacancy occurred, call a by-election in such district which shall be held no later than ninety (90) days after the date of the call, and the person elected in such by-election shall hold the office until the term of his predecessor has expired. "If within sixty (60) days following the date such vacancy arises, the party to which the legislator of the vacant office belonged has not presented a candidate to fill such office, the office shall be deemed to be that of an independent legislator, to the effects of holding the by-election to fill it." Article 5.007 provides: "All electors entitled to vote within the geographic district in which the by-election is to be held, pursuant to the call issued by the Governor to such effect, shall vote in a by-election." 3 On March 22, 1981, while the case was pending before the Supreme Court of Puerto Rico, the Popular Democratic Party held a primary election in which only its members were permitted to participate. From a field of four candidates, the Party's members selected Juan Corujo Collazo. Pursuant to the Supreme Court's mandate, Corujo Collazo's name was presented to the Election Commission, and on July 6, 1981, he was sworn in as the new Representative from District 31. 4 In 22 States, legislative vacancies are filled by appointment, with the appointee serving either until the next general election or until expiration of the term of the previous incumbent. Alaska Stat.Ann. § 15.40.320 et seq. (1975) (unless term expires or election held before next legislative session convenes); Colo.Rev.Stat. § 1-12-103 (1980); Haw.Rev.Stat. §§ 17-3, 17-4 (1976 and Supp.1981); Idaho Code § 59-904A (1976); Ill.Rev.Stat., ch. 46, ¶ 25-6 (1980); Ind.Code § 2-2.1-2-1 et seq. (Cum.Supp.1981); Kan.Stat.Ann. § 25-312(1981); Md.Const., Art. III, § 13; Mont.Code Ann. § 5-2-401 et seq. (1981); Neb.Rev.Stat. § 32-1042(3) (1978); Nev.Const., Art. IV, § 12 (unless biennial or regular election held between time of vacancy and next legislative session); N.M.Stat.Ann. §§ 2-7-9B,2-8-9B (1978); N.C.Gen.Stat. § 163-11 (Cum.Supp.1981); Ohio Const., Art. II, § 11; Ore.Rev.Stat. §§ 171.051, 171.060 (1981) (unless the legislature is not in session; a general election will be held within 90 days; and no special session of the legislature will be convened before such election); S.D.Const., Art. III, § 10; Tenn.Code Ann. § 2-14-201 et seq. (1979) (if less than 12 months remain before next general election); Utah Code Ann. § 20-1-5 (Supp.1981); Vt.Stat.Ann., Tit. 17, § 2623 (Supp.1981); Wash.Const., Amdt. 52, Art. 2, § 15; W.Va.Code § 3-10-5 (1979); Wyo.Stat. §§ 22-18-111(a)(ii), (iii) (1977). Like Puerto Rico, five of these States—Colorado, Illinois, Indiana, Maryland, and North Carolina—confer the appointment power on the political party to which the previous incumbent belonged. Nine more States Alaska, Idaho, Montana, Ohio, Oregon, Utah, Washington, West Virginia, and Wyoming—require that the appointee be selected from a list submitted by the political party, or that the appointee be chosen or confirmed by elected officials affiliated with the party. Another two States—Hawaii and Nevada—simply require that the appointee be a member of the party to which his or her predecessor belonged. 5 The current procedure for filling legislative vacancies is similar to that prescribed by a 1938 amendment to Puerto Rico's Organic Act, which remained in effect until Puerto Rico assumed Commonwealth status in 1952. In the 1938 amendment, Congress mandated that vacancies in the Puerto Rico Legislature be filled by the Governor "upon the recommendation of the central committee of the political party of which such senator or representative was a member." Act of June 1, 1938, ch. 308, § 30, 52 Stat. 595. Title 48 U.S.C. §§ 891, 892 presently provide that vacancies in the elective office of Resident Commissioner to the United States are to be filled by appointment of the Governor with the advice and consent of the Puerto Rico Senate. 6 We have never found it necessary to resolve the precise question whether the guarantee of equal protection is provided to Puerto Ricans under the Equal Protection Clause of the Fourteenth Amendment or the Due Process Clause of the Fifth Amendment. See Examining Board v. Flores de Otero, 426 U.S., at 601, 96 S.Ct., at 2280-2281. 7 With regard to Members of the United States House of Representatives, Art. I, § 2, cl. 4, provides: "When vacancies happen in the Representation from any State, the Executive Authority thereof shall issue Writs of Election to fill such Vacancies." The Seventeenth Amendment provides: "When vacancies happen in the representation of any State in the Senate, the executive authority of such State shall issue writs of election to fill such vacancies: Provided, That the legislature of any State may empower the executive thereof to make temporary appointments until the people fill the vacancies by election as the legislature may direct." 8 The source of this purported right is somewhat unclear. Appellants contend that Art. I, § 2, cl. 1, of the Constitution which provides that those eligible to vote for Members of the United States House of Representatives "shall have the Qualifications requisite for the Electors of the most numerous Branch of the State Legislature"—contemplates that state legislators will be popularly elected. See also U.S.Const., Amdt. 17. Moreover, appellants contend that a popularly elected legislature is an essential element of a "Republican Form of Government," U.S.Const., Art. IV, § 4. See 48 U.S.C. § 731c, requiring Puerto Rico to provide a republican form of government. However, this seems largely irrelevant, since Puerto Rico has in fact established a legislature "whose members shall be elected by direct vote at each general election," P.R.Const., Art. III, § 1. See also Art. II, § 2, guaranteeing "equal, direct, and . . . universal suffrage. . . ." 9 In Sailors, we expressly left open the question "whether a State may constitute a local legislative body through the appointive rather than the elective process." 387 U.S., at 109-110, 87 S.Ct., at 1552-53 (emphasis added). However, we need not consider whether, as urged by appellants, a state or the Commonwealth of Puerto Rico is constitutionally barred from abolishing its elected legislative branch of government; that question is not presented. See n.8, supra. 10 Appellants contend that Article 5.006 "discriminates" between voters in districts in which a vacancy occurs and those in which the elected representative or senator serves out his term, because only the former are denied the opportunity to be represented by an elected legislator. Obviously, a statute designed to deal with the occasional problem of legislative vacancies will affect only those districts in which vacancies actually arise. However, such a statute is not for this reason rendered invalid under equal protection principles. A vacancy in the legislature is an unexpected, unpredictable event, and a statute providing that all such vacancies be filled by appointment does not have a special impact on any discrete group of voters or candidates. Cf. Bullock v. Carter, 405 U.S. 134, 92 S.Ct. 849, 31 L.Ed.2d 92 (1972); Williams v. Rhodes, 393 U.S. 23, 89 S.Ct. 5, 21 L.Ed.2d 24 (1968). Appellants' equal protection argument adds nothing to their basic assertion of an absolute constitutional right to elect representatives to a state or commonwealth legislature. 11 In Valenti, the vacancy was created by the death of Senator Robert F. Kennedy on June 6, 1968. Under New York law, since the vacancy arose less than 60 days prior to New York's regular spring primary in an even-numbered year, an election to fill the vacancy would not be held until the general election in the next even-numbered year, i.e., November 1970. The Governor was empowered to make an interim appointment, effective until December 1, 1970. See Valenti v. Rockefeller, 292 F.Supp. 851, 853 (SDNY 1968) (three-judge District Court). 12 See Garcia v. Barcelo, 671 F.2d 1, 6 (CA1 1982): "One might argue, as a matter of form, that appointment by a governor is indeed more 'democratic' because the governor is himself elected. Yet in practice this is not likely to be so when the governor and former representative are of different parties. In that case the party difference is likely to produce successors of different parties. In such circumstances, we see how the framers of a state constitution might conclude that party selection is more likely to reflect the will of the voters than selection by the governor, for it was the former representative's party, not that of the governor, that won the prior seat. Such a judgment, reflecting a knowledge of political practice, seems perfectly consistent with the basic democratic role of the modern political party—translating the individual wills of myriad voters into a practically achievable program administered by a government that can be held responsible for its performance at the polls." 13 Two devices in Puerto Rico's Constitution ensure representation of minority parties in the Puerto Rico Legislative Assembly. First, 11 of 27 senators and 11 of 51 representatives are elected "at-large," and each voter may vote for only one candidate for senator or representative at-large. Art. III, § 3. Second, if any one party elects more than two-thirds of the members of either house of the legislature, the number of members in that house is increased by declaring elected a sufficient number of minority-party candidates to bring the total number of minority-party members to 9 in the Senate and to 17 in the House. Art. III, § 7. Appellees maintain that "the Commonwealth's unique guarantee of minority party representation . . . is and has been particularly important in Puerto Rico, far beyond its importance in any State of the Union, in order to provide a democratic forum and an outlet for the radically different views of the various political parties as to the ultimate status of Puerto Rico. . . ." Brief for Appellees 14. 14 Puerto Rico is in no sense unique in maintaining continuity of party representation between elections; 16 States have chosen to require that legislative vacancies be filled by appointment of a person affiliated with the same party as the previous incumbent, or by designation of that party, see n.4, supra.

12

457 U.S. 31 102 S.Ct. 2211 72 L.Ed.2d 652 Delbert Lee TIBBS, Petitionerv.FLORIDA. No. 81-5114. Argued March 2, 1982. Decided June 7, 1982. Syllabus Held : Where the Florida Supreme Court's reversal of petitioner's murder and rape convictions at a jury trial was based on the weight of the evidence, a retrial is not barred by the Double Jeopardy Clause of the Fifth Amendment as made applicable to the States by the Due Process Clause of the Fourteenth Amendment. Pp. 39-47. (a) A reversal of a conviction based on the weight of the evidence, unlike a reversal based on insufficient evidence where the Double Jeopardy Clause precludes a retrial, Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1; Greene v. Massey, 437 U.S. 19, 98 S.Ct. 2151, 57 L.Ed.2d 15, does not mean that acquittal was the only proper verdict. Instead, the appellate court sits as a "thirteenth juror" and disagrees with the jury's resolution of the conflicting testimony. Just as a deadlocked jury does not result in an acquittal barring retrial under the Double Jeopardy Clause, an appellate court's disagreement with the jurors' weighing of the evidence does not require the special deference accorded verdicts of acquittal. Moreover, a reversal based on the weight of the evidence can occur only after the State has presented sufficient evidence to support conviction and has persuaded the jury to convict. The reversal simply affords the defendant a second opportunity to seek an acquittal. Giving him this second chance does not amount to governmental oppression of the sort against which the Double Jeopardy Clause was intended to protect. Pp. 39-44. (b) There is no merit to petitioner's arguments that a distinction between the weight and sufficiency of the evidence is unworkable and will undermine the Burks rule by encouraging appellate judges to base reversals on the weight, rather than the sufficiency, of the evidence. Pp. 44-45. 397 So.2d 1120, affirmed. Louis R. Beller, Miami Beach, Fla., for petitioner. Michael A. Palecki, Tampa, Fla., for respondent. Justice O'CONNOR delivered the opinion of the Court. 1 We granted certiorari to decide whether the Double Jeopardy Clause1 bars retrial after a state appellate court sets aside a conviction on the ground that the verdict was against "the weight of the evidence." After examining the policies supporting the Double Jeopardy Clause, we hold that a reversal based on the weight, rather than the sufficiency, of the evidence permits the State to initiate a new prosecution. 2 * In 1974, Florida indicted petitioner Delbert Tibbs for the first-degree murder of Terry Milroy, the felony murder of Milroy, and the rape of Cynthia Nadeau. Nadeau, the State's chief trial witness, testified that she and Milroy were hitchhiking from St. Petersburg to Marathon, Fla., on February 3, 1974. A man in a green truck picked them up near Fort Myers and, after driving a short way, turned off the highway into a field. He asked Milroy to help him siphon gas from some farm machinery, and Milroy agreed. When Nadeau stepped out of the truck a few minutes later, she discovered the driver holding a gun on Milroy. The driver told Milroy that he wished to have sex with Nadeau, and ordered her to strip. After forcing Nadeau to engage in sodomy, the driver agreed that Milroy could leave. As Milroy started to walk away, however, the assailant shot him in the shoulder. When Milroy fell to the ground, pleading for his life, the gunman walked over and taunted, "Does it hurt, boy? You in pain? Does it hurt, boy?" Tr. 508. Then, with a shot to the head, he killed Milroy. 3 This deed finished, the killer raped Nadeau. Fearing for her life, she suggested that they should leave together and that she "would be his old lady." Id., at 510. The killer seemed to agree and they returned to the highway in the truck. After driving a short distance, he stopped the truck and ordered Nadeau to walk directly in front of it. As soon as her feet hit the ground, however, she ran in the opposite direction. The killer fled with the truck, frightened perhaps by an approaching car. When Nadeau reached a nearby house, the occupants let her in and called the police. 4 That night, Nadeau gave the police a detailed description of the assailant and his truck. Several days later a patrolman stopped Tibbs, who was hitchhiking near Ocala, Fla., because his appearance matched Nadeau's description. The Ocala Police Department photographed Tibbs and relayed the pictures to the Fort Myers police. When Nadeau examined these photos, she identified Tibbs as the assailant.2 Nadeau subsequently picked Tibbs out of a lineup and positively identified him at trial as the man who murdered Milroy and raped her.3 5 Tibbs' attorney attempted to show that Nadeau was an unreliable witness. She admitted during cross-examination that she had tried "just about all" types of drugs and that she had smoked marihuana shortly before the crimes occurred. Id., at 526, 545-546. She also evidenced some confusion about the time of day that the assailant had offered her and Milroy a ride. Finally, counsel suggested through questions and closing argument that Nadeau's former boyfriend had killed Milroy and that Nadeau was lying to protect her boyfriend. Nadeau flatly denied these suggestions.4 6 In addition to these attempts to discredit Nadeau, Tibbs testified in his own defense. He explained that he was college educated, that he had published a story and a few poems, and that he was hitchhiking through Florida to learn more about how people live. He claimed that he was in Daytona Beach, across the State from Fort Myers, from the evening of February 1, 1974, through the morning of February 6. He also testified that he did not own a green truck, and that he had not driven any vehicle while in Florida. Finally, he denied committing any of the crimes charged against him. 7 Two Salvation Army officers partially corroborated Tibbs' story. These officers produced a card signed by Tibbs, indicating that he had slept at the Daytona Beach Salvation Army Transit Lodge on the evening of February 1, 1974. Neither witness, however, had seen Tibbs after the morning of February 2. Tibbs' other witnesses testified to his good reputation as a law-abiding citizen and to his good reputation for veracity. 8 On rebuttal, the State produced a card, similar to the one introduced by Tibbs, showing that Tibbs had spent the night of February 4 at the Orlando Salvation Army Transit Lodge. This evidence contradicted Tibbs' claim that he had remained in Daytona Beach until February 6, as well as his sworn statements that he had been in Orlando only once, during the early part of January 1974, and that he had not stayed in any Salvation Army lodge after February 1. After the State presented this rebuttal evidence, Tibbs took the stand to deny both that he had been in Orlando on February 4 and that the signature on the Orlando Salvation Army card was his. 9 The jury convicted Tibbs of first-degree murder and rape. Pursuant to the jury's recommendation, the judge sentenced Tibbs to death. On appeal, the Florida Supreme Court reversed. Tibbs v. State, 337 So.2d 788 (1976) (Tibbs I ). A plurality of three justices, while acknowledging that "the resolution of factual issues in a criminal trial is peculiarly within the province of a jury," id., at 791, identified six weaknesses in the State's case.5 First, except for Nadeau's testimony, the State introduced no evidence placing Tibbs in or near Fort Myers on the day of the crimes. Second, although Nadeau gave a detailed description of the assailant's truck, police never found the vehicle. Third, police discovered neither a gun nor car keys in Tibbs' possession. Fourth, Tibbs cooperated fully with the police when he was stopped and arrested. Fifth, the State introduced no evidence casting doubt on Tibbs' veracity.6 Tibbs, on the other hand, produced witnesses who attested to his good reputation. Finally, several factors undermined Nadeau's believability. Although she asserted at trial that the crimes occurred during daylight, other evidence suggested that the events occurred after nightfall when reliable identification would have been more difficult. Nadeau, furthermore, had smoked marihuana shortly before the crimes and had identified Tibbs during a suggestive photograph session.7 These weaknesses left the plurality in "considerable doubt that Delbert Tibbs [was] the man who committed the crimes for which he ha[d] been convicted." Id., at 790. Therefore, the plurality concluded that the "interests of justice" required a new trial. Ibid.8 10 Justice Boyd concurred specially, noting that " '[t]he test to be applied in determining the adequacy of a verdict is whether a jury of reasonable men could have returned that verdict.' " Id., at 792 (quoting Griffis v. Hill, 230 So.2d 143, 145 (Fla.1969)). Apparently applying that standard, Justice Boyd found the State's evidence deficient. He concluded that "the weakness of the evidence presented in the trial court might well require that [Tibbs] be released from incarceration without further litigation," but "reluctantly concur[red]" in the plurality's decision to order a new trial because he understood Florida law to permit retrial. 337 So.2d, at 792.9 11 On remand, the trial court dismissed the indictment, concluding that retrial would violate the double jeopardy principles articulated in Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978), and Greene v. Massey, 437 U.S. 19, 98 S.Ct. 2151, 57 L.Ed.2d 15 (1978).10 An intermediate appellate court disagreed and remanded the case for trial. 370 So.2d 386 (Fla.App.1979). The Florida Supreme Court affirmed the latter decision, carefully elaborating the difference between a reversal stemming from insufficient evidence and one prompted by the weight of the evidence. 397 So.2d 1120 (1981) (per curiam ) (Tibbs II ). As the court explained, a conviction rests upon insufficient evidence when, even after viewing the evidence in the light most favorable to the prosecution, no rational factfinder could have found the defendant guilty beyond a reasonable doubt. A reversal based on the weight of the evidence, on the other hand, draws the appellate court into questions of credibility. The "weight of the evidence" refers to "a determination [by] the trier of fact that a greater amount of credible evidence supports one side of an issue or cause than the other." Id., at 1123.11 12 The Florida Supreme Court then classified Tibbs I as a reversal resting on the weight of the evidence. Nadeau's testimony, if believed by the jury, was itself "legally sufficient to support Tibbs' conviction under Florida law." 397 So.2d, at 1126. In deciding to upset Tibbs' conviction, the court in Tibbs had stressed those "aspects of Nadeau's testimony which cast serious doubt on her believability," 397 So.2d, at 1126, an approach that bespoke a reweighing of the evidence. "Only by stretching the point . . .," the court concluded in Tibbs II, "could we possibly use an 'insufficiency' analysis to characterize our previous reversal of Tibbs' convictions." Ibid.12 13 Having found that it could not "fairly conclude . . . that Tibbs' convictions were reversed on the grounds of evidentiary insufficiency," id., at 1127, the Florida Supreme Court held that Greene and Burks do not bar retrial. Those decisions, the court believed, as well as United States v. DiFrancesco, 449 U.S. 117, 101 S.Ct. 426, 66 L.Ed.2d 328 (1980), interpret the Double Jeopardy Clause to preclude retrial after reversal of a conviction only when the appellate court has set the conviction aside on the ground that the evidence was legally insufficient to support conviction. Other reversals, including those based on the weight of the evidence or made in the "interests of justice," do not implicate double jeopardy principles.13 We granted certiorari to review this interpretation of the Double Jeopardy Clause. 454 U.S. 963, 102 S.Ct. 502, 70 L.Ed.2d 378 (1981). II 14 In 1896, this Court ruled that a criminal defendant who successfully appeals a judgment against him "may be tried anew . . . for the same offence of which he had been convicted." United States v. Ball, 163 U.S. 662, 672, 16 S.Ct. 1192, 1195, 41 L.Ed. 300. This principle, that the Double Jeopardy Clause "imposes no limitations whatever upon the power to retry a defendant who has succeeded in getting his first conviction set aside," North Carolina v. Pearce, 395 U.S. 711, 720, 89 S.Ct. 2072, 2078, 23 L.Ed.2d 656 (1969), has persevered to the present. See United States v. DiFrancesco, supra, at 131, 101 S.Ct., at 434; United States v. Scott, 437 U.S. 82, 89-92, 98 S.Ct. 2187, 2193-94, 57 L.Ed.2d 65 (1978). Two considerations support the rule. First, the Court has recognized that society would pay too high a price "were every accused granted immunity from punishment because of any defect sufficient to constitute reversible error in the proceedings leading to conviction." United States v. Tateo, 377 U.S. 463, 466, 84 S.Ct. 1587, 1589, 12 L.Ed.2d 448 (1964). Second, the Court has concluded that retrial after reversal of a conviction is not the type of governmental oppression targeted by the Double Jeopardy Clause. United States v. Scott, supra, at 91, 98 S.Ct., at 2193. See generally United States v. DiFrancesco, supra, at 131, 101 S.Ct., at 434.14 15 Burks v. United States and Greene v. Massey carved a narrow exception from the understanding that a defendant who successfully appeals a conviction is subject to retrial. In those cases, we held that the Double Jeopardy Clause precludes retrial "once the reviewing court has found the evidence legally insufficient" to support conviction. Burks, 437 U.S., at 18, 98 S.Ct., at 2150; Greene, 437 U.S., at 24, 98 S.Ct., at 2154. This standard, we explained, "means that the government's case was so lacking that it should not have even been submitted to the jury." Burks, 437 U.S., at 16, 98 S.Ct., at 2149 (emphasis in original). A conviction will survive review, we suggested, whenever "the evidence and inferences therefrom most favorable to the prosecution would warrant the jury's finding the defendant guilty beyond a reasonable doubt." Ibid. See also Greene, supra, at 25, 98 S.Ct., at 2154. In sum, we noted that the rule barring retrial would be "confined to cases where the prosecution's failure is clear." Burks, supra, at 17, 98 S.Ct., at 2150. 16 So defined, the exception recognized in Burks and Greene rests upon two closely related policies. First, the Double Jeopardy Clause attaches special weight to judgments of acquittal.15 A verdict of not guilty, whether rendered by the jury or directed by the trial judge, absolutely shields the defendant from retrial.16 A reversal based on the insufficiency of the evidence has the same effect because it means that no rational factfinder could have voted to convict the defendant. 17 Second, Burks and Greene implement the principle that "[t]he Double Jeopardy Clause forbids a second trial for the purpose of affording the prosecution another opportunity to supply evidence which it failed to muster in the first proceeding." Burks, supra, at 11, 98 S.Ct., at 2147. This prohibition, lying at the core of the Clause's protections, prevents the State from honing its trial strategies and perfecting its evidence through successive attempts at conviction. Repeated prosecutorial sallies would unfairly burden the defendant and create a risk of conviction through sheer governmental perseverance. See Green v. United States, 355 U.S. 184, 187-188, 78 S.Ct. 221, 223, 2 L.Ed.2d 199 (1957); United States v. DiFrancesco, 449 U.S., at 130, 101 S.Ct., at 433. For this reason, when a reversal rests upon the ground that the prosecution has failed to produce sufficient evidence to prove its case, the Double Jeopardy Clause bars the prosecutor from making a second attempt at conviction. 18 As we suggested just last Term, these policies do not have the same force when a judge disagrees with a jury's resolution of conflicting evidence and concludes that a guilty verdict is against the weight of the evidence. See Hudson v. Louisiana, 450 U.S. 40, 44-45, n. 5, 101 S.Ct. 970, 972-973, n. 5, 67 L.Ed.2d 30 (1981). A reversal on this ground, unlike a reversal based on insufficient evidence, does not mean that acquittal was the only proper verdict. Instead, the appellate court sits as a "thirteenth juror" and disagrees with the jury's resolution of the conflicting testimony. This difference of opinion no more signifies acquittal than does a disagreement among the jurors themselves. A deadlocked jury, we consistently have recognized, does not result in an acquittal barring retrial under the Double Jeopardy Clause.17 Similarly, an appellate court's disagreement with the jurors' weighing of the evidence does not require the special deference accorded verdicts of acquittal. 19 A reversal based on the weight of the evidence, moreover, can occur only after the State both has presented sufficient evidence to support conviction and has persuaded the jury to convict. The reversal simply affords the defendant a second opportunity to seek a favorable judgment.18 An appellate court's decision to give the defendant this second chance does not create "an unacceptably high risk that the Government, with its superior resources, [will] wear down [the] defendant" and obtain conviction solely through its persistence. United States v. DiFrancesco, supra, at 130, 98 S.Ct., at 433.19 20 While an appellate ruling based on the weight of the evidence thus fails to implicate the policies supporting Burks and Greene, it does involve the usual principles permitting retrial after a defendant's successful appeal. Just as the Double Jeopardy Clause does not require society to pay the high price of freeing every defendant whose first trial was tainted by prosecutorial error, it should not exact the price of immunity for every defendant who persuades an appellate panel to overturn an error-free conviction and give him a second chance at acquittal. Giving the defendant this second opportunity, when the evidence is sufficient to support the first verdict, hardly amounts to "governmental oppression of the sort against which the Double Jeopardy Clause was intended to protect." United States v. Scott, 437 U.S., at 91, 98 S.Ct., at 2193. 21 Petitioner Tibbs resists these arguments on the grounds that a distinction between the weight and the sufficiency of the evidence is unworkable and that such a distinction will undermine the Burks rule by encouraging appellate judges to base reversals on the weight, rather than the sufficiency, of the evidence. We find these arguments unpersuasive for two reasons. First, trial and appellate judges commonly distinguish between the weight and the sufficiency of the evidence.20 We have no reason to believe that today's decision will erode the demonstrated ability of judges to distinguish legally insufficient evidence from evidence that rationally supports a verdict. 22 Second, our decision in Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979), places some restraints on the power of appellate courts to mask reversals based on legally insufficient evidence as reversals grounded on the weight of the evidence. We held in Jackson that the Due Process Clause forbids any conviction based on evidence insufficient to persuade a rational factfinder of guilt beyond a reasonable doubt. The Due Process Clause, in other words, sets a lower limit on an appellate court's definition of evidentiary sufficiency.21 This limit, together with our belief that state appellate judges faithfully honor their obligations to enforce applicable state and federal laws, persuades us that today's ruling will not undermine Burks. In sum, we conclude that the Double Jeopardy Clause does not prevent an appellate court from granting a convicted defendant an opportunity to seek acquittal through a new trial.22 III 23 We turn, finally, to apply the above principles to the present case. A close reading of Tibbs I suggests that the Florida Supreme Court overturned Tibbs' conviction because the evidence, although sufficient to support the jury's verdict, did not fully persuade the court of Tibbs' guilt. The plurality based its review on a Florida rule directing the court in capital cases to "review the evidence to determine if the interests of justice require a new trial, whether the insufficiency of the evidence is a ground of appeal or not." See n.8, supra. References to the "interests of justice" and the justices' own "considerable doubt" of Tibbs' guilt mark the plurality's conclusions.23 Those conclusions, moreover, stem from the justices' determination that Tibbs' testimony was more reliable than that of Nadeau. This resolution of conflicting testimony in a manner contrary to the jury's verdict is a hallmark of review based on evidentiary weight, not evidentiary sufficiency. 24 Any ambiguity in Tibbs I, finally, was resolved by the Florida Supreme Court in Tibbs II. Absent a conflict with the Due Process Clause, see n.21, supra, that court's construction of its prior opinion binds this Court.24 In Tibbs II, of course, the court unequivocably held that Tibbs I was "one of those rare instances in which reversal was based on evidentiary weight." 397 So.2d, at 1126 (per curiam ). Thus, we conclude that Tibbs' successful appeal of his conviction rested upon a finding that the conviction was against the weight of the evidence, not upon a holding that the evidence was legally insufficient to support the verdict. Under these circumstances, the Double Jeopardy Clause does not bar retrial. Accordingly, the judgment of the Florida Supreme Court is 25 Affirmed. 26 Justice WHITE, with whom Justice BRENNAN, Justice MARSHALL, and Justice BLACKMUN join, dissenting. 27 As our cases in this area indicate, the meaning of the Double Jeopardy Clause is not always readily apparent. See, e.g., Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978) (overruling Bryan v. United States, 338 U.S. 552, 70 S.Ct. 317, 94 L.Ed. 335 (1950), Sapir v. United States, 348 U.S. 373, 75 S.Ct. 422, 99 L.Ed. 426 (1955), and Forman v. United States, 361 U.S. 416, 80 S.Ct. 481, 4 L.Ed.2d 412 (1960)); United States v. Scott, 437 U.S. 82, 98 S.Ct. 2187, 57 L.Ed.2d 65 (1978) (overruling United States v. Jenkins, 420 U.S. 358, 95 S.Ct. 1006, 43 L.Ed.2d 250 (1975)). For this reason, we should begin with a clear understanding of what is at stake in this case. 28 To sustain the convictions in this case, the prosecution was required to convince the Florida Supreme Court not only that the evidence was sufficient under the federal constitutional standard announced in Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979), but also that as a matter of state law, the verdict was not against the weight of the evidence. The Florida Supreme Court found the verdict to be against the weight of the evidence, thus holding that as a matter of state law the prosecution failed to present evidence adequate to sustain the convictions. Were the State to present this same evidence again, we must assume that once again the state courts would reverse any conviction that was based upon it.* The State was not prevented from presenting its best case because of some incorrect procedural ruling by the trial court; rather, the State had a full opportunity to present its case, but that case was not adequate as a matter of state law. If the State presents no new evidence, the defendant has no new or additional burden to meet in successfully presenting a defense: He may stand on, i.e., repeat, what he has already presented. Thus, the only point of any second trial in this case is to allow the State to present additional evidence to bolster its case. If it does not have such evidence, reprosecution can serve no purpose other than harassment. The majority holds that reprosecution under these circumstances does not offend the double jeopardy provision of the Constitution. I do not agree. 29 The majority concedes, as it must under Burks, supra, that if the State's evidence failed to meet the federal due process standard of evidentiary sufficiency, the Double Jeopardy Clause would bar reprosecution. The majority fails to explain why the State should be allowed another try where its proof has been held inadequate on state-law grounds, when it could not do so were it inadequate on federal-law grounds. In both cases the State has failed to present evidence adequate to sustain the conviction. The interests of the State in overcoming the evidentiary insufficiencies of its case would seem to be exactly the same in the two cases; the interests of the defendant in avoiding a second trial would also seem to be exactly the same in each case. Yet the majority holds that the Double Jeopardy Clause leads to different results in the two instances. 30 The majority offers two arguments in its attempt to distinguish the two cases. First, it emphasizes that the Double Jeopardy Clause "attaches special weight to judgments of acquittal." But in neither of the situations posited has there been a judgment of acquittal by the initial factfinder. In each instance, a reviewing court decides that, as a matter of law, the decision of the factfinder cannot stand. Second, the majority thinks it to be of some significance that when the evidence is determined to be insufficient as a matter of federal law, then no rational factfinder could have voted to convict on that basis. On the other hand, when the conviction is reversed on the basis of the state-law rule applying a "weight of the evidence" test, that "does not mean that acquittal was the only proper verdict." Ante, at 42. The constraints of the Double Jeopardy Clause, however, do not depend upon a determination that an "acquittal was the only proper verdict." The fact remains that the State failed to prove the defendant guilty in accordance with the evidentiary requirements of state law. 31 The majority opinion rests finally on a mischaracterization of the appellate court's ruling: "The reversal simply affords the defendant a second opportunity to seek a favorable judgment." Ante, at 43. But as I described above, it is not the defendant who has the burden of coming up with a new case on retrial; it is the prosecution. The defendant has already demonstrated that a conviction based on the State's case, as so far developed, is "against the weight of the evidence." 32 Having concluded that the majority opinion fails to justify the distinction it draws, I too turn to "the policies supporting the Double Jeopardy Clause," ante, at 32, to determine whether this distinction is relevant. I do not believe it necessary to look beyond the articulation of those policies in the majority opinion itself to conclude that it is not: 33 "Burks and Greene [v. Massey, 437 U.S. 19, 98 S.Ct. 2151, 57 L.Ed.2d 15 (1978)] implement the principle that '[t]he Double Jeopardy Clause forbids a second trial for the purpose of affording the prosecution another opportunity to supply evidence which it failed to muster in the first proceeding.' This prohibition, lying at the core of the Clause's protections, prevents the State from honing its trial strategies and perfecting its evidence through successive attempts at conviction. Repeated prosecutorial sallies would unfairly burden the defendant and create a risk of conviction through sheer governmental perseverance." Ante, at 41 (citations omitted). 34 These same policy considerations are at stake when a conviction is reversed on state-law grounds going to the adequacy of the evidence. The relevant question is whether the reversal is " 'due to a failure of proof at trial' where the State received a 'fair opportunity to offer whatever proof it could assemble.' " Hudson v. Louisiana, 450 U.S. 40, 43, 101 S.Ct. 970, 972, 67 L.Ed.2d 30 (1981) (quoting Burks, 437 U.S., at 16, 98 S.Ct., at 2149). That the proof fails on state-law, rather than federal-law, grounds is immaterial to these policy considerations. Thus, the relevant distinction is between reversals based on evidentiary grounds and those based on procedural grounds: Only in the latter case can the State proceed to retrial without offending the deeply ingrained principle that "the State with all its resources and power should not be allowed to make repeated attempts to convict an individual for an alleged offense." Green v. United States, 355 U.S. 184, 187, 78 S.Ct. 221, 223, 2 L.Ed.2d 199 (1957). 35 It must also be noted that judges having doubts about the sufficiency of the evidence under the Jackson standard may prefer to reverse on the weight of the evidence, since retrial would not be barred. If done recurringly, this would undermine Jackson, Burks, and Greene. But under Burks and Greene, retrial is foreclosed by the Double Jeopardy Clause if the evidence fails to satisfy the Jackson standard. Hence, the Jackson issue cannot be avoided; if retrial is to be had, the evidence must be found to be legally sufficient, as a matter of federal law, to sustain the jury verdict. That finding must accompany any reversal based on the weight of the evidence if retrial is contemplated. The upshot may be that appellate judges will not be inclined to proclaim the evidence in a case to be legally sufficient, yet go on to disagree with the jury and the trial court by reversing on weight-of-the-evidence grounds. Indeed, in this case, the Florida Supreme Court declared that prospect to be an anomaly and a mistake and proclaimed that it would never again put itself in this position. 36 With all due respect, I dissent. 1 "[N]or shall any person be subject for the same offence to be twice put in jeopardy of life or limb. . . ." U.S.Const., Amdt. 5. The Clause applies to the States through the Due Process Clause of the Fourteenth Amendment. Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969). 2 The State's witnesses conceded that, at the time of this identification, Nadeau saw only photographs of Tibbs; she did not have the opportunity to pick his picture out of a photographic array. An officer explained, however, that Nadeau had viewed photographs of single suspects on three or four other occasions and had not identified the killer on any of those occasions. Nadeau also had examined several books of photographs without making an identification. We do not pass upon any possible due process questions raised by the State's identification procedures, see generally Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972); Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968), because Tibbs' challenge to retrial rests solely upon double jeopardy grounds. 3 The State's remaining witnesses included law enforcement agents, a man who had driven Milroy and Nadeau to Fort Myers, the houseowner who had called the police for Nadeau, acquaintances of Milroy, a doctor who had examined Nadeau shortly after the crimes, and the doctor who had performed the autopsy on Milroy. The doctors confirmed that Nadeau had had intercourse on the evening of February 3 and that Milroy had died that evening from a bullet wound in the head. The other witnesses confirmed that Nadeau and Milroy had been hitchhiking through Fort Myers on February 3 and that Nadeau had arrived at a house, in a hysterical condition, that evening. A Florida prisoner, sentenced to life imprisonment for rape, also testified for the State. This prisoner claimed that he had met Tibbs while Tibbs was in jail awaiting trial and that Tibbs had confessed the crime to him. The defense substantially discredited this witness on cross-examination, revealing inconsistencies in his testimony and suggesting that he had testified in the hope of obtaining leniency from the State. 4 The results of two polygraph examinations, described in a report read to the jury, indicated that Nadeau was "truthful as to the fact that a black male driving a green pickup truck had picked them up and that this black male had murdered Terry Milroy," Tr. 302. The polygraphs also suggested that Nadeau was truthful when she identified Tibbs as the assailant. Id., at 303. Tibbs challenged the admissibility of these polygraphs during his first appeal. See Tibbs v. State, 337 So.2d 788, 796 (Fla.1976) (Roberts, J., dissenting). The justices who voted to reverse Tibbs' conviction, however, did not reach the issue and we express no opinion on this matter of state law. 5 The plurality completely discounted the testimony of the convicted rapist who recounted Tibbs' alleged confession. See n.3, supra. This testimony, the justices concluded, appeared "to be the product of purely selfish considerations." 337 So.2d, at 790. 6 The plurality opinion summarily dismissed the effect of the rebuttal evidence showing that Tibbs was in Orlando on February 4. A "superficial comparison" of the signature on the Orlando transit card with Tibbs' own signature, the plurality found, supported Tibbs' claim that he had not signed the card. Moreover, evidence that Tibbs was in Orlando on February 4 still did not place him in Fort Myers on February 3. Id., at 790, n.1. 7 See n.2, supra. 8 At the time of Tibbs' first appeal, Florida Appellate Rule 6.16(b) (1962) provided in part: "Upon an appeal from the judgment by a defendant who has been sentenced to death the appellate court shall review the evidence to determine if the interests of justice require a new trial, whether the insufficiency of the evidence is a ground of appeal or not." The substance of this Rule has been recodified as Florida Appellate Rule 9.140(f). 9 At two points, Justice Boyd stated that he "concur[red] in the majority opinion." 337 So.2d, at 792. However, because we are uncertain what weight Florida attaches to special concurrences of this sort and because Justice Boyd's views differed from those of the other justices voting to reverse, we have chosen to designate the lead opinion a "plurality" opinion. Three justices dissented from the court's disposition of Tibbs' appeal. They declared that "the evidence in the record before us does not reveal that the ends of justice require that a new trial be awarded," id., at 796-797, and rejected Tibbs' other assignments of error. 10 We decided Burks and Greene after the Florida Supreme Court reversed Tibbs' conviction, but before he could be retried. We have applied Burks to prosecutions that were not yet final on the date of that decision. See Hudson v. Louisiana, 450 U.S. 40, 101 S.Ct. 970, 67 L.Ed.2d 30 (1981). 11 Other courts similarly have explained the difference between evidentiary weight and evidentiary sufficiency. In United States v. Lincoln, 630 F.2d 1313 (CA8 1980), for example, the court declared: "The court reviewing the sufficiency of the evidence, whether it be the trial or appellate court, must apply familiar principles. It is required to view the evidence in the light most favorable to the verdict, giving the prosecution the benefit of all inferences reasonably to be drawn in its favor from the evidence. The verdict may be based in whole or in part on circumstantial evidence. The evidence need not exclude every reasonable hypothesis except that of guilt . . . ." Id., at 1316. "When a motion for new trial is made on the ground that the verdict is contrary to the weight of the evidence, the issues are far different . . . . The district court need not view the evidence in the light most favorable to the verdict; it may weigh the evidence and in so doing evaluate for itself the credibility of the witnesses. If the court concludes that, despite the abstract sufficiency of the evidence to sustain the verdict, the evidence preponderates sufficiently heavily against the verdict that a serious miscarriage of justice may have occurred, it may set aside the verdict, grant a new trial, and submit the issues for determination by another jury." Id., at 1319. See generally 2 C. Wright, Federal Practice and Procedure § 553 (1969). 12 Elsewhere in its opinion, the Florida Supreme Court ruled that Florida appellate courts no longer may reverse convictions on the ground that the verdict was against the weight of the evidence. 397 So.2d, at 1125. This ruling does not diminish the importance of the issue before us. Courts in other jurisdictions sometimes rely upon the weight of the evidence to overturn convictions. For example, some federal courts have interpreted Rule 33 of the Federal Rules of Criminal Procedure, which authorizes a new trial "if required in the interest of justice," to permit the trial judge to set aside a conviction that is against the weight of the evidence. E.g., United States v. Lincoln, supra, at 1319; United States v. Indelicato, 611 F.2d 376, 387 (CA1 1979); United States v. Turner, 490 F.Supp. 583, 593 (ED Mich. 1979), affirmance order, 633 F.2d 219 (CA6 1980), cert. denied, 450 U.S. 912, 101 S.Ct. 1351, 67 L.Ed.2d 336 (1981); United States v. Felice, 481 F.Supp. 79, 90-91 (ND Ohio 1978). 13 Three justices dissented from the court's decision to permit Tibbs' retrial. Chief Justice Sundberg suggested that the reversal in Tibbs I must have rested upon a finding of evidentiary insufficiency, because the Florida Supreme Court lacked authority to reweigh the evidence. He also rejected the majority's distinction between evidentiary weight and evidentiary sufficiency, proposing that the Double Jeopardy Clause should bar retrial whenever an appellate court reverses "for a substantive lack of evidence to support the verdict." 397 So.2d, at 1128. Justice England merely stated that he would discharge Tibbs "in the interest of justice." Id., at 1130. Justice Boyd concluded that Tibbs I had rested on a finding of evidentiary insufficiency and, accordingly, that Tibbs "should be forever discharged from the accusations made against him." 397 So.2d, at 1131. 14 The rule also appears to coincide with the intent of the Fifth Amendment's drafters. James Madison's proposed version of the Double Jeopardy Clause provided that "[n]o person shall be subject, except in cases of impeachment, to more than one punishment or one trial for the same offence." 1 Annals of Cong. 434 (1789). Several Representatives objected that this language might prevent a defendant from seeking a new trial after conviction. Representative Sherman, for example, observed that "[i]f the [defendant] was acquitted on the first trial, he ought not to be tried a second time; but if he was convicted on the first, and any thing should appear to set the judgment aside, he was entitled to a second, which was certainly favorable to him." Id., at 753. Madison's supporters explained that the language would not prevent a convicted defendant from seeking a new trial, and the House approved Madison's proposal. Ibid. The Senate later substituted the language appearing in the present Clause. S.Jour., 1st Cong., 1st Sess., 71, 77 (1820 ed.). See generally United States v. Wilson, 420 U.S. 332, 340-342, 95 S.Ct. 1013, 1020-21, 43 L.Ed.2d 232 (1975); Sigler, A History of Double Jeopardy, 7 Am. J. Legal Hist. 283, 304-306 (1963). 15 See United States v. DiFrancesco, 449 U.S. 117, 129, 101 S.Ct. 426, 433, 66 L.Ed.2d 328 (1980); United States v. Scott, 437 U.S. 82, 91, 98 S.Ct. 2187, 2193, 57 L.Ed.2d 65 (1978); Arizona v. Washington, 434 U.S. 497, 503, 98 S.Ct. 824, 829, 54 L.Ed.2d 717 (1978); United States v. Martin Linen Supply Co., 430 U.S. 564, 571, 97 S.Ct. 1349, 1354, 51 L.Ed.2d 642 (1977); Fong Foo v. United States, 369 U.S. 141, 143, 82 S.Ct. 671, 672, 7 L.Ed.2d 629 (1962) (per curiam ). 16 See, e.g., United States v. Martin Linen Supply Co., supra; United States v. Ball, 163 U.S. 662, 666-671, 16 S.Ct. 1192, 1193-1195, 41 L.Ed. 300 (1896). 17 See, e.g., Arizona v. Washington, supra, at 509, 98 S.Ct., at 832; United States v. Sanford, 429 U.S. 14, 16, 97 S.Ct. 20, 21, 50 L.Ed.2d 17 (1976) (per curiam ); Johnson v. Louisiana, 406 U.S. 356, 401-402, 92 S.Ct. 1620, 1652-53, 32 L.Ed.2d 152 (1972) (MARSHALL, J., dissenting); Downum v. United States, 372 U.S. 734, 735-736, 83 S.Ct. 1033, 1033-1034, 10 L.Ed.2d 100 (1963); Wade v. Hunter, 336 U.S. 684, 689, 69 S.Ct. 834, 837, 93 L.Ed. 974 (1949); Keerl v. Montana, 213 U.S. 135, 29 S.Ct. 469, 53 L.Ed. 734 (1909); Dreyer v. Illinois, 187 U.S. 71, 84-86, 23 S.Ct. 28, 32-33, 47 L.Ed. 79 (1902); Logan v. United States, 144 U.S. 263, 298, 12 S.Ct. 617, 628, 36 L.Ed. 429 (1892); United States v. Perez, 9 Wheat. 579, 2 U.S. 579, 6 L.Ed. 165 (1824). Our decisions also make clear that disagreements among jurors or judges do not themselves create a reasonable doubt of guilt. As Justice WHITE, writing for the Court in Johnson v. Louisiana, supra, explained, "[t]hat rational men disagree is not in itself equivalent to a failure of proof by the State, nor does it indicate infidelity to the reasonable-doubt standard." 406 U.S., at 362, 92 S.Ct., at 1624. 18 The dissent suggests that a reversal based on the weight of the evidence necessarily requires the prosecution to introduce new evidence on retrial. Once an appellate court rules that a conviction is against the weight of the evidence, the dissent reasons, it must reverse any subsequent conviction resting upon the same evidence. We do not believe, however, that jurisdictions endorsing the "weight of the evidence" standard apply that standard equally to successive convictions. In Florida, for example, the highest state court once observed that, although "[t]here is in this State no limit to the number of new trials that may be granted in any case, . . . it takes a strong case to require an appellate court to grant a new trial in a case upon the ground of insufficiency of conflicting evidence to support a verdict when the finding has been made by two juries." Blocker v. State, 92 Fla. 878, 893, 110 So. 547, 552 (1926) (en banc). The weight of the evidence rule, moreover, often derives from a mandate to act in the interests of justice. See nn. 8 and 12, supra. Although reversal of a first conviction based on sharply conflicting testimony may serve the interests of justice, reversal of a second conviction based on the same evidence may not. See United States v. Weinstein, 452 F.2d 704, 714, n. 14 (CA2 1971) ("We do not join in the . . . forecast that the granting of a new trial would doom the defendant and the Government to an infinite regression. . . . [I]f a third jury were to find [the defendant] guilty, we should suppose any judge would hesitate a long time before concluding that the interests of justice required still another trial"), cert. denied sub nom. Grunberger v. United States, 406 U.S. 917, 92 S.Ct. 1766, 32 L.Ed.2d 116 (1972). While the interests of justice may require an appellate court to sit once as a thirteenth juror, that standard does not compel the court to repeat the role. 19 A second chance for the defendant, of course, inevitably affords the prosecutor a second try as well. It is possible that new evidence or advance understanding of the defendant's trial strategy will make the State's case even stronger during a second trial than it was at the first. It is also possible, however, that the passage of time and experience of defense counsel will weaken the prosecutor's presentation. In this case, for example, more than eight years have elapsed since the crimes. Nadeau's ability to recall the events of February 3, 1974, may have diminished significantly, and a jury may be less willing to credit her identification of a man she saw almost a decade ago. When the State has secured one conviction based on legally sufficient evidence, it has everything to lose and little to gain by retrial. Thus, the type of "second chance" that the State receives when a court rests reversal on evidentiary weight does not involve the overreaching prohibited by the Double Jeopardy Clause. 20 See, e.g., United States v. Lincoln, 630 F.2d, at 1319; United States v. Weinstein, supra, at 714-716; United States v. Shipp, 409 F.2d 33, 36-37 (CA4), cert. denied, 396 U.S. 864, 90 S.Ct. 140, 24 L.Ed.2d 117 (1969); Dorman v. State, 622 P.2d 448, 453-454 (Alaska 1981); Ridley v. State, 236 Ga. 147, 149, 223 S.E.2d 131, 132 (1976); State v. McGranahan, --- R.I. ---, ---, 415 A.2d 1298, 1301-1303 (1980); Tyacke v. State, 65 Wis.2d 513, 521, 223 N.W.2d 595, 599 (1974). 21 The evidence in this case clearly satisfied the due process test of Jackson v. Virginia. As we stressed in Jackson, the reviewing court must view "the evidence in the light most favorable to the prosecution." 443 U.S., at 319, 99 S.Ct., at 2789. The trier of fact, not the appellate court, holds "the responsibility . . . fairly to resolve conflicts in the testimony, to weigh the evidence, and to draw reasonable inferences from basic facts to ultimate facts." Ibid. In this case, Nadeau provided eyewitness testimony to the crimes. If the jury believed her story, the State's presentation was more than sufficient to satisfy due process. 22 We note that a contrary rule, one precluding retrial whenever an appellate court rests reversal on evidentiary weight, might prompt state legislatures simply to forbid those courts to reweigh the evidence. Rulemakers willing to permit a new trial in the face of a verdict supported by legally sufficient evidence may be less willing to free completely a defendant convicted by a jury of his peers. Acceptance of Tibbs' double jeopardy theory might also lead to restrictions on the authority of trial judges to order new trials based on their independent assessment of evidentiary weight. Although Tibbs limits his argument to appellate reversals, his contentions logically apply to a trial judge's finding that a conviction was against the weight of the evidence. Cf. Hudson v. Louisiana, 450 U.S. 40, 101 S.Ct. 970, 67 L.Ed.2d 642 (1981) (applying Burks v. United States, 437 U.S. 1, 98 S.Ct. 2141, 57 L.Ed.2d 1 (1978), to trial judge's postverdict ruling that evidence was insufficient to support conviction). Endorsement of Tibbs' theory, therefore, might only serve to eliminate practices that help shield defendants from unjust convictions. 23 At one point, the opinion does refer to " 'evidence which is not sufficient to convince a fair and impartial mind of the guilt of the accused beyond a reasonable doubt.' " 337 So.2d, at 791 (quoting McNeil v. State, 104 Fla. 360, 361-362, 139 So. 791, 792 (1932)). This reference, however, occurs in a lengthy quotation from an earlier Florida decision. When read in context, it does not appear that the plurality actually applied this standard to the evidence in Tibbs' case. Moreover, the quotation containing this sufficiency language also speaks of evidence that is "not satisfactory" to the appellate court and that is not "substantial in character." Ibid. This language, in line with the remainder of Tibbs I, evidences a weighing of the evidence. 24 In Greene v. Massey, 437 U.S. 19, 98 S.Ct. 2151, 57 L.Ed.2d 15 (1978), we recognized that the meaning attached to an ambiguous prior reversal is a matter of state law. In that case, we remanded a double jeopardy issue to the Court of Appeals for the Fifth Circuit, directing the court to consider the effect under state law of several peculiarities in the state court's opinion. Id., at 25-26, and nn. 8-10, 98 S.Ct., at 2155, and nn. 8-10. We even suggested that the Court of Appeals might "direct further proceedings in the District Court or . . . certify unresolved questions . . . to the Florida Supreme Court" to resolve these problems of state law. Id., at 27, 98 S.Ct., at 2155. * Only Chief Justice Sundberg, concurring in part and dissenting in part, reached this issue below: "Since the same evidence must be used, an appellate court would have no choice but once again to reverse a conviction because of our reversal under identical circumstances." 397 So.2d 1120, 1130 (1981). Because the majority concluded that it would not in the future reverse convictions on grounds of evidentiary weight, it is not clear whether that court, were it presented with the exact same evidence in a Tibbs III, would follow its new rule and affirm or again reverse on "law of the case" grounds. I agree with the majority, however, that the peculiar procedural posture of this case does not affect our consideration of the issue because other jurisdictions, including the Federal Government, make use of a similar rule with respect to evidentiary weight.

01

457 U.S. 132 102 S.Ct. 2355 72 L.Ed.2d 728 Barbara BLUM, etc., Appellantv.Jeanne BACON, etc., et al. No. 81-770. Argued April 28, 1982. Decided June 14, 1982. Syllabus New York's Emergency Assistance (EA) Program, which is federally funded under the Social Security Act (SSA), precludes the furnishing of EA cash to persons receiving or eligible for Aid to Families with Dependent Children (AFDC) or of EA in any form to replace a lost or stolen AFDC grant. Appellees, who had been denied EA under these state provisions, brought a class action in Federal District Court to enjoin enforcement of the provisions, alleging that they conflicted with the SSA and violated equal protection. Ultimately, on remand after its decision invalidating the state provisions under the Supremacy Clause had been vacated by the Court of Appeals, the District Court invalidated the no-cash provision as a violation of equal protection but upheld the loss-or-theft provision. On A Second Appeal, The Court of appeals held that both provisions violated the Equal Protection Clause of the Fourteenth Amendment. Held : Because the New York provisions conflict with a valid federal regulation promulgated by the Secretary of Health, Education, and Welfare (Secretary) which proscribes inequitable treatment of individuals or groups under an EA program, they are invalid under the Supremacy Clause. Pp. 2359-2364. (a) Reliance on the SSA to find the New York provisions invalid is not foreclosed by Quern v. Mandley, 436 U.S. 725. While Quern emphasized that a State retains considerable flexibility in determining which emergencies to cover under its EA plan, it was not suggested that the Secretary was stripped of all authority to review a plan that arbitrarily or inequitably excluded a class of recipients. P. 2360. (b) The Secretary's decision to apply the "equitable treatment" regulation so as to forbid a State to exclude AFDC recipients from its EA program is eminently reasonable and deserves judicial deference, especially where the legislative history leaves no doubt that AFDC recipients were expected to be included in a state EA program. Pp. 2360-2363. 2nd Cir., 648 F.2d 801, affirmed. Robert S. Hammer, New York City, for appellant. Martin A. Schwartz, New York City, for appellees. Justice MARSHALL delivered the opinion of the Court. 1 New York has established an Emergency Assistance Program that receives substantial federal funding under Title IV-A of the Social Security Act (Act), 42 U.S.C. § 603(a)(5). The program excludes recipients of Aid to Families with Dependent Children (AFDC) from emergency assistance in the form of cash. It also excludes public assistance recipients (including AFDC recipients) from reimbursement for lost or stolen grants, even though it provides such reimbursement to other public benefit recipients. The United States Court of Appeals for the Second Circuit held that New York's treatment of AFDC recipients is not inconsistent with the federal Act and regulations but violates the Equal Protection Clause. Because we conclude that the New York law is invalid under the Act, we affirm without reaching the equal protection issue. 2 * Appellee Jeanne Bacon has two minor children and depends entirely on an AFDC grant to support her family. On June 1, 1977, while she was shopping, her wallet and food stamps were stolen. She promptly reported the theft to the police and to the New York Department of Social Services (DSS). She requested emergency assistance (EA) under the State's federally funded Emergency Assistance Program, explaining that she had no money to purchase food and other essential items for her household for the month. DSS denied her request on the basis of a recent state law which precludes the furnishing of any cash EA to persons receiving or eligible for AFDC, N.Y.Soc.Serv.Law §§ 350-j(2)(c) and (3) (McKinney Supp.1981) (the "no-cash" provision), or of EA in any form to replace a lost or stolen public assistance grant, including an AFDC grant. § 350-j(2)(e) (the "loss-or-theft" provision).1 Appellee Gertrude Parrish suffered a similar fate. An AFDC mother, she lost her food and AFDC funds when her apartment was broken into and ransacked. She applied for EA, and DSS denied her request on the same basis as it denied relief to appellee Bacon. The other named appellees, Linda Selders and Freddie Mae Goodwine, also were denied EA after they cashed their AFDC checks and suffered the loss of their money.2 3 Appellees brought this class action to enjoin enforcement of the state law insofar as it denies EA pursuant to the no-cash provision and the loss-or-theft provision.3 Appellees argued that the law conflicts with the Act and violates equal protection because it arbitrarily discriminates against AFDC recipients: it provides cash EA to all eligible recipients other than AFDC recipients, and provides EA for lost or stolen public benefit grants to all public benefit recipients (such as recipients of social security and Supplemental Security Income) other than those on public assistance (including AFDC recipients). 4 The United States District Court for the Southern District of New York granted summary judgment in favor of appellees on the ground that the state provisions impermissibly narrowed the eligibility standards imposed on state EA programs by § 406(e) of the Act, 42 U.S.C. § 606(e),4 and were invalid under the Supremacy Clause. Bacon v. Toia, 437 F.Supp. 1371 (1977). The United States Court of Appeals for the Second Circuit affirmed. Bacon v. Toia, 580 F.2d 1044 (1978). Shortly thereafter, this Court decided Quern v. Mandley, 436 U.S. 725, 98 S.Ct. 2068, 56 L.Ed.2d 658 (1978), in which we held that § 406(e) imposes permissive, not mandatory, standards on participating States. The Court of Appeals granted a motion for rehearing, vacated the judgment of the District Court, and remanded the case for further consideration in light of Quern. On remand, the District Court changed its prior decision and held that the New York law was not inconsistent with the federal Act. In a subsequent opinion, the District Court invalidated the no-cash provision as a violation of equal protection but upheld the loss-or-theft provision. Bacon v. Toia, 493 F.Supp. 865 (1980). On the second appeal, the Court of Appeals agreed with the District Court that our decision in Quern foreclosed a finding that the law violates the Supremacy Clause. The Court of Appeals concluded, however, that both the no-cash and loss-or-theft provisions violate equal protection. 648 F.2d 801 (1981). We noted probable jurisdiction. 1122 U.S. 454, 102 S.Ct. 969, 71 L.Ed.2d 109. II 5 Where a party raises both statutory and constitutional arguments in support of a judgment, ordinarily we first address the statutory argument in order to avoid unnecessary resolution of the constitutional issue. See Califano v. Yamasaki, 442 U.S. 682, 692-693, 99 S.Ct. 2545, 2553, 61 L.Ed.2d 176 (1979); Hagans v. Lavine, 415 U.S. 528, 543, 94 S.Ct. 1372, 1382, 39 L.Ed.2d 577 (1974).5 We conclude that this case may be resolved on statutory grounds. As we explain below, the New York no-cash and loss-or-theft rules conflict with valid federal regulations promulgated by the Secretary of Health, Education, and Welfare (Secretary) (now the Secretary of Health and Human Services) which proscribe inequitable treatment under the EA program. Thus, New York's rules are invalid under the Supremacy Clause. 6 Before reviewing the federal regulations that we find to be dispositive of this case, we first address appellant's claim that reliance on the Act is foreclosed by our decision in Quern v. Mandley, supra. In that case, we carefully reviewed the nature and scope of the EA program and examined one aspect of its relationship to the AFDC program.6 Under Title IV-A of the Act, state public assistance plans approved by the Secretary are eligible for federal financial assistance. AFDC is a major categorical aid program funded under the Act—indeed, it is "the core of the Title IV-A system." Id., at 728, 98 S.Ct., at 2071. States are required, as a condition of federal funding under the AFDC program, to make assistance available to all persons who meet statutory eligibility criteria. Id., at 740, 98 S.Ct., at 2077; 42 U.S.C. §§ 602(a)(10), 606(a). The EA program is a supplement to such categorical assistance programs as AFDC. It permits federal reimbursement to States which choose to provide for temporary emergency assistance in their Title IV-A plans. 42 U.S.C. § 603(a)(5). In contrast to AFDC, the EA program establishes much broader eligibility standards and is not limited to persons eligible for AFDC. 42 U.S.C. § 606(e). 7 Plaintiffs in Quern made the broad claim that a State participating in the federal EA program may not limit eligibility for EA more narrowly than the federal eligibility standards in § 406(e). The state plan at issue provided emergency assistance only to certain AFDC families who were without shelter and to applicants presumptively eligible for AFDC who were in immediate need of clothing or household furnishings. We rejected the plaintiffs' broad claim and held that unlike the AFDC program, § 406(e) establishes only permissive, not mandatory, eligibility standards. 8 Quern did not address the statutory issue before us today whether the complete and automatic exclusion of AFDC recipients from a State's EA program is inconsistent with the Act and applicable regulations. The Court had no occasion to consider the question, since the EA program in that case included only AFDC recipients. In addition, the only pertinent federal regulations in Quern undermined the plaintiffs' claims and supported the State's rules. See 436 U.S., at 743-744, n. 19, 98 S.Ct., at 2078-2079, n. 19; 45 CFR § 233.120 (1981). Here, on the other hand, the Secretary has promulgated a regulation inconsistent with New York's no-cash and loss-or-theft rules. See 45 CFR § 233.10 (1981); infra, at 2360-2362.7 In short, although we emphasized in Quern that a State retains considerable flexibility in determining which emergencies to cover under its EA plan, we hardly suggested that the Secretary had been stripped of all authority to review a plan that arbitrarily or inequitably excluded a class of recipients. B 9 The Secretary, who is charged with administering federal funding for EA under the Act, has promulgated the following regulation applicable to state plans under Title IV-A, including EA programs: 10 "(a) State plan requirements. A State plan under title I, IV-A, X, XIV, or XVI, of the Social Security Act must: "(1) Specify the groups of individuals, based on reasonable classifications, that will be included in the program, and all the conditions of eligibility that must be met by the individuals in the groups. The groups selected for inclusion in the plan and the eligibility conditions imposed must not exclude individuals or groups on an arbitrary or unreasonable basis, and must not result in inequitable treatment of individuals or groups in the light of the provisions and purposes of the public assistance titles of the Social Security Act." 45 CFR § 233.10 (1981). 11 The Secretary has also issued regulations exclusively addressed to the EA program. 45 CFR § 233.120 (1981).8 12 On the authority of these regulations, the Secretary has specifically required the inclusion of AFDC recipients in any EA program, and has disapproved New York's EA plan because it excludes AFDC recipients as a class. Shortly after this Court's decision in Quern, the Office of Family Assistance of the Social Security Administration issued Action Transmittal SSA-AT-78-44 (OFA) addressed to state agencies administering approved public assistance programs. The Transmittal explains that after Quern, "States remain free, under Federal policy to develop their own definition of the kind of emergencies they will meet under this program." App. 173a. Nevertheless, "[a] State Plan must clearly specify that AFDC recipients are included in its EA program. Other categories of needy families with children may be included at State option; these categories must be specified in the plan." Id., at 174a (emphasis added). In an amicus brief filed at the invitation of the Court of Appeals below, the Secretary confirmed that New York's exclusion of AFDC recipients through its no-cash and loss-or-theft provisions violates federal regulations, in particular the "equitable treatment" regulation, 45 CFR § 233.10 (1981). App. to Motion to Affirm 12a-17a. As the Secretary interpreted that regulation, the discrimination in New York's program is not justified by, or tailored to, the purposes of the EA program. Ibid.9 13 We agree that New York's law is invalid under the equitable-treatment regulation insofar as it automatically excludes AFDC recipients from the EA program. The regulation, and the Secretary's decision to apply it to strike down New York's no-cash and loss-or-theft rules, clearly deserve judicial deference. We have often noted that the interpretation of an agency charged with the administration of a statute is entitled to substantial deference. See, e.g., FEC v. Democratic Senatorial Campaign Committee, 454 U.S. 27, 102 S.Ct. 38, 70 L.Ed.2d 23 (1981); Quern 436 U.S., at 738, 98 S.Ct., at 2076. In light of the strong support in the legislative history for the Secretary's conclusion that the automatic exclusion of AFDC recipients from an EA program is inequitable in light of the purposes of the EA program, we find such deference particularly appropriate in this case. C 14 In 1967, Congress thoroughly revised the Social Security Act, including many of its public assistance provisions. The House and Senate Committee Reports concerning the portion of the revision that would ultimately become the EA program10 make it unmistakably clear that AFDC recipients were expected to benefit from the program. The initial House Report states: 15 "Your committee understands that the process of determining eligibility and authorizing payments frequently precludes the meeting of emergency needs when a crisis occurs. In the event of eviction, or when utilities are turned off, or when an alcoholic parent leaves children without food, immediate action is necessary. It frequently is unavailable under State programs today. 16 When a child is suddenly deprived of his parents by their accidental death or when the agency finds that the conditions in the home are contrary to the child's welfare, the normal methods of payment have to be suspended while new arrangements and court referrals are made. 17 "To encourage public welfare agencies to move promptly and with maximum effectiveness in such situations, the bill contains an offer to the States of 50-percent participation in emergency assistance payments. . . . The eligible families involved are those with children under 21 who either are or have recently been living with close relatives. The families do not have to be receiving or eligible upon application to receive AFDC (although they are generally of the same type), but they must be without available resources. . . . 18 "Assistance might be in any form. . . . The provision is broad enough that emergencies can be met in migrant families as well as those meeting residence requirements of the State's AFDC program. Its utilization would be optional with the States." H.R.Rep. No. 544, 90th Cong., 1st Sess., 109 (1967) (emphasis added). 19 This passage leaves the obvious implication that persons who are eligible for AFDC benefits would receive EA. The Senate Report is almost identical, except for an explanation of the changes in the Senate bill. S.Rep. No. 744, 90th Cong., 1st Sess., 165-166 (1967), U.S.Code Cong. & Admin.News 1967, p. 2834.11 Indeed, in an earlier summary of this provision, the Senate Report describes EA as one of a series of amendments that "would set up new protections for the children in AFDC families." Id., at 146, U.S.Code Cong. & Admin.News 1967, p. 2982.12 20 The House and Senate debates on this portion of the Social Security Amendments, although abbreviated, buttress our understanding of congressional intent. Senator Long, floor manager of the bill, repeated the Senate Report's characterization of EA as one of several changes that would establish "new protections for the children in AFDC families." 113 Cong.Rec. 32592 (1967). Comments by other legislators reveal a similar understanding.13 Testimony by witnesses and statements introduced at the Senate hearings on the bill are also illuminating. Many statements assume that AFDC would be covered, and some reveal the belief that EA would be principally an AFDC program.14 III 21 The Secretary's decision to apply the "equitable treatment" regulation so as to forbid a State to exclude AFDC recipients from its EA program is eminently reasonable and deserves judicial deference. The regulation explicitly forbids the "inequitable treatment of individuals or groups in the light of the provisions and purposes of the public assistance titles of the Social Security Act." 45 CFR § 233.10 (1981). AFDC recipients are "the core of the Title IV-A system," Quern, 436 U.S., at 728, 98 S.Ct., at 2071, and the principal group of beneficiaries under federally assisted state welfare programs. Moreover, the legislative history reviewed above leaves no doubt that AFDC recipients were expected to be included in a state EA program receiving federal financial assistance. 22 Because New York's no-cash15 and loss-or-theft rules conflict with a valid federal regulation, they are invalid under the Supremacy Clause. See Chrysler Corp. v. Brown, 441 U.S. 281, 295-296, 99 S.Ct. 1705, 1714, 60 L.Ed.2d 208 (1979). In light of our disposition of this Supremacy Clause claim, we do not address appellees' equal protection argument.16 23 The judgment of the Court of Appeals is affirmed. 24 It is so ordered. 1 New York Soc.Serv.Law § 350-j (McKinney Supp.1981) provides in pertinent part: "2. For purposes of this section, the term 'emergency assistance' means aid, care and services authorized during a period not in excess of thirty days in any twelve month period to meet the emergency needs of a child or the household in which he is living, in the following circumstances: "(a) where the child is under twenty-one years of age; and "(b) the child is living with, or within the previous six months has lived with, one or more persons specified in subdivision b of section three hundred forty-nine of this chapter; and "(c) in cases of applications for grants of cash assistance, such child or such household is not categorically eligible for or receiving aid to dependent children ; and "(d) such emergency needs resulted from a catastrophic occurrence or from a situation which threatens family stability and which has caused the destitution of the child and/or home; and "(e) such occurrence or situation could not have been foreseen by the applicant, was not under his control, and, in the case of a person receiving public assistance, did not result from the loss, theft or mismanagement of a regular public assistance grant ; and "(f) the emergency grant being applied for will not replace or duplicate a public assistance grant already made under section one hundred thirty-one-a of this chapter. "3. Emergency assistance to needy families with children shall be provided to the extent of items of need and services set forth in sections one hundred thirty-one and one hundred thirty-one-a of this chapter . . . . Such emergency assistance, but not including cash grants, may be furnished to a family eligible for aid to dependent children only in the form of emergency services, and so long as federal aid remains available, for emergency fuel grants in the form of vendor restricted payments." (emphasis added). 2 The record indicates that appellees have standing to challenge the no-cash as well as the loss-or-theft provision. In the case of appellees Bacon, Parrish, and Goodwine, DSS purported to rely on the loss-or-theft provision in denying EA. However, Parrish was denied EA for lost food as well as lost money. The no-cash provision, not the loss-or-theft provision, would appear to justify denial of EA for the lost food, since that loss did not result from the theft of her public assistance grant. In the case of appellee Selders, DSS denied EA for her lost cash pursuant to an administrative memorandum that outlines both the no-cash and the loss-or-theft exclusions for AFDC recipients. App. 34a-37a, 51a. It is therefore unclear which exclusion the State purported to rely upon in denying assistance. However, Selders lost a food stamp voucher as well as the cash from her AFDC grant. DSS refused to provide the cash equivalent of the voucher. Although DSS replaced the voucher, Selders was unable to make use of the replacement because she had no cash. Bacon v. Toia, 437 F.Supp. 1371, 1376-1377 (SDNY 1977). Accordingly, it was the State's no-cash rule that prevented her from obtaining effective EA. Id., at 1384-1385. Selders alone was denominated a representative of the subclass of plaintiffs challenging the no-cash rule. Id., at 1381. In light of these facts and appellant's concession that DSS relied upon both the no-cash and the loss-or-theft provisions in denying assistance, Brief for Appellant 6-7; Tr. of Oral Arg. 4, 17, we conclude that appellees have standing to challenge both provisions. 3 Appellees also sought to enjoin enforcement of the statutory provision denying replacement or duplication of a public assistance grant already made. N.Y.Soc.Serv.Law § 350-j(2)(f) (McKinney Supp.1981). They no longer seek relief with respect to that provision because the District Court determined that the provision does not preclude EA in the event of a true emergency. Bacon v. Toia, 493 F.Supp. 865, 875 (SDNY 1980). 4 Section 406(e) of the Act provides in part: "(1) The term 'emergency assistance to needy families with children' means any of the following, furnished for a period not in excess of 30 days in any 12-month period, in the case of a needy child under the age of 21 who is . . . living with any of the relatives specified in subsection (a)(1) of this section in a place of residence maintained by one or more of such relatives as his or their own home, but only where such child is without available resources, the payments, care, or services involved as necessary to avoid destitution of such child or to provide living arrangements in a home for such child, and such destitution or need for living arrangements did not arise because such child or relative refused without good cause to accept employment or training for employment— "(A) money payments, payments in kind, or such other payments as the State agency may specify with respect to, or medical care or any other type of remedial care recognized under State law on behalf of, such child or any other member of the household in which he is living, and "(B) such services as may be specified by the Secretary; "but only with respect to a State whose State plan approved under section 602 of this title includes provision for such assistance. "(2) Emergency assistance as authorized under paragraph (1) may be provided . . . to migrant workers with families in the State or in such part or parts thereof as the State shall designate." 81 Stat. 893. 5 Appellant argues that the statutory issue is not properly before us. It is well accepted, however, that without filing a cross-appeal or cross-petition, an appellee may rely upon any matter appearing in the record in support of the judgment below. See Massachusetts Mutual Life Ins. Co. v. Ludwig, 426 U.S. 479, 96 S.Ct. 2158, 48 L.Ed.2d 784 (1976) (per curiam ); Dayton Board of Education v. Brinkman, 433 U.S. 406, 419, 97 S.Ct. 2766, 2775, 53 L.Ed.2d 851 (1977); see also R. Stern & E. Gressman, Supreme Court Practice 478 (5th ed. 1978). Moreover, the statutory issue was raised and decided in both the District Court and the Court of Appeals. Appellant also asserts that the appellees were required to file a cross-appeal because they seek to modify the judgment below. Acceptance of the appellees' statutory argument will allegedly result in such a modification because "the injunction granted below on the basis of constitutional right would be modified to one based upon statute or regulation." Reply of Appellant to Motion to Affirm 8. Appellant cites no authority for this novel view that an affirmance which does not alter the relief ordered in the judgment below "modifies" the judgment simply because the affirmance rests on a different legal basis than the court below adopted. 6 For a general discussion of the EA and AFDC programs, see Quern v. Mandley, 436 U.S., at 728-729, 735-736, 739, 742-743, 98 S.Ct., at 2071, 2074-2075, 2076, 2078 (1978). 7 In a footnote, the Court reported that some States have narrowed eligibility for EA programs by excluding AFDC recipients if the emergency need is one theoretically covered by the basic assistance grant. Id., at 739-740, n. 16, 98 S.Ct., at 2076-2077, n. 16. This descriptive statement is not, of course, an expression of opinion, much less a holding, with respect to the issue before us. 8 These regulations were adopted pursuant to § 1102 of the Act, 42 U.S.C. § 1302, which provides the Secretary with authority to "make . . . such rules and regulations, not inconsistent with this chapter, as may be necessary to the efficient administration of the functions with which [he] is charged under this chapter." We have described this provision as creating "broad rule-making powers." Thorpe v. Housing Authority, 393 U.S. 268, 277, n. 28, 89 S.Ct. 518, 523, n. 28, 21 L.Ed.2d 474 (1969). The Secretary has also argued that 45 CFR § 233.10(a)(1) (1981) is authorized pursuant to § 402(a)(5), 42 U.S.C. § 602(a)(5), under which a state plan under Title IV-A "must . . . provide such methods of administration . . . as are found by the Secretary to be necessary for the proper and efficient operation of the plan." See App. to Motion to Affirm 8a. This section applies to all portions of a state plan under Title IV-A, including the EA program. See Quern, supra, at 741-742, 98 S.Ct., at 2077-2078. 9 It appears that the Secretary has consistently taken the position that automatic exclusions of AFDC recipients from an EA program violate the "equitable treatment" regulation, 45 CFR § 233.10 (1981). In an amicus brief filed at the invitation of the court in Ingerson v. Pratt, Civ. Action No. 76-3255-S (Mass., Nov. 14, 1979), the Secretary stated that the regulation applies to EA programs. The Office of Family Assistance later determined that Massachusetts' "recipient status" rule violates the regulation because it provides that the EA benefits available to AFDC recipients will be affected by the amount received in the AFDC grant, whereas the EA benefits available to non-AFDC recipients will not be affected by the amount of assistance received. Documentation of the U.S. Dept. of HEW relating to Ingerson v. Pratt, File of the Clerk of this Court in No. 81-770. The District Court has invalidated this rule (as well as another state rule) based on the "equitable treatment" regulation. Ingerson v. Pratt, Civ. Action No. 76-3755-S (Mass., Feb. 8, 1982). 10 The EA program, a small part of the 1967 Social Security Amendments, originated with the President's limited proposal to extend temporary assistance to migratory workers. See H.R. 5710, 90th Cong., 1st Sess., 122-123 (1967). The assistance "would be in an amount consistent with what the individuals would receive if they were eligible under a public assistance plan in the State in which they are living." House Committee on Ways and Means, Section-by-Section Analysis of H.R. 5710, 90th Cong., 1st Sess., 9 (Comm. Print 1967). In the House, the Ways and Means Committee adopted a much broader proposal, very similar to the version ultimately enacted. See H.R. 12080, 90th Cong., 1st Sess., 137-139 (1967). After House passage, the Senate made three changes—increasing the amount of time that emergency assistance would be available, denying assistance if a child or relative refused without good cause to accept employment, and more explicitly protecting migrant workers. In conference, the second and third changes were accepted. H.R.Conf.Rep. No. 1030, 90th Cong., 1st Sess., 60 (1967), U.S.Code Cong. & Admin.News 1967, p. 2834. Congress adopted the conference bill. 11 The Senate Report also inserts "AFDC" prior to the word "eligibility" in the first sentence quoted in the text above—yet another indication that those eligible for or receiving AFDC were presumed to be covered. 12 Other portions of the legislative documents indicate that the EA program was viewed as an extension of the AFDC program. For example, the House bill initially contained the title, "Emergency Assistance for Certain Needy Families with Dependent Children." H.R. 12080, 90th Cong., 1st Sess., 137 (1967) (emphasis added). The bill as enacted placed the EA provisions within Subchapter IV-A of the Social Security Act, 42 U.S.C. § 601 et seq., entitled "Aid to Families With Dependent Children." 42 U.S.C. §§ 603(a)(5), 606(e). Moreover, numerous historical documents place discussion of the EA program within an "AFDC" category. See, e.g., H.R.Rep. No. 544, 90th Cong., 1st Sess., 97 (1967); Senate Committee on Finance, Social Security Amendments of 1967: Comparison of H.R. 12080, As Passed by the House of Representatives with Existing Law, 90th Cong., 1st Sess., 37-38 (Comm. Print 1967); S.Rep. No. 744, 90th Cong., 1st Sess., 4, 165-166 (1967); Senate Committee on Finance, The Social Security Amendments of 1967, Brief Summary of Major Provisions and Detailed Comparison with Prior Law, 90th Cong., 1st Sess., 68-69 (1968). 13 Congress expressed concern that categorical aid programs (which include AFDC) were often too inflexible to afford immediate emergency relief. The implicit assumption is that if a state EA plan covered emergency needs that were not promptly met under these programs, persons receiving or eligible for aid under the programs would receive EA. Thus, Senator Ribicoff emphasized that welfare agencies need to have "flexibility" in dealing with emergencies; and Senator Curtis explained that "[f]or a period of 30 days, emergency assistance can be paid in cases where [EA recipients] cannot meet other qualifications." 113 Cong.Rec. 32853, 36319 (1967). 14 Hearings on H.R. 12080 before the Senate Committee on Finance, 90th Cong., 1st Sess., 1034 (1967) (statement of Commissioner, Louisiana Dept. of Public Welfare); id., at App. A8 (statement of Commissioner, Alabama Dept. of Pensions and Security); id., at App. A125 (statement of Governor of Hawaii); id., at App. A289 (statement of Rhode Island Dept. of Social Welfare). 15 Appellant asserts that the no-cash rule does not discriminate against AFDC recipients at all. On its face, of course, the rule excludes AFDC as well as all other public assistance recipients and thus violates the regulation. But appellant claims that AFDC recipients may obtain the same emergency benefits through special grants under other provisions of the state welfare law. Both lower courts rejected this claim, concluding that the special grant provisions are more limited than EA provisions. 648 F.2d, at 807-808; 493 F.Supp., at 872-873. In particular, the Court of Appeals, after reviewing relevant state cases, determined that the state EA program would provide grants in cases of loss caused by burglary or public auction following eviction, and would provide food and other immediate living expenses, while special grants would not cover these items. 648 F.2d at 808. We have no reason to question the conclusion of the lower courts given their familiarity with state law. See Bishop v. Wood, 426 U.S. 341, 346, 96 S.Ct. 2074, 2078, 48 L.Ed.2d 684 (1976). Of course, we do not suggest that a State may not choose to provide AFDC special grants for emergencies as an alternative to including AFDC recipients within the EA program, if the special grants are provided as promptly and for the same emergencies as the EA grants. Cf. Quern, 436 U.S., at 734-739, 98 S.Ct., at 2074-2076. 16 We do not reach the question whether the loss-or-theft provision in N.Y.Soc.Serv.Law § 350-j (McKinney Supp.1981) is invalid as applied to public assistance recipients other than those receiving AFDC. The question presented in appellant's jurisdictional statement refers only to the discrimination against AFDC recipients; thus, appellant challenges the judgment below only insofar as it requires the granting of EA to AFDC recipients.

12

457 U.S. 147 102 S.Ct. 2364 72 L.Ed.2d 740 GENERAL TELEPHONE COMPANY OF the SOUTHWEST, Petitionerv.Mariano S. FALCON. No. 81-574. Argued April 26, 1982. Decided June 14, 1982. Syllabus After being denied a promotion by petitioner employer, respondent Mexican-American filed a charge with the Equal Employment Opportunity Commission, alleging that he had been passed over for promotion because of his national origin and that petitioner's promotion policy operated against Mexican-Americans as a class. Subsequently, respondent received a right-to-sue letter from the Commission, and he then brought a class action in Federal District Court under Title VII of the Civil Rights Act of 1964. Without conducting an evidentiary hearing, the District Court certified a class consisting of Mexican-American employees of petitioner and Mexican-American applicants who had not been hired. As to liability, the court held that petitioner had discriminated against respondent in its promotion practices but not in its hiring practices, and with respect to the class found that petitioner had discriminated against Mexican-Americans in its hiring practices but not in its promotion practices. Both parties appealed, and the Court of Appeals, rejecting petitioner's argument that the class had been defined too broadly, held that the District Court's class certification was proper under the Fifth Circuit's rule permitting any victim of racial discrimination in employment to maintain an "across-the-board" attack on all unequal employment practices allegedly followed by the employer pursuant to a policy of racial discrimination. On the merits, the Court of Appeals upheld respondent's promotion claim, but held that the District Court's findings were insufficient to support recovery on behalf of the class. Subsequently, this Court vacated the Court of Appeals' judgment and remanded the case for further consideration in light of Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207. The Court of Appeals then vacated its judgment as to respondent's promotion claim but reinstated its approval of the District Court's class certification. Held: The District Court erred in permitting respondent to maintain a class action on behalf of both employees who were denied promotion and applicants who were denied employment. Pp. 155-161. (a) An individual litigant seeking to maintain a class action under Title VII must meet Federal Rule of Civil Procedure 23(a)'s specified "prerequisites of numerosity, commonality, typicality, and adequacy of representation." General Telephone Co. v. EEOC, 446 U.S. 318, 330, 100 S.Ct. 1698, 1706, 64 L.Ed.2d 319. These requirements effectively "limit the class claims to those fairly encompassed by the named plaintiff's claim." Ibid. P. 155-157. (b) There can be no disagreement with the proposition underlying the Fifth Circuit's "across-the-board" rule—that racial discrimination is by definition class discrimination. But the allegation that such discrimination has occurred neither determines whether a class action may be maintained in accordance with Rule 23 nor determines the class that may be certified. Here, to bridge the gap between respondent's promotion claim and the existence of a class of persons who have suffered the same injury as respondent—so that respondent's claim and the class claims share common questions of law or fact and respondent's claim is typical of the class claims—respondent must prove much more than the validity of his own claim. Respondent's complaint provided an insufficient basis for concluding that the adjudication of his claim would require the decision of any common question concerning petitioner's failure to hire more Mexican-Americans. Without any specific presentation identifying the questions of law or fact that were common to the claims of respondent and of the class members he sought to represent, it was error for the District Court to presume that respondent's claim was typical of other claims against petitioner by Mexican-American employees and applicants. Pp. 157-159. (c) As the District Court's bifurcated findings on liability demonstrate, the individual and class claims might as well have been tried separately. Thus, it is clear that the maintenance of the action as a class action did not advance "the efficiency and economy of litigation which is a principal purpose of the procedure." American Pipe & Construction Co. v. Utah, 414 U.S. 538, 553, 94 S.Ct. 756, 766, 38 L.Ed.2d 713. P. 159. (d) The District Court's error, and the error inherent in the "across-the-board" rule, is the failure to evaluate carefully the legitimacy of the named plaintiff's plea that he is a proper class representative under Rule 23(a). P. 160 5th Cir., 647 F.2d 633, reversed and remanded. Noyes Thompson Powers, Washington, D. C., for petitioner. Frank P. Hernandez, Dallas, Tex., for respondent. Justice STEVENS delivered the opinion of the Court. 1 The question presented is whether respondent Falcon, who complained that petitioner did not promote him because he is a Mexican-American, was properly permitted to maintain a class action on behalf of Mexican-American applicants for employment whom petitioner did not hire. 2 * In 1969 petitioner initiated a special recruitment and training program for minorities. Through that program, respondent Falcon was hired in July 1969 as a groundman, and within a year he was twice promoted, first to lineman and then to lineman-in-charge. He subsequently refused a promotion to installer-repairman. In October 1972 he applied for the job of field inspector; his application was denied even though the promotion was granted several white employees with less seniority. 3 Falcon thereupon filed a charge with the Equal Employment Opportunity Commission stating his belief that he had been passed over for promotion because of his national origin and that petitioner's promotion policy operated against Mexican-Americans as a class. Falcon v. General Telephone Co. of Southwest, 626 F.2d 369, 372, n. 2 (CA5 1980). In due course he received a right-to-sue letter from the Commission and, in April 1975, he commenced this action under Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp.IV), in the United States District Court for the Northern District of Texas. His complaint alleged that petitioner maintained "a policy, practice, custom, or usage of: (a) discriminating against [Mexican-Americans] because of national origin and with respect to compensation, terms, conditions, and privileges of employment, and (b) . . . subjecting [Mexican-Americans] to continuous employment discrimination."1 Respondent claimed that as a result of this policy whites with less qualification and experience and lower evaluation scores than respondent had been promoted more rapidly. The complaint contained no factual allegations concerning petitioner's hiring practices. 4 Respondent brought the action "on his own behalf and on behalf of other persons similarly situated, pursuant to Rule 23(b)(2) of the Federal Rules of Civil Procedure."2 The class identified in the complaint was "composed of Mexican-American persons who are employed, or who might be employed, by GENERAL TELEPHONE COMPANY at its place of business located in Irving, Texas, who have been and who continue to be or might be adversely affected by the practices complained of herein."3 5 After responding to petitioner's written interrogatories,4 respondent filed a memorandum in favor of certification of "the class of all hourly Mexican American employees who have been employed, are employed, or may in the future be employed and all those Mexican-Americans who have applied or would have applied for employment had the Defendant not practiced racial discrimination in its employment practices." App. 46-47. His position was supported by the ruling of the United States Court of Appeals for the Fifth Circuit in Johnson v. Georgia Highway Express, Inc., 417 F.2d 1122 (1969), that any victim of racial discrimination in employment may maintain an "across the board" attack on all unequal employment practices alleged to have been committed by the employer pursuant to a policy of racial discrimination. Without conducting an evidentiary hearing, the District Court certified a class including Mexican-American employees and Mexican-American applicants for employment who had not been hired.5 6 Following trial of the liability issues, the District Court entered separate findings of fact and conclusions of law with respect first to respondent and then to the class. The District Court found that petitioner had not discriminated against respondent in hiring, but that it did discriminate against him in its promotion practices. App. to Pet. for Cert. 35a, 37a. The court reached converse conclusions about the class, finding no discrimination in promotion practices, but concluding that petitioner had discriminated against Mexican-Americans at its Irving facility in its hiring practices. Id., at 39a-40a.6 7 After various post-trial proceedings, the District Court ordered petitioner to furnish respondent with a list of all Mexican-Americans who had applied for employment at the Irving facility during the period between January 1, 1973, and October 18, 1976. Respondent was then ordered to give notice to those persons advising them that they might be entitled to some form of recovery. Evidence was taken concerning the applicants who responded to the notice, and backpay was ultimately awarded to 13 persons, in addition to respondent Falcon. The total recovery by respondent and the entire class amounted to $67,925.49, plus costs and interest.7 8 Both parties appealed. The Court of Appeals rejected respondent's contention that the class should have encompassed all of petitioner's operations in Texas, New Mexico, Oklahoma, and Arkansas.8 On the other hand, the court also rejected petitioner's argument that the class had been defined too broadly. For, under the Fifth Circuit's across-the-board rule, it is permissible for "an employee complaining of one employment practice to represent another complaining of another practice, if the plaintiff and the members of the class suffer from essentially the same injury. In this case, all of the claims are based on discrimination because of national origin." 626 F.2d, at 375.9 The court relied on Payne v. Travenol Laboratories, Inc., 565 F.2d 895 (1978), cert. denied, 439 U.S. 835, 99 S.Ct. 118, 58 L.Ed.2d 131, in which the Fifth Circuit stated: 9 "Plaintiffs' action is an 'across the board' attack on unequal employment practices alleged to have been committed by Travenol pursuant to a policy of racial discrimination. As parties who have allegedly been aggrieved by some of those discriminatory practices, plaintiffs have demonstrated a sufficient nexus to enable them to represent other class members suffering from different practices motivated by the same policies." 565 F.2d, at 900, quoted in 626 F.2d, at 375. 10 On the merits, the Court of Appeals upheld respondent's claim of disparate treatment in promotion,10 but held that the District Court's findings relating to disparate impact in hiring were insufficient to support recovery on behalf of the class.11 After this Court decided Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207, we vacated the judgment of the Court of Appeals and directed further consideration in the light of that opinion. General Telephone Co. of Southwest v. Falcon, 450 U.S. 1036, 101 S.Ct. 1752, 68 L.Ed.2d 234. The Fifth Circuit thereupon vacated the portion of its opinion addressing respondent's promotion claim but reinstated the portions of its opinion approving the District Court's class certification. 647 F.2d 633 (1981). With the merits of both respondent's promotion claim and the class hiring claims remaining open for reconsideration in the District Court on remand, we granted certiorari to decide whether the class action was properly maintained on behalf of both employees who were denied promotion and applicants who were denied employment. II 11 The class-action device was designed as "an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only." Califano v. Yamasaki, 442 U.S. 682, 700-701, 99 S.Ct. 2545, 2557-2558, 61 L.Ed.2d 176. Class relief is "peculiarly appropriate" when the "issues involved are common to the class as a whole" and when they "turn on questions of law applicable in the same manner to each member of the class." Id., at 701, 99 S.Ct., at 2557. For in such cases, "the class-action device saves the resources of both the courts and the parties by permitting an issue potentially affecting every [class member] to be litigated in an economical fashion under Rule 23." Ibid. 12 Title VII of the Civil Rights Act of 1964, as amended, authorizes the Equal Employment Opportunity Commission to sue in its own name to secure relief for individuals aggrieved by discriminatory practices forbidden by the Act. See 42 U.S.C. § 2000e-5(f)(1). In exercising this enforcement power, the Commission may seek relief for groups of employees or applicants for employment without complying with the strictures of Rule 23. General Telephone Co. of Northwest v. EEOC, 446 U.S. 318, 100 S.Ct. 1698, 64 L.Ed.2d 319. Title VII, however, contains no special authorization for class suits maintained by private parties. An individual litigant seeking to maintain a class action under Title VII must meet "the prerequisites of numerosity, commonality, typicality, and adequacy of representation" specified in Rule 23(a). Id., at 330, 100 S.Ct., at 1706. These requirements effectively "limit the class claims to those fairly encompassed by the named plaintiff's claims." Ibid. 13 We have repeatedly held that "a class representative must be part of the class and 'possess the same interest and suffer the same injury' as the class members." East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 403, 97 S.Ct. 1891, 1896, 52 L.Ed.2d 453 (quoting Schlesinger v. Reservists Committee to Stop the War, 418 U.S. 208, 216, 94 S.Ct. 2925, 2929-2930, 41 L.Ed.2d 706.) In East Texas Motor Freight, a Title VII action brought by three Mexican-American city drivers, the Fifth Circuit certified a class consisting of the trucking company's black and Mexican-American city drivers allegedly denied on racial or ethnic grounds transfers to more desirable line-driver jobs. We held that the Court of Appeals had "plainly erred in declaring a class action." 431 U.S., at 403, 97 S.Ct., at 1896. Because at the time the class was certified it was clear that the named plaintiffs were not qualified for line-driver positions, "they could have suffered no injury as a result of the allegedly discriminatory practices, and they were, therefore, simply not eligible to represent a class of persons who did allegedly suffer injury." Id., at 403-404, 97 S.Ct., at 1897. 14 Our holding in East Texas Motor Freight was limited; we noted that "a different case would be presented if the District Court had certified a class and only later had it appeared that the named plaintiffs were not class members or were otherwise inappropriate class representatives." Id., at 406, n. 12, 97 S.Ct., at 1898, n. 12. We also recognized the theory behind the Fifth Circuit's across-the-board rule, noting our awareness "that suits alleging racial or ethnic discrimination are often by their very nature class suits, involving classwide wrongs," and that "[c]ommon questions of law or fact are typically present." Id., at 405, 97 S.Ct., at 1898. In the same breath, however, we reiterated that "careful attention to the requirements of Fed.Rule Civ.Proc. 23 remains nonetheless indispensable" and that the "mere fact that a complaint alleges racial or ethnic discrimination does not in itself ensure that the party who has brought the lawsuit will be an adequate representative of those who may have been the real victims of that discrimination." Id., at 405-406, 97 S.Ct., at 1898. 15 We cannot disagree with the proposition underlying the across-the-board rule—that racial discrimination is by definition class discrimination.12 But the allegation that such discrimination has occurred neither determines whether a class action may be maintained in accordance with Rule 23 nor defines the class that may be certified. Conceptually, there is a wide gap between (a) an individual's claim that he has been denied a promotion on discriminatory grounds, and his otherwise unsupported allegation that the company has a policy of discrimination, and (b) the existence of a class of persons who have suffered the same injury as that individual, such that the individual's claim and the class claims will share common questions of law or fact and that the individual's claim will be typical of the class claims.13 For respondent to bridge that gap, he must prove much more than the validity of his own claim. Even though evidence that he was passed over for promotion when several less deserving whites were advanced may support the conclusion that respondent was denied the promotion because of his national origin, such evidence would not necessarily justify the additional inferences (1) that this discriminatory treatment is typical of petitioner's promotion practices, (2) that petitioner's promotion practices are motivated by a policy of ethnic discrimination that pervades petitioner's Irving division, or (3) that this policy of ethnic discrimination is reflected in petitioner's other employment practices, such as hiring, in the same way it is manifested in the promotion practices. These additional inferences demonstrate the tenuous character of any presumption that the class claims are "fairly encompassed" within respondent's claim. 16 Respondent's complaint provided an insufficient basis for concluding that the adjudication of his claim of discrimination in promotion would require the decision of any common question concerning the failure of petitioner to hire more Mexican-Americans. Without any specific presentation identifying the questions of law or fact that were common to the claims of respondent and of the members of the class he sought to represent,14 it was error for the District Court to presume that respondent's claim was typical of other claims against petitioner by Mexican-American employees and applicants. If one allegation of specific discriminatory treatment were sufficient to support an across-the-board attack, every Title VII case would be a potential companywide class action. We find nothing in the statute to indicate that Congress intended to authorize such a wholesale expansion of class-action litigation.15 17 The trial of this class action followed a predictable course. Instead of raising common questions of law or fact, respondent's evidentiary approaches to the individual and class claims were entirely different. He attempted to sustain his individual claim by proving intentional discrimination. He tried to prove the class claims through statistical evidence of disparate impact. Ironically, the District Court rejected the class claim of promotion discrimination, which conceptually might have borne a closer typicality and commonality relationship with respondent's individual claim, but sustained the class claim of hiring discrimination. As the District Court's bifurcated findings on liability demonstrate, the individual and class claims might as well have been tried separately. It is clear that the maintenance of respondent's action as a class action did not advance "the efficiency and economy of litigation which is a principal purpose of the procedure." American Pipe & Construction Co. v. Utah, 414 U.S. 538, 553, 94 S.Ct. 756, 766, 38 L.Ed.2d 713. We do not, of course, judge the propriety of a class certification by hindsight. The District Court's error in this case, and the error inherent in the across-the-board rule, is the failure to evaluate carefully the legitimacy of the named plaintiff's plea that he is a proper class representative under Rule 23(a). As we noted in Coopers & Lybrand v. Livesay, 437 U.S. 463, 98 S.Ct. 2454, 57 L.Ed.2d 351, "the class determination generally involves considerations that are 'enmeshed in the factual and legal issues comprising the plaintiff's cause of action.' " Id., at 469, 98 S.Ct., at 2458 (quoting Mercantile Nat. Bank v. Langdeau, 371 U.S. 555, 558, 83 S.Ct. 520, 522, 9 L.Ed.2d 523). Sometimes the issues are plain enough from the pleadings to determine whether the interests of the absent parties are fairly encompassed within the named plaintiff's claim, and sometimes it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question. Even after a certification order is entered, the judge remains free to modify it in the light of subsequent developments in the litigation.16 For such an order, particularly during the period before any notice is sent to members of the class, "is inherently tentative." 437 U.S., at 469, n. 11, 98 S.Ct., at 2458 n. 11. This flexibility enhances the usefulness of the class-action device; actual, not presumed, conformance with Rule 23(a) remains, however, indispensable. III 18 The need to carefully apply the requirements of Rule 23(a) to Title VII class actions was noticed by a member of the Fifth Circuit panel that announced the across-the-board rule. In a specially concurring opinion in Johnson v. Georgia Highway Express, Inc., 417 F.2d, at 1125-1127, Judge Godbold emphasized the need for "more precise pleadings," id., at 1125, for "without reasonable specificity the court cannot define the class, cannot determine whether the representation is adequate, and the employer does not know how to defend," id., at 1126. He termed as "most significant" the potential unfairness to the class members bound by the judgment if the framing of the class is overbroad. Ibid. And he pointed out the error of the "tacit assumption" underlying the across-the-board rule that "all will be well for surely the plaintiff will win and manna will fall on all members of the class." Id., at 1127. With the same concerns in mind, we reiterate today that a Title VII class action, like any other class action, may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied. 19 The judgment of the Court of Appeals affirming the certification order is reversed, and the case is remanded for further proceedings consistent with this opinion. 20 It is so ordered. 21 Chief Justice BURGER, concurring in part and dissenting in part. 22 I agree with the Court's decision insofar as it states the general principles which apply in determining whether a class should be certified in this case under Rule 23. However, in my view it is not necessary to remand for further proceedings since it is entirely clear on this record that no class should have been certified in this case. I would simply reverse the Court of Appeals and remand with instructions to dismiss the class claim. 23 As the Court notes, the purpose of Rule 23 is to promote judicial economy by allowing for litigation of common questions of law and fact at one time. Califano v. Yamasaki, 442 U.S. 682, 701, 99 S.Ct. 2545, 2557-2558, 61 L.Ed.2d 176 (1979). We have stressed that strict attention to the requirements of Rule 23 is indispensable in employment discrimination cases. East Texas Motor Freight System Inc. v. Rodriguez, 431 U.S. 395, 405-406, 97 S.Ct. 1891, 1897-1898, 52 L.Ed.2d 453 (1977). This means that class claims are limited to those " 'fairly encompassed by the named plaintiff's claims.' " Ante, at 2370, quoting General Telephone Co. of Northwest v. EEOC, 446 U.S. 318, 330, 100 S.Ct. 1698, 1706, 64 L.Ed.2d 319 (1980). 24 Respondent claims that he was not promoted to a job as field inspector because he is a Mexican-American. To be successful in his claim, which he advances under the "disparate treatment" theory, he must convince a court that those who were promoted were promoted not because they were better qualified than he was, but, instead, that he was not promoted for discriminatory reasons. The success of this claim depends on evaluation of the comparative qualifications of the applicants for promotion to field inspector and on analysis of the credibility of the reasons for the promotion decisions provided by those who made the decisions. Respondent's class claim on behalf of unsuccessful applicants for jobs with petitioner, in contrast, is advanced under the "adverse impact" theory. Its success depends on an analysis of statistics concerning petitioner's hiring patterns.* 25 The record in this case clearly shows that there are no common questions of law or fact between respondent's claim and the class claim; the only commonality is that respondent is a Mexican-American and he seeks to represent a class of Mexican-Americans. See ante, at 153, and n. 9. We have repeatedly held that the bare fact that a plaintiff alleges racial or ethnic discrimination is not enough to justify class certification. Ante, at 157; East Texas Motor Freight, supra, at 405-406, 97 S.Ct., at 1897-1898. Accordingly, the class should not have been certified. 26 Moreover, while a judge's decision to certify a class is not normally to be evaluated by hindsight, ante, at 160, since the judge cannot know what the evidence will show, there is no reason for us at this stage of these lengthy judicial proceedings not to proceed in light of the evidence actually presented. The Court properly concludes that the Court of Appeals and the District Court failed to consider the requirements of Rule 23. In determining whether to reverse and remand or to simply reverse, we can and should look at the evidence. The record shows that there is no support for the class claim. Respondent's own statistics show that 7.7% of those hired by petitioner between 1972 and 1976 were Mexican-American while the relevant labor force was 5.2% Mexican-American. Falcon v. General Telephone Company of Southwest, 626 F.2d 369, 372, 381, n. 16 (1980). Petitioner's unchallenged evidence shows that it hired Mexican-Americans in numbers greater than their percentage of the labor force even though Mexican-Americans applied for jobs with petitioner in numbers smaller than their percentage of the labor force. Id., at 373, n. 4. This negates any claim of Falcon as a class representative. 27 Like so many Title VII cases, this case has already gone on for years, draining judicial resources as well as resources of the litigants. Rather than promoting judicial economy, the "across-the-board" class action has promoted multiplication of claims and endless litigation. Since it is clear that the class claim brought on behalf of unsuccessful applicants for jobs with petitioner cannot succeed, I would simply reverse and remand with instructions to dismiss the class claim. 1 App. 14. In paragraph VI of the complaint, respondent alleged: "The Defendant has established an employment, transfer, promotional, and seniority system, the design, intent, and purpose of which is to continue and preserve, and which has the effect of continuing and preserving, the Defendant's policy, practice, custom and usage of limiting the employment, transfer, and promotional opportunities of Mexican-American employees of the company because of national origin." Id., at 15. 2 Id., at 13. Rule 23 provides, in part: "(a) Prerequisites to a Class Action. One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class. "(b) Class Actions Maintainable. An action may be maintained as a class action if the prerequisites of subdivision (a) are satisfied, and in addition: * * * * * "(2) the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole. . . ." 3 App. 13-14. The paragraph of the complaint in which respondent alleged conformance with the requirements of Rule 23 continued: "There are common questions of law and fact affecting the rights of the members of this class who are, and who continue to be, limited, classified, and discriminated against in ways which deprive and/or tend to deprive them of equal employment opportunities and which otherwise adversely affect their status as employees because of national origin. These persons are so numerous that joinder of all members is impracticable. A common relief is sought. The interests of said class are adequately represented by Plaintiff. Defendant has acted or refused to act on grounds generally applicable to the Plaintiff." Id., at 14. 4 Petitioner's Interrogatory No. 8 stated: "Identify the common questions of law and fac[t] which affect the rights of the members of the purported class." Id., at 26. Respondent answered that interrogatory as follows: "The facts which affect the rights of the members of the class are the facts of their employment, the ways in which evaluations are made, the subjective rather than objective manner in which recommendations for raises and transfers and promotions are handled, and all of the facts surrounding the employment of Mexican-American persons by General Telephone Company. The questions of law specified in Interrogatory No. 8 call for a conclusion on the part of the Plaintiff." Id., at 34. 5 The District Court's pretrial order of February 2, 1976, provided, in part: "The case is to proceed as a class action and the Plaintiff is to represent the class. The class is to be made up of those employees who are employed and employees who have applied for employment in the Irving Division of the Defendant company, and no other division. * * * * * "Plaintiff and Defendant are to hold further negotiations to see if there is a possibility of granting individual relief to the Plaintiff, MARIANO S. FALCON." App. to Pet. for Cert. 48a-49a. The District Court denied subsequent motions to decertify the class both before and after the trial. 6 The District Court ordered petitioner to accelerate its affirmative-action plan by taking specified steps to more actively recruit and promote Mexican-Americans at its Irving facility. See id., at 41a-45a. 7 Respondent's individual recovery amounted to $1,040.33. A large share of the class award, $28,827.50, represented attorney's fees. Most of the remainder resulted from petitioner's practice of keeping all applications active for only 90 days; the District Court found that most of the applications had been properly rejected at the time they were considered, but that petitioner could not justify the refusal to extend employment to disappointed applicants after an interval of 90 days. See 463 F.Supp. 315 (1978). 8 The Court of Appeals held that the District Court had not abused its discretion since each of petitioner's divisions conducted its own hiring and since management of the broader class would be much more difficult. Falcon v. General Telephone Co. of Southwest, 626 F.2d 369, 376 (CA5 1980). 9 The court continued: "While similarities of sex, race or national origin claims are not dispositive in favor of finding that the prerequisites of Rule 23 have been met, they are an extremely important factor in the determination, that can outweigh the fact that the members of the plaintiff class may be complaining about somewhat different specific discriminatory practices. In addition here, the plaintiff showed more than an alliance based simply on the same type of discriminatory claim. He also showed a similarity of interests based on job location, job function and other considerations." Id., at 375-376 (citations omitted). The court did not explain how job location, job function, and the unidentified other considerations were relevant to the Rule 23(a) determination. 10 The District Court found that petitioner's proffered reasons for promoting the whites, rather than respondent, were insufficient and subjective. The Court of Appeals held that respondent had made out a prima facie case under the test set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802, 93 S.Ct. 1817, 1824, 36 L.Ed.2d 668, and that the District Court's conclusion that petitioner had not rebutted that prima facie case was not clearly erroneous. In so holding, the Court of Appeals relied on its earlier opinion in Burdine v. Texas Dept. of Community Affairs, 608 F.2d 563 (1979). Our opinion in Burdine had not yet been announced. The Court of Appeals disposed of a number of other contentions raised by both parties, and reserved others pending the further proceedings before the District Court on remand. Among the latter issues was petitioner's objection to the District Court's theory for computing the class backpay awards. See n. 7, supra. 11 The District Court's finding was based on statistical evidence comparing the number of Mexican-Americans in the company's employ, and the number hired in 1972 and 1973, with the percentage of Mexican-Americans in the Dallas-Fort Worth labor force. See App. to Pet. for Cert. 39a. Since recovery had been allowed for the years 1973 through 1976 based on statistical evidence pertaining to only a portion of that period, and since petitioner's evidence concerning the entire period suggested that there was no disparate impact, the Court of Appeals ordered further proceedings on the class hiring claims. 626 F.2d, at 380-382. 12 See Hall v. Werthan Bag Corp., 251 F.Supp. 184, 186 (MD Tenn.1966). 13 The commonality and typicality requirements of Rule 23(a) tend to merge. Both serve as guideposts for determining whether under the particular circumstances maintenance of a class action is economical and whether the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence. Those requirements therefore also tend to merge with the adequacy-of-representation requirement, although the latter requirement also raises concerns about the competency of class counsel and conflicts of interest. In this case, we need not address petitioner's argument that there is a conflict of interest between respondent and the class of rejected applicants because an enlargement of the pool of Mexican-American employees will decrease respondent's chances for promotion. See General Telephone Co. of Northwest v. EEOC, 446 U.S. 318, 331, 100 S.Ct. 1698, 1706-1707, 64 L.Ed.2d 319 ("In employment discrimination litigation, conflicts might arise, for example, between employees and applicants who were denied employment and who will, if granted relief, compete with employees for fringe benefits or seniority. Under Rule 23, the same plaintiff could not represent these classes"); see also East Texas Motor Freight System, Inc. v. Rodriguez, 431 U.S. 395, 404-405, 97 S.Ct. 1891, 1897-1898, 52 L.Ed.2d 453. 14 See n. 4, supra. 15 If petitioner used a biased testing procedure to evaluate both applicants for employment and incumbent employees, a class action on behalf of every applicant or employee who might have been prejudiced by the test clearly would satisfy the commonality and typicality requirements of Rule 23(a). Significant proof that an employer operated under a general policy of discrimination conceivably could justify a class of both applicants and employees if the discrimination manifested itself in hiring and promotion practices in the same general fashion, such as through entirely subjective decisionmaking processes. In this regard it is noteworthy that Title VII prohibits discriminatory employment practices, not an abstract policy of discrimination. The mere fact that an aggrieved private plaintiff is a member of an identifiable class of persons of the same race or national origin is insufficient to establish his standing to litigate on their behalf all possible claims of discrimination against a common employer. 16 "As soon as practicable after the commencement of an action brought as a class action, the court shall determine by order whether it is to be so maintained. An order under this subdivision may be conditional, and may be altered or amended before the decision on the merits." Fed.Rule Civ.Proc. 23(c)(1). * There is no allegation that those who made the hiring decisions are the same persons who determined who was promoted to field inspector. Thus there is no claim that the same person or persons who made the challenged decisions were motivated by prejudice against Mexican-Americans, and that this prejudice manifested itself in both the hiring decisions and the decisions not to promote respondent.

89

457 U.S. 55 102 S.Ct. 2309 72 L.Ed.2d 672 Ronald M. ZOBEL and Patricia L. Zobel, Appellants,v.Thomas WILLIAMS, Commissioner of Revenue, and Alaska. No. 80-1146. Argued Oct. 7, 1981. Decided June 14, 1982. Syllabus After Alaska amended its Constitution to establish a Permanent Fund into which the State must deposit at least 25% of its mineral income each year, the state legislature in 1980 enacted a dividend program to distribute annually a portion of the Fund's earnings directly to the State's adult residents. Under the plan, each adult resident receives one dividend unit for each year of residency subsequent to 1959, the first year of Alaska's statehood. Appellants, residents of Alaska since 1978, brought an action in an Alaska state court challenging the statutory dividend distribution plan as violative of, inter alia, their right to equal protection guarantees. The trial court granted summary judgment in appellants' favor, but the Alaska Supreme Court reversed and upheld the statute. Held : The Alaska dividend distribution plan violates the guarantees of the Equal Protection Clause of the Fourteenth Amendment. Pp. 58-65. (a) Rather than imposing any threshold waiting period for entitlement to dividend benefits or establishing a test of bona fides of state residence, the dividend statute creates fixed, permanent distinctions between an ever-increasing number of classes of concededly bona fide residents based on how long they have lived in the State. Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532; Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306; Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274; and Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600, distinguished. When a state distributes benefits unequally, the distinctions it makes are subject to scrutiny under the Equal Protection Clause, and generally a law will survive that scrutiny if the distinctions rationally further a legitimate state purpose. Pp. 58-61. (b) Alaska has shown no valid state interests that are rationally served by the distinctions it makes between citizens who established residence before 1959 and those who have become residents since then. Neither the State's claimed interest in creating a financial incentive for individuals to establish and maintain residence in Alaska nor its claimed interest in assuring prudent management of the Permanent Fund is rationally related to such distinctions. And the State's interest in rewarding citizens for past contributions is not a legitimate state purpose. Alaska's reasoning could open the door to state apportionment of other rights, benefits, and services according to length of residency, and would permit the states to divide citizens into expanding numbers of permanent classes. Such a result would be clearly impermissible. Pp. 61-64. 619 P.2d 448, reversed and remanded. Mark A. Sandberg, Anchorage, Alaska, for appellants. Avrum M. Gross, Atty. Gen., Juneau, Alaska, for appellees. Chief Justice BURGER delivered the opinion of the Court. 1 The question presented on this appeal is whether a statutory scheme by which a State distributes income derived from its natural resources to the adult citizens of the State in varying amounts, based on the length of each citizen's residence, violates the equal protection rights of newer state citizens. The Alaska Supreme Court sustained the constitutionality of the statute. 619 P.2d 448 (1980). We stayed the distribution of dividend funds, 449 U.S. 989, 101 S.Ct. 524, 66 L.Ed.2d 286 (1980), and noted probable jurisdiction, 450 U.S. 908, 101 S.Ct. 1344, 67 L.Ed.2d 331 (1981). We reverse. 2 * The 1967 discovery of large oil reserves on state-owned land in the Prudhoe Bay area of Alaska resulted in a windfall to the State. The State, which had a total budget of $124 million in 1969, before the oil revenues began to flow into the state coffers, received $3.7 billion in petroleum revenues during the 1981 fiscal year.1 This income will continue, and most likely grow for some years in the future. Recognizing that its mineral reserves, although large, are finite and that the resulting income will not continue in perpetuity, the State took steps to assure that its current good fortune will bring long-range benefits. To accomplish this, Alaska in 1976 adopted a constitutional amendment establishing the Permanent Fund into which the State must deposit at least 25% of its mineral income each year. Alaska Const., Art. IX, § 15. The amendment prohibits the legislature from appropriating any of the principal of the Fund but permits use of the Fund's earnings for general governmental purposes. 3 In 1980, the legislature enacted a dividend program to distribute annually a portion of the Fund's earnings directly to the State's adult residents. Under the plan, each citizen 18 years of age or older receives one dividend unit for each year of residency subsequent to 1959, the first year of statehood. The statute fixed the value of each dividend unit at $50 for the 1979 fiscal year; a one-year resident thus would receive one unit, or $50, while a resident of Alaska since it became a State in 1959 would receive 21 units, or $1,050. The value of a dividend unit will vary each year depending on the income of the Permanent Fund and the amount of that income the State allocates for other purposes. The State now estimates that the 1985 fiscal year dividend will be nearly four times as large as that for 1979. 4 Appellants, residents of Alaska since 1978, brought this suit in 1980 challenging the dividend distribution plan as violative of their right to equal protection guarantees and their constitutional right to migrate to Alaska, to establish residency there and thereafter to enjoy the full rights of Alaska citizenship on the same terms as all other citizens of the State. The Superior Court for Alaska's Third Judicial District granted summary judgment in appellants' favor, holding that the plan violated the rights of interstate travel and equal protection. A divided Alaska Supreme Court reversed and upheld the statute.2 II 5 The Alaska dividend distribution law is quite unlike the durational residency requirements we examined in Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532 (1975); Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306 (1974); Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972); and Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969). Those cases involved laws which required new residents to reside in the State a fixed minimum period to be eligible for certain benefits available on an equal basis to all other residents.3 The asserted purpose of the durational residency requirements was to assure that only persons who had established bona fide residence received rights and benefits provided for residents. 6 The Alaska statute does not impose any threshold waiting period on those seeking dividend benefits; persons with less than a full year of residency are entitled to share in the distribution. Alaska Stat.Ann. § 43.23.010 (Supp.1981).4 Nor does the statute purport to establish a test of the bona fides of state residence. Instead, the dividend statute creates fixed, permanent distinctions between an ever-increasing number of perpetual classes of concededly bona fide residents, based on how long they have been in the State. 7 Appellants established residence in Alaska two years before the dividend law was passed. The distinction they complain of is not one which the State makes between those who arrived in Alaska after the enactment of the dividend distribution law and those who were residents prior to its enactment. Appellants instead challenge the distinctions made within the class of persons who were residents when the dividend scheme was enacted in 1980. The distinctions appellants attack include the preference given to persons who were residents when Alaska became a State in 1959 over all those who have arrived since then, as well as the distinctions made between all bona fide residents who settled in Alaska at different times during the 1959 to 1980 period.5 8 When a state distributes benefits its unequally, the distinctions it makes are subject to scrutiny under the Equal Protection Clause of the Fourteenth Amendment.6 Generally, a law will survive that scrutiny if the distinction it makes rationally furthers a legitimate state purpose. Some particularly invidious distinctions are subject to more rigorous scrutiny. Apellants claim that the distinctions made by the Alaska law should be subjected to the higher level of scrutiny applied to the durational residency requirements in Shapiro v. Thompson, supra, and Memorial Hospital v. Maricopa County, supra. The State, on the other hand, asserts that the law need only meet the minimum rationality test. In any event, if the statutory scheme cannot pass even the minimal test proposed by the State, we need not decide whether any enhanced scrutiny is called for. 9 The State advanced and the Alaska Supreme Court accepted three purposes justifying the distinctions made by the dividend program: (a) creation of a financial incentive for individuals to establish and maintain residence in Alaska; (b) encouragement of prudent management of the Permanent Fund; and (c) apportionment of benefits in recognition of undefined "contributions of various kinds, both tangible and intangible, which residents have made during their years of residency," 619 P.2d at 458.7 10 As the Alaska Supreme Court apparently realized, the first two state objectives—creating a financial incentive for individuals to establish and maintain Alaska residence, and assuring prudent management of the Permanent Fund and the State's natural and mineral resources—are not rationally related to the distinctions Alaska seeks to make between newer residents and those who have been in the State since 1959.8 Assuming, arguendo, that granting increased dividend benefits for each year of continued Alaska residence might give some residents an incentive to stay in the State in order to reap increased dividend benefits in the future, the State's interest is not in any way served by granting greater dividends to persons for their residency during the 21 years prior to the enactment.9 11 Nor does the State's purpose of furthering the prudent management of the Permanent Fund and the State's resources support retrospective application of its plan to the date of statehood. On this score the State's contention is straightforward: 12 "[A]s population increases, each individual share in the income stream is diluted. The income must be divided equally among increasingly large numbers of people. If residents believed that twenty years from now they would be required to share permanent fund income on a per capita basis with the large population that Alaska will no doubt have by then, the temptation would be great to urge the legislature to provide immediately for the highest possible percentage return on the investments of the permanent fund principal, which would require investments in riskier ventures." Id., at 462. 13 The State similarly argues that equal per capita distribution would encourage rapacious development of natural resources. Ibid. Even if we assume that the state interest is served by increasing the dividend for each year of residency beginning with the date of enactment, is it rationally served by granting greater dividends in varying amounts to those who resided in Alaska during the 21 years prior to enactment? We think not. 14 The last of the State's objectives—to reward citizens for past contributions—alone was relied upon by the Alaska Supreme Court to support the retrospective application of the law to 1959. However, that objective is not a legitimate state purpose. A similar "past contributions" argument was made and rejected in Shapiro v. Thompson, 394 U.S., at 632-633, 89 S.Ct., at 1330: 15 "Appellants argue further that the challenged classification may be sustained as an attempt to distinguish between new and old residents on the basis of the contributions they have made to the community through the payment of taxes. . . . Appellants' reasoning would . . . permit the State to apportion all benefits and services according to the past tax [or intangible] contributions of its citizens. The Equal Protection Clause prohibits such an apportionment of state services." (Emphasis added.) 16 Similarly, in Vlandis v. Kline, 412 U.S. 441, 93 S.Ct. 2230, 37 L.Ed.2d 63 (1973), we noted that "apportion[ment of] tuition rates on the basis of old and new residency . . . would give rise to grave problems under the Equal Protection Clause of the Fourteenth Amendment." Id., at 449-450, and n.6, 93 S.Ct., at 2234-2235, and n.6.10 17 If the states can make the amount of a cash dividend depend on length of residence, what would preclude varying university tuition on a sliding scale based on years of residence—or even limiting access to finite public facilities, eligibility for student loans, for civil service jobs, or for government contracts by length of domicile? Could states impose different taxes based on length of residence? Alaska's reasoning could open the door to state apportionment of other rights, benefits, and services according to length of residency.11 It would permit the states to divide citizens into expanding numbers of permanent classes.12 Such a result would be clearly impermissible.13 B 18 We need not consider whether the State could enact the dividend program prospectively only. Invalidation of a portion of a statute does not necessarily render the whole invalid unless it is evident that the legislature would not have enacted the legislation without the invalid portion. Buckley v. Valeo, 424 U.S. 1, 108, 96 S.Ct. 612, 677, 46 L.Ed.2d 659 (1976); United States v. Jackson, 390 U.S. 570, 585, 88 S.Ct. 1209, 1218, 20 L.Ed.2d 138 (1968); Champlin Refining Co. v. Corporation Comm'n of Oklahoma, 286 U.S. 210, 234, 52 S.Ct. 559, 564, 76 L.Ed. 1062 (1932). Here, we need not speculate as to the intent of the Alaska Legislature; the legislation expressly provides that invalidation of any portion of the statute renders the whole invalid: 19 "Sec. 4. If any provision enacted in sec. 2 of this Act [which included the dividend distribution plan in its entirety] is held to be invalid by the final judgment, decision or order of a court of competent jurisdiction, then that provision is nonseverable, and all provisions enacted in sec. 2 of this Act are invalid and of no force or effect." 1980 Alaska Sess. Laws, ch. 21, § 4. 20 However, it is of course for the Alaska courts to pass on the severability clause of the statute. III 21 The only apparent justification for the retrospective aspect of the program, "favoring established residents over new residents," is constitutionally unacceptable. Vlandis v. Kline, supra, at 450, 93 S.Ct., at 2235. In our view Alaska has shown no valid state interests which are rationally served by the distinction it makes between citizens who established residence before 1959 and those who have become residents since then. 22 We hold that the Alaska dividend distribution plan violates the guarantees of the Equal Protection Clause of the Fourteenth Amendment. Accordingly, the judgment of the Alaska Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. 23 Reversed and remanded. 24 Justice BRENNAN, with whom Justice MARSHALL, Justice BLACKMUN, and Justice POWELL join, concurring. 25 I join the opinion of the Court, and agree with its conclusion that the retrospective aspects of Alaska's dividend-distribution law are not rationally related to a legitimate state purpose. I write separately only to emphasize that the pervasive discrimination embodied in the Alaska distribution scheme gives rise to constitutional concerns of somewhat larger proportions than may be evident on a cursory reading of the Court's opinion. In my view, these concerns might well preclude even the prospective operation of Alaska's scheme. 26 * I agree with Justice O'CONNOR that these more fundamental defects in the Alaska dividend-distribution law are, in part, reflected in what has come to be called the "right to travel."1 That right—or, more precisely, the federal interest in free interstate migration—is clearly, though indirectly, affected by the Alaska dividend-distribution law, and this threat to free interstate migration provides an independent rationale for holding that law unconstitutional. At the outset, however, I note that the frequent attempts to assign the right to travel some textual source in the Constitution seem to me to have proved both inconclusive and unnecessary. Justice O'CONNOR plausibly argues, post, at 78-81, that the right predates the Constitution and was carried forward in the Privileges and Immunities Clause of Art. IV. But equally plausible, I think, is the argument that the right resides in the Commerce Clause, see Edwards v. California, 314 U.S. 160, 173, 62 S.Ct. 164, 166, 86 L.Ed. 119 (1941), or in the Privileges and Immunities Clause of the Fourteenth Amendment, see id., at 177-178, 62 S.Ct., at 168-169 (Douglas, J., concurring). In any event, in light of the unquestioned historic recognition of the principle of free interstate migration, and of its role in the development of the Nation, we need not feel impelled to "ascribe the source of this right to travel interstate to a particular constitutional provision." Shapiro v. Thompson, 394 U.S. 618, 630, 89 S.Ct. 1322, 1329, 22 L.Ed.2d 600 (1969). It suffices that: 27 " 'The constitutional right to travel from one State to another . . . occupies a position fundamental to the concept of our Federal Union. It is a right that has been firmly established and repeatedly recognized. 28 " '. . . [T]he right finds no explicit mention in the Constitution. The reason, it has been suggested, is that a right so elementary was conceived from the beginning to be a necessary concomitant of the stronger Union the Constitution created. In any event, freedom to travel throughout the United States has long been recognized as a basic right under the Constitution.' " Id., at 630-631, 89 S.Ct., at 1329, quoting United States v. Guest, 383 U.S. 745, 757-758, 86 S.Ct. 1170, 1177-1178, 16 L.Ed.2d 239 (1966). 29 As is clear from our cases, the right to travel achieves its most forceful expression in the context of equal protection analysis. But if, finding no citable passage in the Constitution to assign as its source, some might be led to question the independent vitality of the principle of free interstate migration, I find its unmistakable essence in that document that transformed a loose confederation of States into one Nation. A scheme of the sort adopted by Alaska is inconsistent with the federal structure even in its prospective operation. 30 A State clearly may undertake to enhance the advantages of industry, economy, and resources that make it a desirable place in which to live. In addition, a State may make residence within its boundaries more attractive by offering direct benefits to its citizens in the form of public services, lower taxes than other States offer, or direct distributions of its munificence. Through these means, one State may attract citizens of other States to join the numbers of its citizenry. That is a healthy form of rivalry: It inheres in the very idea of maintaining the States as independent sovereigns within a larger framework, and it is fully—indeed, necessarily—consistent with the Framers' further idea of joining these independent sovereigns into a single Nation. But a State cannot compound its offer of direct benefits in the inventive manner exemplified by the Alaska distribution scheme: For if each State were free to reward its citizens incrementally for their years of residence, so that a citizen leaving one State would thereby forfeit his accrued seniority, only to have to begin building such seniority again in his new State of residence, then the mobility so essential to the economic progress of our Nation, and so commonly accepted as a fundamental aspect of our social order, would not long survive. II 31 The Court today reaffirms the important principle that, at least with respect to a durational-residency discrimination, a State's desire "to reward citizens for past contributions" is clearly "not a legitimate state purpose." Ante, at 63. I do not think it "odd," post, at 72, that the Court disclaims reliance on the "right to travel" as the source of this limitation on state power. In my view, the acknowledged illegitimacy of that state purpose has a different heritage—it reflects not the structure of the Federal Union but the idea of constitutionally protected equality. See Shapiro v. Thompson, supra, at 632-633, 89 S.Ct., at 1330 ("The Equal Protection Clause prohibits such an apportionment of state services"); Vlandis v. Kline, 412 U.S. 441, 450, n.6, 93 S.Ct. 2230, 2235, n.6, 37 L.Ed.2d 63 (1973). The Constitution places the recently naturalized immigrant from a foreign land on an equal footing with those citizens of a State who are able to trace their lineage back for many generations within the State's borders. The 18-year-old native resident of a State is as much a citizen as the 55-year-old native resident. But the Alaska plan discriminates against the recently naturalized citizen, in favor of the Alaska citizen of longer duration; it discriminates against the 18-year-old native resident, in favor of all residents of longer duration. If the Alaska plan were limited to discriminations such as these, and did not purport to apply to migrants from sister States, interstate travel would not be noticeably burdened—yet those discriminations would surely be constitutionally suspect. 32 The Fourteenth Amendment guarantees the equal protection of the law to anyone who may be within the territorial jurisdiction of a State. That Amendment does not suggest by its terms that equal treatment might be denied a person depending upon how long that person has been within the jurisdiction of the State. The Fourteenth Amendment does, however, expressly recognize one elementary basis for distinguishing between persons who may be within a State's jurisdiction at any particular time—by setting forth the requirements for state citizenship. But it is significant that the Citizenship Clause of the Fourteenth Amendment expressly equates citizenship only with simple residence.2 That Clause does not provide for, and does not allow for, degrees of citizenship based on length of residence.3 And the Equal Protection Clause would not tolerate such distinctions. In short, as much as the right to travel, equality of citizenship is of the essence in our Republic. As the Court notes, States may not "divide citizens into expanding numbers of permanent classes." Ante, at 64. 33 It is, of course, elementary that the Constitution does not bar the States from making reasoned distinctions between citizens: Insofar as those distinctions are rationally related to the legitimate ends of the State they present no constitutional difficulty, as our equal protection jurisprudence attests. But we have never suggested that duration of residence vel non provides a valid justification for discrimination. To the contrary, discrimination on the basis of residence must be supported by a valid state interest independent of the discrimination itself. To be sure, allegiance and attachment may be rationally measured by length of residence—length of residence may, for example, be used to test the bona fides of citizenship—and allegiance and attachment may bear some rational relationship to a very limited number of legitimate state purposes. Cf. Chimento v. Stark, 353 F.Supp. 1211 (NH), summarily aff'd, 414 U.S. 802, 94 S.Ct. 125, 38 L.Ed.2d 39 (1973) (7-year citizenship requirement to run for Governor); U.S.Const., Art. I, § 2, cl. 2, § 3, cl. 3; Art. II, § 1, cl. 5. But those instances in which length of residence could provide a legitimate basis for distinguishing one citizen from another are rare. 34 Permissible discriminations between persons must bear a rational relationship to their relevant characteristics. While some imprecision is unavoidable in the process of legislative classification, the ideal of equal protection requires attention to individual merit, to individual need. In almost all instances, the business of the State is not with the past, but with the present: to remedy continuing injustices, to fill current needs, to build on the present in order to better the future. The past actions of individuals may be relevant in assessing their present needs; past actions may also be relevant in predicting current ability and future performance. In addition, to a limited extent, recognition and reward of past public service have independent utility for the State, for such recognition may encourage other people to engage in comparably meritorious service. But even the idea of rewarding past public service offers scarce support for the "past contribution" justification for durational-residence classifications since length of residence has only the most tenuous relation to the actual service of individuals to the State. 35 Thus, the past-contribution rationale proves much too little to provide a rational predicate for discrimination on the basis of length of residence. But it also proves far too much, for "it would permit the State to apportion all benefits and services according to the past . . . contributions of its citizens." Shapiro v. Thompson, 394 U.S., at 632-633, 89 S.Ct., at 1330. In effect, then, the past-contribution rationale is so far-reaching in its potential application, and the relationship between residence and contribution to the State so vague and insupportable, that it amounts to little more than a restatement of the criterion for discrimination that it purports to justify. But while duration of residence has minimal utility as a measure of things that are, in fact, constitutionally relevant, resort to duration of residence as the basis for a distribution of state largesse does closely track the constitutionally untenable position that the longer one's residence, the worthier one is of the State's favor. In my view, it is difficult to escape from the recognition that underlying any scheme of classification on the basis of duration of residence, we shall almost invariably find the unstated premise that "some citizens are more equal than others." We rejected that premise and, I believe, implicitly rejected most forms of discrimination based upon length of residence, when we adopted the Equal Protection Clause. 36 Justice O'CONNOR, concurring in the judgment. 37 The Court strikes Alaska's distribution scheme, purporting to rely solely upon the Equal Protection Clause of the Fourteenth Amendment. The phrase "right to travel" appears only fleetingly in the Court's analysis, dismissed with an observation that "right to travel analysis refers to little more than a particular application of equal protection analysis." Ante, at 60, n.6. The Court's reluctance to rely explicitly on a right to travel is odd, because its holding depends on the assumption that Alaska's desire "to reward citizens for past contributions . . . is not a legitimate state purpose." Ante, at 63. Nothing in the Equal Protection Clause itself, however, declares this objective illegitimate. Instead, as a full reading of Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969), and Vlandis v. Kline, 412 U.S. 441, 93 S.Ct. 2230, 37 L.Ed.2d 63 (1973), reveals, the Court has rejected this objective only when its implementation would abridge an interest in interstate travel or migration. 38 I respectfully suggest, therefore, that the Court misdirects its criticism when it labels Alaska's objective illegitimate. A desire to compensate citizens for their prior contributions is neither inherently invidious nor irrational. Under some circumstances, the objective may be wholly reasonable.1 Even a generalized desire to reward citizens for past endurance, particularly in a State where years of hardship only recently have produced prosperity, is not innately improper. The difficulty is that plans enacted to further this objective necessarily treat new residents of a State less favorably than the longer term residents who have past contributions to "reward." This inequality, as the Court repeatedly has recognized, conflicts with the constitutional purpose of maintaining a Union rather than a mere "league of States." See Paul v. Virginia, 8 Wall. 168, 180, 19 L.Ed. 357 (1869). The Court's task, therefore, should be (1) to articulate this constitutional principle, explaining its textual sources, and (2) to test the strength of Alaska's objective against the constitutional imperative. By choosing instead to declare Alaska's purpose wholly illegitimate, the Court establishes an uncertain jurisprudence. What makes Alaska's purpose illegitimate? Is the purpose illegitimate under all circumstances? What other state interests are wholly illegitimate? Will an "illegitimate" purpose survive review if it becomes "important" or "compelling"?2 These ambiguities in the Court's analysis prompt me to develop my own approach to Alaska's scheme. 39 Alaska's distribution plan distinguishes between long-term residents and recent arrivals. Stripped to its essentials, the plan denies non-Alaskans settling in the State the same privileges afforded longer term residents. The Privileges and Immunities Clause of Art. IV, which guarantees "[t]he Citizens of each State . . . all Privileges and Immunities of Citizens in the several States," addresses just this type of discrimination.3 Accordingly, I would measure Alaska's scheme against the principles implementing the Privileges and Immunities Clause. In addition to resolving the particular problems raised by Alaska's scheme, this analysis supplies a needed foundation for many of the "right to travel" claims discussed in the Court's prior opinions. 40 * Our opinions teach that Art. IV's Privileges and Immunities Clause "was designed to insure to a citizen of State A who ventures into State B the same privileges which the citizens of State B enjoy." Toomer v. Witsell, 334 U.S. 385, 395, 68 S.Ct. 1156, 1161, 92 L.Ed. 1460 (1948). The Clause protects a nonresident who enters a State to work, Hicklin v. Orbeck, 437 U.S. 518, 98 S.Ct. 2482, 57 L.Ed.2d 397 (1978), to hunt commercial game, Toomer, supra, or to procure medical services, Doe v. Bolton, 410 U.S. 179, 93 S.Ct. 739, 35 L.Ed.2d 201 (1973).4 A fortiori, the Privileges and Immunities Clause should protect the "citizen of State A who ventures into State B" to settle there and establish a home. 41 In this case, Alaska forces nonresidents settling in the State to accept a status inferior to that of oldtimers. In its first year of operation, the distribution scheme would have given $1,050 to an Alaskan who had lived in the State since statehood. A resident of 10 years would have received $500, while a one-year resident would have received only $50. In effect, therefore, the State told its citizens: "Your status depends upon the date on which you established residence here. Those of you who migrated to the State cannot share its bounty on the same basis as those who were here before you." Surely this scheme imposes one of the "disabilities of alienage" prohibited by Art. IV's Privileges and Immunities Clause. See Paul v. Virginia, supra, at 180. 42 It could be argued that Alaska's scheme does not trigger the Privileges and Immunities Clause because it discriminates among classes of residents, rather than between residents and nonresidents. This argument, however, misinterprets the force of Alaska's distribution system. Alaska's scheme classifies citizens on the basis of their former residential status, imposing a relative burden on those who migrated to the State after 1959. Residents who arrived in Alaska after that date have a less valuable citizenship right than do the oldtimers who preceded them. Citizens who arrive in the State tomorrow will receive an even smaller claim on Alaska's resources. The fact that this discrimination unfolds after the nonresident establishes residency does not insulate Alaska's scheme from scrutiny under the Privileges and Immunities Clause. Each group of citizens who migrated to Alaska in the past, or chooses to move there in the future, lives in the State on less favorable terms than those who arrived earlier. The circumstance that some of the disfavored citizens already live in Alaska does not negate the fact that "the citizen of State A who ventures into [Alaska]" to establish a home labors under a continuous disability.5 43 If the Privileges and Immunities Clause applies to Alaska's distribution system, then our prior opinions describe the proper standard of review. In Baldwin v. Montana Fish and Game Comm'n, 436 U.S. 371, 98 S.Ct. 1852, 56 L.Ed.2d 354 (1978), we held that States must treat residents and nonresidents "without unnecessary distinctions" when the nonresident seeks to "engage in an essential activity or exercise a basic right." Id., at 387, 98 S.Ct., at 1862. On the other hand, if the nonresident engages in conduct that is not "fundamental" because it does not "bea[r] upon the vitality of the Nation as a single entity," the Privileges and Immunities Clause affords no protection. Id., at 387, 383, 98 S.Ct., at 1862, 1860. 44 Once the Court ascertains that discrimination burdens an "essential activity," it will test the constitutionality of the discrimination under a two-part test. First, there must be " 'something to indicate that non-citizens constitute a peculiar source of the evil at which the statute is aimed.' " Hicklin v. Orbeck, supra, at 525-526, 98 S.Ct., at 2487 (quoting Toomer v. Witsell, supra, at 398, 68 S.Ct., at 1163). Second, the Court must find a "substantial relationship" between the evil and the discrimination practiced against the noncitizens. 437 U.S., at 527, 98 S.Ct., at 2488. 45 Certainly the right infringed in this case is "fundamental." Alaska's statute burdens those nonresidents who choose to settle in the State.6 It is difficult to imagine a right more essential to the Nation as a whole than the right to establish residence in a new State. Just as our federal system permits the States to experiment with different social and economic programs, New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386, 76 L.Ed. 747 (1932) (Brandeis, J., dissenting), it allows the individual to settle in the State offering those programs best tailored to his or her tastes.7 Alaska's encumbrance on the right of nonresidents to settle in that State, therefore, must satisfy the dual standard identified in Hicklin. 46 Alaska has not shown that its new residents are the "peculiar source" of any evil addressed by its disbursement scheme. The State does not argue that recent arrivals constitute a particular source of its population turnover problem. Indeed, the State urges that it has a special interest in persuading young adults, who have grown to maturity in the State, to remain there. Brief for Appellees 35, n.24. Nor is there any evidence that new residents, rather than old, will foolishly deplete the State's mineral and financial resources. Finally, although Alaska argues that its scheme compensates residents for their prior tangible and intangible contributions to the State, nonresidents are hardly a peculiar source of the "evil" of partaking in current largesse without having made prior contributions. A multitude of native Alaskans—including children and paupers—may have failed to contribute to the State in the past. Yet the State does not dock paupers for their prior failures to contribute, and it awards every person over the age of 18 dividends equal to the number of years that person has lived in the State. 47 Even if new residents were the peculiar source of these evils, Alaska has not chosen a cure that bears a "substantial relationship" to the malady. As the dissenting judges below observed, Alaska's scheme gives the largest dividends to residents who have lived longest in the State. The dividends awarded to new residents may be too small to encourage them to stay in Alaska. The size of these dividends appears to give new residents only a weak interest in prudent management of the State's resources. As a reward for prior contributions, finally, Alaska's scheme is quite ill-suited. While the phrase "substantial relationship" does not require mathematical precision, it demands at least some recognition of the fact that persons who have migrated to Alaska may have contributed significantly more to the State, both before and after their arrival, than have some natives. 48 For these reasons, I conclude that Alaska's disbursement scheme violates Art. IV's Privileges and Immunities Clause. I thus reach the same destination as the Court, but along a course that more precisely identifies the evils of the challenged statute. II 49 The analysis outlined above might apply to many cases in which a litigant asserts a right to travel or migrate interstate.8 To historians, this would come as no surprise. Article IV's Privileges and Immunities Clause has enjoyed a long association with the rights to travel and migrate interstate. 50 The Clause derives from Art. IV of the Articles of Confederation. The latter expressly recognized a right of "free ingress and regress to and from any other State," in addition to guaranteeing "the free inhabitants of each of these states . . . [the] privileges and immunities of free citizens in the several States."9 While the Framers of our Constitution omitted the reference to "free ingress and regress," they retained the general guaranty of "privileges and immunities." Charles Pinckney, who drafted the current version of Art. IV, told the Convention that this Article was "formed exactly upon the principles of the 4th article of the present Confederation." 3 M. Farrand, Records of the Federal Convention of 1787, p. 112 (1934). Commentators, therefore, have assumed that the Framers omitted the express guaranty merely because it was redundant, not because they wished to excise the right from the Constitution.10 51 Early opinions by the Justices of this Court also traced a right to travel or migrate interstate to Art. IV's Privileges and Immunities Clause. In Corfield v. Coryell, 6 F.Cas. 546, 552 (No. 3,230) (CC ED Pa.1823), for example, Justice Washington explained that the Clause protects the "right of a citizen of one state to pass through, or to reside in any other state." Similarly, in Paul v. Virginia, 8 Wall., at 180, the Court found that one of the "undoubt[ed]" effects of the Clause was to give "the citizens of each State . . . the right of free ingress into other States, and egress from them. . . ." See also Ward v. Maryland, 12 Wall. 418, 430, 20 L.Ed. 449 (1871). Finally, in United States v. Wheeler, 254 U.S. 281, 297-298, 41 S.Ct. 133, 135, 65 L.Ed. 270 (1920), the Court found that the Clause fused two distinct concepts: (1) "the right of citizens of the States to reside peacefully in, and to have free ingress into and egress from" their own States, and (2) the right to exercise the same privileges in other States. 52 History, therefore, supports assessment of Alaska's scheme, as well as other infringements of the right to travel, under the Privileges and Immunities Clause. This Clause may not address every conceivable type of discrimination that the Court previously has denominated a burden on interstate travel. I believe, however, that application of the Privileges and Immunities Clause to controversies involving the "right to travel" would at least begin the task of reuniting this elusive right with the constitutional principles it embodies. Because I believe that Alaska's distribution scheme violates the Privileges and Immunities Clause of Art. IV, I concur in the Court's judgment insofar as it reverses the judgment of the Alaska Supreme Court. 53 Justice REHNQUIST, dissenting. 54 Alaska's dividend distribution scheme represents one State's effort to apportion unique economic benefits among its citizens. Although the wealth received from the oil deposits of Prudhoe Bay may be quite unlike the economic resources enjoyed by most States, Alaska's distribution of that wealth is in substance no different from any other State's allocation of economic benefits. The distribution scheme being in the nature of economic regulation, I am at a loss to see the rationality behind the Court's invalidation of it as a denial of equal protection. This Court has long held that state economic regulations are presumptively valid, and violate the Fourteenth Amendment only in the rarest of circumstances: 55 "When local economic regulation is challenged solely as violating the Equal Protection Clause, this Court consistently defers to legislative determinations as to the desirability of particular statutory discriminations. See, e.g., Lehnhausen v. Lake Shore Auto Parts Co., 410 U.S. 356 [93 S.Ct. 1001, 35 L.Ed.2d 351] (1973). Unless a classification trammels fundamental personal rights or is drawn upon inherently suspect distinctions such as race, religion, or alienage, our decisions presume the constitutionality of the statutory discriminations and require only that the classification challenged be rationally related to a legitimate state interest. States are accorded wide latitude in the regulation of their local economies under their police powers, and rational distinctions may be made with substantially less than mathematical exactitude." New Orleans v. Dukes, 427 U.S. 297, 303, 96 S.Ct. 2513, 2516, 49 L.Ed.2d 511 (1976). See also Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 101 S.Ct. 715, 66 L.Ed.2d 659 (1981); United States Railroad Retirement Board v. Fritz, 449 U.S. 166, 101 S.Ct. 453, 66 L.Ed.2d 368 (1980); Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 96 S.Ct. 2488, 49 L.Ed.2d 220 (1976). 56 Despite the highly deferential approach which we invariably have taken toward state economic regulations, the Court today finds the retroactive aspect of the Alaska distribution scheme violative of the Fourteenth Amendment. The Court concludes that the State's first two justifications are not rationally related to the retroactive portion of the distribution scheme, and that the third justification—the reward of citizens for their past contributions—is not a legitimate state objective. But the illegitimacy of a State's recognizing the past contributions of its citizens has been established by the Court only in certain cases considering an infringement of the right to travel,1 and the majority itself rightly declines to apply the strict scrutiny analysis of those right-to-travel cases. See ante, at 60-61. The distribution scheme at issue in this case impedes no person's right to travel to and settle in Alaska; if anything, the prospect of receiving annual cash dividends would encourage immigration to Alaska. The State's third justification cannot, therefore, be dismissed simply by quoting language about its legitimacy from right-to-travel cases which have no relevance to the question before us. 57 So understood, this case clearly passes equal protection muster. There can be no doubt that the state legislature acted rationally when it concluded that dividends retroactive to the year of statehood would "recognize the 'contributions of various kinds, both tangible and intangible,' which residents have made during their years of state residency." 619 P.2d 448, 458 (Alaska 1980). Nor can there be any doubt that Alaska, perhaps more than any other State in the Union, has good reason for recognizing such contributions.2 Because the distribution scheme is thus rationally based, I dissent from its invalidation under the guise of equal protection analysis.3 In striking down the Alaskan scheme, the Court seems momentarily to have forgotten "the principle that the Fourteenth Amendment gives the federal courts no power to impose upon the States their views of what constitutes wise economic or social policy." Dandridge v. Williams, 397 U.S. 471, 486, 90 S.Ct. 1153, 1162, 25 L.Ed.2d 491 (1970). 1 Alaska Dept. of Revenue, Revenue Sources FY 1981-1983 (Sept.1981). (Includes General Fund unrestricted petroleum revenues of $3.3 billion and petroleum revenues directly deposited in the Permanent Fund in the amount of $400 million. An additional $900 million was transferred from the General Fund to the Permanent Fund in the 1981 fiscal year.) The 1980 census reports that Alaska's adult population is 270,265; per capita 1981 oil revenues amount to $13,632 for each adult resident. Petroleum revenues now amount to 89% of the State's total government revenue. Ibid. 2 The infusion of Permanent Fund earnings into state general revenues also led the Alaska Legislature to enact a statute giving residents a one-third exemption from state income taxes for each year of residence; this operated to exempt entirely anyone with three or more years of residency. The Alaska Supreme Court, again by a 3-2 vote, held that this statute violated the State Constitution's equal protection clause. Williams v. Zobel, 619 P.2d 422 (1980). Chief Justice Rabinowitz, the only justice in the majority in both cases, found that the tax exemption statute, but not the dividend distribution plan, could "be perceived as a penalty imposed on a person who chooses to exercise his or her right to move into Alaska." 619 P.2d, at 458. 3 In the durational residency cases, we examined state laws which imposed waiting periods on access to divorce courts, Sosna v. Iowa; eligibility for free nonemergency medical care, Memorial Hospital v. Maricopa County; voting rights, Dunn v. Blumstein; and welfare assistance, Shapiro v. Thompson. 4 Section 43.23.010(b) provides: "For each year, an individual is eligible to receive payment of the permanent fund dividends for which he is entitled under this section if he "(1) is at least 18 years of age; and "(2) is a state resident during all or part of the year for which the permanent fund dividend is paid." The remainder of § 43.23.010 establishes the number of dividend units residents are entitled to receive and the method of payment. Section 43.23.010(f) provides that a resident entitled to benefits under subsection (b) who was a resident for less than a full year is entitled to a dividend prorated on the basis of the number of months of state residence. 5 The Alaska statute does not simply make distinctions between native-born Alaskans and those who migrate to Alaska from other states; it does not discriminate only against those who have recently exercised the right to travel, as did the statute involved in Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969). The Alaska statute also discriminates among long-time residents and even native-born residents. For example, a person born in Alaska in 1962 would have received $100 less than someone who was born in the State in 1960. Of course the native Alaskan born in 1962 would also receive $100 less than the person who moved to the State in 1960. The statute does not involve the kind of discrimination which the Privileges and Immunities Clause of Art. IV was designed to prevent. That Clause "was designed to insure to a citizen of State A who ventures into State B the same privileges which the citizens of State B enjoy." Toomer v. Witsell, 334 U.S. 385, 395, 68 S.Ct. 1156, 1161, 92 L.Ed. 1460 (1948). The Clause is thus not applicable to this case. 6 The Alaska courts considered whether the dividend distribution law violated appellants' constitutional right to travel. The right to travel and to move from one state to another has long been accepted, yet both the nature and the source of that right have remained obscure. See Jones v. Helms, 452 U.S. 412, 417-419, and nn.12 and 13, 101 S.Ct. 2434, 2439-2440, and nn.12 and 13, 69 L.Ed.2d 118 (1981); Shapiro v. Thompson, supra, at 629-631, 89 S.Ct., at 1328-1329; United States v. Guest, 383 U.S. 745, 757-759, 86 S.Ct. 1170, 1177-1178, 16 L.Ed.2d 239 (1966). See also Z. Chafee, Three Human Rights in the Constitution of 1787, pp. 188-193 (1956). In addition to protecting persons against the erection of actual barriers to interstate movement, the right to travel, when applied to residency requirements, protects new residents of a state from being disadvantaged because of their recent migration or from otherwise being treated differently from longer term residents. In reality, right to travel analysis refers to little more than a particular application of equal protection analysis. Right to travel cases have examined, in equal protection terms, state distinctions between newcomers and longer term residents. See Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306 (1974); Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972); Shapiro v. Thompson, supra. This case also involves distinctions between residents based on when they arrived in the State and is therefore also subject to equal protection analysis. 7 These purposes were enumerated in the first section of the Act creating the dividend distribution plan, 1980 Alaska Sess.Laws, ch. 21, § 1(b): "(b) The purposes of this Act are "(1) to provide a mechanism for equitable distribution to the people of Alaska of at least a portion of the state's energy wealth derived from the development and production of the natural resources belonging to them as Alaskans; "(2) to encourage persons to maintain their residence in Alaska and to reduce population turnover in the state; and "(3) to encourage increased awareness and involvement by the residents of the state in the management and expenditure of the Alaska permanent fund (art. IX, sec. 15, state constitution)." Thus we need not speculate as to the objectives of the legislature. 8 In response to the argument that the objectives of stabilizing population and encouraging prudent management of the Permanent Fund and of the State's natural resources did not justify the application of the dividend program to the years 1959 to 1980, the Alaska Supreme Court maintained that the retrospective aspect of the program was justified by the objective of rewarding state citizens for past contributions. 619 P.2d, at 461-462, n.37. See also dissenting opinion of Justice Dimond, id., at 469-471. 9 In fact, newcomers seem more likely to become dissatisfied and to leave the State than well-established residents; it would thus seem that the State would give a larger, rather than a smaller, dividend to new residents if it wanted to discourage emigration. The separation of residents into classes hardly seems a likely way to persuade new Alaskans that the State welcomes them and wants them to stay. Of course, the State's objective of reducing population turnover cannot be interpreted as an attempt to inhibit migration into the State without encountering insurmountable constitutional difficulties. See Shapiro v. Thompson, 394 U.S., at 629, 89 S.Ct., at 1328. 10 Even if the objective of rewarding past contributions were valid, it would be ironic to apply that rationale here. As Representative Randolph noted during debate in the state legislature on the dividend statute: "The pipeline is the entity that has allowed us all this latitude to do all the things we're considering doing, not only today but throughout the session. And without . . . newcomers, we couldn't have built that pipeline. Without their skill, without their ability, without their money, the pipeline wouldn't be there. So I get a little bit tired of—and I've got a hunch an awful lot of people who have been here five or six or seven or ten years, whatever we knock off as newcomers, get a little bit tired of being chastized and penalized and discriminated against for having not been born here or not have been here 30 or 40 or 50 years." 11 Apportionment would thus be prohibited only when it involves "fundamental rights" and services deemed to involve "basic necessities of life." See Memorial Hospital v. Maricopa County, 415 U.S., at 259, 94 S.Ct., at 1082. 12 "Such a power in the States could produce nothing but discord and mutual irritation, and they very clearly do not possess it." Passenger Cases, 7 How. 283, 492, 12 L.Ed. 702 (1849) (Taney, C.J., dissenting). 13 Starns v. Malkerson, 326 F.Supp. 234 (Minn.1970), summarily aff'd, 401 U.S. 985, 91 S.Ct. 1231, 28 L.Ed.2d 527 (1971), cannot be read as a contrary decision of this Court. First, summary affirmance by this Court is not to be read as an adoption of the reasoning supporting the judgment under review. Fusari v. Steinberg, 419 U.S. 379, 391, 95 S.Ct. 533, 540, 42 L.Ed.2d 521 (1975) (concurring opinion). See also Colorado Springs Amusements, Ltd. v. Rizzo, 428 U.S. 913, 920-921, 96 S.Ct. 3228, 3232, 49 L.Ed.2d 1222 (1976) (BRENNAN, J., dissenting); Edelman v. Jordan, 415 U.S. 651, 671, 94 S.Ct. 1347, 1359, 39 L.Ed.2d 662 (1974). Moreover, as we pointed out in Vlandis v. Kline, 412 U.S. 441, 452-453, n.9, 93 S.Ct. 2230, 2236-2237, n.9, 37 L.Ed.2d 63 (1973), we considered the Minnesota one-year residency requirement examined in Starns a test of bona fide residence, not a return on prior contributions to the commonwealth. 1 What is notably at stake in this case, and what clearly must be taken into account in determining the constitutionality of this legislative scheme, is the national interest in a fluid system of interstate movement. It may be that national interests are not always easily translated into individual rights, but where the "right to travel" is involved, our cases leave no doubt that it will trigger intensified equal protection scrutiny. See, e.g., Memorial Hospital v. Maricopa County, 415 U.S. 250, 94 S.Ct. 1076, 39 L.Ed.2d 306 (1974); Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972); Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969). As the Court notes, the "right to travel" is implicated not only by "actual barriers to interstate movement," but also by "state distinctions between newcomers and longer term residents." Ante, at 60, n.6. 2 "[A] citizen of the United States can, of his own volition, become a citizen of any State of the Union by a bona fide residence therein, with the same rights as other citizens of that State." Slaughter-House Cases, 16 Wall. 36, 80, 21 L.Ed. 394 (1873). See id., at 112-113 (Bradley, J., dissenting) ("A citizen of the United States has a perfect constitutional right to go to and reside in any State he chooses, and to claim citizenship therein, and an equality of rights with every other citizen"). 3 The American aversion to aristocracy developed long before the Fourteenth Amendment and is, of course, reflected elsewhere in the Constitution. See Art. I, § 9, cl. 8 ("No Title of Nobility shall be granted by the United States"). See also Virginia Declaration of Rights (1776), in R. Rutland, The Birth of the Bill of Rights, App. A (1955) ("no man, or set of men, are entitled to exclusive or separate emoluments or privileges from the community, but in consideration of publick services"). 1 A State, for example, might choose to divide its largesse among all persons who previously have contributed their time to volunteer community organizations. If the State graded its dividends according to the number of years devoted to prior community service, it could be said that the State intended "to reward citizens for past contributions." Alternatively, a State might enact a tax credit for citizens who contribute to the State's ecology by building alternative fuel sources or establishing recycling plants. If the State made this credit retroactive, to benefit those citizens who launched these improvements before they became fashionable, the State once again would be rewarding past contributions. The Court's opinion would dismiss these objectives as wholly illegitimate. I would recognize them as valid goals and inquire only whether their implementation infringed any constitutionally protected interest. 2 The Court's conclusion that Alaska's scheme lacks a rational basis masks a puzzling aspect of its analysis. By refusing to extend any legitimacy to Alaska's objective, the Court implies that a program designed to reward prior contributions will never survive equal protection scrutiny. For example, the programs described in n.1, supra, could not survive the Court's analysis even if the State demonstrated a compelling interest in rewarding volunteer activity or promoting conservation measures. The Court's opinion, although purporting to apply a deferential standard of review, actually insures that any governmental program depending upon a "past contributions" rationale will violate the Equal Protection Clause. 3 While the Clause refers to "Citizens," this Court has found that "the terms 'citizen' and 'resident' are 'essentially interchangeable' . . . for purposes of analysis of most cases under the Privileges and Immunities Clause." Hicklin v. Orbeck, 437 U.S. 518, 524, n.8, 98 S.Ct. 2482, 2486, n.8, 57 L.Ed.2d 397 (1978) (quoting Austin v. New Hampshire, 420 U.S. 656, 662, n.8, 95 S.Ct. 1191, 1195, n.8, 43 L.Ed.2d 530 (1975)). This opinion, therefore, will refer to "nonresidents" of Alaska, as well as to "noncitizens" of that State. It is settled that the Privileges and Immunities Clause does not protect corporations. See Paul v. Virginia, 8 Wall. 168, 19 L.Ed. 357 (1869). The word "Citizens" suggests that the Clause also excludes aliens. See, e.g., id., at 177 (dictum); L. Tribe, American Constitutional Law § 6-33, p. 411, n.18 (1978). Any prohibition of discrimination aimed at aliens or corporations must derive from other constitutional provisions. 4 See generally Ward v. Maryland, 12 Wall. 418, 430, 20 L.Ed. 449 (1871) (The Clause "plainly and unmistakably secures and protects the right of a citizen of one State to pass into any other State of the Union for the purpose of engaging in lawful commerce, trade, or business, without molestation; to acquire personal property; [and] to take and hold real estate. . ."). 5 See Note, A Constitutional Analysis of State Bar Residency Requirements under the Interstate Privileges and Immunities Clause of Article IV, 92 Harv.L.Rev. 1461, 1464-1465, n.17 (1979) (labeling contrary argument "technical"). As the Court points out, ante, at 59-60, n.5, Alaska's plan differentiates even among native Alaskans, by tying their benefits to date of birth. If the scheme merely distributed benefits on the basis of age, without reference to the date beneficiaries established residence in Alaska, I doubt it would violate the Privileges and Immunities Clause. Under those circumstances, a 25-year-old Texan establishing residence in Alaska would acquire the same privileges of citizenship held by a 25-year-old native Alaskan. The scheme would not treat the citizen who moves to the State differently from citizens who already reside there. The Court does not explain whether it would find such an age-based scheme objectionable. 6 The "burden" imposed on nonresidents is relative to the benefits enjoyed by residents. It is immaterial, for purposes of the Privileges and Immunities Clause, that the nonresident may enjoy a benefit in the new State that he lacked completely in his former State. The Clause addresses only differences in treatment; it does not judge the quality of treatment a State affords citizens and noncitizens. 7 See also Baldwin v. G.A.F. Seelig, Inc., 294 U.S. 511, 523, 55 S.Ct. 497, 500, 79 L.Ed. 1032 (1935) (the Constitution "was framed upon the theory that the peoples of the several states must sink or swim together, and that in the long run prosperity and salvation are in union and not division"); Paul v. Virginia, 8 Wall., at 180 ("Indeed, without some provision of the kind removing from the citizens of each State the disabilities of alienage in the other States, and giving them equality of privilege with citizens of those States, the Republic would have constituted little more than a league of States; it would not have constituted the Union which now exists"); Edwards v. California, 314 U.S. 160, 173, 62 S.Ct. 164, 166, 86 L.Ed. 119 (1941) (Constitution prohibits "attempts on the part of any single State to isolate itself from difficulties common to all of them by restraining the transportation of persons and property across its borders"). 8 Any durational residency requirement, for example, treats nonresidents who have exercised their right to settle in a State differently from longer term residents. This is not to say, however, that all such requirements would fail scrutiny under the Privileges and Immunities Clause. The durational residency requirement upheld in Sosna v. Iowa, 419 U.S. 393, 95 S.Ct. 553, 42 L.Ed.2d 532 (1975) (one year to obtain divorce), for example, would have survived under the analysis outlined above. In Sosna the State showed that nonresidents were a peculiar source of the evil addressed by its durational residency requirement. Those persons could misrepresent their attachment to Iowa and obtain divorces that would be susceptible to collateral attack in other States. Iowa adopted a reasonable response to this problem by requiring nonresidents to demonstrate their bona fide residency for one year before obtaining a divorce. I am confident that the analysis developed in Hicklin v. Orbeck, 437 U.S. 518, 98 S.Ct. 2482, 57 L.Ed.2d 397 (1978), will adequately identify other legitimate durational residency requirements. 9 Even before adoption of the Articles, a few of the Colonies explicitly protected freedom of movement. The Rhode Island Charter gave members of that Colony the right "to passe and repasse with freedome, into and through the rest of the English Collonies, upon their lawful and civill occasions." Z. Chafee, Three Human Rights in the Constitution of 1787, p. 177 (1956). The Massachusetts Body of Liberties provided: "Every man of or within this Jurisdiction shall have free libertie, not with standing any Civill power, to remove both himselfe and his familie at their pleasure out of the same, provided there be no legall impediment to the contrarie." Id., at 178. Massachusetts showed some of the same liberality to foreigners entering the Colony: "If any people of other Nations professing the true Christian Religion shall flee to us from the Tiranny or oppression of their persecutors, or from famyne, warres, or the like necessary and compulsarie cause, They shall be entertayned and succoured among us, according to that power and prudence god shall give us." Ibid. These attitudes contrasted with the more restrictive views prevailing in 17th-century Europe. See generally id., at 163-171. 10 See, e.g., id., at 185; Note, The Right to Travel and Exclusionary Zoning, 26 Hastings L.J. 849, 858-859 (1975); Comment, The Right to Travel: In Search of a Constitutional Source, 55 Neb.L.Rev. 117, 119-120, n. 14 (1975); Comment, A Strict Scrutiny of the Right to Travel, 22 UCLA L.Rev. 1129, 1130, n.7 (1975). See also Austin v. New Hampshire, 420 U.S., at 661, 95 S.Ct., at 1195 (footnotes omitted) (Article IV of the Articles of Confederation was "carried over into the comity article of the Constitution in briefer form but with no change of substance or intent, unless it was to strengthen the force of the Clause in fashioning a single nation"); United States v. Wheeler, 254 U.S. 281, 294, 41 S.Ct. 133, 134, 65 L.Ed. 270 (1920) ("the text of Article IV, § 2, of the Constitution, makes manifest that it was drawn with reference to the corresponding clause of the Articles of Confederation and was intended to perpetuate its limitations; and . . . that view has been so conclusively settled as to leave no room for controversy"). 1 The Court relies upon Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969), and Vlandis v. Kline, 412 U.S. 441, 93 S.Ct. 2230, 37 L.Ed.2d 63 (1973), in holding that Alaska may not justify its dividend distribution scheme by a desire to reward its citizens for their past contributions. In Shapiro, however, the Court found that the classification at issue "touche[d] on the fundamental right of interstate movement" and therefore could be justified only if it promoted a "compelling state interest." 394 U.S., at 638, 89 S.Ct., at 1333 (emphasis in original). Similarly, Vlandis concerned the right to move to and establish residency in Connecticut, and noted only in dicta that rewarding citizens for their past contributions was an impermissible state objective. See 412 U.S., at 449-450, and n. 6, 93 S.Ct., at 2234-2235, and n.6. Although I have expressed my disagreement with this holding even in the right-to-travel cases, see Memorial Hospital v. Maricopa County, 415 U.S. 250, 286-287, 94 S.Ct. 1076, 1095-1096, 39 L.Ed.2d 306 (1974) (REHNQUIST, J., dissenting); Vlandis v. Kline, supra, at 468-469, 93 S.Ct., at 2244 (same), there is no need to rely upon that dissenting position here. The majority does not analyze this as a right-to-travel case. Compare ante, at 60-61, with Memorial Hospital v. Maricopa County, supra, at 261-262, 94 S.Ct., at 1083-1084, and Shapiro v. Thompson, supra, at 634, 638, 89 S.Ct., at 1331, 1333. 2 As the Alaska Supreme Court noted, those who have lived in Alaska from the year of its statehood have borne unusual expenses and hardships: " 'A government such as the one embodied in the Alaska constitution, . . . with its complete range of governmental services, was expensive for a State with limited sources of taxation. Alaska could only boast a couple of pulp mills. . . . The State's business enterprises were small and catered mostly to local needs. In addition, Alaska's population was modest and hardly amounted to more than that of a medium-sized city in the continental United States. " 'Accordingly, revenues were small. Yet, the demands were great. The State government had to provide all the governmental services and social overhead required by modern American society. For instance, it would have been relatively simple to build a few roads, furnish normal police protection, and establish the customary school facilities. But nothing was normal in Alaska; it was and remains a land of superlatives. Subarctic engineering is relatively new, but the State would have to face the problem of permafrost conditions that frequently cause the roadtop to buckle and heave. Police protection would have to be provided for an area one-fifth the size of the forty-eight United States but with very few roads available. Flying would become a way of life for law enforcement officials as well as other Alaskans—an expensive way of life. "Bush schools" scattered along the Aleutian chain, through the Yukon Valley, and on the Seaward Peninsula and the islands of southeastern Alaska were expensive to maintain. It was not until the discovery of oil on a large scale that the picture changed.' " 619 P.2d, at 462, n.37 (quoting C. Naske, An Interpretive History of Alaskan Statehood 169-170 (1973)). 3 I also disagree with the suggestion of Justice O'CONNOR that the Alaska distribution scheme contravenes the Privileges and Immunities Clause of Art. IV of the Constitution. That Clause assures that nonresidents of a State shall enjoy the same privileges and immunities as residents enjoy: "It was designed to insure to a citizen of State A who ventures into State B the same privileges which the citizens of State B enjoy." Toomer v. Witsell, 334 U.S. 385, 395, 68 S.Ct. 1156, 1161, 92 L.Ed. 1460 (1948). We long ago held that the Clause has no application to a citizen of the State whose laws are complained of. "The constitutional provision there alluded to did not create those rights, which it called privileges and immunities of citizens of the States. It threw around them in that clause no security for the citizen of the State in which they were claimed or exercised. Nor did it profess to control the power of the State governments over the rights of its own citizens." Slaughter-House Cases, 16 Wall. 36, 77, 21 L.Ed. 394 (1873).

12

457 U.S. 102 102 S.Ct. 2339 72 L.Ed.2d 707 UNITED STEELWORKERS OF AMERICA, AFL-CIO-CLC, Petitionerv.Edward SADLOWSKI, Jr., et al. No. 81-395. Argued March 31, 1982. Decided June 14, 1982. Rehearing Denied Oct. 4, 1982. See 459 U.S. 899, 103 S.Ct. 199. Syllabus Petitioner union amended its constitution to include an "outsider rule" which prohibits candidates for union office from accepting campaign contributions from nonmembers and creates a committee to enforce the rule, the committee's decisions being final and binding. Respondents, including a union member who had been an unsuccessful candidate for union office before adoption of the outsider rule and had received much of the financial support for his campaign from sources outside the union, filed suit against petitioner in Federal District Court, claiming that the rule prohibited nonmember contributions to finance campaign-related litigation and thus violated § 101(a)(4) of the Labor-Management Reporting and Disclosure Act of 1959 (LMRDA), which provides that a union may not limit the rights of its members to institute an action in any court or administrative agency. The District Court found for respondents. The Court of Appeals affirmed, agreeing that the outsider rule violated § 101(a)(4). It also accepted respondents' argument, first raised on appeal, that the rule violated the "freedom of speech and assembly" provision of § 101(a)(2) of the LMRDA giving every union member the right to assemble freely with other members and to express at union meetings his views about candidates in union elections or any business properly before the meeting. The Court of Appeals rejected petitioner's argument that the outsider rule was protected by § 101(a)(2)'s proviso, which gives a union authority to adopt "reasonable" rules regarding its members' responsibilities. Held : 1. Petitioner's outsider rule does not violate § 101(a)(2). Although it may interfere with rights Congress intended to protect, it is rationally related to a legitimate and protected purpose, and thus is sheltered by § 101(a)(2)'s proviso. Pp. 108-119. (a) In light of the legislative history, § 101(a)(2) cannot be read as incorporating the entire body of First Amendment law so as to require that the scope of protections afforded union members by the statute coincide with the protections afforded by the Constitution as to a political election candidate's freedom to receive campaign contributions. Union rules are valid under the statute so long as they are reasonable; they need not pass the stringent tests applied in the First Amendment context. Pp. 108-111. (b) Congress adopted the freedom of speech and assembly provision of § 101(a)(2) in order to promote union democracy, particularly through fostering vigorous debate during election campaigns. Although petitioner's outsider rule does affect rights protected by the statute and may limit somewhat the ability of insurgent union members to wage an effective campaign against incumbent officers, as a practical matter the impact may not be substantial. The record shows that challengers have been able to defeat incumbents or administration-backed candidates, despite the absence of financial support from nonmembers. Pp. 111-115. (c) Petitioner's purpose in adopting the outsider rule was to ensure that nonmembers would not unduly influence union affairs and that the union leadership would remain responsive to the membership. The policies underlying the LMRDA show that this is a legitimate purpose that Congress meant to protect. Nor is the rule invalid on the asserted ground that it is not rationally related to that purpose because the union could have simply established contribution ceilings, or need not have limited contributions by relatives and friends, or could have simply required that candidates reveal the sources of their funds. Petitioner had a reasonable basis for its decision to impose a broad ban seeking to eradicate the threat of outside influence. Pp. 115-119. 2. Petitioner's outsider rule does not violate § 101(a)(4)'s right-to-sue provision. The rule simply does not apply where a member uses funds from outsiders to finance litigation. Neither the rule's language nor the debates leading up to its passage indicate that petitioner intended the rule to apply in such context. Moreover, petitioner's rule-enforcement committee issued an opinion stating that the rule's limitations "do not apply to the financing of lawsuits by non-members for the purpose of asserting the legal rights of candidates or other union members in connection with elections." Pp. 119-121. 207 U.S.App.D.C. 189, 645 F.2d 1114, reversed and remanded. Michael H. Gottesman, Washington, D. C., for petitioner. Joseph L. Rauh, Jr., Washington, D. C., for respondents. Justice MARSHALL delivered the opinion of the Court. 1 In this case, we confront the question whether § 101(a)(2) of the Labor-Management Reporting and Disclosure Act of 1959 (LMRDA), 73 Stat. 522, 29 U.S.C. § 411(a)(2), precludes the membership of a union from adopting a rule that prohibits candidates for union office from accepting campaign contributions from nonmembers. The United States Court of Appeals for the District of Columbia Circuit held that such a rule violated § 101(a)(2). 207 U.S.App.D.C. 189, 645 F.2d 1114 (1981). We granted certiorari, 454 U.S. 962, 102 S.Ct. 500, 70 L.Ed.2d 376 (1981), and now reverse. 2 * A. 3 Petitioner United Steelworkers of America (USWA), a labor organization with 1,300,000 members, conducts elections for union president and other top union officers every four years. The elections for these officers are decided by referendum vote of the membership. In the 1977 election, which was hotly contested, two candidates ran for president: respondent Edward Sadlowski, Jr., the Director of USWA's largest District, and Lloyd McBride, another District Director.1 Both Sadlowski and McBride headed a slate of candidates for the other top union positions. 4 McBride was endorsed by the incumbent union leadership, and received substantial financial support from union officers and staff. Sadlowski, on the other hand, received much of his financial support from sources outside the union. During the campaign, the question whether candidates should accept contributions from persons who were not members of the union was vigorously debated. The McBride slate contended that outsider participation in USWA elections was dangerous for the union. App. 27, n. 2, 298. See also id., at 129, 398; see generally id., at 40-48. McBride ultimately defeated Sadlowski by a fairly wide margin—57% to 43%. The other candidates on the McBride slate won by similar margins. 5 After the elections, union members continued to debate the question whether outsider participation in union campaigns was desirable. This debate was finally resolved in 1978, when USWA held its biennial Convention. The Convention, which consists of approximately 5,000 delegates elected by members of USWA's local unions, is USWA's highest governing body. At the 1978 Convention, several local unions submitted resolutions recommending amendment of the USWA Constitution to include an "outsider rule" prohibiting campaign contributions by nonmembers. The union's International Executive Board also recommended a ban on nonmember contributions. Acting on the basis of these recommendations, the Convention's Constitution Committee proposed to the Convention that it adopt an outsider rule. After a debate on the floor of the Convention, the delegates, by a margin of roughly 10 to 1, voted to include such a rule in the Constitution. Id., at 35-36, 81-105. 6 The outsider rule, Article V, § 27, of the USWA Constitution (1978), provides in pertinent part: 7 "Sec. 27. No candidate (including a prospective candidate) for any position set forth in Article IV, Section 1, and supporter of a candidate may solicit or accept financial support, or any other direct or indirect support of any kind (except an individual's own volunteered personal time) from any non-member."2 8 Section 27 confers authority upon the International Executive Board to adopt regulations necessary to implement the provision. It also creates a Campaign Contribution Administrative Committee, consisting of three "distinguished, impartial" nonmembers to administer and enforce the provision. The Committee may order a candidate to cease and desist from conduct that breaches § 27, and may declare a candidate disqualified. Its decisions are final and binding. B 9 In October 1979, Sadlowski and several other individuals3 filed suit against USWA in the United States District Court for the District of Columbia. They claimed, inter alia, that the outsider rule violated the "right to sue" provision of Title I of the LMRDA, § 101(a)(4), 73 Stat. 522, 29 U.S.C. § 411(a)(4), because it would prohibit a candidate from accepting nonmember contributions to finance campaign-related litigation. Both sides moved for summary judgment. The District Court found that the rule violated § 101(a)(4). 507 F.Supp. 623, 625 (1981). The District Court further decided to invalidate the rule in toto, because the portion of the rule that "limits meaningful access to the courts . . . cannot be separated or isolated from the rule in its entirety." Ibid. 10 The United States Court of Appeals for the District of Columbia Circuit affirmed. 207 U.S.App.D.C. 189, 645 F.2d 1114 (1981). The court agreed that Article V, § 27, violated the right-to-sue provision. However, it chose not to decide whether this violation alone justified an injunction restraining enforcement of the entire rule. It accepted respondents' argument, first raised on appeal, that the outsider rule also violated the § 101(a)(2) "freedom of speech and assembly" provision, and that this violation justified the injunction. The Court of Appeals reasoned that the statutory goal of union democracy could be achieved only if "effective challenges can be made to the often-entrenched union leadership." 207 U.S.App.D.C., at 197, 645 F.2d, at 1122. But effective challenges are possible only if insurgent candidates can solicit contributions from outsiders. "Even without contribution limitations, challengers to the union leadership face substantial barriers, especially the electoral power of the union staff." Id., at 196, 645 F.2d, at 1121. The court rejected the union's argument that even if the rule interfered with rights protected by the statute, it was protected by the proviso to § 101(a)(2), which gives a union authority to adopt "reasonable" rules regarding the responsibilities of its members. Id., at 198, 645 F.2d, at 1123. 11 To buttress its analysis, the Court of Appeals relied heavily on its understanding of First Amendment jurisprudence. It stated that § 101(a)(2) places "essentially the same limits on labor unions with respect to outside campaign contributions that the First Amendment would if it applied to labor unions." Id., at 195, 645 F.2d, at 1120. Citing Buckley v. Valeo, 424 U.S. 1, 19, 96 S.Ct. 612, 634, 46 L.Ed.2d 659 (1976) (per curiam), the Court of Appeals suggested that contribution rules that prevent candidates for political office from amassing the resources necessary for effective advocacy are unconstitutional. By analogy, since the outsider rule would interfere with effective advocacy in union campaigns, it must violate § 101(a)(2). 207 U.S.App.D.C., at 197, 645 F.2d, at 1122. II 12 Section 101(a)(2) is contained in Title I of the LMRDA, the "Bill of Rights of Members of Labor Organizations." See 29 U.S.C. §§ 411-415. It provides: 13 "Freedom of Speech and Assembly.—Every member of any labor organization shall have the right to meet and assemble freely with other members; and to express any views, arguments, or opinions; and to express at meetings of the labor organization his views, upon candidates in an election of the labor organization or upon any business properly before the meeting, subject to the organization's established and reasonable rules pertaining to the conduct of meetings: Provided, That nothing herein shall be construed to impair the right of a labor organization to adopt and enforce reasonable rules as to the responsibility of every member toward the organization as an institution and to his refraining from conduct that would interfere with its performance of its legal or contractual obligations." 73 Stat. 522. 14 We must decide whether this statute is violated by a union rule that prohibits candidates for union office from accepting campaign contributions from individuals who are not members of the union. A. 15 At the outset, we address respondents' contention that this case can be resolved simply by reference to First Amendment law. Respondents claim that § 101(a)(2) confers upon union members rights equivalent to the rights established by the First Amendment. They further argue that in the context of a political election, a rule that placed substantial restrictions on a candidate's freedom to receive campaign contributions would violate the First Amendment. Thus, a rule that substantially restricts contributions in union campaigns must violate § 101(a)(2). We are not persuaded by this argument. In light of the legislative history, we do not believe that § 101(a)(2) should be read as incorporating the entire body of First Amendment law, so that the scope of protections afforded by the statute coincides with the protections afforded by the Constitution. 16 The legislation that ultimately evolved into Title I of the LMRDA was introduced on the floor of the Senate by Senator McClellan. The Senate Committee on Labor and Public Welfare had reported out a bill containing provisions that were the forerunners of Titles II through VI of the LMRDA. These provisions focused on specific aspects of union affairs: they established disclosure requirements and rules governing union trusteeships and elections. See S. 1555, 86th Cong., 1st Sess. (1959), U.S.Code Cong. & Admin.News 1959, p. 2318, 1 NLRB, Legislative History of the Labor-Management Reporting and Disclosure Act of 1959, pp. 338-396 (1959) (Leg.Hist.); see also Finnegan v. Leu, 456 U.S. 431, 435-436, 102 S.Ct. 1867, 1870-1871, 72 L.Ed.2d 239 (1982). Senator McClellan and other legislators feared that the bill did not go far enough because it did not provide general protection to union members who spoke out against the union leadership. Senator McClellan therefore proposed an amendment that he described as a "Bill of Rights" for union members. This amendment, which contained the forerunner of § 101(a)(2), as well as the forerunners of other Title I provisions, was designed to guarantee every union member equal voting rights, rights of free speech and assembly, and a right to sue. 105 Cong.Rec. 6469-6493 (1959), 2 Leg.Hist. 1096-1119. 17 Senator McClellan hoped that the amendment would "bring to the conduct of union affairs and to union members the reality of some of the freedoms from oppression that we enjoy as citizens by virtue of the Constitution of the United States." 105 Cong.Rec. 6472 (1959), 2 Leg.Hist. 1098. He further stated: "[T]he rights which I desire to have spelled out in the bill are not now defined in the bill. Such rights are basic. They ought to be basic to every person, and they are, under the Constitution of the United States." 105 Cong.Rec. 6478 (1959), 2 Leg.Hist. 1104-1105. Senator McClellan explained the freedom of assembly provision, in particular, as follows: 18 "That [provision] gives union members the right to assemble in groups, if they like, and to visit their neighbors and to discuss union affairs, and to say what they think, or perhaps discuss what should be done to straighten out union affairs, or perhaps discuss the promotion of a union movement, or perhaps a policy in which they believe. They would be able to do all of that without being punished for doing it, as is actually happening today." 105 Cong.Rec. 6477 (1959), 2 Leg.Hist. 1104. 19 Other Senators made similar statements. See 105 Cong.Rec. 6483 (1959) (Sen. Curtis); id., at 6488 (Sen. Goldwater); id., at 6489 (Sen. Mundt); id., at 6490 (Sen. Dirksen); id., at 6726 (Sen. Javits); 2 Leg.Hist. 1109, 1115, 1116, 1238. 20 The McClellan amendment was adopted by a vote of 47-46. 105 Cong.Rec. 6492 (1959), 2 Leg.Hist. 1119. Shortly thereafter, Senator Kuchel offered a substitute for the McClellan amendment. This substitute added the proviso that now appears in § 101(a)(2), which preserves the union's right to adopt reasonable rules governing the responsibilities of its members. It was designed to remove "the extremes raised by the [McClellan] amendment," 105 Cong.Rec. 6722 (1959), 2 Leg.Hist. 1234 (Sen. Cooper), and to assure that the amendment would not "unduly harass and obstruct legitimate unionism." 105 Cong.Rec. 6721 (1959), 2 Leg.Hist. 1233 (Sen. Church). The Kuchel amendment was approved by a vote of 77-14. See 105 Cong.Rec. 6717-6727 (1959), 2 Leg.Hist. 1229-1239. The legislation was then taken up in the House of Representatives. The House bill, which contained a "Bill of Rights" identical to that adopted by the Senate, was quickly approved. H.R. 8400, 86th Cong., 1st Sess. (1959), 1 Leg.Hist. 628-633. 21 This history reveals that Congress modeled Title I after the Bill of Rights, and that the legislators intended § 101(a)(2) to restate a principal First Amendment value—the right to speak one's mind without fear of reprisal. However, there is absolutely no indication that Congress intended the scope of § 101(a)(2) to be identical to the scope of the First Amendment. Rather, Congress' decision to include a proviso covering "reasonable" rules refutes that proposition. First Amendment freedoms may not be infringed absent a compelling governmental interest. Even then, any government regulation must be carefully tailored, so that rights are not needlessly impaired. Brown v. Hartlage, 456 U.S. 45, 53-54, 102 S.Ct. 1523, 1529, 71 L.Ed.2d 732 (1982). Union rules, by contrast, are valid under § 101(a)(2) so long as they are reasonable; they need not pass the stringent tests applied in the First Amendment context. B 22 To determine whether a union rule is valid under the statute, we first consider whether the rule interferes with an interest protected by the first part of § 101(a)(2). If it does, we then determine whether the rule is "reasonable" and thus sheltered by the proviso to § 101(a)(2). In conducting these inquiries, we find guidance in the policies that underlie the LMRDA in general and Title I in particular. First Amendment principles may be helpful, although they are not controlling. We must look to the objectives Congress sought to achieve, and avoid " 'placing great emphasis upon close construction of the words.' " Wirtz v. Glass Bottle Blowers, 389 U.S. 463, 468, and n. 6, 88 S.Ct. 643, 646, and n. 6, 19 L.Ed.2d 705 (1968) (quoting Cox, Internal Affairs of Labor Unions Under the Labor Reform Act of 1959, 58 Mich.L.Rev. 819, 852 (1960)); Hall v. Cole, 412 U.S. 1, 11 & n. 17, 93 S.Ct. 1943, 1949 & n. 17, 36 L.Ed.2d 702 (1973).4 The critical question is whether a rule that partially interferes with a protected interest is nevertheless reasonably related to the protection of the organization as an institution. 23 Applying this form of analysis here, we conclude that the outsider rule is valid. Although it may limit somewhat the ability of insurgent union members to wage an effective campaign, an interest deserving some protection under the statute, it is rationally related to the union's legitimate interest in reducing outsider interference with union affairs. 24 (1) 25 An examination of the policies underlying the LMRDA indicates that the outsider rule may have some impact on interests that Congress intended to protect under § 101(a)(2). Congress adopted the freedom of speech and assembly provision in order to promote union democracy. See supra, at 109-111; see also S.Rep. No. 187, 86th Cong., 1st Sess., 2 (1959), 1 Leg.Hist. 398; H.R.Rep.No. 741, 86th Cong., 1st Sess., 2 (1959), 1 Leg.Hist. 760. It recognized that democracy would be assured only if union members are free to discuss union policies and criticize the leadership without fear of reprisal. Congress also recognized that this freedom is particularly critical, and deserves vigorous protection, in the context of election campaigns. For it is in elections that members can wield their power, and directly express their approval or disapproval of the union leadership. See S.Rep.No. 187, supra, at 2-5, 7, 1 Leg.Hist. 398-401, 403; H.R. .No. 741, supra, at 1-7, 15-16, 1 Leg.Hist. 759-765, 773-774.5 26 The interest in fostering vigorous debate during election campaigns may be affected by the outsider rule. If candidates are not permitted to accept contributions from persons outside the union, their ability to criticize union policies and to mount effective challenges to union leadership may be weakened. Restrictions that limit access to funds may reduce the number of issues discussed, the attention that is devoted to each issue, and the size of the audience reached. Cf. Buckley v. Valeo, 424 U.S., at 65-66, 96 S.Ct., at 657 (per curiam ) (First Amendment freedom of expression and association may be "diluted if it does not include the right to pool money through contributions, for funds are often essential if 'advocacy' is to be truly or optimally 'effective' ").6 27 Although the outsider rule does affect rights protected by the statute, as a practical matter the impact may not be substantial. Respondents, as well as the Court of Appeals, suggest that incumbents have a large advantage because they can rely on their union staff during election campaigns. Challengers cannot counter this power simply by seeking funds from union members; the rank and file cannot provide sufficient support. Thus, they must be permitted to seek funds from outsiders. In fact, however, the rank and file probably can provide support. The USWA is a very large union whose members earn sufficient income to make campaign contributions. See App. 118-120. Requiring candidates to rely solely on contributions from members will not unduly limit their ability to raise campaign funds. Uncontradicted record evidence7 discloses that challengers have been able to defeat incumbents or administration-backed candidates, despite the absence of financial support from nonmembers. See id., at 25, 118-119.8 28 In addition, although there are undoubtedly advantages to incumbency, see Hall v. Cole, 412 U.S., at 13, 93 S.Ct., at 631, respondents and the Court of Appeals may overstate those advantages. Staff employees are forbidden by § 401(g) of the LMRDA, 29 U.S.C. § 481(g), and by internal USWA rules to campaign on union time or to use union funds, facilities, or equipment for campaign purposes. App. 110-117; see 29 CFR § 452.76 (1981). Staff officers have a contractual right to choose whether or not to participate in any USWA campaign without being subjected to discipline or reprisal for their decision. See App. 107-110, 115-117, 228, 384-385. Indeed, USWA elections have frequently involved challenges to incumbents by members of the staff. Many of these challenges have been successful. Id., at 108, 201-216. 29 The impact of the outsider rule on rights protected under § 101(a)(2) is limited in another important respect. The union has stated that the rule would not prohibit union members who are not involved in a campaign from using outside funds to address particular issues. That is, members could solicit funds from outsiders in order to focus the attention of the rank and file on a specific problem. The fact that union members remain free to seek funds for this purpose will serve as a counter to the power of entrenched leadership, and ensures that debate on issues that are important to the membership will never be stifled. 30 (2) 31 Although the outsider rule may implicate rights protected by § 101(a)(2), it serves a legitimate purpose that is clearly protected under the statute. The union adopted the rule because it wanted to ensure that nonmembers do not unduly influence union affairs. USWA feared that officers who received campaign contributions from nonmembers might be beholden to those individuals and might allow their decisions to be influenced by considerations other than the best interests of the union. The union wanted to ensure that the union leadership remained responsive to the membership. See id., at 210; see also id., at 61-62, 81-97, 275, 303, 304.9 An examination of the policies underlying the LMRDA reveals that this is a legitimate purpose that Congress meant to protect. 32 Evidence that Congress regarded the desire to minimize outsider influence as a legitimate purpose is provided by the history to Title I. On the Senate floor, Senator McClellan argued that a bill of rights for union members was necessary because some unions had been "invaded" or "infiltrated" by outsiders who had no interest in the members but rather had seized control for their own purposes. 105 Cong.Rec. 6469-6474 (1959), 2 Leg.Hist. 1097-1100. He stated that the strongest support for the bill of rights provisions "should come from traditional union leaders. It will protect them from the assaults of those who would capture their unions." 105 Cong.Rec. 6472 (1959), 2 Leg.Hist. 1098. And he stated: 33 "[Infiltration could be ended] by placing the ultimate power in the hands of the members, where it rightfully belongs, so that they may be ruled by their free consent, [and] may bring about a regeneration of union leadership. I believe the unions should be returned to those whom they were designed to serve; they should not be left to the hands of those who act as masters. The union must be returned to their members, to whom they rightfully belong." 105 Cong.Rec. 6472 (1959), 2 Leg.Hist. 1099. 34 It is true that Senator McClellan was particularly concerned about infiltration of unions by racketeers: he described situations in which "thugs and hoodlums" had taken over unions so that they could exploit the members for pecuniary gain. 105 Cong.Rec. 6471 (1959), 2 Leg.Hist. 1097. However, his statements also indicate a more general desire to ensure that union members, and not outsiders, control the affairs of their union. 35 Additional evidence that Congress regarded the union's desire to maintain control over its own affairs as legitimate is provided by the history of other sections of the LMRDA. In drafting Titles II through VI, Congress was guided by the general principle that unions should be left free to "operate their own affairs, as far as possible." S.Rep.No. 1684, 85th Cong., 2d Sess., 4-5 (1958). It believed that only essential standards should be imposed by legislation, and that in establishing those standards, great care should be taken not to undermine union self-government. Given certain minimum standards, "individual members are fully competent to regulate union affairs." Ibid. Thus, for example, in Title IV, which regulates the conduct of union elections, Congress simply set forth certain minimum standards. So long as unions conform with these standards, they are free "to run their own elections." Wirtz v. Glass Bottle Blowers, 389 U.S., at 471, 88 S.Ct., at 648. Congress' desire to permit unions to regulate their own affairs and to minimize governmental intervention suggests that it would have endorsed union efforts to reduce outsider influence. 36 Indeed, specific provisions contained in Title IV provide support for our conclusion that the outsider rule serves a legitimate and protected purpose. Section 401(g), 29 U.S.C. § 481(g), prohibits the use of employer as well as union funds in election campaigns. This ban reflects a desire to minimize the danger that employers will influence the outcome of union elections. A union rule that seeks to reduce the influence of outsiders other than employers is clearly consistent with that goal. See also § 403 of Title IV of the LMRDA, 29 U.S.C. § 483 (authorizing unions to establish their own election rules).10 37 Respondents argue that even if the desire to reduce outside influence is a legitimate purpose, the rule is not rationally related to that purpose. They contend, first, that the union could simply have established contribution ceilings, rather than placing an absolute ban on nonmember contributions. However, USWA feared not only that a few individual nonmembers would make large contributions, but also that outsiders would solicit many like-minded persons for small contributions which, when pooled, would have a substantial impact on the election. This fear appears to have been reasonable. In the 1977 election, Sadlowski received a significant percentage of his campaign funds from individuals who made contributions after receiving mail solicitations signed by prominent nonmembers. App. 128-129, 350-353. 38 Respondents also contend that even if the union was justified in limiting contributions by true outsiders, it need not have limited contributions by relatives and friends. Again, however, the USWA had a reasonable basis for its decision to impose a broad ban. An exception for family members and friends might have created a loophole that would have made the rule unenforceable: true outsiders could simply funnel their contributions through relatives and friends. See id., at 32. Cf. Buckley v. Valeo, 424 U.S., at 53, n. 59, 96 S.Ct., at 651, n. 59 (Congress could constitutionally subject family members to the same limitations as nonfamily members). 39 Finally, respondents contend that USWA could simply have required that candidates for union office reveal the sources of their funds. But a disclosure rule, by itself, would not have solved the problem. Candidates who received such funds might still be beholden to outsiders. A disclosure requirement ensures only that union members know about this possibility when they cast their votes. It does not eradicate the threat of outside influence.11 III 40 As an alternative basis for sustaining the result below, respondents ask this Court to hold that the outsider rule impermissibly encroaches upon a union member's right, guaranteed by § 101(a)(4) of the LMRDA, to institute legal proceedings, and that the appropriate remedy for this violation is an injunction striking down the rule in toto. However, unlike the District Court and the Court of Appeals, we do not believe that the union's rule violates the right-to-sue provision. 41 Section 101(a)(4) provides that a union may not "limit the right of any member thereof to institute an action in any court, or in a proceeding before any administrative agency." 29 U.S.C. § 411(a)(4). The outsider rule would clearly violate this provision if it prohibited union members from accepting financial or other support from nonmembers for the purpose of conducting campaign-related litigation. In our view, however, the outsider rule simply does not apply where a member uses funds from outsiders to finance litigation. 42 The language of the rule contains no reference to litigation. In addition, the debates leading up to the passage of the rule do not contain any indication that the union intended the rule to apply in this context. But what is most persuasive, the Campaign Contribution Administrative Committee12—which was given authority to make final and binding interpretations of the outsider rule—has issued an opinion concerning the impact of the outsider rule on the right to sue. In this opinion, it holds that "the limitations imposed by Section 27 do not apply to the financing of lawsuits by non-members for the purpose of asserting the legal rights of candidates or other union members in connection with elections."13 App. 455; see also id., at 456-458.14 43 The Court of Appeals expressed concern about a regulation contained in the USWA's Elections Manual which provides that although the outsider rule "does not prohibit the candidate's use of financial support or services from non-members to pay fees for legal or accounting services performed in . . . securing . . . legal rights of candidates," it does prohibit "[a]ctivities which are designed to extract political gain from legal proceedings." Id., at 495. According to the Court of Appeals, the reference to "activities" might include steps in the legal proceedings themselves, and might prohibit outside assistance to finance a lawsuit even if it was brought in good faith, if it was designed to extract political gain. 207 U.S.App.D.C., at 194, 645 F.2d, at 1119. USWA has explained, however, that this language is intended to cover only nonlitigation activities that in some way refer to litigation, such as mailing a flyer announcing a legal victory, or some information learned during discovery.15 See id., at 193-194, 645 F.2d, at 1118-1119.16 IV 44 We hold that USWA's rule prohibiting candidates for union office from accepting campaign contributions from nonmembers does not violate § 101(a)(2). Although it may interfere with rights Congress intended to protect, it is rationally related to a legitimate and protected purpose, and thus is sheltered by the proviso to § 101(a)(2). We reverse the decision below and remand for further proceedings consistent with this opinion. 45 It is so ordered. 46 Justice WHITE, with whom Chief Justice BURGER, Justice BRENNAN, and Justice BLACKMUN join, dissenting. 47 The question before us is what Congress intended when in 1959 it passed § 101(a)(2), the Bill of Rights provision of the LMRDA. That question is best answered by identifying the problem that Congress intended to solve by adopting the provision. The answer, in turn, is not at all difficult to discover. 48 After long and careful examination and hearings dealing with the labor union movement, Congress found that too often unions were run by entrenched, corrupt leaders who maintained themselves and discouraged challenge by any means available, including violence and threats.1 As Senator McClellan explained: "[T]he records of our committee's investigations show over and over again that a rank-and-file member dare not risk any opposition to a corrupt or autocratic leadership. If he does so, he may be beaten, his family threatened, his property destroyed or damaged, and he may be forced out of his job—all of these things can happen and have happened." 105 Cong.Rec. 6472 (1959), 2 NLRB, Legislative History of the Labor-Management Reporting and Disclosure Act of 1959, p. 1098 (1958) (Leg.Hist.). And again, "Members had better not offer any competition. They had better not seek election. They had better not aspire to the presidency or the secretaryship, or they will be expelled or disciplined." 105 Cong.Rec. 6478 (1959), 2 Leg.Hist. 1104. 49 This was the problem that Congress meant to solve. As Senator McClellan stated, its goal was to end "autocratic rule by placing the ultimate power in the hands of the members, where it rightfully belongs so that they may be ruled by their free consent, may bring about a regeneration of union leadership. I believe the unions should be returned to those whom they were designed to serve; they should not be left to the hands of those who act as masters." 105 Cong.Rec. 6472 (1959), 2 Leg.Hist. 1099. 50 What Congress then did was to guarantee the union member's right to run for election, § 401(e), and to guarantee him freedom of speech and assembly. § 101(a)(2). There is no question, and the Court concedes as much, that the Act created statutory protection for the union member's right effectively to run for union office. Without doubt, § 101(a)(2) was not only aimed at protecting the member who speaks his mind on union affairs, even if critical of the leadership, but was also "specifically designed to protect the union member's right to seek higher office within the union." Hall v. Cole, 412 U.S. 1, 14, 93 S.Ct. 1943, 1951, 36 L.Ed.2d 702 (1973). The LMRDA was a major effort by Congress "to insure union democracy." S.Rep.No. 187, 86th Cong., 1st Sess., 2 (1959). The chosen instrument for curbing the abuses of entrenched union leadership was "free and democratic union elections." Steelworkers v. Usery, 429 U.S. 305, 309, 97 S.Ct. 611, 614, 50 L.Ed.2d 502 (1977). The abuses of "entrenched union leadership" were to be curbed, among other means, by the "check of democratic elections." Wirtz v. Hotel Employees, 391 U.S. 492, 499, 88 S.Ct. 1743, 1748, 20 L.Ed. 763 (1968). These elections were to be modeled on the "political elections in this country." Wirtz v. Hotel Employees, supra, at 504, 88 S.Ct., at 1750; Steelworkers v. Usery, supra, 429 U.S., at 309, 97 S.Ct., at 614. 51 The member's right to run for office and to speak and assemble was to be subject to reasonable union rules, but the reasonableness of a particular rule must surely be judged with reference to the paradigmatic situation that Congress intended to address by guaranteeing free elections: a large union with entrenched, autocratic leadership bent on maintaining itself by fair means or foul. We do not by any means suggest that the USWA had or has the characteristics that led to the enactment of § 101(a)(2), but it is clear that the section should be construed with reference to those unions with the kind of leadership that caused the congressional response. Such a leadership is not only determined to discourage opposition; it also has at its disposal all of the advantages of incumbency for doing so, including the facilities of the union. Those leaders have normally appointed the union staff, the bureaucracy that makes the union run. The staff is dependent upon and totally loyal to the leadership. It amounts to a built-in campaign organization that can be relied upon to make substantial contributions and to solicit others for more. Such a management is in control of the union's communication system and has immediate access to membership lists and to the members themselves. Obviously, even if the incumbents eschew violence, threats, or intimidation, mounting an effective challenge would be a large and difficult endeavor. And if those in office are as unscrupulous as Congress often found them to be, the dimensions of the task facing the insurgent are exceedingly large. But Congress intended to help the members help solve these very difficulties by guaranteeing them the right to run for office and to have free and open elections in the American tradition. 52 It is incredible to me that the union rule at issue in this case can be found to be a reasonable restriction on the right of Edward Sadlowski, Jr., to speak, assemble, and run for union office in a free and democratic election. The scope and stringency of the rule cannot be doubted. It forbids any candidate for union office and his supporters to solicit or accept financial support from any nonmember. The candidate cannot accept contributions from members of his family, relatives, friends, or well-wishers unless they are members of the union. Retired members such as Edward Sadlowski, Sr., may not contribute; neither may members not in good standing. Even a fully secured loan from a non-member with a standard rate of interest is forbidden under the rule. The rule goes even further. It forbids the acceptance of "any other direct or indirect support of any kind from any non-member," except an individual's volunteered personal time.2 The regulations issued under the rule clearly show that the union intends to prohibit, as far as it is within its power to do so, all nonmember contributions on behalf of a member running for union office. These regulations specify: 53 "[W]hen prohibited support is contributed, there will be a presumption that it was accepted by the candidate or his or her supporters, unless they have taken affirmative steps in good faith to dissuade the non-member from providing such support and have taken action to correct the effects of the prohibited support." App. 494.3 54 A candidate unable to rebut this presumption may be disqualified, fined, suspended, or expelled. This is a Draconian rule. How could any candidate "correct the effects of the prohibited support"? The rule thus goes far beyond the limitations on contributions approved in Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), and severely limits expenditures as well. The candidate may actually be denied his statutory right to run for office because nonmembers have exercised their own First Amendment rights. 55 The impact of the rule with respect to Edward Sadlowski, Sr., illustrates the rigor of the rule. It prohibits him from contributing to the campaign of Edward Sadlowski, Jr., even though the elder Sadlowski is the father of the candidate, was a charter member of the USWA, remained a member for 32 years prior to his retirement, and receives a USWA pension, the terms of which are negotiated by USWA's officers. 56 Restrictions such as this are a far cry from the free and open elections that Congress anticipated and are wholly inconsistent with the way elections have been run in this country. The Court has long recognized the close relationship between the ability to solicit funds and the ability to express views. "[W]ithout solicitation, the flow of . . . information and advocacy would likely cease." Village of Schaumburg v. Citizens for Better Environment, 444 U.S. 620, 632, 100 S.Ct. 826, 834, 63 L.Ed.2d 73 (1980). See also Schneider v. State, 308 U.S. 147, 60 S.Ct. 146, 84 L.Ed. 155 (1939); Cantwell v. Connecticut, 310 U.S. 296, 60 S.Ct. 900, 84 L.Ed. 1213 (1940); Virginia Pharmacy Board v. Virginia Citizens Consumer Council, 425 U.S. 748, 761, 96 S.Ct. 1817, 1825, 48 L.Ed.2d 346 (1976); Bates v. State Bar of Arizona, 433 U.S. 350, 363, 97 S.Ct. 2691, 2698, 53 L.Ed.2d 810 (1977). 57 In Thomas v. Collins, 323 U.S. 516, 65 S.Ct. 315, 89 L.Ed. 430 (1945), the Court held that the First Amendment barred enforcement of a state statute requiring a permit before soliciting membership in any labor organization. Solicitation and speech were deemed to be so intertwined that a prior permit could not be required. The Court conceded that the "collection of funds" might be subject to reasonable regulation, but concluded that such regulation "must be done and the restriction applied, in such a manner as not to intrude upon the rights of free speech and free assembly." Id., at 540-541, 65 S.Ct., at 327. 58 Specifically with regard to elections and campaign financing, the Court observed in Buckley v. Valeo, supra, 424 U.S., at 19, 96 S.Ct., at 634: 59 "A restriction on the amount of money a person or group can spend on political communication during a campaign necessarily reduces the quantity of expression by restricting the number of issues discussed, the depth of their exploration, and the size of the audience reached. This is because virtually every means of communicating ideas in today's mass society requires the expenditure of money. The distribution of the humblest handbill or leaflet entails printing, paper, and circulation costs. Speeches and rallies generally necessitate hiring a hall and publicizing the event." (Footnote omitted.) 60 Thus, as the Court of Appeals recognized in this case " 'contribution restrictions could have a severe impact on political dialogue if the limitations prevented candidates and political committees from amassing the resources necessary for effective advocacy.' " 207 U.S.App.D.C. 189, 197, 645 F.2d 1114, 1122 (1981), quoting 424 U.S., at 21, 96 S.Ct., at 636. 61 It goes without saying that running for office in a union with 1.3 million members spread throughout the United States and Canada requires a substantial war chest if the campaign is to be effective and to have any reasonable chance of succeeding. Attempting to unseat the incumbents of union office is a substantial undertaking. As we noted in Steelworkers v. Usery, 429 U.S., at 311, 97 S.Ct., at 615, there is no permanent opposition party within the union. There is only a one-party system consisting of the union's incumbent officers and hired staff all controlled from the top down. "[T]he full-time officers collectively, under the direction of the top officer, constitute the sole political machine for the preservation of their offices and power." J. Edelstein & M. Warner, Comparative Union Democracy 39 (1979). The union involved in this case has some 30 elected positions, its president appoints more than 1,500 office and field staff, and salaries and expenses for union personnel in 1978 totalled over $37 million. App. 141. 62 Thus, in the best of circumstances, the role of the challenger is very difficult. And if one keeps in mind that Congress intended to give the challenger a fair chance even in a union controlled by unscrupulous leaders with an iron grip on the staff and a willingness to employ means both within and without the law, it is wholly unrealistic to confine the challenger to financial support garnered within the union. Surely, Congress never intended that a union should be permitted to impose such a limitation. As Clyde Summers, a recognized authority in this field, stated in this case on behalf of Sadlowski: 63 "Opposition candidates customarily finance their campaigns in the first instance out of their own pockets and out of loans or gifts from friends. They get contributions from sympathetic union members, but at the beginning they may have few open supporters and they do not have a large organization to solicit contributions. They have to do enough publicizing and campaigning to make themselves appear as a viable candidate before they begin to get support from any substantial number of members. Even then, the individual contributions of members is inevitably small. Seldom is it enough to mount a really substantial campaign, and it is almost never enough to match the resources of the incumbents." Id., at 156. 64 "In my opinion, the practical effect of prohibiting all contributions to union election campaigns except those made by union members would be to gravely damage if not destroy the possibility of democratic elections in unions, particularly in large local unions and in international unions. . . . 65 "If opposition groups are barred from getting any help from the outside, they can, in most situations, have no hopes of mounting an effective campaign." Id., at 160.4 66 In addressing itself to union elections, Congress forbade union and employer contributions, but went no further in restricting contributions or expenditures to or on behalf of union candidates for office. The majority emphasizes that Congress was concerned about the control of unions by outsiders and asserts that the challenged rule serves the congressional purpose. It is true, as Senator McClellan explained, that "impositions and abuses . . . have been perpetrated upon the working people of many of our States by the thugs who have muscled into positions of power in labor unions and who masquerade as labor leaders and as friends of working people. . . ." 105 Cong.Rec. 6470 (1959), 2 Leg.Hist. 1097. But the remedy which he proposed and which was adopted was to end "autocratic rule by placing the ultimate power in the hands of the members," 105 Cong.Rec. 6472 (1959), 2 Leg.Hist. 1099, and by giving them sufficient statutory protection to participate in a fair election to unseat an entrenched leadership. 67 Yet the majority somehow finds the absolute, unbending, no-contribution rule to be a reasonable regulation of a member's right to seek office and of the free and open elections that Congress anticipated. This, in spite of the availability of other means to satisfy the union's legitimate concerns about outsiders controlling their affairs through those whose campaigns they have financed. A requirement of disclosure of all contributions, together with a ceiling on contributions, would avoid outside corruption without trampling on the rights of members to raise reasonable sums for election campaigns. Such rules would honor both purposes of the legislation: protecting against outside influence and empowering members to express their views and to challenge established leadership. As I see it, the rule at issue contradicts the values the statute was designed to protect and thwarts its purpose. 68 I respectfully dissent. 1 The USWA is divided into 25 districts, which are headed by District Directors. District Directors are elected every four years by referendum vote of the members within each district. App. 7. 2 The offices set forth in Article IV, Section 1 of the USWA Constitution are International President, International Secretary, International Treasurer, International Vice President (Administration), International Vice President (Human Affairs), District Director for the 25 Districts, and a National Director for Canada. 3 Other plaintiffs included Joseph Samargia, a USWA member and potential candidate for union office; Edward Sadlowski, Sr., a retired union member who campaigned for his son in 1977; Leonard S. Rubenstein, a nonmember who made contributions to Sadlowski, Jr., during the 1977 campaign; and James Miller, a nonmember who donated legal services during the 1977 campaign. Samargia alleged that he might run for union office in the 1981 elections. Sadlowski, Sr., Rubenstein, and Miller alleged that they might wish to contribute services or funds in future USWA elections. App. 5-6, 16. Each of these individuals is also a respondent here. 4 Neither the language contained in the first part of § 101(a)(2), which describes the "right to meet and assemble freely," nor the language contained in the proviso, which states that unions may adopt "reasonable rules as to the responsibility of every member toward the organization as an institution," should be read narrowly. As we have already indicated, it seems clear that Congress intended the first part of § 101(a)(2) to be given a flexible interpretation. See supra, at 109-111; see also infra, at 112-113. And Congress adopted the proviso in order to ensure that the scope of the statute was limited by a general rule of reason. It indicated that the courts are to play a role in the determination of reasonableness. See 105 Cong.Rec. 6719 (1959) (Sen. Kuchel), 2 Leg.Hist. 1231; 105 Cong.Rec. 6726 (1959) (Sen. Javits), 2 Leg.Hist. 1238. See also supra, at 110. See generally 105 Cong.Rec. 6717-6727 (1959), 2 Leg.Hist. 1229-1239. 5 See also Hall v. Cole, 412 U.S. 1, 14, 93 S.Ct. 1943, 1951, 36 L.Ed.2d 702 (1973) ("Title I of the LMRDA was specifically designed to protect the union member's right to seek higher office within the union"). Cf. Wirtz v. Glass Bottle Blowers, 389 U.S. 463, 470, 88 S.Ct. 643, 648, 19 L.Ed.2d 705 (1968) (Title IV designed to ensure free and democratic elections). 6 In several First Amendment cases, we have protected contribution and solicitation of the financial support necessary to further effective advocacy. See, e.g., Citizens Against Rent Control v. Berkeley, 454 U.S. 290, 102 S.Ct. 434, 70 L.Ed.2d 492 (1981); Village of Schaumburg v. Citizens for a Better Environment, 444 U.S. 620, 100 S.Ct. 826, 63 L.Ed.2d 73 (1980); First National Bank of Boston v. Bellotti, 435 U.S. 765, 98 S.Ct. 1407, 55 L.Ed.2d 707 (1978). These cases are not directly analogous, however. Contribution limitations potentially infringe the First Amendment rights of contributors as well as candidates. Buckley v. Valeo, 424 U.S., at 24-25, 96 S.Ct., at 637 (per curiam ). Here, the nonmember contributors have no right of expression protected by the statute. 7 This case is here on cross-motions for summary judgment. We reach a conclusion opposite to that reached by the Court of Appeals—that the outsider rule is valid. In making this decision, we have assumed that all of the evidence submitted by respondents is true. In addition, we have relied on evidence submitted by the union only when it is uncontradicted. Here, to support their claim that incumbents have a large advantage in union elections, respondents have submitted numerous affidavits. We do not intend to deny the existence of this advantage. For the purposes of our decision in this case, we think it sufficient to observe that there is uncontradicted evidence demonstrating that effective campaigns have been mounted by nonincumbents—and that the interference with interests protected by § 101(a)(2) is only partial. 8 USWA has submitted evidence suggesting that the adoption of the outsider rule did not have an adverse effect on the 1981 election campaigns. A nonincumbent candidate for District Director in a relatively small district has testified that he had raised in excess of $30,000 from rank-and-file members as early as 15 months before the election. App. 121-123, 217-218. Respondents have not submitted any opposing evidence. 9 Respondents allege that the rule was forced upon the union members by high union officers, who wanted to ensure that they were insulated from effective challenges in future elections. However, the record does not support respondents' claims. The outsider rule was adopted through democratic processes, and was favored by an overwhelming majority of the delegates to the 1978 Convention. See supra, at 105. These delegates had been elected by the rank and file. See App. 301-302. 10 Section 403 provides: "No labor organization shall be required by law to conduct elections of officers with greater frequency or in a different form or manner than is required by its own constitution or bylaws, except as otherwise provided by [Title IV]." 73 Stat. 534. The union argues that the outsider rule can be justified solely on the basis of this provision, since the rule is otherwise consistent with Title IV. We are not persuaded by this argument. Section 403 must be interpreted in light of the provisions of Title I, which were adopted precisely because Congress feared that Titles II through VI did not provide sufficient protection to union members. Thus, even if the rule satisfies § 403, it must also satisfy Title I. 11 Respondents also contend that the outsider rule is underinclusive, because it does not apply to local union elections. As USWA explains in an unrebutted affidavit, however, an outsider rule for local union elections was considered and rejected because outsiders generally have little interest in influencing local campaigns, and enforcing an outsider rule in such elections would be an administrative burden. App. 30-32. 12 The three Committee members included former Secretary of Labor W. Willard Wirtz; David Lewis, professor at Carleton University; and Eric Springer, former Director of Compliance of the United States Equal Employment Opportunity Commission and Chairman of the Commission on Human Relations in Pittsburgh, Pa. Id., at 37-38. 13 The opinion further stated that it was "confined to services which are in fact legal services customarily performed by lawyers. The Committee recognizes the possibility that any ruling which it makes in general terms and in response to a broad inquiry may be misconstrued or distorted in an attempt to rationalize political activities as 'legal services.' It will deal with these questions whenever they arise on a case-by-case basis." Id., at 458. 14 The Committee left open the question whether the outsider rule would apply to a lawsuit that is not a bona fide attempt to secure an adjudication of legal rights, but, rather, is motivated solely by a desire to promote a candidate's political campaign. Ibid. The USWA has urged the Committee not to impose such a ban unless it is clear that the rule would not deter bona fide lawsuits. Id., at 245-246. It is arguable that such a rule might violate § 101(a)(4) if it had the unmistakable effect of deterring bona fide lawsuits by individuals who feared that the Committee might misjudge their motives and impose sanctions. However, the speculative possibility that the Committee will in the future apply the rule in a manner that deters bona fide lawsuits does not justify striking down the rule in toto at this time. 15 The Court of Appeals refused to accept this construction. 207 U.S.App.D.C., at 194, 645 F.2d, at 1119. However, it is consistent with the language of the regulation and is also supported by the Committee's opinion. See n. 14, supra, and accompanying text. 16 Respondents also argue that the decision below can be affirmed on the ground that the outsider rule violates the First Amendment because it interferes with members' and nonmembers' constitutional rights of free speech and free association. However, the union's decision to adopt an outsider rule does not involve state action. See Steelworkers v. Weber, 443 U.S. 193, 200, 99 S.Ct. 2721, 2726, 61 L.Ed.2d 480 (1979). 1 The Court of Appeals in this case summarized these findings: "Prior to the enactment of the LMRDA in 1959 the Select Senate Committee ferreted out widespread corruption, dictatorship and racketeering in a number of large international unions. The Committee found that the President of the Bakery and Confectionary [sic] Workers' International Union of America had 'railroaded through changes in the union constitution which destroyed any vestigial pretenses of union democracy.' Select Committee Report [S.Rep.No. 1417, 85th Cong., 2d Sess.] 129 [1958]. It reported that Dave Beck, General President of the International Brotherhood of Teamsters 'shamefully enriched himself at [the] expense [of the union members] and that in the final instance he capitulated to the forces within the union who promoted the interests of racketeers and hoodlums.' Id. at 84. The Committee likewise found Teamster officials joining with others to take over illegal gambling operations with an 'underworld combine,' id. at 38-39, and the top officers of the United Textile Workers of America avariciously misappropriating union funds, id. at 159. 'Democracy [was] virtually nonexistent' in the International Union of Operating Engineers because the union was ruthlessly dominated through 'violence, intimidation and other dictatorial practices.' Id. at 437. Practices in the Teamsters 'advanced the cause of union dictatorship.' Id. at 444. The Committee cited other similar instances of widespread abuses in its 462-page Report." 207 U.S.App.D.C. 189, 199, 645 F.2d 1114, 1124 (1981) (footnote omitted). 2 The regulations specify: " 'Financial Support' means a direct or indirect contribution where the purpose, object or foreseeable effect of the support is to influence the election of a candidate. Financial support includes, but is not limited to: "1. Contributions of money, securities, or any material thing of value; "2. Payments to or subscription for fund raising events of any kind (e.g. raffles, dinners, beer or cocktail parties and so forth); "3. Discounts in the price or cost of goods or services, except to the extent that commercially established discounts are generally available to the customers of the supplier; "4. Extensions of credit, loans, and other similar forms of finance, except when obtained in the regular course of business of a commercial lending institution and on such terms and conditions as are regularly required by such institutions; and "5. The payment for the personal services of another person, or for the use of building or office space, equipment or supplies, or advertisements through the media." App. 492. The regulations also explain that "[e]xamples of indirect support from nonmembers would include the contribution of cash to a member who in turn makes a contribution to a candidate; the donation of travel expenses, printing services, office supplies, office space, or of clerical, secretarial, or professional services used by a non-member in conjunction with his or her own volunteered service to a candidate; the distribution of election materials with the aid of a volunteer's paid staff; and the procuring of discounts." Id., at 494. 3 The regulation continues: "In such cases where prohibited support has been contributed, it is the candidate's obligation to contact immediately the non-member contributor, reject the prohibited support, return the contribution, insist that such support be discontinued, and take whatever action on his own or her own, or as directed by the Committee, may be necessary to eliminate any impact on the election. Full reports must be made to the Committee promptly." Id., at 495. 4 The following is a summary of other relevant views presented by Mr. Summers, id., at 152-160: Incumbents in unions elections have four crucial advantages: "First, and most important, they have control of the paid staff of representatives or business agents who provide the back-bone of the incumbents' political organization. The paid staff owe their jobs to the officers, take orders from the officers, and can be dismissed by the officers. . . . Second, the incumbent officers have control of the union newspaper. There is no such thing as a free and independent press within the union. . . . Third, the incumbents have ready access to members." They have immediate access to names, addresses and telephone numbers of union members. "The law requires equal access to membership lists but there is no practical equality when the incumbent administration includes the secretary treasurer of the union. Fourth, the officers have access to legal services, at the union's expenses." "These advantages are critical when one considers the financing of union election campaigns. . . . [T]he incumbent officers have a paid built-in campaign organization in the paid staff representatives. . . . In short, the incumbents can run a campaign with little or no money," while the opposition must have substantial funds even to get started. For incumbents, the largest single source is the paid staff, and it is quite unrealistic to expect opposition candidates to obtain substantial support from staff representatives who are contributing to those to whom they owe their jobs. Incumbents also raise funds from union members, and in doing so they have a marked advantage over insurgent candidates. Incumbents also raise funds through testimonial dinners given in their honor. Opposition candidates cannot successfully match this effort. "Union candidates' acceptance of money and other help from sources outside the union is a common and accepted practice. . . . Up until the last two years, no one, and I would emphasize no one, seriously suggested that there was anything inappropriate about union candidates soliciting financial support from non-members. . . . Union constitutions placed no such restrictions on such contributions. . . . Only within the last two years has the Steelworkers placed such a restriction in its constitution and this seems to be part of an effort of the administration to void any effective challenge by an opposition candidate in the future."

78

457 U.S. 164 102 S.Ct. 2335 72 L.Ed.2d 755 State of CALIFORNIAv.State of TEXAS et al. No. 88, Orig. June 14, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1131, 103 S.Ct. 14. PER CURIAM. 1 In this motion, California seeks leave to file a complaint against Texas under this Court's original jurisdiction. The proposed complaint asks us to decide whether Howard Hughes was domiciled in California or Texas at the time of his death. The decision about domicile could determine which State is entitled to levy death taxes on the estate. 2 This motion renews the one which California made in November 1977. At that time, we denied leave to file. Cali- fornia v. Texas, 437 U.S. 601, 98 S.Ct. 3107, 57 L.Ed.2d 464 (1978). Following the suggestion of four Justices who concurred in California v. Texas, the estate then sought a determination of Hughes' domicile by filing an interpleader action under 28 U.S.C. § 1335 in Federal District Court. This motion for leave to file a complaint accompanied the petition for certiorari in Cory v. White, 457 U.S. 85, 102 S.Ct. 2325, 72 L.Ed.2d 694, in which California taxing officials requested review of the decision of the Fifth Circuit holding that the Federal Interpleader Act provided a jurisdictional basis for resolving the dispute. 3 We granted certiorari in Cory v. White, 452 U.S. 904, 101 S.Ct. 3028, 69 L.Ed.2d 404 (1981), and today have held that the Federal Interpleader Act, 28 U.S.C. § 1335, does not give a federal district court jurisdiction to resolve inconsistent death tax claims by the officials of two States. See 457 U.S., at 91, 102 S.Ct., at 2329. We reached that decision because the suit is barred by the Eleventh Amendment under Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed.d 268 (1937). We now also conclude that California's motion for leave to file should be granted. 4 First, California's bill of complaint states a "controversy" between California and Texas within the exclusive jurisdiction of this Court under 28 U.S.C. § 1251(a). It is undisputed that each State's authority to impose a death tax on the intangibles owned by a decedent depends on the decedent's having been a domiciliary of that State. Also, it is the law of each State that an individual has but one domicile. Thus only one State is entitled to impose death taxes; the outcome of this action would determine which State is privileged to tax. The other would be barred from doing so. It is apparent, therefore, that California and Texas are asserting inconsistent claims and are undeniably adversaries in this action. 5 Moreover, in its Memorandum in Support of Motion to File Bill of Complaint 6, California asserts: 6 "The effective rate of tax in California on all amounts in excess of $400,000 is 24% (see CAL.REV. & TAX CODE § 13406(g)); the effective rate of tax in Texas (including the so-called 'pick-up tax') on amounts exceeding $1,000,000 is approximately 16% (see TEX. TAX CODE ANN. ARTS. 14.05, 14.12); and the federal estate tax on amounts in excess of $10,000,000 is 77%, less a credit of 16% for state death taxes (see 26 U.S.C. §§ 2001, 2011). The combined marginal rate of tax is therefore 101%." (Footnote omitted). 7 California adds that interest on the unpaid taxes will further deplete the estate. Although these allegations have not been proved, they are sufficient under Texas v. Florida, 306 U.S. 398, 59 S.Ct. 563, 83 L.Ed. 817 (1939), to characterize this case as a "controversy" between two States within the meaning of 28 U.S.C. § 1251(a).1 8 In Texas v. Florida, supra, this Court, raising the issue sua sponte, held that it had original jurisdiction over a suit "brought to determine the true domicile of decedent as the basis of rival claims of four states for death taxes upon his estate." 306 U.S., at 401, 58 S.Ct., at 565. None of the States had reduced its claims to judgments, but all conceded that the estate was insufficient to satisfy the total amount of taxes claimed. The Court compared the suit to a bill in the nature of interpleader, which permits a plaintiff threatened with rival claimants to the same debt or legal duty to bring an interpleader action before the institution of the independent suits. On the basis of this analogy, the Court concluded: 9 "When, by appropriate procedure, a court possessing equity powers is . . . asked to prevent the loss which might otherwise result from the independent prosecution of rival but mutually exclusive claims, a justiciable issue is presented for adjudication which because it is a recognized subject of the equity procedure which we have inherited from England, is a 'case' or 'controversy,' within the meaning of the Constitutional provision; and when the case is one prosecuted between states, which are the rival claimants, and the risk of loss is shown to be real and substantial, the case is within the original jurisdiction of this Court conferred by the Judiciary Article." Id., at 407-408, 58 S.Ct., at 568. 10 As Justice Stewart wrote when California first petitioned this Court to resolve its dispute with Texas over Hughes' estate: "The facts alleged in the complaint now before us are indistinguishable in all material respects from those on which jurisdiction was based in Texas v. Florida." California v. Texas, 437 U.S., at 606, 98 S.Ct., at 3110 (concurring opinion). We agree.2 Thus, this dispute is a controversy between two States within our original jurisdiction under 28 U.S.C. § 1251(a). 11 Second, it is appropriate to exercise our jurisdiction in this case. A determination that this Court has original jurisdiction over a case, of course, does not require us to exercise that jurisdiction. We have imposed prudential and equitable limitations upon the exercise of our original jurisdiction. As we explained in Illinois v. City of Milwaukee, 406 U.S. 91, 93-94, 92 S.Ct. 1385, 1387-1388, 31 L.Ed.2d 712 (1972): 12 "We construe 28 U.S.C. § 1251(a)(1), as we do Art. III, § 2, cl.2, to honor our original jurisdiction but to make it obligatory only in appropriate cases. And the question of what is appropriate concerns, of course, the seriousness and dignity of the claim; yet beyond that it necessarily involves the availability of another forum where there is jurisdiction over the named parties, where the issues tendered may be litigated, and where appropriate relief may be had. We incline to a sparing use of our original jurisdiction so that our increasing duties with the appellate docket will not suffer." 13 At the time we decided California v. Texas, it seemed to several Members of the Court that statutory interpleader might obviate the need to exercise original jurisdiction. Justice BRENNAN, for example, explained: 14 "If we have jurisdiction at all, that jurisdiction does not attach until it can be shown that the two States may possibly be able to obtain conflicting adjudications of domicile. That showing has not been made at this time in this case, since it may well be possible for the Hughes estate to obtain a judgment under the Federal Interpleader Statute, 28 U.S.C. § 1335, from a United States district court, which would be binding on both California and Texas. In this event, the precondition for our original jurisdiction would be lacking. Accordingly, I would deny California's motion, at least until such time as it is shown that such a statutory interpleader action cannot or will not be brought." 437 U.S., at 601-602, 98 S.Ct., at 3107-3108. 15 Our decision in Cory v. White has now shown that such a statutory interpleader action cannot be brought. Thus, the precondition for the exercise of original jurisdiction has been met. 16 There were several other uncertainties that affected the case when we denied California's earlier motion. At that time, Texas urged that the controversy was not ripe because of the pending claim of the Howard Hughes Medical Institute that a "lost will" left the entire estate to it and the contention that the so-called "Mormon Will" was valid. A jury has since rejected the "Mormon Will," the Nevada Supreme Court and the Texas Probate Court the "lost will." Another changed circumstance is the expiration of a conditional settlement agreement between California and the estate. Texas had argued because of this allegedly collusive agreement, the case was not a justiciable case or controversy. 17 We conclude that our original jurisdiction is properly invoked under Texas v. Florida, and we grant California leave to file its bill of complaint. The defendants shall have 60 days to answer. 18 It is so ordered. 19 Justice POWELL, with whom Justice MARSHALL, Justice REHNQUIST, and Justice STEVENS join, dissenting. 20 In Cory v. White, 457 U.S. 85, 89, 102 S.Ct. 2325, 2328, 72 L.Ed.2d 694, the Court today reaffirms the holding of Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed. 268 (1937), that "inconsistent determinations by the courts of two states as to the domicile of a taxpayer [do] not raise a substantial federal constitutional question." Under Worcester County there is no constitutional bar to both Texas and California taxing the Hughes estate on the ground that he was a domiciliary. 21 Having reaffirmed the authority of Worcester County, the Court concludes that "California and Texas are asserting inconsistent claims and are undeniably adversaries in [the interpleader action]." Ante, at 165. But its own premises will not support this conclusion. If both States legally can tax the Hughes estate, a controversy between them would arise only if both were to obtain money judgments against the estate and, further, if the estate then were to prove insufficient to satisfy both claims. Yet it is no more clear today than it was in 1978, when we unanimously decided California v. Texas, 437 U.S. 601, 98 S.Ct. 3107, 57 L.Ed.2d 464 (1978), that this situation ever will occur. Thus, under the Court's own assumptions, there is no ripe controversy between the States, and no basis for our consideration of the original complaint in No. 88, Original. 22 As if discomfited by the logic of its position, the Court argues that the jurisdictional allegations here at least are "no more speculative," ante, at 166, n. 1, then those in Texas v. Florida, 306 U.S. 398, 59 S.Ct. 563, 83 L.Ed. 817 (1939). Yet as Justice Stewart argued persuasively in our 1978 decision in California v. Texas, supra, it is inescapable that Texas v. Florida was wrongly decided. See 437 U.S., at 606, 611-612, 98 S.Ct., at 3110, 3113-3114 (Stewart, J., concurring). The mere possibility of inconsistent state determinations of domicile, resulting in a still more remote possibility of the estate's being insufficient to satisfy the competing claims, simply does not give rise to a case or controversy in the constitutional sense. "The necessity that the plaintiff who seeks to invoke judicial power stand to profit in some personal interest remains an Art. III requirement. A federal court cannot ignore this requirement without overstepping its assigned role in our system of adjudicating only actual cases and controversies." Simon v. Eastern Kentucky Welfare Rights Org., 426 U.S. 26, 39, 96 S.Ct. 1917, 1924, 48 L.Ed.2d 450 (1976). See Valley Forge Christian College v. Americans United for Separation of Church and State, 454 U.S. 464, 472, 102 S.Ct. 752, 758, 70 L.Ed.2d 700 (1982); Warth v. Seldin, 422 U.S. 490, 508, 95 S.Ct. 2197, 2210, 45 L.Ed.2d 343 (1975). 23 Nor is the Court entitled to base its finding of original jurisdiction on an "analogy" between the original action and "a bill in the nature of interpleader." Ante, at 167. Under the Interpleader Act, the stakeholder is the "plaintiff." 28 U.S.C. § 1335. Having been notified of claims by two or more "claimants," the stakeholder normally would have standing to litigate the validity of each of the individual claims. The presence of these justiciable controversies between stakeholder and claimants satisfies the "case or controversy" requirement of Art. III. Interpleader jurisdiction merely provides for convenient resolution in a single forum. Interpleader jurisdiction thus is irrelevant to the question whether there is an independently justiciable controversy "between" States. 1 Texas asserts that California has not demonstrated the jurisdictional prerequisite of showing a "threatened injury" of "serious magnitude and imminent." Brief in Opposition to Motion for Leave to File 6, quoting Alabama v. Arizona, 291 U.S. 286, 292, 54 S.Ct. 399, 401, 78 L.Ed. 798 (1934). Texas explains that the true value of the estate is subject to dispute and litigation and that the estate can fully satisfy all potential death tax claims against it even under California's own valuation. The Court in Texas v. Florida, however, required only that "[t]he risk that decedent's estate might constitutionally be subjected to conflicting tax assessments in excess of its total value and that the right of complainant or some other state to collect the tax might thus be defeated was a real one." 306 U.S., at 410, 59 S.Ct., at 569. The claims before us here are no more speculative than the ones there. As that case recognized, to bring an interpleader suit, "[a] plaintiff need not await actual institution of independent suits; it is enough if he shows that conflicting claims are asserted and that the consequent risk of loss is substantial." Id., at 406, 58 S.Ct., at 567. Thus, California's allegations are sufficient to present a controversy within the meaning of 28 U.S.C. § 1251(a). Despite the suggestion that we do so, we decline to overrule Texas v. Florida. 2 As in Texas v. Florida, the idiosyncratic pattern of the decedent's life provides a basis for more than one State's claims. Hughes spent much of his time in California and many of his business activities were based there. He was, however, born in Texas and long continued to use Texas as his mailing address and sometimes stated that Texas was his domicile. Indeed, a jury in Texas probate proceedings has already found Hughes to have been a domiciliary of Texas at the time of his death. The administrator of Hughes' estate timely perfected an appeal of that judgment. Brief for Respondent Lummis in Cory v. White, O.T.1981, No. 80-1556, p. 5. The Texas Court of Civil Appeals stayed the appeal of the Texas domicile judgment pending the outcome of the federal interpleader action. Id., at 7.

89

457 U.S. 85 102 S.Ct. 2325 72 L.Ed.2d 694 Kenneth CORY, Controller of the State of California, et al., Petitioners,v.Mark WHITE, Attorney General of the State of Texas, et al. No. 80-1556. Argued Jan. 18, 1982. Decided June 14, 1982. Syllabus Both Texas and California assert the right to levy state death taxes on the estate of Howard Hughes, the taxing officials of each State claiming that Hughes was domiciled in their State at the time of his death. The administrator of the estate filed an action in Federal District Court under the Federal Interpleader Act, alleging that the respective state officials were seeking to tax the estate on the basis of inconsistent claims. The District Court dismissed the action for lack of subject-matter jurisdiction because of the failure to satisfy the Act's requirement that there be diversity of citizenship between at least two adverse parties. The Court of Appeals reversed, holding that the requisite diversity was present between the administrator and the County Treasurer of Los Angeles County. The court rejected the State's claim that although the suit was nominally against state officials, it was in effect a suit against two sovereign States barred by the Eleventh Amendment. Held : The Eleventh Amendment bars the statutory interpleader action. Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed. 268. Contrary to the Court of Appeals' view, Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662, did not overrule Worcester County Trust Co. Pp. 89-91. 629 F.2d 397 (5th Cir. 1980), reversed. Jerome B. Falk, Jr., San Francisco, Cal., for petitioners. O. Clayton Lilienstern, Houston, Tex., for respondents Lummis, et al. Rick Harrison, Austin, Tex., for respondents White and Bullock. Justice WHITE delivered the opinion of the Court. 1 In this case, both Texas and California assert the right to levy state death taxes on the estate of Howard Hughes. The laws of each State impose an inheritance tax on the real and tangible personal property located within its borders, and upon the intangible personalty, wherever situated, of a person domiciled in the State at the time of death. Under the laws of Texas and California, an individual has but one domicile at any time. Taxing officials in each State assert that Howard Hughes was domiciled in their State at the time of his death. The issue before us is whether the Federal Interpleader Act, 28 U.S.C. § 1335, provides a jurisdictional basis for resolution of inconsistent death tax claims by the officials of two States. 2 * This case is the sequel to California v. Texas, 437 U.S. 601, 98 S.Ct. 3107, 57 L.Ed.2d 464 (1978). There, California petitioned for leave to file a complaint against Texas under this Court's original jurisdiction. At that time, we denied the motion. In concurring opinions, however, four Justices suggested that a determination of Hughes' domicile might be obtained in federal district court pursuant to the Federal Interpleader Act, 28 U.S.C. § 1335.1 3 Three weeks after the decision in California v. Texas, the administrator of the estate filed a statutory interpleader action in the United States District Court for the Western District of Texas. 491 F.Supp. 5. Asserting that the officials of the two States were seeking to tax the estate on the basis of inconsistent claims that each of their respective States was Howard Hughes' domicile at death, it requested the District Court to adjudicate the issue of domicile. The District Court entered a temporary restraining order prohibiting the California and Texas taxing officials from pursuing domicile-based inheritance tax claims in any other forum, including their own state courts. 4 The District Court then dismissed for lack of subject-matter jurisdiction for failure to satisfy the requirement of § 1335 that there be diversity of citizenship between at least two adverse claimants. It found that the administrator was not a claimant. Among the claimants, it held that the County Treasurer for Los Angeles County was a citizen of California for diversity purposes, citing Moor v. County of Alameda, 411 U.S. 693, 93 S.Ct. 1785, 36 L.Ed.2d 596 (1973). The court ruled, however, that the State of Texas, rather than its taxing officials, was the opposing claimant and that because a State is not a citizen of itself for diversity purposes, Postal Telegraph Cable Co. v. Alabama, 155 U.S. 482, 15 S.Ct. 192, 39 L.Ed. 231 (1894), the action did not involve two or more adverse claimants of diverse citizenship as required by the statute. 5 The Court of Appeals for the Fifth Circuit reversed the order of dismissal. Lummis v. White, 629 F.2d 397 (1980). In addition to the County Treasurer, it found the administrator of the estate, a citizen of Nevada, to be a claimant for the purposes of statutory interpleader. It recognized that Treinies v. Sunshine Mining Co., 308 U.S. 66, 60 S.Ct. 44, 84 L.Ed. 85 (1939), held that a citizenship of a disinterested stakeholder could not be considered in determining interpleader jurisdiction. Reasoning, however, that here the administrator's legal duty of preserving the estate's assets from the double death tax liability and his assertion that Hughes was domiciled in Nevada, which has no state death tax, made the administrator an interested stakeholder, the court further held that the citizenship of an interested stakeholder may be considered for purposes of establishing diversity under § 1335. The requisite diversity—between the administrator and the County Treasurer of Los Angeles County—was therefore present. 6 The Court of Appeals went on to reject the States' claim that although the suit was nominally against state officials, it was in effect a suit against two sovereign States barred by the Eleventh Amendment. Recognizing that Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed. 268 (1937), had squarely held that an interpleader action in all critical respects similar to this one was barred by the Eleventh Amendment, the Court of Appeals, relying on the concurring views of four Justices in California v. Texas, held that Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662 (1974), had silently, but effectively, overruled Worcester, and that the Eleventh Amendment as interpreted in Edelman did not bar the interpleader action. 7 The California officials petitioned for certiorari and, at the same time, filed a new motion seeking leave to file a complaint against Texas under this Court's original jurisdiction. Because of the troubling issues involving federal-court jurisdiction in such disputes, we granted certiorari. 452 U.S. 904, 101 S.Ct. 3028, 69 L.Ed.2d 404. II 8 In Worcester County Trust Co. v. Riley, supra, the States of California and Massachusetts each claimed to be the domicile of a decedent and to have the right to assess death taxes on his entire intangible estate. A federal interpleader action followed, the estate naming as defendant the revenue officers of California and Massachusetts. This Court unanimously held that the case was in reality a suit against the States and that it was barred by the Eleventh Amendment. In arriving at this conclusion, the Court applied the accepted rules (1) that "a suit nominally against individuals, but restraining or otherwise affecting their action as state officers, may be in substance a suit against the state, which the Constitution forbids," 302 U.S., at 296, 58 S.Ct., at 186, and (2) that "generally, suits to restrain action of state officials can, consistently with the constitutional prohibition, be prosecuted only when the action sought to be restrained is without the authority of state law or contravenes the statutes or Constitution of the United States." Id., at 297, 58 S.Ct., at 187. The Court held that there could be no credible claim of a violation of federal law since it was clear from prior cases that inconsistent determinations by the courts of two States as to the domicile of a taxpayer did not raise a substantial federal constitutional question. The Court also concluded that the claim that the officials were acting without authority under state law was insufficient. Hence, "[s]ince the proposed action is the performance of a duty imposed by the statute of the state upon state officials through whom alone a state can act, restraint of their action, which the bill of complaint prays, is restraint of state action, and the suit is in substance one against the State which the Eleventh Amendment forbids." Id., at 299-300, 58 S.Ct., at 187-188. 9 The Court of Appeals' opinion that Edelman v. Jordan had overruled Worcester rested on a passage in the Edelman opinion that it interpreted as limiting the bar of the Eleventh Amendment to suits "by private parties seeking to impose a liability which must be paid from public funds in the state treasury." 415 U.S., at 663, 94 S.Ct., at 1356. Because the interpleader plaintiff, the administrator of the estate, had sought only prospective relief, the appellate court held that the Eleventh Amendment did not bar his suit. 10 We are unpersuaded by this view of Edelman. That case involved a suit against state officials claiming that their administration of a particular federal-state program was contrary to federal regulations and the Constitution. Among other things, the plaintiffs sought a judgment for benefits that had not been paid them. The case was against individual officers who allegedly were violating federal law, and it therefore arguably fell outside the reach of the Eleventh Amendment under Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908). Edelman held, however, that the case was in effect a suit against the State itself because a judgment payable from state funds was demanded. It was correctly noted that Ford Motor Co. v. Department of Treasury of Indiana, 323 U.S. 459, 65 S.Ct. 347, 89 L.Ed. 389 (1945), was authority for this result. 11 Edelman did not hold, however, that the Eleventh Amendment never applies unless a judgment for money payable from the state treasury is sought.2 It would be a novel proposition indeed that the Eleventh Amendment does not bar a suit to enjoin the State itself simply because no money judgment is sought. The Eleventh Amendment reads: "The Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State. . . ." Thus, the Eleventh Amendment by its terms clearly applies to a suit seeking an injunction, a remedy available only from equity. To adopt the suggested rule, limiting the strictures of the Eleventh Amendment to a suit for a money judgment, would ignore the explicit language and contradict the very words of the Amendment itself. Edelman did not embrace, much less imply, any such proposition. 12 Neither did Edelman deal with a suit naming a state officer as defendant, but not alleging a violation of either federal or state law. Thus, there was no occasion in the opinion to cite or discuss the unanimous opinion in Worcester that the Eleventh Amendment bars suits against state officers unless they are alleged to be acting contrary to federal law or against the authority of state law. Edelman did not hold that suits against state officers who are not alleged to be acting against federal or state law are permissible under the Eleventh Amendment if only prospective relief is sought. Whether or not that would be the preferable rule, Edelman v. Jordan did not adopt it. 13 Furthermore, if that were to be the law, Worcester must in major part be overruled. We are unwilling, however, to overrule that decision and narrow the scope of the Eleventh Amendment to the extent that action would entail. We hold that the Eleventh Amendment bars the statutory interpleader sought in this case. The judgment of the Court of Appeals is 14 Reversed. 15 Justice BRENNAN, concurring in the judgment. 16 In California v. Texas, 437 U.S. 601, 98 S.Ct. 3107, 57 L.Ed.2d 464 (1978), I joined in the judgment of the Court denying California's motion for leave to file an original complaint. I was of the view that California's motion should be denied, "at least until such time as it is shown that . . . a statutory interpleader action cannot or will not be brought." Id., at 602, 98 S.Ct., at 3108. I also stated that I was "not so sure as" Justice Stewart and Justice POWELL that Texas v. Florida, 306 U.S. 398, 59 S.Ct. 563, 83 L.Ed. 817 (1939), had been wrongly decided. 437 U.S., at 601, 98 S.Ct., at 3108. See id., at 606, 98 S.Ct., at 3110 (Stewart, J., concurring); id., at 615, 98 S.Ct., at 3116 (POWELL, J., concurring). 17 Substantially for the reasons set forth in the opinion of the Court, it is now clear to me that so long as Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed. 268 (1937), remains good law, an interpleader suit in the district court is not a practical solution to the problem of potential double taxation presented in cases such as these. As Justice POWELL persuasively argues in Part III of his dissenting opinion, later cases, construing the Due Process Clause, have undermined Worcester County's holding that the unfairness of double taxation on the basis of conflicting determinations of domicile does not rise to constitutional dimensions. And Justice POWELL is surely correct in observing that "[t]he threat of multiple taxation based solely on domicile simply is incompatible with the structural principles of a federal system recognizing as 'fundamental' a constitutional right to travel." Post, at 2334. 18 But if Worcester County is not to be overruled, and interpleader is not available to provide relief from the possibility of duplicative taxation of this estate, I think it appropriate under Texas v. Florida, supra, to exercise our original jurisdiction to decide the present controversy. I agree with Professor Chafee, quoted post, at 101, that "[s]omewhere within [the] federal system we should be able to find remedies for the frictions which that system creates." Where such a remedy exists—even if only in the narrow class of cases falling within the holding of Texas v. Florida —it should be employed. The exercise of the Court's original jurisdiction in circumstances such as this is both just and prudent, and very likely in accordance with the Framer's original intent. 19 Justice POWELL, with whom Justice MARSHALL and Justice STEVENS join, dissenting. 20 The Court today decides two cases arising from the same set of facts, the instant case and California v. Texas, 457 U.S. 164, 102 S.Ct. 2335, 72 L.Ed.2d 755. Both cases involve the efforts of officials of California and Texas to tax the intangible property of the late Howard Hughes. Each State asserts its right to tax the Hughes estate on the basis of Hughes' domicile. Yet both recognize that Hughes could have had only one domicile at the time of his death. 21 In order to avoid multiple taxation that all agree would be unfair, the administrator of the Hughes estate invoked the Federal Interpleader Act1 as a means of litigating Hughes' domicile in one federal proceeding. The administrator alleged in his complaint, however, that Hughes was not a domiciliary of either California or Texas, but rather of the State of Nevada. App. 10.2 22 In the instant case, the Court holds today that this interpleader action is barred by the Eleventh Amendment. The Court does not dispute that multiple taxation based on domicile is unfair. Nor does it deny that the burden of multiple taxation ordinarily would fall, not on one of the claiming States, but solely on the heirs to an estate. But the Court opinion does not address these issues directly. Rigidly applying an aged and indefensible precedent, the Court denies the administrator and heirs of an estate any federal forum in which to resolve incompatible claims of domicile. 23 Having held in this case that there is no legal bar to both California and Texas taxing the Hughes estate on the basis of domicile, the Court surprisingly concludes in today's decision in California v. Texas, 457 U.S. 164, 102 S.Ct. 2335, 72 L.Ed.2d 755, that there presently exists a justiciable controversy "between" those two States as to which actually was Hughes' domicile.3 But these two cases—both decided today and both arising from the same set of facts—cannot be reconciled. Under the holding in the case that there is no federal prohibition against two States taxing the Hughes estate on the basis of domicile, the mere assertion of claims by the two States cannot suffice to establish a controversy "between" them. In finding that there is a case ripe for decision, the Court must rely on a double contingency: first, that both States might win judgments in their own courts that Hughes was a domiciliary subject to estate taxation; and second, that in such a case the Hughes estate might not be large enough to satisfy both claims. This is too speculative a foundation to support the conclusion that there is a case or controversy appropriately within our original jurisdiction. 24 In my view the Court's decisions in these cases rest on a misconception of the rights and obligations created by our federal system, both in its constitutional and in its statutory aspects. Accordingly I dissent. 25 * The issues before the Court today are substantially identical to those presented in California v. Texas, 437 U.S. 601, 98 S.Ct. 3107, 57 L.Ed.2d 464 (1978). In that case the Court unanimously denied California's motion for leave to file an original complaint. The Court's one-sentence order did not explain our decision to decline to exercise our exclusive original jurisdiction over controversies "between" States. Justice Stewart, however, in an opinion that Justice STEVENS and I joined, stated fully his reasons for agreeing that there existed no case or controversy between States.4 He argued cogently that California's complaint "contain[ed] the seeds of two distinct lawsuits": 26 "One is a dispute between two States as to the proper division of a finite sum of money. The other is a suit in the nature of interpleader to settle the question of a decedent's domicile for purposes of the taxes to be imposed upon his estate. But the suit in the nature of interpleader is not within the original and exclusive jurisdiction of this Court because it is not a dispute between States. And the dispute between the States, if indeed it is justiciable at all, is certainly not yet a case or controversy within the constitutional meaning of that term." Id., at 610-611, 98 S.Ct., at 3113. 27 No material fact has changed since 1978. On the premises of the Court's opinion, there still is no justiciable controversy between Texas and California. See California v. Texas, 457 U.S., at 170, 102 S.Ct. 2338 (POWELL, J., dissenting).5 There is, however, a ripe dispute about the estate's tax liability to the two States—a dispute of the kind for which federal interpleader jurisdiction ought to be available. II 28 In our 1978 decision in California v. Texas, supra, four Justices of this Court suggested that the administrator of the Hughes estate might invoke the Federal Interpleader Act to protect the estate from taxation based on inconsistent claims of domicile. Contradicting the clear message conveyed by our decision in that case, the Court today finds interpleader unavailable on the ground that a suit against the state taxing officials is barred by the Eleventh Amendment. 29 The concurring opinions in California v. Texas, supra, all proposed that the administrator of the Hughes estate might invoke the "fiction" of Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908), as interpreted in Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662 (1974), to bring an interpleader action naming as defendants the taxing officials of Texas and California. The Court today holds otherwise. According to the Court, it is the lawful function of the state officials to litigate Hughes' domicile. There is accordingly no colorable claim that they are acting in excess of their authority under state law; no constitutional violation is alleged; and Edelman v. Jordan is read narrowly to retain the Eleventh Amendment bar to injunctive suits against state officials not acting unlawfully, even in this case in which no money damages are sought from the state treasury. 30 There can be no doubt that Edelman will admit of a broader construction. The plain language of that decision asserts that the Eleventh Amendment bars only suits "by private parties seeking to impose a liability which must be paid from public funds in the state treasury," id., at 663, 94 S.Ct., at 1356, and not actions that may have "fiscal consequences to state treasuries . . . [that are] the necessary result of compliance with decrees which by their terms [are] prospective in nature," id., at 667-668, 94 S.Ct., at 1358. Thus, at least in a case such as this, in which the very controversy is the result of our federal system, I continue to believe that resort to federal interpleader is not proscribed by the Eleventh Amendment as construed by Edelman v. Jordan. 31 In rejecting this interpretation of Edelman, the Court relies at the last on Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed. 268 (1937). If this broader view of Edelman "were to be the law," the Court reasons, "Worcester must in major part be overruled." Ante, at 91. In light of Edelman, however, it must be recognized that the law has changed since 1937, and that the legal assumptions on which Worcester County rested no longer are uniformly valid. See California v. Texas, 437 U.S., at 601, 98 S.Ct., at 3108 (BRENNAN, J., concurring). If Worcester County cannot be defended on the basis of its internal logic and adherence to constitutional principles, this Court should not be bound by it. III 32 The Court today continues to reason from the premise, accepted by Worcester County, that multiple taxation on the basis of domicile does not offend the Constitution—even in a case in which both of the taxing States concede that a person may have but one domicile.6 In my view this premise is wrong. As an alternative to the approach that I embraced in California v. Texas, I now would be prepared to overrule Worcester County on this point and to hold that multiple taxation on the basis of domicile—at least insofar as "domicile" is treated as indivisible, so that a person can be the domiciliary of but one State—is incompatible with the structure of our federal system. 33 As Justice Stewart demonstrated in California v. Texas, the Court's conclusion in Texas v. Florida, 306 U.S. 398, 59 S.Ct. 563, 83 L.Ed. 817 (1939)—that there was a controversy between States, identifiable by analogy to a suit in the nature of interpleader—can be explained only by its concern for "the plight of the estate, which was indeed confronted with a 'substantial likelihood' of multiple and inconsistent tax claims." 437 U.S., at 606, 98 S.Ct., at 3111. Yet this focus of concern found no justification in the principles actually stated in Texas v. Florida, and it finds no justification in the principles on which the Court rests today. If the Constitution and laws provide no direct remedy to a decedent's estate faced with multiple taxation on the basis of domicile, there is no principled reason to protect the estate, before the fact, against the bare possibility that multiple taxation may exhaust the estate completely. See 437 U.S., at 611, 98 S.Ct., at 3113.7 In my view, however, such taxation is not only unfair but offensive to the Due Process Clause of the Fourteenth Amendment. B 34 Our decisions consistently have recognized that state taxation must be rationally related to " 'values connected with the taxing state.' " Moorman Mfg. Co. v. Bair, 437 U.S. 267, 273, 98 S.Ct. 2340, 2344, 57 L.Ed.2d 197 (1978), quoting Norfolk & Western R. Co. v. Missouri State Tax Comm'n, 390 U.S. 317, 325, 88 S.Ct. 995, 1001, 19 L.Ed.2d 1201 (1968). As framed by Justice Frankfurter in Wisconsin v. J. C. Penney Co., 311 U.S. 435, 444, 61 S.Ct. 246, 250, 85 L.Ed. 267 (1940): 35 "Th[e] test is whether property was taken without due process of law, or, if paraphrase we must, whether the taxing power exerted by the state bears fiscal relation to protection, opportunities and benefits given by the state. The simple but controlling question is whether the state has given anything for which it can ask return." 36 Under these principles tangible property generally may be taxed only by the State where it is located. Curry v. McCanless, 307 U.S. 357, 364, 59 S.Ct. 900, 904, 83 L.Ed. 1339 (1939).8 Physical presence also is required to justify a state succession tax on the transfer of real property occasioned by the death of the owner. Treichler v. Wisconsin, 338 U.S. 251, 70 S.Ct. 1, 94 L.Ed. 27 (1949); Frick v. Pennsylvania, 268 U.S. 473, 492, 45 S.Ct. 603, 605, 69 L.Ed. 1058 (1925). 37 In contrast with real property, intangible personal property is not physically located in any particular place, at least in any simple sense.9 Moreover, there may be more than one State that has a significant connection with intangible property—for example, the State in which a trust's assets are administered and the State in which the trustee is domiciled. See Curry v. McCanless, supra. Recognizing these differences, this Court has upheld the multiple taxation of intangible property. The decisions in which the Court has done so have not, however, undermined the fundamental principle that a State's levy of a tax must be connected rationally with the values on which the tax is imposed or with protections that the State has afforded. 38 In this case both California and Texas—as most States recognize that a person can have but one domicile. And it would appear settled that domicile provides the only adequate basis for taxation of intangible property in a decedent's estate, not located in the State or otherwise dependent on the protection of its laws. See Curry v. McCanless, supra, 307 U.S., at 365-366, 59 S.Ct., at 904-905; cf. Complete Auto Transit, Inc. v. Brady, 430 U.S. 274, 286-288, 97 S.Ct. 1076, 1082-1084, 51 L.Ed.2d 326 (1977) (defining Commerce Clause limits on state taxation in terms of connections to and benefits conferred by the taxing State). Here neither State alleges an entitlement to tax the Hughes estate on any other basis. From these premises it follows that multiple taxation based solely on conflicting determinations of domicile not only is unfair, but that taxation on this basis by at least one of the States must lack the only predicate asserted to justify its levy under the Due Process Clause.10 39 It is, of course, true that in 1937 Worcester County Trust Co. v. Riley, 302 U.S. 292, 58 S.Ct. 185, 82 L.Ed.2d 268, held that this admitted unfairness did not offend the Constitution. But Worcester County § holding on this point already has been undermined, not only by intervening decisions reiterating due process limits on state taxation of intangible property, see Norfolk & Western R. Co. v. Missouri State Tax Comm'n, supra, 390 U.S., at 323-326, 88 S.Ct., at 999-1002, and of income, see, e.g., Mobil Oil Corp. v. Commissioner of Taxes, 445 U.S. 425, 436-442, 100 S.Ct. 1223, 1230-1234, 63 L.Ed.2d 510 (1980), but also by cases in which this Court has recognized a fundamental right to travel. See, e.g., Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274 (1972); Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969). It is only by moving from State to State that a taxpayer risks incurring multiple taxation based on conflicting determinations of domicile. While no single State can be charged with creating this risk, the fact of its existence cannot be defended. The threat of multiple taxation based solely on domicile simply is incompatible with the structural principles of a federal system recognizing as "fundamental" a constitutional right to travel. C 40 By holding that multiple taxation based on domicile is prohibited by the Due Process Clause, the Court could lay the basis for resolution of disputes such as this one under the interpleader jurisdiction of the federal district courts. By alleging that state taxing officials threatened the estate with multiple liability, an administrator would state a colorable claim that the relevant state officers were acting outside of constitutional limits and thus that they were acting in their individual capacities under Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908). The Eleventh Amendment thus would not bar the suit under Ex parte Young and Edelman v. Jordan, and the interpleader requirement of competing claimants would be satisfied. 41 Professor Zechariah Chafee, the father of the federal interpleader statute, argued: "It is our federal system which creates the possibility of double taxation. Somewhere within that federal system we should be able to find remedies for the frictions which that system creates." Federal Interpleader Since the Act of 1936, 49 Yale L.J. 377, 388 (1940). 42 In my view the Due Process Clause provides the right to be free of multiple taxation of intangibles based on domicile. The Federal Interpleader Act provides the remedy. 43 As the Court holds otherwise, I respectfully dissent. 1 The Federal Interpleader Act, 28 U.S.C. § 1335, provides: "(a) The district courts shall have original jurisdiction of any civil action of interpleader or in the nature of interpleader filed by any person, firm, or corporation, association, or society having in his or its custody or possession money or property of $500 or more, or having issued a note, bond, certificate, policy of insurance, or other instrument of value or amount of $500 or more, or providing for the delivery or payment or the loan of money or property of such amount or value, or being under any obligation written or unwritten to the amount of $500 or more, if "(1) Two or more adverse claimants, of diverse citizenship as defined in section 1332 of this title, are claiming or may claim to be entitled to such money or property, or to any one or more of the benefits arising by virtue of any note, bond, certificate, policy or other instrument or arising by virtue of any such obligation; and if (2) the plaintiff has deposited such money or property or has paid the amount of or the loan or other value of such instrument or the amount due under such obligation into the registry of the court, there to abide the judgment of the court, or has given bond payable to the clerk of the court in such amount and with such surety as the court or judge may deem proper, conditioned upon the compliance by the plaintiff with the future order or judgment of the court with respect to the subject matter of the controversy. "(b) Such an action may be entertained although the titles or claims of the conflicting claimants do not have a common origin, or are not identical, but are adverse to and independent of one another." 2 The dissent mischaracterizes Edelman as asserting that the Eleventh Amendment bars "only" suits seeking money damages. Post, at 96. Edelman recognized the rule "that a suit by private parties seeking to impose a liability which must be paid from public funds in the state treasury is barred by the Eleventh Amendment," 415 U.S., at 663, 94 S.Ct., at 1356, but never asserted that such suits were the only ones so barred. 1 28 U.S.C. § 1335. 2 Nevada imposes no estate tax and therefore has not appeared as a party. 3 As a result of this decision, the Hughes heirs apparently will not suffer unfair double taxation. Other heirs of other estates presumably will not be so fortunate. 4 Justice BRENNAN also filed a concurring opinion tentatively accepting Justice Stewart's conclusion and stating that he would "deny California's motion, at least until such time as it is shown that . . . a statutory interpleader action cannot or will not be brought." 437 U.S., at 601, 602, 98 S.Ct., at 3108. I too filed a concurring opinion. Id., at 615, 98 S.Ct., at 3115-3116. 5 The Court's main ground for distinguishing the situation in 1978 from the situation today seems to be that "it seemed to several Members of the Court [in 1978] that statutory interpleader might obviate the need to exercise our original jurisdiction." California v. Texas, 457 U.S., at 168, 102 S.Ct., at 2337. Yet this argument simply is unresponsive to the question whether there is an actual case or controversy for which our original jurisdiction properly can be invoked. The Court notes that "several other uncertainties" have disappeared. Post, at 169, 102 S.Ct., at 2338. But its arguments are makeweights. Until the States have obtained conflicting judgments in their own courts, there is no ripe "dispute between two States as to the proper division of [the] finite sum of money" comprising the Hughes estate. California v. Texas, 437 U.S. 601, 610, 98 S.Ct. 3107, 3113, 57 L.Ed.2d 464 (1978) (Stewart, J., concurring). See post, at 170, 102 S.Ct., at 2338 (POWELL, J., dissenting). 6 Worcester County must be viewed in the context of a constitutional history that is hardly one of settled consistency. Only seven years before the Court decided Worcester County, in Farmers Loan & Trust Co. v. Minnesota, 280 U.S. 204, 50 S.Ct. 98, 74 L.Ed. 371 (1930), this Court had overruled Blackstone v. Miller, 188 U.S. 189, 23 S.Ct. 277, 47 L.Ed. 439 (1903), and held that the Due Process Clause forbids the multiple taxation of intangibles. For a time Farmers Loan & Trust Co. appeared to have established that only the single State of a person's domicile could tax intangible property in a decedent's estate. See First National Bank v. Maine, 284 U.S. 312, 52 S.Ct. 174, 76 L.Ed. 313 (1932). But the Court then reached the contrary conclusion in Worcester County, finding that inconsistent state-court adjudications of domicile and consequent assessment of estate taxes did not violate the Due Process Clause. First National Bank v. Maine, supra, then squarely was overruled by State Tax Comm'n v. Aldrich, 316 U.S. 174, 181, 62 S.Ct. 1008, 1011, 86 L.Ed. 1358 (1942), which held that multiple taxation of intangibles did not per se offend the Constitution. 7 "If it is unfair to subject an estate to two domicile-based taxes when all agree that it is possible to have only one domicile, that unfairness is just as great, if not greater, when a decedent's estate is able to pay the taxes to both States." 437 U.S., at 611, 98 S.Ct., at 3113. 8 "When we speak of the jurisdiction to tax land or chattels as being exclusively in the state where they are physically located, we mean no more than that the benefit and protection of laws enabling the owner to enjoy the fruits of his ownership . . . are so narrowly restricted to the state in whose territory the property is physically located as to set practical limits to taxation by others. Other states have been said to be without jurisdiction and so without constitutional power to tax tangibles if, because of their location elsewhere, those states can afford no substantial protection to the rights taxed. . . ." 307 U.S., at 364, 59 S.Ct., at 904. 9 See Curry v. McCanless, 307 U.S. 357, 365-366, 59 S.Ct. 900, 905, 83 L.Ed. 1339 (1939): "Very different considerations, both theoretical and practical, apply to the taxation of intangibles, that is, rights which are not related to physical things. Such rights are but relationships between persons, natural or corporate, which the law recognizes by attaching to them certain sanctions enforceable in courts. The power of government over them and the protection which it gives them cannot be exerted through control of a physical thing. They can be made effective only through control over and protection afforded to those persons whose relationships are the origin of the rights." 10 See Chafee, Federal Interpleader Since the Act of 1936, 49 Yale L.J. 377, 383-384 (1940) (footnotes omitted): "[T]here are two types of double taxation. In one kind, the same property or person is taxed in two states on two different theories. . . . In the other kind of double taxation, a single theory is applied in both states to tax the same person or property, but the two state governments disagree on a vital issue of fact. The Worcester County Trust Co. case falls into this class. Both states had the same law, that a death tax is levied only at the decedent's domicile and that a man has only one domicile. The only dispute was, where was that domicile? "It is rather surprising that almost all the attacks on double taxation . . . have been directed at the first kind, because the second kind seems more unjust. . . . [I]t is highly unfair for both state governments to tell the taxpayer, 'You have to pay only one tax,' and then make him pay twice. The injustice of the situation is clearly brought out by the fact that the courts of each state regard the other state as acting unlawfully, and yet neither state gives the taxpayer any remedy."

89

457 U.S. 172 102 S.Ct. 2923 72 L.Ed.2d 762 State of TEXAS, plaintiff,v.State of OKLAHOMA No. 85 Supreme Court of the United States June 14, 1982 1 On bill of complaint. DECREE 2 The motion for entry of judgment by consent of plaintiff and defendant, with the deletion of paragraph 10 thereof, and as amended with respect to paragraph 7, is granted. IT IS ORDERED, ADJUDGED, AND DECREED: 3 1. This judgment determines the boundary line between Texas and Oklahoma along the South bank of the Red River in Grayson County, Texas, from a point on said River as it existed prior to the construction of the Texoma Reservoir Dam (Denison Dam) approximately 1973 feet West of the center line of said Dam, with its meanders, to a point on said River approximately 6103 feet East of the center line of said Dam, upon the Complaint, Answer and agreement of Counsel for Texas and Oklahoma. 4 2. The source of the boundary line between Texas and Oklahoma from the 100th meridian of longitude to the eastern border of Oklahoma (which encompasses the boundary determined by this judgment) lies in the Treaty of 1819, 8 Stat. 252 (1821), which was construed by the Court in United States v. Texas, 162 U.S. 1, 16 S.Ct. 725, 40 L.Ed. 867 (1895), to be the south bank of the Red River. The Court later confirmed this definition of the boundary in Oklahoma v. Texas, 256 U.S. 70, 41 S.Ct. 420, 65 L.Ed. 831 (1921), and in a later partial decree therein, 261 U.S. 340, 341-342, 43 S.Ct. 376, 377, 67 L.Ed. 687 (1923), defined the South bank as: 5 ". . . the water-washed and relatively permanent elevation or acclivity, commonly called a cut bank, along the southerly side of the river, which separates its bed from the adjacent upland, whether valley or hill, and usually serves to confine the waters within the bed, and to preserve the course of the river. 6 ". . . The boundary between the two states is on and along that bank at the mean level attained by the waters of the river when they reach and wash the bank without overflowing it. 7 ". . . At exceptional places where there is no well-defined cut bank, but only a gradual incline from the sand bed of the river to the upland, the boundary is a line over such incline, conforming to the mean level of the waters, when at other places in that vicinity they reach and wash the cut bank without overflowing it." 8 3. As a result of the partial decree in 1923 and other partial decrees arising from the same controversy, a boundary commission was established to take evidence, find facts and report to the Court. Said Commission filed a documentary report, 269 U.S. 536, 46 S.Ct. 104, 70 L.Ed. 399 (1925), styled as the Third Report of the Boundary Commissioners. At page 41 of said Report appears the following entry: VII. 9 "IN GRAYSON COUNTY, TEXAS OPPOSIVE 10 MARSHALL AND BRYAN COUNTIES, 11 OKLAHOMA. 12 "Public Hearing: Sherman, Texas, May 7, 1925 13 "We found no avulsive changes in the position of the Red River in this County and make no surveys." 14 4. In 1939 the United States Army Corps of Engineers made surveys of certain tracts of land in Grayson County, Texas, known as Tract T-2-1, Tract T-2-2 and Tract T-2-4, whose northern boundaries coincide with the South bank of the Red River and the boundary determined by this judgment. Said surveys were made in anticipation of the condemnation of said Tracts for purposes of the construction of the Texoma Dam Facility and Reservoir pursuant to an Act of Congress, 52 Stat. 1215 (1938). 15 5. On September 7, 1940, a judgment was entered in the United States District Court for the Eastern District of Texas making final the award of special commissioners as to Tract T-2-1. On November 23, 1940, judgment was entered in the United States District Court for the Eastern District of Texas making final the acquisition by the United States of Tract T-2-4. On December 28, 1939, a warranty deed was executed by P. O. Brack to the United States as to Tract T-2-2. 16 6. On June 23, 1980, the Plaintiff and Defendant together retained a Registered Public Surveyor of the State of Texas to make a reenactment survey of Tract T-2-1, Tract T-2-2 and Tract T-2-4. On the basis of such survey it has been determined that, when tying back to the original South bank of the Red River, the northern boundaries of Tracts T-2-1, T-2-2 and T-2-4 (coinciding with the South bank of the Red River) as established by the U. S. Army Corps of Engineers in its survey of 1939 were correct at the time that survey was made, which was prior to the construction of the Texoma Reservoir Dam Facility. A certified copy of the plat of the survey commissioned by the States with the surveyor's certified explanation thereof is filed with the Clerk of this Court. 17 7. The boundary between Texas and Oklahoma determined by this judgment coincides with the boundaries of Tracts T-2-1, T-2-2 and T-2-4 and is described as follows, with all bearings from the above referenced judgment and deed descriptions, and all distances in feet: 18 BEGINNING at a point, same being the centerline of Shawnee Creek and the Northeast corner of Tract T-2-4 as acquired by the United States in a Judgment as recorded in Volume 420 Page 556 of the Deed Records of Grayson County, Texas. 19 THENCE up the south bank of the Red River as it existed prior to the construction of the Texoma Reservoir and Denison Dam, with its meanders: South 78x08' West a distance of 528.9 feet; South 86x10' West a distance of 1370.50 feet; South 89x06' West a distance of 484.0 feet; North 88x22' West a distance of 447.2 feet; North 85x37' West passing the Northeast corner of Tract T-2-2 as recorded in Volume 417 Page 23 of the Deed Records of Grayson County, Texas and the total distance of 1675.30 feet; North 76x21' West a distance of 413.7 feet; North 86x01' West a distance of 170.10 feet to the Northwest corner of Tract T-2-2; 20 THENCE North 76x01' West a distance of 394.50 feet; North 87x34' West a distance of 1198.90 feet; North 68x27' West a distance of 362.20 feet; North 55x54' West a distance of 1030.80 feet to a point, same being the Northwest corner of Tract T-2-1 as acquired by the United States in a Judgment as recorded in Volume 417 Page 123 of the Deed Records of Grayson County, Texas. 21 8. The boundary line delineated hereinabove is depicted by a line marked "STATE LINE" on the plat of the survey commissioned by the States and filed with the Clerk of this Court. 22 9. The construction of the Texoma Reservoir and Denison Dam did not alter the boundary between Texas and Oklahoma as the South bank of the Red River as it existed prior to such construction in any manner whatsoever. 23 10. The cost of this action shall be equally divided between the two States.

1011

457 U.S. 255 102 S.Ct. 2421 72 L.Ed.2d 824 Ralph HATHORN, et al., Petitionersv.Mrs. Bobby LOVORN, et al. No. 81-451. Argued April 27, 1982. Decided June 15, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1131, 103 S.Ct. 15. Syllabus A 1964 Mississippi statute provides that boards of trustees of municipal separate school districts in the State shall consist of five members, and that in any county in which a district embraces the entire county "in which Highways 14 and 15 intersect," one trustee shall be elected from each supervisors district. The Louisville School District is coextensive with Winston County, Miss., which is the only county in which Highways 14 and 15 intersect. Since 1960, the Louisville mayor and city aldermen appointed three of the five members of the District's Board of Trustees, and Winston County voters residing outside Louisville elected the other two members. The county officials never implemented the 1964 statute. Respondent Winston County voters filed an action against petitioner local officials in Mississippi Chancery Court seeking to enforce the 1964 statute. The court dismissed the complaint on the ground that the statute violated the state constitutional bar against local legislation. The Mississippi Supreme Court reversed and remanded, striking only the statute's reference to Highways 14 and 15 and upholding the remainder of the statute. The Supreme Court without comment denied petitioners' petition for rehearing in which they argued for the first time that the Chancery Court could not implement the reformed statute until the change had been precleared under § 5 of the Voting Rights Act of 1965. On remand, the Chancery Court ordered an election pursuant to the redacted statute under procedures prescribed by the court, but directed petitioners to submit the election plan to the United States Attorney General for preclearance under § 5 of the Voting Rights Act. The Attorney General subsequently objected to the plan, and the Chancery Court ultimately concluded that its order would remain in force subject to compliance with the Voting Rights Act. Respondents once again appealed to the Supreme Court, which held that its prior decision was the law of the case and that the Chancery Court improperly conditioned the election on compliance with the Voting Rights Act. Held: 1. The Mississippi Supreme Court's decision did not rest on independent and adequate state grounds so as to bar this Court's review of the federal issue. Where the state court's first decision did not appear final when rendered, the court's subsequent reliance on the law of the case does not prevent this Court from reviewing federal questions determined in the first appeal. Nor does the fact that petitioners' reliance upon the Voting Rights Act issue for the first time in their petition for rehearing may have been untimely under a Mississippi procedural rule constitute an independent and adequate state ground barring this Court's review of the federal question, where it appears that, if Mississippi still follows such a rule, it does not do so "strictly or regularly." Pp. 261-265. 2. The Mississippi courts had the power to decide whether § 5 of the Voting Rights Act applied to the change in election procedures sought by respondents, and must withhold further implementation of the disputed change until the parties demonstrate compliance with § 5. Both the language and purposes of the Act refute the notion that a state court asked to implement a change in the State's voting laws cannot inquire whether the change is subject to § 5 but must ignore that circumstance and enter a decree violating federal law. Section 14(b) of the Act, which provides that no court other than the District Court for the District of Columbia shall have jurisdiction to enter a declaratory judgment pursuant to § 5 governs only declaratory judgments approving proposed voting procedure changes. And nothing in the provisions of § 5, requiring an action under that section to be heard by a three-judge federal district court, or in the provisions of § 12(f) of the Act, giving federal district courts jurisdiction of proceedings under that section, negates the presumption that, at least when the issue arises collaterally, state courts have the power to decide whether a proposed change in election procedures requires preclearance under § 5. Granting state courts such power helps to insure compliance with the preclearance scheme. Pp. 265-271. Miss., 399 So.2d 1356, reversed and remanded. James C. Mayo, Louisville, Miss., for petitioners. William Bradford Reynolds, Washington, D. C., for United States, as amicus curiae, by special leave of Court. Laurel G. Weir, Philadelphia, Miss., for respondents. Justice O'CONNOR delivered the opinion of the Court. 1 We granted certiorari to decide whether a state court may order implementation of a change in election procedure over objections that the change is subject to preclearance under § 5 of the Voting Rights Act of 1965.1 2 * Since 1960, the Louisville School District has been coextensive with Winston County, Miss. Until last December, the Louisville mayor and city aldermen appointed three of the five members of the District's Board of Trustees, and Winston County voters residing outside Louisville elected the other two members. 3 In 1964, the Mississippi Legislature enacted a statute providing in part: "The boards of trustees of all municipal separate school districts, either with or without added territory, shall consist of five (5) members, each to be chosen for a term of five (5) years, but so chosen that the term of office of one (1) member shall expire each year. . . . [I]n any county in which a municipal separate school district embraces the entire county in which Highways 14 and 15 intersect, one (1) trustee shall be elected from each supervisors district." 1964 Miss.Gen.Laws, ch. 391, p. 563, codified, as amended, in Miss.Code Ann. § 37-7-203(1) (Supp.1981). 4 Winston County is the only Mississippi county in which Highways 14 and 15 intersect. Officials in that county never implemented § 37-7-203(1) because they believed the statute's reference to Highways 14 and 15 violated a state constitutional prohibition against local, private, or special legislation.2 5 In 1975, five Winston County voters filed an action in the Chancery Court of Winston County,3 seeking to enforce the neglected 1964 state statute.4 These plaintiffs, respondents here, named numerous Louisville and Winston County officials as defendants. The Chancery Court dismissed respondents' complaint, holding that the statute violated Mississippi's constitutional bar against local legislation. The Mississippi Supreme Court reversed, striking only the specific reference to Highways 14 and 15 and upholding the remaining requirement that, "in any county in which a municipal separate school district embraces the entire county," each supervisors district must elect one trustee. Lovorn v. Hathorn, 365 So.2d 947 (1979) (en banc). The court then "remanded to the chancery court for further proceedings not inconsistent with [its] opinion." Id., at 952. 6 The local officials, petitioners here, filed a petition for rehearing, in which they argued for the first time that the Chancery Court could not implement the reformed statute until the change had been precleared under § 5 of the Voting Rights Act. The Mississippi Supreme Court denied the petition without comment, and this Court denied a petition for a writ of certiorari. Hathorn v. Lovorn, 441 U.S. 946, 99 S.Ct. 2167, 60 L.Ed.2d 1049 (1979). 7 On remand, the Chancery Court ordered an election pursuant to the redacted statute. The court set out detailed procedures governing the election, including the requirement that "[i]f no candidate receives a majority of the vote cast at any of said elections . . ., a runoff election shall be held . . . between the two candidates receiving the highest vote [in the first election]." Record 143. The court derived the latter requirement from Miss.Code Ann. § 37-7-217 (Supp.1981), which mandates runoffs in elections conducted under § 37-7-203(1). See Miss.Code Ann. § 37-7-209 (Supp.1981). The Chancery Court also agreed with petitioners' claim that the changes in election procedure fell within § 5 of the Voting Rights Act, and directed petitioners to submit the election plan to the United States Attorney General for preclearance. Record 141, 146-147.5 8 Upon review of petitioners' submission, the Attorney General objected to the proposed change in election procedure "insofar as it incorporate[d] a majority vote requirement." App. to Pet. for Cert. A-8. Because of the substantial black population in Winston County,6 an apparent pattern of racially polarized voting in the county, and the historical absence of blacks from various local governing boards, the Attorney General concluded that the runoff procedure could have a discriminatory effect. Ibid.7 9 Respondents attempted to overcome this obstacle by both joining the Attorney General as a defendant and persuading the Chancery Court to hold the election without the runoff procedure. The court, however, refused to join the Attorney General and held that state law unambiguously required runoff elections. Buffeted by apparently conflicting state and federal statutes, the Chancery Court concluded that its decree calling for an election would "remain in force subject to compliance with the Federal Voters Rights Act [sic] as previously ordered by this Court." Record 342. 10 Failing to obtain an election from the Chancery Court, respondents once again appealed to the Mississippi Supreme Court. That court observed that its "prior decision, which the United States Supreme Court declined to reverse or alter in any respect, became and is the law of the case." Carter v. Luke, 399 So.2d 1356, 1358 (1981). The court explained that because the prior decision upheld a statute referring to the statute requiring runoffs, and because both parties had agreed during oral argument to abide by the runoff procedure, the Chancery Court properly enforced the law requiring runoffs and improperly conditioned the election on compliance with the Voting Rights Act. Accordingly, the Mississippi Supreme Court reversed the portion of the Chancery Court's decree referring to the Voting Rights Act and "remanded with directions for the lower court to call and require the holding of an election." Ibid. We granted certiorari to decide whether the Mississippi Supreme Court properly ordered the election without insuring compliance with federal law. 454 U.S. 1122, 102 S.Ct. 969, 71 L.Ed.2d 109 (1981).8 II 11 Before addressing the federal question raised by the Mississippi Supreme Court's decision, we must consider respondents' assertion that the lower court decision rests upon two adequate and independent state grounds. First, respondents contend that the state court's reliance upon the law of the case bars review of the federal question. It has long been established, however, that "[w]e have jurisdiction to consider all of the substantial federal questions determined in the earlier stages of [state proceedings], . . . and our right to re-examine such questions is not affected by a ruling that the first decision of the state court became the law of the case. . . ." Reece v. Georgia, 350 U.S. 85, 87, 76 S.Ct. 167, 169, 100 L.Ed. 77 (1955). See also Davis v. O'Hara, 266 U.S. 314, 321, 45 S.Ct. 104, 107, 69 L.Ed. 303 (1924); United States v. Denver & Rio Grande R. Co., 191 U.S. 84, 93, 24 S.Ct. 33, 86, 48 L.Ed. 106 (1903). Because we cannot review a state court judgment until it is final,9 a contrary rule would insulate interlocutory state court rulings on important federal questions from our consideration. 12 In this case the Mississippi Supreme Court's first decision plainly did not appear final at the time it was rendered. The court's remand "for further proceedings not inconsistent with [its] opinion," 365 So.2d, at 952 (en banc), together with its failure to address expressly the Voting Rights Act issue, suggested that the Chancery Court could still consider the federal issue on remand. Indeed, the Chancery Court interpreted its mandate in precisely this manner.10 Under these circumstances, the Mississippi Supreme Court's subsequent reliance on the law of the case cannot prevent us from reviewing federal questions determined in the first appeal.11 13 Respondents also argue that the Mississippi Supreme Court pretermitted consideration of the Voting Rights Act because petitioners' reliance upon the issue in a petition for rehearing was untimely. We have recognized that the failure to comply with a state procedural rule may constitute an independent and adequate state ground barring our review of a federal question.12 Our decisions, however, stress that a state procedural ground is not "adequate" unless the procedural rule is "strictly or regularly followed." Barr v. City of Columbia, 378 U.S. 146, 149, 84 S.Ct. 1734, 1736, 12 L.Ed.2d 766 (1964). State courts may not avoid deciding federal issues by invoking procedural rules that they do not apply evenhandedly to all similar claims. Even if we construe the Mississippi Supreme Court's denial of petitioners' petition for rehearing as the silent application of a procedural bar, we cannot conclude that the state court consistently relies upon this rule. 14 Respondents cite two cases indicating that the Mississippi Supreme Court will consider an issue raised for the first time in a petition for rehearing "[o]nly in exceptional cases." New & Hughes Drilling Co. v. Smith, 219 So.2d 657, 661 (Miss.1969); Rigdon v. General Box Co., 249 Miss. 239, 246, 162 So.2d 863, 864 (1964). Although these opinions may summarize the court's practice prior to 1969, we have been unable to find any more recent decisions repeating or applying the rule.13 On the contrary, the Mississippi Supreme Court now regularly grants petitions for rehearing without mentioning any restrictions on its authority to consider issues raised for the first time in the petitions.14 15 One particular decision by the Mississippi Supreme Court, decided only last year, demonstrates that the court does not consistently preclude consideration of issues raised for the first time on rehearing. In Quinn v. Branning, 404 So.2d 1018 (1981), the court held that part of a criminal statute violated the State Constitution's prohibition against local legislation. Striking the offensive language, the court approved the rest of the statute and affirmed the underlying conviction. The defendant then petitioned for rehearing, pointing out that the affidavit against him did not allege a crime under the reformed statute. The court agreed with this contention, granted the petition in part, and reversed the conviction, all without mentioning the rule against consideration of new issues on rehearing. The striking similarity between Quinn and this case, both involving issues that the parties could have foreseen but that arose with urgency only after the court upheld part of a challenged statute, persuades us that the Mississippi Supreme Court is not "strictly or regularly" following a procedural rule precluding review of issues raised for the first time in a petition for rehearing. The denial of rehearing in this case, although not appearing sufficiently final to permit our immediate review, must have rested either upon a substantive rejection of petitioners' federal claim or upon a procedural rule that the state court applies only irregularly.15 Thus, there are no independent and adequate state grounds barring our review of the federal issue. III 16 Respondents do not dispute that the change in election procedures ordered by the Mississippi courts is subject to preclearance under § 5.16 They urge, however, that the Voting Rights Act deprives state courts of the power even to decide whether § 5 applies to a proposed change in voting procedures.17 Under their analysis of the Act, a state court asked to implement a change in the State's voting laws could not inquire whether the change was subject to § 5. Even if the change plainly fell within § 5, the court would have to ignore that circumstance and enter a decree violating federal law. Both the language and purposes of the Voting Rights Act refute this notion. 17 Only last Term we summarized the principles governing state court jurisdiction to decide federal issues. Gulf Offshore Co. v. Mobil Oil Corp., 453 U.S. 473, 101 S.Ct. 2870, 69 L.Ed.2d 784 (1981). We begin, in every case, "with the presumption that state courts enjoy concurrent jurisdiction" over those claims. Id., at 478, 101 S.Ct., at 2875. Only "an explicit statutory directive, [an] unmistakable implication from legislative history, or . . . a clear incompatibility between state-court jurisdiction and federal interests" will rebut the presumption. Ibid. Most important for our purposes, even a finding of exclusive federal jurisdiction over claims arising under a federal statute usually "will not prevent a state court from deciding a federal question collaterally." Id., at 483, n. 12, 101 S.Ct., at 2878, n. 12.18 18 Respondents rest their jurisdictional argument on three sections of the Act. Section 14(b) provides that "[n]o court other than the District Court for the District of Columbia . . . shall have jurisdiction to issue any declaratory judgment pursuant to . . . section 5. . . ." 79 Stat. 445, 42 U.S.C. § 1973l (b). We have already held, however, that this provision governs only declaratory judgments approving proposed changes in voting procedure. Other courts may decide the distinct question of whether a proposed change is subject to the Act. See Allen v. State Board of Elections, 393 U.S. 544, 557-560, 89 S.Ct. 817, 22 L.Ed.2d 1 (1969); McDaniel v. Sanchez, 452 U.S. 130, 101 S.Ct. 2224, 62 L.Ed.2d 724 (1981). 19 Sections 5 and 12(f) of the Act provide somewhat stronger support for respondents' claim. Section 5 provides that "[a]ny action under this section shall be heard and determined by a court of three judges in accordance with the provisions of section 2284 of title 28 of the United States Code," 79 Stat. 439, 42 U.S.C. § 1973c, while § 12(f) declares that "[t]he district courts of the United States shall have jurisdiction of proceedings instituted pursuant to this section." 79 Stat. 444, 42 U.S.C. § 1973j(f).19 It is possible that these sections grant the federal courts exclusive jurisdiction over "action[s] under" § 5 or "proceedings instituted pursuant" to § 12.20 We need not resolve that question in this case, however, because respondents' state suit fell within neither of these categories. Instead, respondents' initial suit was an action to compel compliance with a forgotten state law.21 Nothing in § 5 or § 12 negates the presumption that, at least when the issue arises collaterally, state courts may decide whether a proposed change in election procedure requires preclearance under § 5. 20 The policies of the Act support the same result.22 The Voting Rights Act "implemented Congress' firm intention to rid the country of racial discrimination in voting." Allen v. State Board of Elections, supra, at 548, 89 S.Ct., at 822. Fearing that covered jurisdictions would exercise their ingenuity to devise new and subtle forms of discrimination, Congress prohibited those jurisdictions from implementing any change in voting procedure without obtaining preclearance under § 5. Granting state courts the power to decide, as a collateral matter, whether § 5 applies to contemplated changes in election procedures will help insure compliance with the preclearance scheme.23 Approval of this limited jurisdiction also avoids placing state courts in the uncomfortable position of ordering voting changes that they suspect, but cannot determine, should be precleared under § 5. Accordingly, we hold that the Mississippi courts had the power to decide whether § 5 applied to the change sought by respondents. 21 If the Mississippi courts had the power to make this determination, then it is clear that they also had the duty to do so. "State courts, like federal courts, have a constitutional obligation . . . to uphold federal law." Stone v. Powell, 428 U.S. 465, 494, n. 35, 96 S.Ct. 3037, 3052, 49 L.Ed.2d 1067 (1976) (citing Martin v. Hunter's Lessee, 1 Wheat. 304, 341-344, 4 L.Ed. 97 (1816)). Section 5 declares that whenever a covered jurisdiction shall "enact or seek to administer any . . . standard, practice, or procedure with respect to voting different from that in force or effect on November 1, 1964," see n. 1, supra, it must obtain either preclearance from the Attorney General or a declaratory judgment from the United States District Court for the District of Columbia. Our opinions repeatedly note that failure to follow either of these routes renders the change unenforceable. See, e.g., Dougherty County Board of Education v. White, 439 U.S. 32, 46, 99 S.Ct. 368, 376, 58 L.Ed.2d 269 (1978); United States v. Board of Supervisors, 429 U.S. 642, 645, 97 S.Ct. 833, 834, 51 L.Ed.2d 106 (1977) (per curiam ). When a party to a state proceeding asserts that § 5 renders the contemplated relief unenforceable, therefore, the state court must examine the claim and refrain from ordering relief that would violate federal law.24 IV 22 Our holding mandates reversal of the lower court judgment. Under our analysis, the change in election procedure is subject to § 5, see n. 16, supra, and the Mississippi courts may not further implement that change until the parties comply with § 5. At this time, however, we need not decide whether petitioners are entitled to any additional relief. The United States has initiated a federal suit challenging the change at issue here, see n. 8, supra, and we agree with the Solicitor General that the District Court entertaining that suit should address the problem of relief in the first instance. As we noted in Perkins v. Matthews, 400 U.S. 379, 395-397, 91 S.Ct. 431, 440-441, 27 L.Ed.2d 476 (1971), a local district court is in a better position than this Court to fashion relief, because the district court "is more familiar with the nuances of the local situation" and has the opportunity to hear evidence. Id., at 397, 91 S.Ct., at 441. In this case, the District Court for the Northern District of Mississippi will be better able to decide whether a special election is necessary, whether a more moderate form of interim relief will satisfy § 5,25 or whether new elections are so imminent that special relief is inappropriate. We hold only that the Mississippi courts must withhold further implementation of the disputed change in election procedures until the parties demonstrate compliance with § 5. Accordingly, the judgment of the Mississippi Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. 23 So ordered. 24 Justice POWELL concurs in the judgment. 25 Justice REHNQUIST, dissenting. 26 The provisions of §§ 5, 12(f), and 14(b) of the Voting Rights Act, referred to in the opinion of the Court, ante, at 265-268, convince me that Congress did not intend the state courts to play a role in the enforcement of that Act. In Gulf Offshore Co. v. Mobil Oil Corp., 453 U.S. 473, 101 S.Ct. 2870, 69 L.Ed. 784 (1981), upon which the Court heavily relies for its contrary conclusion, we said: 27 "The factors generally recommending exclusive federal-court jurisdiction over an area of federal law include the desirability of uniform interpretation, the expertise of federal judges in federal law, and the assumed greater hospitality of federal courts to peculiarly federal claims." Id., at 483-484, 101 S.Ct., at 2878-2880 (footnotes omitted). 28 It seems to me that each of these factors counsels in favor of exclusive federal-court jurisdiction, and I do not understand the Court to contend otherwise. 29 From a practical point of view, I think the Court's decision is bound to breed conflicts between the state courts and the federal district courts sitting within the States, each of which may now determine whether or not a particular voting change must be precleared with the Attorney General before being enforced in a covered jurisdiction. Indeed, the precursor of such conflict may well be found in the Court's concluding observations that the District Court for the Northern District of Mississippi, in which the United States has pending a suit pertaining to the change involved in this case, should proceed to make determinations under the Voting Rights Act before the state court whose judgment we are reviewing renders further remedy in this case. Exactly what is to be left to the States under this construction is more than a little problematical. 30 I do not think that the goals of the Voting Rights Act will be materially advanced by the Court's somewhat tortured effort to make the state courts a third line of enforcement for the Act, after the District Court for the District of Columbia and other federal district courts. The principal effect of today's decision will be to enable one or the other of parties such as those involved in this case, neither of whom were intended to be primary beneficiaries of the Voting Rights Act, to employ the Act as another weapon in their arsenal of litigation strategies. 1 Section 5 provides in relevant part: "Whenever a [covered] State or political subdivision . . . shall enact or seek to administer any voting qualification or prerequisite to voting, or standard, practice, or procedure with respect to voting different from that in force or effect on November 1, 1964, . . . such State or subdivision may institute an action in the United States District Court for the District of Columbia for a declaratory judgment that such qualification, prerequisite, standard, practice, or procedure does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, and unless and until the court enters such judgment no person shall be denied the right to vote for failure to comply with such qualification, prerequisite, standard, practice, or procedure: Provided, That such qualification, prerequisite, standard, practice, or procedure may be enforced without such proceeding if the qualification, prerequisite, standard, practice, or procedure has been submitted by the chief legal officer or other appropriate official of such State or subdivision to the Attorney General and the Attorney General has not interposed an objection within sixty days after such submission, or upon good cause shown, to facilitate an expedited approval within sixty days after such submission, the Attorney General has affirmatively indicated that such objection will not be made. . . ." 79 Stat. 439, as amended, 42 U.S.C. § 1973c. Section 4 of the Act, 79 Stat. 438, as amended, 42 U.S.C. § 1973b, defines covered jurisdictions. 2 Mississippi Const., Art. 4, § 90, provides: "The legislature shall not pass local, private, or special laws in any of the following enumerated cases, but such matters shall be provided for only by general laws, viz.: * * * * * "(p ) Providing for the management or support of any private or common school, incorporating the same, or granting such school any privileges." 3 The voters initially filed their suit in the United States District Court for the Northern District of Mississippi. That court stayed federal proceedings to give the Mississippi courts an opportunity to construe the state statute at issue. Record 320. In 1979, pursuant to a notice of voluntary dismissal by stipulation, the court dismissed the federal action without prejudice. Id., at 323. 4 The voters also charged that the electoral system then in force violated the constitutional principle of one person/one vote. This issue is not before us. 5 As we have explained on numerous occasions, covered jurisdictions may satisfy § 5 by submitting proposed changes to the Attorney General. If the Attorney General objects to the proposal, the jurisdiction may either request reconsideration or seek a declaratory judgment from the United States District Court for the District of Columbia. A covered jurisdiction, of course, also may seek a declaratory judgment in the first instance, omitting submission to the Attorney General. See generally Blanding v. DuBose, 454 U.S. 393, 102 S.Ct. 715, 70 L.Ed.2d 576 (1982); Allen v. State Board of Elections, 393 U.S. 544, 548-550, 89 S.Ct. 817, 822-823, 22 L.Ed.2d 1 (1969). 6 At that time, the Attorney General noted, blacks constituted approximately 39% of the Winston County population but were not a majority in any of the districts from which trustees were to be elected. 7 The Attorney General also observed that the Louisville School District appears to be the only countywide district in which Mississippi requires runoff elections. 8 Shortly before petitioners filed their petition for certiorari, the Chancery Court set an election for December 5, 1981. That court, the Mississippi Supreme Court, and this Court denied motions to stay the election. See 454 U.S. 1070, 102 S.Ct. 623, 70 L.Ed.2d 607 (1981). On December 1, the United States filed suit in the United States District Court for the Northern District of Mississippi, seeking to enjoin implementation of the voting change involved in this case. The District Court refused to issue a temporary restraining order and has not taken any other action. The December 5 election was held as scheduled. Although the record does not reflect the results of the election, the United States has informed us that a runoff election was held. Brief for United States as Amicus Curiae 10, n. 12. 9 28 U.S.C. § 1257; O'Dell v. Espinoza, 456 U.S. 430, 102 S.Ct. 1865, 72 L.Ed.2d 237 (1982); Market Street R. Co. v. Railroad Comm'n of California, 324 U.S. 548, 551, 65 S.Ct. 770, 772, 89 L.Ed. 1171 (1945). 10 The Chancellor, in fact, noted that it "would have been impossible to have submitted to the Attorney General for approval until this Court had set up the mechanics of the election, for until that was done, the Attorney General would not have the data necessary to either approve or disapprove." Record 90-91. 11 Nor, of course, does our previous denial of petitioners' petition for a writ of certiorari preclude us from examining questions decided during the first state appeal. It is "well-settled . . . that denial of certiorari imparts no implication or inference concerning the court's view of the merits." Hughes Tool Co. v. Trans World Airlines, Inc., 409 U.S. 363, 366, n. 1, 93 S.Ct. 647, 650, 34 L.Ed.2d 577 (1973). 12 E.g., Michigan v. Tyler, 436 U.S. 499, 512, n. 7, 98 S.Ct. 1942, 1951, n. 7, 56 L.Ed.2d 486 (1978); New York Times Co. v. Sullivan, 376 U.S. 254, 264, n. 4, 84 S.Ct. 710, 717, 11 L.Ed.2d 686 (1964). 13 In New & Hughes Drilling Co. itself, the Mississippi Supreme Court permitted an exception to the alleged rule barring review of questions raised for the first time on rehearing. A case decided the same year as New & Hughes Drilling Co. is the most recent decision we have found that might have actually applied the procedural rule described by respondents. See Leake County Cooperative v. Dependents of Barrett, 226 So.2d 608, 614-616 (Miss.1969). Even that decision, however, may have rested upon a special rule involving waiver of defects in venue. Neither the Mississippi Code nor the Rules of the Supreme Court of Mississippi embody the alleged prohibition against presentation of new issues in petitions for rehearing. Under these circumstances, it is difficult to know whether the Mississippi Supreme Court still adheres to the rule, applying it silently, or whether the court has abandoned the rule. 14 See, e.g., Cortez v. Brown, 408 So.2d 464 (1981) (en banc); Cash v. Illinois Central Gulf R. Co., 388 So.2d 871 (1980) (en banc); McKee v. McKee, 382 So.2d 287 (1980) (en banc); City of Jackson v. Capital Reporter Publishing Co., 373 So.2d 802 (1979) (en banc); Realty Title Guaranty Co. v. Howard, 355 So.2d 657 (1977) (en banc); Couch v. Martinez, 357 So.2d 107 (1978) (en banc); Foster v. Foster, 344 So.2d 460 (1977) (en banc); McCrory v. State, 342 So.2d 897 (1977) (en banc); Daniels v. State, 341 So.2d 918 (1977) (en banc); Mississippi State Highway Comm'n v. Gresham, 323 So.2d 100, 103 (1975) (en banc); Powers v. Malley, 302 So.2d 262, 264 (1974). In Mississippi State Highway Comm'n v. Gresham, supra, the court expressly noted that its disposition depended upon a fact mentioned for the first time in the petition for rehearing. In several other decisions, the type of question considered on rehearing suggests that it was raised for the first time by the party petitioning for that relief. E.g., Cortez v. Brown, supra; City of Jackson v. Capital Reporter Publishing Co., supra; Powers v. Malley, supra. These decisions, however, do not expressly acknowledge the novelty of the points raised on rehearing. 15 Respondents also contend that our decisions establish a general rule against review of questions presented for the first time in a petition for rehearing. We have recognized that, under many circumstances, "[q]uestions first presented to the highest State court on a petition for rehearing come too late for consideration here." Radio Station WOW, Inc. v. Johnson, 326 U.S. 120, 128, 65 S.Ct. 1475, 1480, 89 L.Ed. 2092 (1945). At the same time, however, we have explained that this bar does not apply if "the State court exerted its jurisdiction in such a way that the case could have been brought here had the questions been raised prior to the original disposition." Ibid. In this case we conclude that the Mississippi Supreme Court's first judgment on appeal either decided the federal question on the merits, although in a manner that did not appear final, or avoided the federal question by invoking an inconsistently applied procedural rule. If petitioners had made their claim prior to the court's original disposition, either of these circumstances would have permitted us to review the federal question. 16 Mississippi plainly is one of the jurisdictions covered by the statute. South Carolina v. Katzenbach, 383 U.S. 301, 318, 86 S.Ct. 803, 813, 15 L.Ed.2d 769 (1966); 30 Fed.Reg. 9897 (1965). The Louisville School District Board of Trustees, like all political entities within the State, accordingly must comply with § 5's strictures. See Dougherty County Board of Education v. White, 439 U.S. 32, 46, 99 S.Ct. 368, 376, 58 L.Ed.2d 269 (1978); United States v. Board of Commissioners of Sheffield, 435 U.S. 110, 98 S.Ct. 965, 55 L.Ed.2d 148 (1978). It is immaterial that the change sought by respondents derives from a statute that predates the Voting Rights Act, because § 5 comes into play whenever a covered jurisdiction departs from an election procedure that was "in fact 'in force or effect' . . . on November 1, 1964." Perkins v. Matthews, 400 U.S. 379, 395, 91 S.Ct. 431, 440, 27 L.Ed.2d 476 (1971) (emphasis in original). Finally, the presence of a court decree does not exempt the contested change from § 5. We held only last Term that § 5 applies to any change "reflecting the policy choices of the elected representatives of the people," even if a judicial decree constrains those choices. McDaniel v. Sanchez, 452 U.S. 130, 153, 101 S.Ct. 2224, 2238, 68 L.Ed.2d 724 (1981). Although McDaniel involved a reapportionment plan drafted pursuant to a federal court's order, its interpretation of § 5 is equally instructive here. When state or local officials comply with a court order to enforce a state statute, there is no doubt that their actions "reflec[t] the policy choices of . . . elected representatives." Indeed, if § 5 did not encompass this situation, covered jurisdictions easily could evade the statute by declining to implement new state statutes until ordered to do so by state courts. Cf. McDaniel v. Sanchez, supra, at 151, 101 S.Ct., at 2237 (noting that "if covered jurisdictions could avoid the normal preclearance procedure by awaiting litigation challenging a refusal to redistrict after a census is completed, [§ 5] might have the unintended effect of actually encouraging delay in making obviously needed changes in district boundaries"). In light of McDaniel, we conclude that a state court decree directing compliance with a state election statute contemplates "administ[ration]" of the state statute within the meaning of § 5. 17 Respondents do not claim that Mississippi law restricts the state courts' power to decide questions related to § 5. 18 We frequently permit state courts to decide "collaterally" issues that would be reserved for the federal courts if the cause of action arose directly under federal law. For example, the state courts may decide a variety of questions involving the federal patent laws. American Well Works Co. v. Layne & Bowler Co., 241 U.S. 257, 36 S.Ct. 585, 60 L.Ed. 987 (1916); New Marshall Engine Co. v. Marshall Engine Co., 223 U.S. 473, 32 S.Ct. 238, 56 L.Ed. 513 (1912); Pratt v. Paris Gas Light & Coke Co., 168 U.S. 255, 18 S.Ct. 62, 42 L.Ed. 458 (1897). Similarly, although state courts lack jurisdiction to entertain suits brought pursuant to § 4 of the Clayton Act, 15 U.S.C. § 15, they often decide issues concerning the federal antitrust laws in other contexts. See, e.g., California Retail Liquor Dealers Assn. v. Midcal Aluminum, Inc., 445 U.S. 97, 100 S.Ct. 937, 63 L.Ed.2d 283 (1980); Bement v. National Harrow Co., 186 U.S. 70, 22 S.Ct. 747, 46 L.Ed. 1058 (1902), quoted with approval in Kaiser Steel Corp. v. Mullins, 455 U.S. 72, 81-82, n. 7, 101 S.Ct. 851, 858, n. 7, 70 L.Ed. 833 (1982). See generally Note, Exclusive Jurisdiction of the Federal Courts in Private Civil Actions, 70 Harv.L.Rev. 509, 510-511 (1957). 19 Section 12(d) authorizes preventive relief against persons "engaged or . . . about to engage in any act or practice prohibited by" designated sections of the Voting Rights Act. 79 Stat. 444, 42 U.S.C. § 1973j(d). 20 At least one state court has ruled that it lacks jurisdiction over claims arising under the Voting Rights Act. Ortiz v. Thompson, 604 S.W.2d 443 (Tex.Civ.App.1980). See also Beatty v. Esposito, 411 F.Supp. 107 (EDNY 1976) (finding that state court lacked jurisdiction to decide § 5 issue, without explaining whether state suit arose under the Voting Rights Act). 21 Respondents also based their suit on the Fourteenth Amendment. See n. 4, supra. 22 Neither the parties nor the United States, appearing as amicus curiae, has cited any legislative history bearing upon state court jurisdiction to decide issues arising under the Voting Rights Act. 23 As respondents point out, state court jurisdiction to decide these collateral issues is not absolutely necessary to effectuate the Act's scheme, because interested parties have the ability to seek relief from a federal district court. Recognition of a limited state power to address § 5 issues, however, furthers the Act's ameliorative purposes by permitting additional tribunals to enforce its commands. It also insures that the question of coverage will be addressed at the earliest possible time, without requiring duplicative lawsuits. We find little force in respondents' claim that, if the state courts possess jurisdiction to decide § 5 issues arising in disputes between private parties, they will frustrate the Attorney General's enforcement of the Act by interpreting the preclearance requirement conservatively. The Attorney General is not bound by the resolution of § 5 issues in cases to which he was not a party. City of Richmond v. United States, 422 U.S. 358, 373-374, n. 6, 95 S.Ct. 2296, 2305, 45 L.Ed.2d 245 (1975). Common notions of collateral estoppel suggest that the state proceedings similarly would not bind other interested persons who did not participate in them. See Restatement (Second) of Judgments § 68 (Tent. Draft No. 4, Apr. 15, 1977). Persons dissatisfied with a state court's collateral resolution of a § 5 issue in proceedings involving other parties, therefore, are likely to be able to litigate the issue anew in federal court. 24 Our holding does not prevent state courts from attempting to accommodate both state and federal interests. A state court, for example, might adopt the approach followed by the Chancery Court in this case, and order the parties to submit the proposed relief to the Attorney General. If the Attorney General registers an objection, the court might then order the parties to seek a declaratory judgment from the District Court for the District of Columbia. 25 For example, since the Attorney General objected only to the runoff procedure, the District Court simply might void the results of any runoff elections, permitting the candidates who gathered a plurality of votes in the general election to take those seats. We, of course, intimate no view on the best form of relief, leaving that matter to the District Court's discretion.

12

457 U.S. 176 102 S.Ct. 2374 72 L.Ed.2d 765 SUMITOMO SHOJI AMERICA, INC., Petitionerv.Lisa M. AVAGLIANO et al. Lisa M. AVAGLIANO, et al., Petitioners v. SUMITOMO SHOJI AMERICA, INC. Nos. 80-2070, 81-24. Argued April 26, 1982. Decided June 15, 1982. Syllabus Petitioner Sumitomo Shoji America, Inc., is a New York corporation and a wholly owned subsidiary of a Japanese general trading company. Past and present female secretarial employees of Sumitomo, who, with one exception, are United States citizens, brought a class action in Federal District Court against Sumitomo, claiming that its alleged practice of hiring only male Japanese citizens to fill executive, managerial, and sales positions violated Title VII of the Civil Rights Act of 1964. Sumitomo moved to dismiss the complaint on the ground that its practices were protected under Art. VIII(1) of the Friendship, Commerce and Navigation Treaty between the United States and Japan. Article VIII(1) provides that the "companies of either Party shall be permitted to engage, within the territories of the other Party, accountants and other technical experts, executive personnel, attorneys, agents and other specialists of their choice." Article XXII(3) of the Treaty defines "companies" as "[c]ompanies constituted under the applicable laws and regulations within the territories of either Party." The District Court refused to dismiss, holding that because Sumitomo was incorporated in the United States, it was not covered by Art. VIII(1), but the court then certified for interlocutory appeal to the Court of Appeals the question whether the terms of the Treaty exempted Sumitomo from Title VII's provisions. The Court of Appeals reversed in part, holding that Art. VIII(1) was intended to cover locally incorporated subsidiaries of foreign companies but that the Treaty language did not insulate Sumitomo's employment practices from Title VII scrutiny. Held: Sumitomo is not a company of Japan and thus is not covered by Art. VIII(1) of the Treaty. Pp. 180-189. (a) Under Art. XXII(3)'s literal language, Sumitomo is a company of the United States, since it was "constituted under the applicable laws and regulations" of New York. As a company of the United States, it cannot invoke the rights provided in Art. VIII(1), which are available only to companies of Japan operating in the United States and to companies of the United States operating in Japan. Where both parties to the Treaty agree with this meaning and such interpretation follows from the clear Treaty language, deference will be given to it, absent extraordinarily strong contrary evidence. Pp. 180-185. (b) Adherence to the Treaty language does not overlook the Treaty's purpose, since the primary purpose of the corporation provisions was to give corporations of each signatory legal status in the territory of the other party and to allow them to conduct business in the other country on a comparable basis with domestic firms. Pp. 185-189. 2d Cir., 638 F.2d 552, vacated and remanded. Abram Chayes, Washington, D. C., for Sumitomo Shoji America, Inc. Lewis M. Steel, New York City, for Avagliano, et al. Lawrence G. Wallace, Washington, D. C., for the United States as amicus curiae by special leave of Court. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to decide whether Article VIII(1) of the Friendship, Commerce and Navigation Treaty between the United States and Japan provides a defense to a Title VII employment discrimination suit against an American subsidiary of a Japanese company. 2 * Petitioner, Sumitomo Shoji America, Inc., is a New York corporation and a wholly owned subsidiary of Sumitomo Shoji Kabushiki Kaisha, a Japanese general trading company or sogo shosha.1 Respondents are past and present female secretarial employees of Sumitomo.2 All but one of the respondents are United States citizens; that one exception is a Japanese citizen living in the United States. Respondents brought this suit as a class action claiming that Sumitomo's alleged practice of hiring only male Japanese citizens to fill executive, managerial, and sales positions violated both 42 U.S.C. § 1981 and Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp.IV).3 Respondents sought both injunctive relief and damages. 3 Without admitting the alleged discriminatory practice, Sumitomo moved under Rule 12(b)(6) of the Federal Rules of Civil Procedure to dismiss the complaint. Sumitomo's motion was based on two grounds: (1) discrimination on the basis of Japanese citizenship does not violate Title VII or § 1981; and (2) Sumitomo's practices are protected under Article VIII(1) of the Friendship, Commerce and Navigation Treaty between the United States and Japan, Apr. 2, 1953, [1953] 4 U.S.T. 2063, T.I.A.S. No. 2863. The District Court dismissed the § 1981 claim, holding that neither sex discrimination nor national origin discrimination are cognizable under that section. 473 F.Supp. 506 (S.D.N.Y.1979). The court refused to dismiss the Title VII claims, however; it held that because Sumitomo is incorporated in the United States it is not covered by Article VIII(1) of the Treaty. The District Court then certified for interlocutory appeal to the Court of Appeals under 28 U.S.C. § 1292(b) the question of whether the terms of the Treaty exempted Sumitomo from the provisions of Title VII. 4 The Court of Appeals reversed in part. 638 F.2d 552 (C.A.2 1981). The court first examined the Treaty's language and its history and concluded that the Treaty parties intended Article VIII(1) to cover locally incorporated subsidiaries of foreign companies such as Sumitomo. The court then held that the Treaty language does not insulate Sumitomo's executive employment practices from Title VII scrutiny. The court concluded that under certain conditions, Japanese citizenship could be a bona fide occupational qualification for high-level employment with a Japanese-owned domestic corporation and that Sumitomo's practices might thus fit within a statutory exception to Title VII.4 The court remanded for further proceedings.5 5 We granted certiorari, 454 U.S. 962, 102 S.Ct. 501, 70 L.Ed.2d 377 (1981), and we vacate and remand. II 6 Interpretation of the Friendship, Commerce and Navigation Treaty between Japan and the United States must, of course, begin with the language of the Treaty itself. The clear import of treaty language controls unless "application of the words of the treaty according to their obvious meaning effects a result inconsistent with the intent or expectations of its signatories." Maximov v. United States, 373 U.S. 49, 54, 83 S.Ct. 1054, 1057, 10 L.Ed.2d 184 (1963). See also The Amiable Isabella, 6 Wheat. (10 U.S.) 1, 72, 5 L.Ed. 191 (1821). 7 Article VIII(1) of the Treaty provides in pertinent part: 8 "[C]ompanies of either Party shall be permitted to engage, within the territories of the other Party, accountants and other technical experts, executive personnel, attorneys, agents and other specialists of their choice." (Emphasis added.)6 9 Clearly Article VIII(1) only applies to companies of one of the Treaty countries operating in the other country. Sumitomo contends that it is a company of Japan, and that Article VIII(1) of the Treaty grants it very broad discretion to fill its executive, managerial, and sales positions exclusively with male Japanese citizens.7 10 Article VIII(1) does not define any of its terms; the definitional section of the Treaty is contained in Article XXII. Article XXII(3) provides: 11 "As used in the present Treaty, the term 'companies' means corporations, partnerships, companies and other associations, whether or not with limited liability and whether or not for pecuniary profit. Companies constituted under the applicable laws and regulations within the territories of either Party shall be deemed companies thereof and shall have their juridical status recognized within the territories of the other Party." (Emphasis added.) 12 Sumitomo is "constituted under the applicable laws and regulations" of New York; based on Article XXII(3), it is a company of the United States, not a company of Japan.8 As a company of the United States operating in the United States, under the literal language of Article XXII(3) of the Treaty, Sumitomo cannot invoke the rights provided in Article VIII(1), which are available only to companies of Japan operating in the United States and to companies of the United States operating in Japan. 13 The Governments of Japan and the United States support this interpretation of the Treaty. Both the Ministry of Foreign Affairs of Japan and the United States Department of State agree that a United States corporation, even when wholly owned by a Japanese company, is not a company of Japan under the Treaty and is therefore not covered by Article VIII(1). The Ministry of Foreign Affairs stated its position to the American Embassy in Tokyo with reference to this case: 14 "The Ministry of Foreign Affairs, as the Office of [the Government of Japan] responsible for the interpretation of the [Friendship, Commerce and Navigation] Treaty, reiterates its view concerning the application of Article 8, Paragraph 1 of the Treaty: For the purpose of the Treaty, companies constituted under the applicable laws . . . of either Party shall be deemed companies thereof and, therefore, a subsidiary of a Japanese company which is incorporated under the laws of New York is not covered by Article 8 Paragraph 1 when it operates in the United States."9 15 The United States Department of State also maintains that Article VIII(1) rights do not apply to locally incorporated subsidiaries.10 Although not conclusive, the meaning attributed to treaty provisions by the Government agencies charged with their negotiation and enforcement is entitled to great weight. Kolovrat v. Oregon, 366 U.S. 187, 194, 81 S.Ct. 922, 926, 6 L.Ed.2d 218 (1961).11 16 Our role is limited to giving effect to the intent of the Treaty parties. When the parties to a treaty both agree as to the meaning of a treaty provision, and that interpretation follows from the clear treaty language, we must, absent extraordinarily strong contrary evidence, defer to that interpretation.12 III 17 Sumitomo maintains that although the literal language of the Treaty supports the contrary interpretation, the intent of Japan and the United States was to cover subsidiaries regardless of their place of incorporation. We disagree. 18 Contrary to the view of the Court of Appeals and the claims of Sumitomo, adherence to the language of the Treaty would not "overlook the purpose of the Treaty." 638 F.2d, at 556. The Friendship, Commerce and Navigation Treaty between Japan and the United States is but one of a series of similar commercial agreements negotiated after World War II.13 The primary purpose of the corporation provisions of the Treaties was to give corporations of each signatory legal status in the territory of the other party, and to allow them to conduct business in the other country on a comparable basis with domestic firms. Although the United States negotiated commercial treaties as early as 1778, and thereafter throughout the 19th century and early 20th century,14 these early commercial treaties were primarily concerned with the trade and shipping rights of individuals. Until the 20th century, international commerce was much more an individual than a corporate affair.15 19 As corporate involvement in international trade expanded in this century, old commercial treaties became outmoded. Because "corporation[s] can have no legal existence out of the boundaries of the sovereignty by which [they are] created," Bank of Augusta v. Earle, 13 Pet. (38 U.S.) 519, 588, 10 L.Ed. 274 (1839), it became necessary to negotiate new treaties granting corporations legal status and the right to function abroad. A series of Treaties negotiated before World War II gave corporations legal status and access to foreign courts,16 but it was not until the postwar Friendship, Commerce and Navigation Treaties that United States corporations gained the right to conduct business in other countries.17 The purpose of the Treaties was not to give foreign corporations greater rights than domestic companies, but instead to assure them the right to conduct business on an equal basis without suffering discrimination based on their alienage. 20 The Treaties accomplished their purpose by granting foreign corporations "national treatment"18 in most respects and by allowing foreign individuals and companies to form locally incorporated subsidiaries. These local subsidiaries are considered for purposes of the Treaty to be companies of the country in which they are incorporated; they are entitled to the rights, and subject to the responsibilities of other domestic corporations. By treating these subsidiaries as domestic companies, the purpose of the Treaty provisions—to assure that corporations of one Treaty party have the right to conduct business within the territory of the other party without suffering discrimination as an alien entity—is fully met. 21 Nor can we agree with the Court of Appeals view that literal interpretation of the Treaty would create a "crazy-quilt pattern" in which the rights of branches of Japanese companies operating directly in the United States would be greatly superior to the right of locally incorporated subsidiaries of Japanese companies. 638 F.2d, at 556. The Court of Appeals maintained that if such subsidiaries were not considered companies of Japan under the Treaty, they, unlike branch offices of Japanese corporations, would be denied access to the legal system, would be left unprotected against unlawful entry and molestation, and would be unable to dispose of property, obtain patents, engage in importation and exportation, or make payments, remittances, and transfers of funds. Ibid. That this is not the case is obvious; the subsidiaries, as companies of the United States, would enjoy all of those rights and more. The only significant advantage branches may have over subsidiaries is that conferred by Article VIII(1). IV 22 We are persuaded, as both signatories agree, that under the literal language of Article XXII(3) of the Treaty, Sumitomo is a company of the United States; we discern no reason to depart from the plain meaning of the Treaty language. Accordingly, we hold that Sumitomo is not a company of Japan and is thus not covered by Article VIII(1) of the Treaty.19 The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. 23 Vacated and remanded. 1 General trading companies have been a unique fixture of the Japanese economy since the Meiji era. These companies each market large numbers of Japanese products, typically those of smaller concerns, and also have a large role in the importation of raw materials and manufactured products to Japan. In addition, the trading companies play a large part in financing Japan's international trade. The largest trading companies—including Sumitomo's parent company—in a typical year account for over 50% of Japanese exports and over 60% of imports to Japan. See Krause & Sekiguchi, Japan and the World Economy, in Asia's New Giant: How the Japanese Economy Works 383, 389-397 (H. Patrick & H. Rosovsky eds. 1976). 2 Respondents have also filed a cross-petition in this case. Thus, the past and present secretaries, generally referred to as respondents, are the respondents in No. 80-2070 and the cross-petitioners in No. 81-24. Sumitomo is the petitioner in No. 80-2070 and the cross-respondent in No. 81-24. 3 Prior to bringing this suit, respondents each filed timely complaints with the Equal Employment Opportunity Commission. The EEOC issued "right to sue" letters to the respondents on October 27, 1977. This suit was filed on November 21, 1977, well within the statutory 90-day period allowed for filing suits after receipt of an EEOC notice of right to sue. 42 U.S.C. § 2000e-5(f)(1). 4 Sumitomo argued in the District Court that discrimination on the basis of national citizenship, as opposed to national origin, was not prohibited by Title VII. The District Court disagreed, however. It relied on Espinoza v. Farah Manufacturing Co., 414 U.S. 86, 92, 94 S.Ct. 334, 338, 38 L.Ed.2d 287 (1973), in which we noted that "Title VII prohibits discrimination on the basis of citizenship whenever it has the purpose or effect of discriminating on the basis of national origin." Although discussed at length in the briefs, this issue is not properly before the Court and we do not reach it. It was not included in the question certified for interlocutory review by the Court of Appeals under 28 U.S.C. § 1292(b), was not decided by the Court of Appeals, and was not set forth or fairly included in the questions presented for review by this Court as required by Rule 21.1(a). 5 In a nearly identical case, a divided panel of the Court of Appeals for the Fifth Circuit came to somewhat contrary results. Spiess v. C. Itoh & Co., 643 F.2d 353 (1981), cert. pending, No. 81-1496. The Fifth Circuit majority agreed with the Second Circuit decision that a locally incorporated subsidiary of a Japanese corporation is covered by Article VIII(1) of the Treaty, but disagreed with the latter court's decision on the effect of the Treaty on Title VII. The court held that the Treaty provision did protect the subsidiary's practices from Title VII liability. In dissent, Judge Reavley disagreed with the majority's initial conclusion. He would have held that under the plain language of the Treaty, locally incorporated subsidiaries are to be considered domestic corporations and are thus not covered by Article VIII(1). 6 Similar provisions are contained in the Friendship, Commerce and Navigation Treaties between the United States and other countries. See, e.g., Article XII(4) of the Treaty with Greece, [1954] 5 U.S.T. 1829, 1857, T.I.A.S. No. 3057 (1951); Article VIII(1) of the Treaty with Israel, [1954] 5 U.S.T. 550, 557, T.I.A.S. No. 551 (1951); Article VIII(1) of the Treaty with the Federal Republic of Germany, [1956] 7 U.S.T. 1839, 1848, T.I.A.S. No. 3593 (1954). These provisions were apparently included at the insistence of the United States; in fact, other countries, including Japan, unsuccessfully fought for their deletion. See, e.g., State Department Airgram No. A-453, dated Jan. 7, 1952, pp. 1, 3, reprinted in App. 130a, 131a, 133a (discussing Japanese objections to Article VIII(1)); Foreign Service Despatch No. 2529, dated Mar. 18, 1954, reprinted in App. 181a, 182a (discussing German objections to Article VIII(1)). According to Herman Walker, Jr., who at the time of the drafting of the Treaty served as Adviser on Commercial Treaties at the State Department, Article VIII(1) and the comparable provisions of other treaties were intended to avoid the effect of strict percentile limitations on the employment of Americans abroad and "to prevent the imposition of ultranationalistic policies with respect to essential executive and technical personnel." Walker, Provisions on Companies in United States Commercial Treaties, 50 Am.J.Int'l L. 373, 386 (1956); Walker, Treaties for the Encouragement and Protection of Foreign Investment: Present United States Practice, 5 Am.J.Comp.L. 229, 234 (1956). According to the State Department, Mr. Walker was responsible for formulation of the postwar form of the Friendship, Commerce and Navigation Treaty and negotiated several of the treaties for the United States. Department of State Airgram A-105, dated Jan. 9, 1976, reprinted in App. 157a. See also Foreign Service Despatch No. 2529, supra, App. 182a (Purpose of Article VIII(1) of Treaty with Germany "is to preclude the imposition of 'percentile' legislation. It gives freedom of choice as among persons lawfully present in the country and occupationally qualified under the local law"). 7 The issues raised by this contention are clearly of widespread importance. As we noted in n. 6, supra, treaty provisions similar to that invoked by Sumitomo are in effect with many other countries. In fact, some treaties contain even more broad language. See, e.g., Article XII(4), Treaty of Friendship, Commerce and Navigation with Greece, [1954] 5 U.S.T., at 1857-1859 ("Nationals and companies of either party shall be permitted to engage, within the territories of the other Party, accountants and other technical experts, executive personnel, attorneys, agents and other employees of their choice. . .") (emphasis added). As of 1979, United States affiliates of foreign corporations employed over 1.6 million workers in this country. Howenstine, Selected Data on the Operations of U. S. Affiliates of Foreign Companies, 1978 and 1979, in Survey of Current Business 35, 36 (U.S. Dept. of Commerce, May 1981). 8 The clear language of Article VII(1) and Article XXII(3) is consistent with other Treaty provisions. For example, Article XVI(2) accords national treatment to "[a]rticles produced by nationals and companies of either Party within the territories of the other Party, or by companies of the latter Party controlled by such nationals and companies. . . ." (Emphasis added.) This provision obviously envisions that companies of one party may be controlled by companies of the other party. If the nationality of a company were determined by the nationality of its controlling entity as Sumitomo proposes, rather than by the place of its incorporation, this provision would make no sense. Several other Treaty provisions would make little sense if American subsidiaries were considered companies of Japan. Articles VII(1), VII(4), and XVI(2) contain clauses dealing with companies or enterprises controlled by companies of either party. If those companies or enterprises were themselves companies of the country of their parents, this separate treatment would be unwarranted. 9 State Department Cable, Tokyo 03300, dated Feb. 26, 1982 (cable from the United States Embassy in Tokyo to the Secretary of State relaying the position of the Ministry of Foreign Affairs of Japan). See also Diplomatic Communication from the Embassy of Japan in Washington to the United States Department of State, dated Apr. 21, 1982 ("The Government of Japan reconfirms its view that a subsidiary of a Japanese company which is incorporated under the laws of New York is not itself covered by article 8., paragraph 1 of the Treaty of Friendship, Commerce and Navigation between Japan and the United States (the FCN Treaty) when it operates in the United States"). 10 Brief for United States as Amicus Curiae 8-22; Letter of James R. Atwood, Deputy Legal Adviser, U.S. Department of State, to Lutz Alexander Prager, Assistant General Counsel, Equal Employment Opportunity Commission, dated Sept. 11, 1979, reprinted in App. 307a. ("On further reflection on the scope of application of the first sentence of Paragraph 1 of Article VIII of the U. S.-Japan FCN, we have established to our satisfaction that it was not the intent of the negotiators to cover locally-incorporated subsidiaries, and that therefore U. S. subsidiaries of Japanese corporations cannot avail themselves of this provision of the treaty"). The Court of Appeals and Sumitomo dismiss the Atwood letter as incorrect, and point to a letter written by a previous State Department Deputy Legal Adviser as taking the contrary view. Letter of Lee R. Marks, Deputy Legal Adviser, U.S. Department of State, to Abner W. Sibal, General Counsel, Equal Employment Opportunity Commission, dated Oct. 17, 1978, reprinted in App. 94a. However neither of these letters is indicative of the state of mind of the Treaty negotiators; they are merely evidence of the later interpretation of the State Department as the agency of the United States charged with interpreting and enforcing the Treaty. However ambiguous the State Department position may have been previously, it is certainly beyond dispute that the Department now interprets the Treaty in conformity with its plain language, and is of the opinion that Sumitomo is not a company of Japan and is not covered by Article VIII(1). That interpretation, and the identical position of the Government of Japan, is entitled to great weight. Kolovrat v. Oregon, 366 U.S. 187, 81 S.Ct. 922, 6 L.Ed.2d 218 (1961). 11 Determining the nationality of a company by its place of incorporation is consistent with prior treaty practice. See Walker, 50 Am.J.Int'l L., supra, n. 6, at 382-383. The place-of-incorporation rule also has the advantage of making determination of nationality a simple matter. On the other hand, application of a control test could certainly make nationality a subject of dispute. 12 We express no view, of course, as to the interpretation of other Friendship, Commerce and Navigation Treaties which, although similarly worded, may have different negotiating histories. 13 See, e.g., Treaties of Friendship, Commerce and Navigation with China, 63 Stat. 1299, T.I.A.S. No. 1871 (1946); Italy, 63 Stat. 2255, T.I.A.S. No. 1965 (1948); Israel, [1954] 5 U.S.T. 550, T.I.A.S. No. 551 (1951); Greece, [1954] 5 U.S.T. 1829, T.I.A.S. No. 3057 (1951); Japan, [1953] 4 U.S.T. 2063, T.I.A.S. No. 2863 (1953); Federal Republic of Germany, [1956] 7 U.S.T. 1839, T.I.A.S. No. 3593 (1954); The Netherlands, [1957] 8 U.S.T. 2043, T.I.A.S. No. 3942 (1956); and Pakistan, [1961] 12 U.S.T. 110, T.I.A.S. No. 4683 (1959). The provisions of several of the treaties are compared in tabular form in Commercial Treaties: Hearing on Treaties of Friendship, Commerce and Navigation with Israel, Ethiopia, Italy, Denmark, Greece, Finland, Germany, and Japan, before the Subcommittee of the Senate Committee on Foreign Relations, 83d Cong., 1st Sess., 7-17 (1953). 14 See, e.g., Treaty of Amity and Commerce with France, 8 Stat. 12, T.S. No. 83 (1778); Treaty of Amity, Commerce and Navigation with Great Britain, 8 Stat. 116, T.S. No. 105 (1794); Treaty of Commerce and Friendship with Sweden and Norway, 8 Stat. 232, T.S. No. 347 (1816); Treaty of Commerce and Navigation with the Netherlands, 8 Stat. 524, T.S. No. 251 (1839); Treaty of Commerce and Navigation with Belgium, 8 Stat. 606, T.S. No. 19 (1845); Treaty of Commerce and Navigation with Italy, 17 Stat. 845, T.S. No. 177 (1871); Treaty of Commerce with Spain, 23 Stat. 750, T.S. No. 337 (1884); Treaty of Commerce with Germany, 31 Stat. 1935, T.S. No. 101 (1900); Treaty of Commerce with China, 33 Stat. 2208, T.S. No. 430 (1903). 15 See Walker, 50 Am.J.Int'l L., supra n. 6, at 374-378. 16 Treaty of Commerce and Consular Rights with Japan, 37 Stat. 1504, T.S. No. 558 (1911); Treaties of Friendship, Commerce and Navigation with Germany, 44 Stat. 2132, T.S. No. 725 (1923); Estonia, 44 Stat. 2379, T.S. No. 736 (1925); Hungary, 44 Stat. 2441, T.S. No. 748 (1925); El Salvador, 46 Stat. 2817, T.S. No. 827 (1926); Honduras, 45 Stat. 2618, T.S. No. 764 (1927); Latvia, 45 Stat. 2641, T.S. No. 765 (1928); Austria, 47 Stat. 1876, T.S. No. 838 (1928); Norway, 47 Stat. 2135, T.S. No. 852 (1928); Poland, 48 Stat. 1507, T.S. No. 862 (1931); Finland, 49 Stat. 2659, T.S. No. 868 (1934); Treaties of Friendship, Commerce and Navigation with Siam, 53 Stat. 1731, T.S. No. 940 (1937); Liberia, 54 Stat. 1739, T.S. No. 956 (1938). These rights given to corporations by these Treaties were quite limited. For example, Article VII of the 1911 Treaty with Japan provided: "Limited liability and other companies and associations . . . already or hereafter to be organized in accordance with the laws of either High Contracting Party and domiciled in the territories of such Party, are authorized, in the territories of the other, to exercise their rights and appear in the courts either as plaintiffs or defendants, subject to the laws of such other Party. "The foregoing stipulation has no bearing upon the question whether a company or association organized in one of the two countries will or will not be permitted to transact its business or industry in the other, this permission remaining always subject to the laws and regulations enacted or established in the respective countries or in any part thereof." 37 Stat. 1506. A similarly limited provision was contained in the other Treaties. 17 The significance of this advance was emphasized in the Senate hearings on an early set of postwar Friendship, Commerce and Navigation Treaties: "Perhaps the most striking advance of the postwar treaties is the cognizance taken of the widespread use of the corporate form of business organization in present-day economic affairs. In the treaties antedating World War II American corporations were specifically assured only small protection against possible discriminatory treatment in foreign countries. In the postwar treaties, however, corporations are accorded essentially the same treaty rights as individuals in such vital matters as the right to do business, taxation on a nondiscriminatory basis, the acquisition and enjoyment of real and personal property, and the application of exchange controls. Furthermore, the citizens and corporations of one country are given substantial rights in connection with forming local subsidiaries under the corporation laws of the other country and controlling and managing the affairs of such local companies." Commercial Treaties: Hearing on Treaties of Friendship, Commerce and Navigation Between the United States and Colombia, Israel, Ethiopia, Italy, Denmark and Greece before a Subcommittee of the Senate Committee on Foreign Relations, 82d Cong., 2d Sess., 4-5 (1952) (opening statement of Harold Linder, Deputy Assistant Secretary of State for Economic Affairs). 18 "National treatment" is defined in Article XXII(1) of the Treaty: "The term 'national treatment' means treatment accorded within the territories of a Party upon terms no less favorable than the treatment accorded therein, in like situations, to nationals, companies, products, vessels or other objects, as the case may be, of such Party." In short, national treatment of corporations means equal treatment with domestic corporations. It is ordinarily the highest level of protection afforded by commercial treaties. In certain areas treaty parties are unwilling to grant full national treatment; in those areas the parties frequently grant "most-favored-nation treatment," which means treatment no less favorable than that accorded to nationals or companies of any third country. See Article XXII(2) of the Treaty. "The most-favored-nation rule can now, therefore, imply or allow the status of alien disability rather than of favor. In applicable situations nowadays, the first-class treatment tends to be national treatment; that which the citizens of the country enjoy." Walker, Modern Treaties of Friendship, Commerce and Navigation, 42 Minn.L.Rev. 805, 811 (1958). 19 We express no view as to whether Japanese citizenship may be a bona fide occupational qualification for certain positions at Sumitomo or as to whether a business necessity defense may be available. There can be little doubt that some positions in a Japanese controlled company doing business in the United States call for great familiarity with not only the language of Japan, but also the culture, customs, and business practices of that country. However, the Court of Appeals found the evidentiary record insufficient to determine whether Japanese citizenship was a bona fide occupational qualification for any of Sumitomo's positions within the reach of Article VIII(1). Nor did it discuss the bona fide occupational qualification exception in relation to respondents' sex discrimination claim or the possibility of a business necessity defense. Whether Sumitomo can support its assertion of a bona fide occupational qualification or a business necessity defense is not before us. See n. 4, supra. We also express no view as to whether Sumitomo may assert any Article VIII(1) rights of its parent.

12

72 L.Ed.2d 777 102 S.Ct. 2414 457 U.S. 191 Victor P. DIEDRICH et ux., Petitionersv.COMMISSIONER OF INTERNAL REVENUE. UNITED MISSOURI BANK OF KANSAS v. COMMISSIONER OF INTERNAL REVENUE. No. 80-2204. Argued Feb. 24, 1982. Decided June 15, 1982. Syllabus ¢s191¢s Held : A donor (such as petitioner husband and wife and petitioner executor's decedent) who makes a gift of property on condition that the donee pay the resulting gift taxes realizes taxable income to the extent that the gift taxes paid by the donee exceed the donor's adjusted basis in the property. Pp. 194-200. (a) The substance, not the form, of the agreed transaction controls in determining whether taxable income was realized. Old Colony Trust Co. v. Commissioner, 279 U.S. 716, 49 S.Ct. 499, 73 L.Ed. 918; Crane v. Commissioner, 331 U.S. 1, 67 S.Ct. 1047, 91 L.Ed. 1301. Pp. 2417-2418. (b) When a donor makes a gift, he incurs a "debt" to the United States for the amount of whatever gift taxes are due, which are as much the donor's legal obligation as his income taxes. When conditional gifts, such as those in question here, are made, the donor realizes an immediate economic benefit by the donee's assumption of the donor's legal obligation to pay the gift taxes. Subjective intent, while relevant in determining whether a gift has been made, is not characteristically a factor in determining whether an individual has realized income. Even if intent were a factor, the donor's intent as to the condition shifting the gift tax obligation to the donee is plainly to relieve the donor of the debt owed to the United States. And the economic benefit realized by the donor is not diminished by the fact that the liability attaches during the course of the donative transfer, such benefit being indistinguishable from the benefit arising from discharge of a pre-existing obligation. Pp. 196-198. (c) Treating the amount by which the gift taxes exceed the donor's adjusted basis in the property as income is consistent with § 1001 of the Internal Revenue Code, which provides that the gain from the disposition of property is the excess of the amount realized over the transferor's adjusted basis in the property. Pp. 198-199. 8th Cir., 643 F.2d 499, affirmed. Norman E. Beal, Kansas City, Mo., for petitioners. Stuart A. Smith, Washington, D. C., for respondent. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to resolve a Circuit conflict as to whether a donor who makes a gift of property on condition that the donee pay the resulting gift tax receives taxable income to the extent that the gift tax paid by the donee exceeds the donor's adjusted basis in the property transferred. 454 U.S. 813, 102 S.Ct. 89, 70 L.Ed.2d 82 (1981). The United States Court of Appeals for the Eighth Circuit held that the donor realized income. 643 F.2d 499 (1981). We affirm. 2 * A. 3 Diedrich v. Commissioner of Internal Revenue 4 In 1972 petitioners Victor and Frances Diedrich made gifts of approximately 85,000 shares of stock to their three children, using both a direct transfer and a trust arrangement. The gifts were subject to a condition that the donees pay the resulting federal and state gift taxes. There is no dispute concerning the amount of the gift tax paid by the donees. The donors' basis in the transferred stock was $51,073; the gift tax paid in 1972 by the donees was $62,992. Petitioners did not include as income on their 1972 federal income tax returns any portion of the gift tax paid by the donees. After an audit the Commissioner of Internal Revenue determined that petitioners had realized income to the extent that the gift tax owed by petitioners but paid by the donees exceeded the donors' basis in the property. Accordingly, petitioners' taxable income for 1972 was increased by $5,959.1 Petitioners filed a petition in the United States Tax Court for redetermination of the deficiencies. The Tax Court held for the taxpayers, concluding that no income had been realized. 39 TCM 433 (1979). B 5 United Missouri Bank of Kansas City v. Commissioner of Internal Revenue 6 In 1970 and 1971 Mrs. Frances Grant gave 90,000 voting trust certificates to her son on condition that he pay the resulting gift tax. Mrs. Grant's basis in the stock was $8,742.60; the gift tax paid by the donee was $232,620.09. As in Diedrich, there is no dispute concerning the amount of the gift tax or the fact of its payment by the donee pursuant to the condition. 7 Like the Diedrichs, Mrs. Grant did not include as income on her 1970 or 1971 federal income tax returns any portion of the amount of the gift tax owed by her but paid by the donee. After auditing her returns, the Commissioner determined that the gift of stock to her son was part gift and part sale, with the result that Mrs. Grant realized income to the extent that the amount of the gift tax exceeded the adjusted basis in the property. Accordingly, Mrs. Grant's taxable income was increased by approximately $112,000.2 Mrs. Grant filed a petition in the United States Tax Court for redetermination of the deficiencies. The Tax Court held for the taxpayer, concluding that no income had been realized. Grant v. Commissioner, 39 TCM 1088 (1980). C 8 The United States Court of Appeals for the Eighth Circuit consolidated the two appeals and reversed, concluding that "to the extent the gift taxes paid by donees" exceeded the donors' adjusted bases in the property transferred, "the donors realized taxable income." 643 F.2d, at 504. The Court of Appeals rejected the Tax Court's conclusion that the taxpayers merely had made a "net gift" of the difference between the fair market value of the transferred property and the gift taxes paid by the donees. The court reasoned that a donor receives a benefit when a donee discharges a donor's legal obligation to pay gift taxes. The Court of Appeals agreed with the Commissioner in rejecting the holding in Turner v. Commissioner, 49 T.C. 356 (1968), aff'd per curiam, 410 F.2d 752 (CA6 1969), and its progeny, and adopted the approach of Johnson v. Commissioner, 59 T.C. 791 (1973), aff'd, 495 F.2d 1079 (CA6), cert. denied, 419 U.S. 1040, 95 S.Ct. 527, 42 L.Ed.2d 317 (1974), and Estate of Levine v. Commissioner, 72 T.C. 780 (1979), aff'd, 634 F.2d 12 (CA2 1980). We granted certiorari to resolve this conflict, 454 U.S. 813, 102 S.Ct. 89, 70 L.Ed.2d 82 (1981), and we affirm. II A. 9 Pursuant to its constitutional authority, Congress has defined "gross income" as income "from whatever source derived," including "[i]ncome from discharge of indebtedness." 26 U.S.C. § 61(12).3 This Court has recognized that "income" may be realized by a variety of indirect means. In Old Colony Trust Co. v. Commissioner, 279 U.S. 716, 49 S.Ct. 499, 73 L.Ed. 918 (1929), the Court held that payment of an employee's income taxes by an employer constituted income to the employee. Speaking for the Court, Chief Justice Taft concluded that "[t]he payment of the tax by the employe[r] was in consideration of the services rendered by the employee and was a gain derived by the employee from his labor." Id., at 729, 49 S.Ct. at 504. The Court made clear that the substance, not the form, of the agreed transaction controls. "The discharge by a third person of an obligation to him is equivalent to receipt by the person taxed." Ibid. The employee, in other words, was placed in a better position as a result of the employer's discharge of the employee's legal obligation to pay the income taxes; the employee thus received a gain subject to income tax. 10 The holding in Old Colony was reaffirmed in Crane v. Commissioner, 331 U.S. 1, 67 S.Ct. 1047, 91 L.Ed. 1301 (1947). In Crane the Court concluded that relief from the obligation of a nonrecourse mortgage in which the value of the property exceeded the value of the mortgage constituted income to the taxpayer. The taxpayer in Crane acquired depreciable property, an apartment building, subject to an unassumed mortgage. The taxpayer later sold the apartment building, which was still subject to the nonrecourse mortgage, for cash plus the buyer's assumption of the mortgage. This Court held that the amount of the mortgage was properly included in the amount realized on the sale, noting that if the taxpayer transfers subject to the mortgage, 11 "the benefit to him is as real and substantial as if the mortgage were discharged, or as if a personal debt in an equal amount had been assumed by another." Id., at 14, 67 S.Ct. at 1054.4 12 Again, it was the "reality," not the form, of the transaction that governed. Ibid. The Court found it immaterial whether the seller received money prior to the sale in order to discharge the mortgage, or whether the seller merely transferred the property subject to the mortgage. In either case the taxpayer realized an economic benefit. B 13 The principles of Old Colony and Crane control.5 A common method of structuring gift transactions is for the donor to make the gift subject to the condition that the donee pay the resulting gift tax, as was done in each of the cases now before us. When a gift is made, the gift tax liability falls on the donor under 26 U.S.C. § 2502(d).6 When a donor makes a gift to a donee, a "debt" to the United States for the amount of the gift tax is incurred by the donor. Those taxes are as much the legal obligation of the donor as the donor's income taxes; for these purposes they are the same kind of debt obligation as the income taxes of the employee in Old Colony, supra. Similarly, when a donee agrees to discharge an indebtedness in consideration of the gift, the person relieved of the tax liability realizes an economic benefit. In short, the donor realizes an immediate economic benefit by the donee's assumption of the donor's legal obligation to pay the gift tax. 14 An examination of the donor's intent does not change the character of this benefit. Although intent is relevant in determining whether a gift has been made, subjective intent has not characteristically been a factor in determining whether an individual has realized income.7 Even if intent were a factor, the donor's intent with respect to the condition shifting the gift tax obligation from the donor to the donee was plainly to relieve the donor of a debt owed to the United States; the choice was made because the donor would receive a benefit in relief from the obligation to pay the gift tax.8 15 Finally, the benefit realized by the taxpayer is not diminished by the fact that the liability attaches during the course of a donative transfer. It cannot be doubted that the donors were aware that the gift tax obligation would arise immediately upon the transfer of the property; the economic benefit to the donors in the discharge of the gift tax liability is indistinguishable from the benefit arising from discharge of a preexisting obligation. Nor is there any doubt that had the donors sold a portion of the stock immediately before the gift transfer in order to raise funds to pay the expected gift tax, a taxable gain would have been realized. 26 U.S.C. § 1001. The fact that the gift tax obligation was discharged by way of a conditional gift rather than from funds derived from a pregift sale does not alter the underlying benefit to the donors. C 16 Consistent with the economic reality, the Commissioner has treated these conditional gifts as a discharge of indebtedness through a part gift and part sale of the gift property transferred. The transfer is treated as if the donor sells the property to the donee for less than the fair market value. The "sale" price is the amount necessary to discharge the gift tax indebtedness; the balance of the value of the transferred property is treated as a gift. The gain thus derived by the donor is the amount of the gift tax liability less the donor's adjusted basis in the entire property. Accordingly, income is realized to the extent that the gift tax exceeds the donor's adjusted basis in the property. This treatment is consistent with § 1001 of the Internal Revenue Code, which provides that the gain from the disposition of property is the excess of the amount realized over the transferor's adjusted basis in the property.9 III 17 We recognize that Congress has structured gift transactions to encourage transfer of property by limiting the tax consequences of a transfer. See,e.g., 26 U.S.C. § 102 (gifts excluded from donee's gross income). Congress may obviously provide a similar exclusion for the conditional gift. Should Congress wish to encourage "net gifts," changes in the income tax consequences of such gifts lie within the legislative responsibility. Until such time, we are bound by Congress' mandate that gross income includes income "from whatever source derived." We therefore hold that a donor who makes a gift of property on condition that the donee pay the resulting gift taxes realizes taxable income to the extent that the gift taxes paid by the donee exceed the donor's adjusted basis in the property.10 18 The judgment of the United States Court of Appeals for the Eighth Circuit is 19 Affirmed. 20 Justice REHNQUIST, dissenting. 21 It is a well-settled principle today that a taxpayer realizes income when another person relieves the taxpayer of a legal obligation in connection with an otherwise taxable transaction. See Crane v. Commissioner, 331 U.S. 1, 67 S.Ct. 1047, 91 L.Ed. 1301 (1947) (sale of real property); Old Colony Trust Co. v. Commissioner, 279 U.S. 716, 49 S.Ct. 499, 73 L.Ed. 918 (1929) (employment compensation). In neither Old Colony nor Crane was there any question as to the existence of a taxable transaction; the only question concerned the amount of income realized by the taxpayer as a result of the taxable transaction. The Court in this case, however, begs the question of whether a taxable transaction has taken place at all when it concludes that "[t]he principles of Old Colony and Crane control" this case. Ante, at 196. 22 In Old Colony, the employer agreed to pay the employee's federal tax liability as part of his compensation. The employee provided his services to the employer in exchange for compensation. The exchange of compensation for services was undeniably a taxable transaction. The only question was whether the employee's taxable income included the employer's assumption of the employee's income tax liability. 23 In Crane, the taxpayer sold real property for cash plus the buyer's assumption of a mortgage. Clearly a sale had occurred, and the only question was whether the amount of the mortgage assumed by the buyer should be included in the amount realized by the taxpayer. The Court rejected the taxpayer's contention that what she sold was not the property itself, but her equity in that property. 24 Unlike Old Colony or Crane, the question in this case is not the amount of income the taxpayer has realized as a result of a concededly taxable transaction, but whether a taxable transaction has taken place at all. Only after one concludes that a partial sale occurs when the donee agrees to pay the gift tax do Old Colony and Crane become relevant in ascertaining the amount of income realized by the donor as a result of the transaction. Nowhere does the Court explain why a gift becomes a partial sale merely because the donor and donee structure the gift so that the gift tax imposed by Congress on the transaction is paid by the donee rather than the donor. 25 In my view, the resolution of this case turns upon congressional intent: whether Congress intended to characterize a gift as a partial sale whenever the donee agrees to pay the gift tax. Congress has determined that a gift should not be considered income to the donee. 26 U.S.C. § 102. Instead, gift transactions are to be subject to a tax system wholly separate and distinct from the income tax. See 26 U.S.C. § 2501 et seq. Both the donor and the donee may be held liable for the gift tax. §§ 2502(d), 6324(b). Although the primary liability for the gift tax is on the donor, the donee is liable to the extent of the value of the gift should the donor fail to pay the tax. I see no evidence in the tax statutes that Congress forbade the parties to agree among themselves as to who would pay the gift tax upon pain of such an agreement being considered a taxable event for the purposes of the income tax. Although Congress could certainly determine that the payment of the gift tax by the donee constitutes income to the donor, the relevant statutes do not affirmatively indicate that Congress has made such a determination. 26 I dissent. 1 Subtracting the stock basis of $51,073 from the gift tax paid by the donees of $62,992, the Commissioner found that petitioners had realized a long-term capital gain of $11,919. After a 50% reduction in long-term capital gain, 26 U.S.C. § 1202, the Diedrichs' taxable income increased by $5,959. 2 The gift taxes were $232,630.09. Subtracting the adjusted basis of $8,742.60, the Commissioner found that Mrs. Grant realized a long-term capital gain of $223,887.49. After a 50% reduction for long-term capital gain, 26 U.S.C. § 1202, Mrs. Grant's taxable income increased by $111,943.75. During pendency of this lawsuit, Mrs. Grant died and the United Missouri Bank of Kansas City, the decedent's executor, was substituted as petitioner. 3 The United States Constitution provides that Congress shall have the power to lay and collect taxes on income "from whatever source derived." Art. I, § 8, cl. 1; Amend. 16. In Helvering v. Bruun, 309 U.S. 461, 469, 60 S.Ct. 631, 634, 84 L.Ed. 864 (1940), the Court noted: "While it is true that economic gain is not always taxable as income, it is settled that the realization of gain need not be in cash derived from the sale of an asset. Gain may occur as a result of exchange of property, payment of the taxpayer's indebtedness, relief from a liability, or other profit realized from the completion of a transaction." (Emphasis supplied.) 4 In Crane the taxpayer received favorable tax treatment for the loan and was allowed depreciation on the property. The Court concluded that the taxpayer could not then later escape taxation after having received these benefits when the loan obligation was assumed by another. Whether income would have been realized in Crane if the value of the property at the time of transfer had been less than the amount of the mortgage need not be considered here. See Crane, 331 U.S., at 14, n. 37, 67 S.Ct., at 1054, n. 37. 5 Although the Commissioner has argued consistently that payment of gift taxes by the donee results in income to the donor, several courts have rejected this interpretation. See, e.g., Turner v. Commissioner, 49 T.C. 356 (1968), aff'd per curiam, 410 F.2d 752 (CA6 1969); Hirst v. Commissioner, 572 F.2d 427 (CA4 1978) (en banc). Cf. Johnson v. Commissioner, 495 F.2d 1079 (CA6), cert. denied, 419 U.S. 1040, 95 S.Ct. 527, 42 L.Ed.2d 317 (1974). It should be noted that the gift tax consequences of a conditional gift will be unaffected by the holding in this case. When a conditional "net" gift is given, the gift tax attributable to the transfer is to be deducted from the value of the property in determining the value of the gift at the time of transfer. See Rev.Rul. 75-72, 1975-1 Cum.Bull. 310 (general formula for computation of gift tax on conditional gift); Rev.Rul. 71-232, 1971-1 Cum.Bull. 275. 6 "The tax imposed by section 2501 shall be paid by the donor." Section 6321 imposes a lien on the personal property of the donor when a tax is not paid when due. The donee is secondarily responsible for payment of the gift tax should the donor fail to pay the tax. 26 U.S.C. § 6324(b). The donee's liability, however, is limited to the value of the gift. Ibid. This responsibility of the donee is analogous to a lien or security. Ibid. See also S.Rep.No. 665, 72d Cong., 1st Sess., 42 (1932); H.R.Rep.No. 708, 72d Cong., 1st Sess., 30 (1932). 7 Several courts have found it highly significant that the donor intended to make a gift. Turner v. Commissioner, supra; Hirst v. Commissioner, supra. It is not enough, however, to state that the donor intended simply to make a gift of the amount which will remain after the donee pays the gift tax. As noted above, subjective intent has not characteristically been a factor in determining whether an individual has realized income. In Commissioner v. Duberstein, 363 U.S. 278, 286, 80 S.Ct. 1190, 1197, 4 L.Ed.2d 1218 (1960), the Court noted that "the donor's characterization of his action is not determinative." See also Minnesota Tea Co. v. Helvering, 302 U.S. 609, 613, 58 S.Ct. 393, 394, 82 L.Ed. 474 (1938) ("A given result at the end of a straight path is not made a different result because reached by following a devious path"). 8 The existence of the "condition" that the gift will be made only if the donee assumes the gift tax consequences precludes any characterization that the payment of the taxes was simply a gift from the donee back to the donor. A conditional gift not only relieves the donor of the gift tax liability, but also may enable the donor to transfer a larger sum of money to the donee than would otherwise be possible due to such factors as differing income tax brackets of the donor and donee. 9 Section 1001 provides: "(a) Computation of gain or loss.—The gain from the sale or other disposition of property shall be the excess of the amount realized therefrom over the adjusted basis provided in section 1011 for determining gain, and the loss shall be the excess of the adjusted basis provided in such section for determining loss over the amount realized. "(b) Amount realized.—The amount realized from the sale or other disposition of property shall be the sum of any money received plus the fair market value of the property (other than money) received. . . . "By treating conditional gifts as a part gift and part sale, income is realized only when highly appreciated property is transferred, for only highly appreciated property will result in a gift tax greater than the adjusted basis." 10 Petitioners argue that even if this Court holds that a donor realizes income on a conditional gift to the extent that the gift tax exceeds the adjusted basis, that holding should be applied prospectively and should not apply to the taxpayers in this case. In this case, however, there was no dispositive Eighth Circuit holding prior to the decision on review. In addition, this Court frequently has applied decisions which have altered the tax law and applied the clarified law to the facts of the case before it. See, e.g., United States v. Estate of Donnelly, 397 U.S. 286, 294-295, 90 S.Ct. 1033, 1038, 25 L.Ed.2d 312 (1970).

1112

457 U.S. 202 102 S.Ct. 2382 72 L.Ed.2d 786 James PLYLER, Superintendent of the Tyler Independent School District and Its Board of Trustees et al., Appellants,v.J. and R. DOE et al. TEXAS, et al., Appellants, v. CERTAIN NAMED AND UNNAMED UNDOCUMENTED ALIEN CHILDREN et al. Nos. 80-1538, 80-1934. Argued Dec. 1, 1981. Decided June 15, 1982. Rehearings Denied Sept. 9, 1982. See 458 U.S. 1131, 103 S.Ct. 14. Syllabus Held : A Texas statute which withholds from local school districts any state funds for the education of children who were not "legally admitted" into the United States, and which authorizes local school districts to deny enrollment to such children, violates the Equal Protection Clause of the Fourteenth Amendment. Pp. 210-230. (a) The illegal aliens who are plaintiffs in these cases challenging the statute may claim the benefit of the Equal Protection Clause, which provides that no State shall "deny to any person within its jurisdiction the equal protection of the laws." Whatever his status under the immigration laws, an alien is a "person" in any ordinary sense of that term. This Court's prior cases recognizing that illegal aliens are "persons" protected by the Due Process Clauses of the Fifth and Fourteenth Amendments, which Clauses do not include the phrase "within its jurisdiction," cannot be distinguished on the asserted ground that persons who have entered the country illegally are not "within the jurisdiction" of a State even if they are present within its boundaries and subject to its laws. Nor do the logic and history of the Fourteenth Amendment support such a construction. Instead, use of the phrase "within its jurisdiction" confirms the understanding that the Fourteenth Amendment's protection extends to anyone, citizen or stranger, who is subject to the laws of a State, and reaches into every corner of a State's territory. Pp. 210-216. (b) The discrimination contained in the Texas statute cannot be considered rational unless it furthers some substantial goal of the State. Although undocumented resident aliens cannot be treated as a "suspect class," and although education is not a "fundamental right," so as to require the State to justify the statutory classification by showing that it serves a compelling governmental interest, nevertheless the Texas statute imposes a lifetime hardship on a discrete class of children not accountable for their disabling status. These children can neither affect their parents' conduct nor their own undocumented status. The deprivation of public education is not like the deprivation of some other governmental benefit. Public education has a pivotal role in maintaining the fabric of our society and in sustaining our political and cultural heritage: the deprivation of education takes an inestimable toll on the social, economic, intellectual, and psychological well-being of the individual, and poses an obstacle to individual achievement. In determining the rationality of the Texas statute, its costs to the Nation and to the innocent children may properly be considered. Pp. 216-224. (c) The undocumented status of these children vel non does not establish a sufficient rational basis for denying them benefits that the State affords other residents. It is true that when faced with an equal protection challenge respecting a State's differential treatment of aliens, the courts must be attentive to congressional policy concerning aliens. But in the area of special constitutional sensitivity presented by these cases, and in the absence of any contrary indication fairly discernible in the legislative record, no national policy is perceived that might justify the State in denying these children an elementary education. Pp. 224-226. (d) Texas' statutory classification cannot be sustained as furthering its interest in the "preservation of the state's limited resources for the education of its lawful residents." While the State might have an interest in mitigating potentially harsh economic effects from an influx of legal immigrants, the Texas statute does not offer an effective method of dealing with the problem. Even assuming that the net impact of illegal aliens on the economy is negative, charging tuition to undocumented children constitutes an ineffectual attempt to stem the tide of illegal immigration, at least when compared with the alternative of prohibiting employment of illegal aliens. Nor is there any merit to the suggestion that undocumented children are appropriately singled out for exclusion because of the special burdens they impose on the State's ability to provide high-quality public education. The record does not show that exclusion of undocumented children is likely to improve the overall quality of education in the State. Neither is there any merit to the claim that undocumented children are appropriately singled out because their unlawful presence within the United States renders them less likely than other children to remain within the State's boundaries and to put their education to productive social or political use within the State. Pp. 2400-2402. No. 80-1538, 5th Cir., 628 F.2d 448, and No. 80-1934, affirmed. Richard L. Arnett, Austin, Tex., for State of Tex. et al. John C. Hardy, Tyler, Tex., for James L. Plyler et al. Peter A. Schey, Los Angeles, Cal., for appellees in No. 80-1934. Peter D. Roos, San Francisco, Cal., for appellees in 80-1538. [Amicus Curiae Information from pages 204-205 intentionally omitted] Justice BRENNAN delivered the opinion of the Court. 1 The question presented by these cases is whether, consistent with the Equal Protection Clause of the Fourteenth Amendment, Texas may deny to undocumented school-age children the free public education that it provides to children who are citizens of the United States or legally admitted aliens. 2 * Since the late 19th century, the United States has restricted immigration into this country. Unsanctioned entry into the United States is a crime, 8 U.S.C. § 1325, and those who have entered unlawfully are subject to deportation, 8 U.S.C. §§ 1251, 1252 (1976 ed. and Supp.IV). But despite the existence of these legal restrictions, a substantial number of persons have succeeded in unlawfully entering the United States, and now live within various States, including the State of Texas. 3 In May 1975, the Texas Legislature revised its education laws to withhold from local school districts any state funds for the education of children who were not "legally admitted" into the United States. The 1975 revision also authorized local school districts to deny enrollment in their public schools to children not "legally admitted" to the country. Tex.Educ.Code Ann. § 21.031 (Vernon Supp.1981).1 These cases involve constitutional challenges to those provisions. No. 80-1538 Plyler v. Doe 4 This is a class action, filed in the United States District Court for the Eastern District of Texas in September 1977, on behalf of certain school-age children of Mexican origin residing in Smith County, Tex., who could not establish that they had been legally admitted into the United States. The action complained of the exclusion of plaintiff children from the public schools of the Tyler Independent School District.2 The Superintendent and members of the Board of Trustees of the School District were named as defendants; the State of Texas intervened as a party-defendant. After certifying a class consisting of all undocumented school-age children of Mexican origin residing within the School District, the District Court preliminarily enjoined defendants from denying a free education to members of the plaintiff class. In December 1977, the court conducted an extensive hearing on plaintiffs' motion for permanent injunctive relief. 5 In considering this motion, the District Court made extensive findings of fact. The court found that neither § 21.031 nor the School District policy implementing it had "either the purpose or effect of keeping illegal aliens out of the State of Texas." 458 F.Supp. 569, 575 (1978). Respecting defendants' further claim that § 21.031 was simply a financial measure designed to avoid a drain on the State's fisc, the court recognized that the increases in population resulting from the immigration of Mexican nationals into the United States had created problems for the public schools of the State, and that these problems were exacerbated by the special educational needs of immigrant Mexican children. The court noted, however, that the increase in school enrollment was primarily attributable to the admission of children who were legal residents. Id., at 575-576. It also found that while the "exclusion of all undocumented children from the public schools in Texas would eventually result in economies at some level," id., at 576, funding from both the State and Federal Governments was based primarily on the number of children enrolled. In net effect then, barring undocumented children from the schools would save money, but it would "not necessarily" improve "the quality of education." Id., at 577. The court further observed that the impact of § 21.031 was borne primarily by a very small subclass of illegal aliens, "entire families who have migrated illegally and—for all practical purposes—permanently to the United States." Id., at 578.3 Finally, the court noted that under current laws and practices "the illegal alien of today may well be the legal alien of tomorrow,"4 and that without an education, these undocumented children, "[a]lready disadvantaged as a result of poverty, lack of English-speaking ability, and undeniable racial prejudices, . . . will become permanently locked into the lowest socio-economic class." Id., at 577. 6 The District Court held that illegal aliens were entitled to the protection of the Equal Protection Clause of the Fourteenth Amendment, and that § 21.031 violated that Clause. Suggesting that "the state's exclusion of undocumented children from its public schools . . . may well be the type of invidiously motivated state action for which the suspect classification doctrine was designed," the court held that it was unnecessary to decide whether the statute would survive a "strict scrutiny" analysis because, in any event, the discrimination embodied in the statute was not supported by a rational basis. Id., at 585. The District Court also concluded that the Texas statute violated the Supremacy Clause.5 Id., at 590-592. 7 The Court of Appeals for the Fifth Circuit upheld the District Court's injunction. 628 F.2d 448 (1980). The Court of Appeals held that the District Court had erred in finding the Texas statute pre-empted by federal law.6 With respect to equal protection, however, the Court of Appeals affirmed in all essential respects the analysis of the District Court, id., at 454-458, concluding that § 21.031 was "constitutionally infirm regardless of whether it was tested using the mere rational basis standard or some more stringent test," id., at 458. We noted probable jurisdiction. 451 U.S. 968, 101 S.Ct. 2044, 68 L.Ed.2d 347 (1981). No. 80-1934 In re Alien Children Education Litigation 8 During 1978 and 1979, suits challenging the constitutionality of § 21.031 and various local practices undertaken on the authority of that provision were filed in the United States District Courts for the Southern, Western, and Northern Districts of Texas. Each suit named the State of Texas and the Texas Education Agency as defendants, along with local officials. In November 1979, the Judicial Panel on Multidistrict Litigation, on motion of the State, consolidated the claims against the state officials into a single action to be heard in the District Court for the Southern District of Texas. A hearing was conducted in February and March 1980. In July 1980, the court entered an opinion and order holding that § 21.031 violated the Equal Protection Clause of the Fourteenth Amendment. In re Alien Children Education Litigation, 501 F.Supp. 544.7 The court held that "the absolute deprivation of education should trigger strict judicial scrutiny, particularly when the absolute deprivation is the result of complete inability to pay for the desired benefit." Id., at 582. The court determined that the State's concern for fiscal integrity was not a compelling state interest, id., at 582-583; that exclusion of these children had not been shown to be necessary to improve education within the State, id., at 583; and that the educational needs of the children statutorily excluded were not different from the needs of children not excluded, ibid. The court therefore concluded that § 21.031 was not carefully tailored to advance the asserted state interest in an acceptable manner. Id., at 583-584. While appeal of the District Court's decision was pending, the Court of Appeals rendered its decision in No. 80-1538. Apparently on the strength of that opinion, the Court of Appeals, on February 23, 1981, summarily affirmed the decision of the Southern District. We noted probable jurisdiction, 452 U.S. 937, 101 S.Ct. 3078, 69 L.Ed.2d 950 (1981), and consolidated this case with No. 80-1538 for briefing and argument.8 II 9 The Fourteenth Amendment provides that "[n]o State shall . . . deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws." (Emphasis added.) Appellants argue at the outset that undocumented aliens, because of their immigration status, are not "persons within the jurisdiction" of the State of Texas, and that they therefore have no right to the equal protection of Texas law. We reject this argument. Whatever his status under the immigration laws, an alien is surely a "person" in any ordinary sense of that term. Aliens, even aliens whose presence in this country is unlawful, have long been recognized as "persons" guaranteed due process of law by the Fifth and Fourteenth Amendments. Shaughnessy v. Mezei, 345 U.S. 206, 212, 73 S.Ct. 625, 629, 97 L.Ed. 956 (1953); Wong Wing v. United States, 163 U.S. 228, 238, 16 S.Ct. 977, 981, 41 L.Ed. 140 (1896); Yick Wo v. Hopkins, 118 U.S. 356, 369, 6 S.Ct. 1064, 1070, 30 L.Ed. 220 (1886). Indeed, we have clearly held that the Fifth Amendment protects aliens whose presence in this country is unlawful from invidious discrimination by the Federal Government. Mathews v. Diaz, 426 U.S. 67, 77, 96 S.Ct. 1883, 1890, 48 L.Ed.2d 478 (1976).9 10 Appellants seek to distinguish our prior cases, emphasizing that the Equal Protection Clause directs a State to afford its protection to persons within its jurisdiction while the Due Process Clauses of the Fifth and Fourteenth Amendments contain no such assertedly limiting phrase. In appellants' view, persons who have entered the United States illegally are not "within the jurisdiction" of a State even if they are present within a State's boundaries and subject to its laws. Neither our cases nor the logic of the Fourteenth Amendment supports that constricting construction of the phrase "within its jurisdiction."10 We have never suggested that the class of persons who might avail themselves of the equal protection guarantee is less than coextensive with that entitled to due process. To the contrary, we have recognized that both provisions were fashioned to protect an identical class of persons, and to reach every exercise of state authority. 11 "The Fourteenth Amendment to the Constitution is not confined to the protection of citizens. It says: 'Nor shall any state deprive any person of life, liberty, or property without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.' These provisions are universal in their application, to all persons within the territorial jurisdiction, without regard to any differences of race, of color, or of nationality; and the protection of the laws is a pledge of the protection of equal laws." Yick Wo, supra, at 369, 6 S.Ct., at 1070 (emphasis added). 12 In concluding that "all persons within the territory of the United States," including aliens unlawfully present, may invoke the Fifth and Sixth Amendments to challenge actions of the Federal Government, we reasoned from the understanding that the Fourteenth Amendment was designed to afford its protection to all within the boundaries of a State. Wong Wing, supra, at 238, 16 S.Ct., at 981.11 Our cases applying the Equal Protection Clause reflect the same territorial theme:12 13 "Manifestly, the obligation of the State to give the protection of equal laws can be performed only where its laws operate, that is, within its own jurisdiction. It is there that the equality of legal right must be maintained. That obligation is imposed by the Constitution upon the States severally as governmental entities,—each responsible for its own laws establishing the rights and duties of persons within its borders." Missouri ex rel. Gaines v. Canada, 305 U.S. 337, 350, 59 S.Ct. 232, 236, 83 L.Ed. 208 (1938). 14 There is simply no support for appellants' suggestion that "due process" is somehow of greater stature than "equal protection" and therefore available to a larger class of persons. To the contrary, each aspect of the Fourteenth Amendment reflects an elementary limitation on state power. To permit a State to employ the phrase "within its jurisdiction" in order to identify subclasses of persons whom it would define as beyond its jurisdiction, thereby relieving itself of the obligation to assure that its laws are designed and applied equally to those persons, would undermine the principal purpose for which the Equal Protection Clause was incorporated in the Fourteenth Amendment. The Equal Protection Clause was intended to work nothing less than the abolition of all caste-based and invidious class-based legislation. That objective is fundamentally at odds with the power the State asserts here to classify persons subject to its laws as nonetheless excepted from its protection. 15 Although the congressional debate concerning § 1 of the Fourteenth Amendment was limited, that debate clearly confirms the understanding that the phrase "within its jurisdiction" was intended in a broad sense to offer the guarantee of equal protection to all within a State's boundaries, and to all upon whom the State would impose the obligations of its laws. Indeed, it appears from those debates that Congress, by using the phrase "person within its jurisdiction," sought expressly to ensure that the equal protection of the laws was provided to the alien population. Representative Bingham reported to the House the draft resolution of the Joint Committee of Fifteen on Reconstruction (H.R. 63) that was to become the Fourteenth Amendment.13 Cong. Globe, 39th Cong., 1st Sess., 1033 (1866). Two days later, Bingham posed the following question in support of the resolution: 16 "Is it not essential to the unity of the people that the citizens of each State shall be entitled to all the privileges and immunities of citizens in the several States? Is it not essential to the unity of the Government and the unity of the people that all persons, whether citizens or strangers, within this land, shall have equal protection in every State in this Union in the rights of life and liberty and property?" Id., at 1090. 17 Senator Howard, also a member of the Joint Committee of Fifteen, and the floor manager of the Amendment in the Senate, was no less explicit about the broad objectives of the Amendment, and the intention to make its provisions applicable to all who "may happen to be" within the jurisdiction of a State: "The last two clauses of the first section of the amendment disable a State from depriving not merely a citizen of the United States, but any person, whoever he may be, of life, liberty, or property without due process of law, or from denying to him the equal protection of the laws of the State. This abolishes all class legislation in the States and does away with the injustice of subjecting one caste of persons to a code not applicable to another. . . . It will, if adopted by the States, forever disable every one of them from passing laws trenching upon those fundamental rights and privileges which pertain to citizens of the United States, and to all persons who may happen to be within their jurisdiction." Id., at 2766 (emphasis added). 18 Use of the phrase "within its jurisdiction" thus does not detract from, but rather confirms, the understanding that the protection of the Fourteenth Amendment extends to anyone, citizen or stranger, who is subject to the laws of a State, and reaches into every corner of a State's territory. That a person's initial entry into a State, or into the United States, was unlawful, and that he may for that reason be expelled, cannot negate the simple fact of his presence within the State's territorial perimeter. Given such presence, he is subject to the full range of obligations imposed by the State's civil and criminal laws. And until he leaves the jurisdiction—either voluntarily, or involuntarily in accordance with the Constitution and laws of the United States—he is entitled to the equal protection of the laws that a State may choose to establish. 19 Our conclusion that the illegal aliens who are plaintiffs in these cases may claim the benefit of the Fourteenth Amendment's guarantee of equal protection only begins the inquiry. The more difficult question is whether the Equal Protection Clause has been violated by the refusal of the State of Texas to reimburse local school boards for the education of children who cannot demonstrate that their presence within the United States is lawful, or by the imposition by those school boards of the burden of tuition on those children. It is to this question that we now turn. III 20 The Equal Protection Clause directs that "all persons similarly circumstanced shall be treated alike." F. S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415, 40 S.Ct. 560, 561, 64 L.Ed. 989 (1920). But so too, "[t]he Constitution does not require things which are different in fact or opinion to be treated in law as though they were the same." Tigner v. Texas, 310 U.S. 141, 147, 60 S.Ct. 879, 882, 84 L.Ed. 1124 (1940). The initial discretion to determine what is "different" and what is "the same" resides in the legislatures of the States. A legislature must have substantial latitude to establish classifications that roughly approximate the nature of the problem perceived, that accommodate competing concerns both public and private, and that account for limitations on the practical ability of the State to remedy every ill. In applying the Equal Protection Clause to most forms of state action, we thus seek only the assurance that the classification at issue bears some fair relationship to a legitimate public purpose. 21 But we would not be faithful to our obligations under the Fourteenth Amendment if we applied so deferential a standard to every classification. The Equal Protection Clause was intended as a restriction on state legislative action inconsistent with elemental constitutional premises. Thus we have treated as presumptively invidious those classifications that disadvantage a "suspect class,"14 or that impinge upon the exercise of a "fundamental right."15 With respect to such classifications, it is appropriate to enforce the mandate of equal protection by requiring the State to demonstrate that its classification has been precisely tailored to serve a compelling governmental interest. In addition, we have recognized that certain forms of legislative classification, while not facially invidious, nonetheless give rise to recurring constitutional difficulties; in these limited circumstances we have sought the assurance that the classification reflects a reasoned judgment consistent with the ideal of equal protection by inquiring whether it may fairly be viewed as furthering a substantial interest of the State.16 We turn to a consideration of the standard appropriate for the evaluation of § 21.031. 22 Sheer incapability or lax enforcement of the laws barring entry into this country, coupled with the failure to establish an effective bar to the employment of undocumented aliens, has resulted in the creation of a substantial "shadow population" of illegal migrants—numbering in the millions—within our borders.17 This situation raises the specter of a permanent caste of undocumented resident aliens, encouraged by some to remain here as a source of cheap labor, but nevertheless denied the benefits that our society makes available to citizens and lawful residents.18 The existence of such an underclass presents most difficult problems for a Nation that prides itself on adherence to principles of equality under law.19 23 The children who are plaintiffs in these cases are special members of this underclass. Persuasive arguments support the view that a State may withhold its beneficence from those whose very presence within the United States is the product of their own unlawful conduct. These arguments do not apply with the same force to classifications imposing disabilities on the minor children of such illegal entrants. At the least, those who elect to enter our territory by stealth and in violation of our law should be prepared to bear the consequences, including, but not limited to, deportation. But the children of those illegal entrants are not comparably situated. Their "parents have the ability to conform their conduct to societal norms," and presumably the ability to remove themselves from the State's jurisdiction; but the children who are plaintiffs in these cases "can affect neither their parents' conduct nor their own status." Trimble v. Gordon, 430 U.S. 762, 770, 97 S.Ct. 1459, 1465, 52 L.Ed.2d 31 (1977). Even if the State found it expedient to control the conduct of adults by acting against their children, legislation directing the onus of a parent's misconduct against his children does not comport with fundamental conceptions of justice. 24 "[V]isiting . . . condemnation on the head of an infant is illogical and unjust. Moreover, imposing disabilities on the . . . child is contrary to the basic concept of our system that legal burdens should bear some relationship to individual responsibility or wrongdoing. Obviously, no child is responsible for his birth and penalizing the . . . child is an ineffectual—as well as unjust—way of deterring the parent." Weber v. Aetna Casualty & Surety Co., 406 U.S. 164, 175, 92 S.Ct. 1400, 1406, 31 L.Ed.2d 768 (1972) (footnote omitted). 25 Of course, undocumented status is not irrelevant to any proper legislative goal. Nor is undocumented status an absolutely immutable characteristic since it is the product of conscious, indeed unlawful, action. But § 21.031 is directed against children, and imposes its discriminatory burden on the basis of a legal characteristic over which children can have little control. It is thus difficult to conceive of a rational justification for penalizing these children for their presence within the United States. Yet that appears to be precisely the effect of § 21.031. 26 Public education is not a "right" granted to individuals by the Constitution. San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 35, 93 S.Ct. 1278, 1298, 36 L.Ed.2d 16 (1973). But neither is it merely some governmental "benefit" indistinguishable from other forms of social welfare legislation. Both the importance of education in maintaining our basic institutions, and the lasting impact of its deprivation on the life of the child, mark the distinction. The "American people have always regarded education and [the] acquisition of knowledge as matters of supreme importance." Meyer v. Nebraska, 262 U.S. 390, 400, 43 S.Ct. 625, 627, 67 L.Ed. 1042 (1923). We have recognized "the public schools as a most vital civic institution for the preservation of a democratic system of government," Abington School District v. Schempp, 374 U.S. 203, 230, 83 S.Ct. 1560, 1575, 10 L.Ed.2d 844 (1963) (BRENNAN, J., concurring), and as the primary vehicle for transmitting "the values on which our society rests." Ambach v. Norwick, 441 U.S. 68, 76, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979). "[A]s . . . pointed out early in our history, . . . some degree of education is necessary to prepare citizens to participate effectively and intelligently in our open political system if we are to preserve freedom and independence." Wisconsin v. Yoder, 406 U.S. 205, 221, 92 S.Ct. 1526, 1536, 32 L.Ed.2d 15 (1972). And these historic "perceptions of the public schools as inculcating fundamental values necessary to the maintenance of a democratic political system have been confirmed by the observations of social scientists." Ambach v. Norwick, supra, 411 U.S., at 77, 99 S.Ct., at 1594. In addition, education provides the basic tools by which individuals might lead economically productive lives to the benefit of us all. In sum, education has a fundamental role in maintaining the fabric of our society. We cannot ignore the significant social costs borne by our Nation when select groups are denied the means to absorb the values and skills upon which our social order rests. 27 In addition to the pivotal role of education in sustaining our political and cultural heritage, denial of education to some isolated group of children poses an affront to one of the goals of the Equal Protection Clause: the abolition of governmental barriers presenting unreasonable obstacles to advancement on the basis of individual merit. Paradoxically, by depriving the children of any disfavored group of an education, we foreclose the means by which that group might raise the level of esteem in which it is held by the majority. But more directly, "education prepares individuals to be self-reliant and self-sufficient participants in society." Wisconsin v. Yoder, supra, 406 U.S., at 221, 92 S.Ct., at 1536. Illiteracy is an enduring disability. The inability to read and write will handicap the individual deprived of a basic education each and every day of his life. The inestimable toll of that deprivation on the social economic, intellectual, and psychological well-being of the individual, and the obstacle it poses to individual achievement, make it most difficult to reconcile the cost or the principle of a status-based denial of basic education with the framework of equality embodied in the Equal Protection Clause.20 What we said 28 years ago in Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), still holds true: 28 "Today, education is perhaps the most important function of state and local governments. Compulsory school attendance laws and the great expenditures for education both demonstrate our recognition of the importance of education to our democratic society. It is required in the performance of our most basic public responsibilities, even service in the armed forces. It is the very foundation of good citizenship. Today it is a principal instrument in awakening the child to cultural values, in preparing him for later professional training, and in helping him to adjust normally to his environment. In these days, it is doubtful that any child may reasonably be expected to succeed in life if he is denied the opportunity of an education. Such an opportunity, where the state has undertaken to provide it, is a right which must be made available to all on equal terms." Id., at 493, 74 S.Ct., at 691. B 29 These well-settled principles allow us to determine the proper level of deference to be afforded § 21.031. Undocumented aliens cannot be treated as a suspect class because their presence in this country in violation of federal law is not a "constitutional irrelevancy." Nor is education a fundamental right; a State need not justify by compelling necessity every variation in the manner in which education is provided to its population. See San Antonio Independent School Dist. v. Rodriguez, supra, at 28-39, 93 S.Ct., at 1293-1300. But more is involved in these cases than the abstract question whether § 21.031 discriminates against a suspect class, or whether education is a fundamental right. Section 21.031 imposes a lifetime hardship on a discrete class of children not accountable for their disabling status. The stigma of illiteracy will mark them for the rest of their lives. By denying these children a basic education, we deny them the ability to live within the structure of our civic institutions, and foreclose any realistic possibility that they will contribute in even the smallest way to the progress of our Nation. In determining the rationality of § 21.031, we may appropriately take into account its costs to the Nation and to the innocent children who are its victims. In light of these countervailing costs, the discrimination contained in § 21.031 can hardly be considered rational unless it furthers some substantial goal of the State. IV 30 It is the State's principal argument, and apparently the view of the dissenting Justices, that the undocumented status of these children vel non establishes a sufficient rational basis for denying them benefits that a State might choose to afford other residents. The State notes that while other aliens are admitted "on an equality of legal privileges with all citizens under non-discriminatory laws," Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 420, 68 S.Ct. 1138, 1143, 92 L.Ed. 1478 (1948), the asserted right of these children to an education can claim no implicit congressional imprimatur.21 Indeed, in the State's view, Congress' apparent disapproval of the presence of these children within the United States, and the evasion of the federal regulatory program that is the mark of undocumented status, provides authority for its decision to impose upon them special disabilities. Faced with an equal protection challenge respecting the treatment of aliens, we agree that the courts must be attentive to congressional policy; the exercise of congressional power might well affect the State's prerogatives to afford differential treatment to a particular class of aliens. But we are unable to find in the congressional immigration scheme any statement of policy that might weigh significantly in arriving at an equal protection balance concerning the State's authority to deprive these children of an education. 31 The Constitution grants Congress the power to "establish an uniform Rule of Naturalization." Art. I., § 8, cl. 4. Drawing upon this power, upon its plenary authority with respect to foreign relations and international commerce, and upon the inherent power of a sovereign to close its borders, Congress has developed a complex scheme governing admission to our Nation and status within our borders. See Mathews v. Diaz, 426 U.S. 67, 96 S.Ct. 1883, 48 L.Ed.2d 478 (1976); Harisiades v. Shaughnessy, 342 U.S. 580, 588-589, 72 S.Ct. 512, 518-519, 96 L.Ed. 586 (1952). The obvious need for delicate policy judgments has counseled the Judicial Branch to avoid intrusion into this field. Mathews, supra, at 81, 96 S.Ct., at 1892. But this traditional caution does not persuade us that unusual deference must be shown the classification embodied in § 21.031. The States enjoy no power with respect to the classification of aliens. See Hines v. Davidowitz, 312 U.S. 52, 61 S.Ct. 399, 85 L.Ed. 581 (1941). This power is "committed to the political branches of the Federal Government." Mathews, 426 U.S., at 81, 96 S.Ct., at 1892. Although it is "a routine and normally legitimate part" of the business of the Federal Government to classify on the basis of alien status, id., at 85, 96 S.Ct., at 1894, and to "take into account the character of the relationship between the alien and this country," id., at 80, 96 S.Ct., at 1891, only rarely are such matters relevant to legislation by a State. See Id., at 84-85, 96 S.Ct., at 1893-1894; Nyquist v. Mauclet, 432 U.S. 1, 7, n. 8, 97 S.Ct. 2120, 2124, n. 8, 53 L.Ed.2d 63 (1977). 32 As we recognized in DeCanas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976), the States do have some authority to act with respect to illegal aliens, at least where such action mirrors federal objectives and furthers a legitimate state goal. In DeCanas, the State's program reflected Congress' intention to bar from employment all aliens except those possessing a grant of permission to work in this country. Id., at 361, 96 S.Ct., at 939. In contrast, there is no indication that the disability imposed by § 21.031 corresponds to any identifiable congressional policy. The State does not claim that the conservation of state educational resources was ever a congressional concern in restricting immigration. More importantly, the classification reflected in § 21.031 does not operate harmoniously within the federal program. 33 To be sure, like all persons who have entered the United States unlawfully, these children are subject to deportation. 8 U.S.C. §§ 1251, 1252 (1976 ed. and Supp.IV). But there is no assurance that a child subject to deportation will ever be deported. An illegal entrant might be granted federal permission to continue to reside in this country, or even to become a citizen. See, e.g., 8 U.S.C. §§ 1252, 1253(h), 1254 (1976 ed. and Supp.IV). In light of the discretionary federal power to grant relief from deportation, a State cannot realistically determine that any particular undocumented child will in fact be deported until after deportation proceedings have been completed. It would of course be most difficult for the State to justify a denial of education to a child enjoying an inchoate federal permission to remain. 34 We are reluctant to impute to Congress the intention to withhold from these children, for so long as they are present in this country through no fault of their own, access to a basic education. In other contexts, undocumented status, coupled with some articulable federal policy, might enhance state authority with respect to the treatment of undocumented aliens. But in the area of special constitutional sensitivity presented by these cases, and in the absence of any contrary indication fairly discernible in the present legislative record, we perceive no national policy that supports the State in denying these children an elementary education. The State may borrow the federal classification. But to justify its use as a criterion for its own discriminatory policy, the State must demonstrate that the classification is reasonably adapted to "the purposes for which the state desires to use it." Oyama v. California, 332 U.S. 633, 664-665, 68 S.Ct. 269, 284, 92 L.Ed. 249 (1948) (Murphy, J., concurring) (emphasis added). We therefore turn to the state objectives that are said to support § 21.031. V 35 Appellants argue that the classification at issue furthers an interest in the "preservation of the state's limited resources for the education of its lawful residents."22 Brief for Appellants 26. Of course, a concern for the preservation of resources standing alone can hardly justify the classification used in allocating those resources. Graham v. Richardson, 403 U.S. 365, 374-375, 91 S.Ct. 1848, 1853, 29 L.Ed.2d 534 (1971). The State must do more than justify its classification with a concise expression of an intention to discriminate. Examining Board v. Flores de Otero, 426 U.S. 572, 605, 96 S.Ct. 2264, 2282, 49 L.Ed.2d 65 (1976). Apart from the asserted state prerogative to act against undocumented children solely on the basis of their undocumented status—an asserted prerogative that carries only minimal force in the circumstances of these cases—we discern three colorable state interests that might support § 21.031. 36 First, appellants appear to suggest that the State may seek to protect itself from an influx of illegal immigrants. While a State might have an interest in mitigating the potentially harsh economic effects of sudden shifts in population,23 § 21.031 hardly offers an effective method of dealing with an urgent demographic or economic problem. There is no evidence in the record suggesting that illegal entrants impose any significant burden on the State's economy. To the contrary, the available evidence suggests that illegal aliens underutilize public services, while contributing their labor to the local economy and tax money to the state fisc. 458 F.Supp., at 578; 501 F.Supp., at 570-571. The dominant incentive for illegal entry into the State of Texas is the availability of employment; few if any illegal immigrants come to this country, or presumably to the State of Texas, in order to avail themselves of a free education.24 Thus, even making the doubtful assumption that the net impact of illegal aliens on the economy of the State is negative, we think it clear that "[c]harging tuition to undocumented children constitutes a ludicrously ineffectual attempt to stem the tide of illegal immigration," at least when compared with the alternative of prohibiting the employment of illegal aliens. 458 F.Supp., at 585. See 628 F.2d, at 461; 501 F.Supp., at 579 and n. 88. 37 Second, while it is apparent that a State may "not . . . reduce expenditures for education by barring [some arbitrarily chosen class of] children from its schools," Shapiro v. Thompson, 394 U.S. 618, 633, 89 S.Ct. 1322, 1330, 22 L.Ed.2d 600 (1969), appellants suggest that undocumented children are appropriately singled out for exclusion because of the special burdens they impose on the State's ability to provide high-quality public education. But the record in no way supports the claim that exclusion of undocumented children is likely to improve the overall quality of education in the State.25 As the District Court in No. 80-1934 noted, the State failed to offer any "credible supporting evidence that a proportionately small diminution of the funds spent on each child [which might result from devoting some state funds to the education of the excluded group] will have a grave impact on the quality of education." 501 F.Supp., at 583. And, after reviewing the State's school financing mechanism, the District Court in No. 80-1538 concluded that barring undocumented children from local schools would not necessarily improve the quality of education provided in those schools. 458 F.Supp., at 577. Of course, even if improvement in the quality of education were a likely result of barring some number of children from the schools of the State, the State must support its selection of this group as the appropriate target for exclusion. In terms of educational cost and need, however, undocumented children are "basically indistinguishable" from legally resident alien children. Id., at 589; 501 F.Supp., at 583, and n. 104. 38 Finally, appellants suggest that undocumented children are appropriately singled out because their unlawful presence within the United States renders them less likely than other children to remain within the boundaries of the State, and to put their education to productive social or political use within the State. Even assuming that such an interest is legitimate, it is an interest that is most difficult to quantify. The State has no assurance that any child, citizen or not, will employ the education provided by the State within the confines of the State's borders. In any event, the record is clear that many of the undocumented children disabled by this classification will remain in this country indefinitely, and that some will become lawful residents or citizens of the United States. It is difficult to understand precisely what the State hopes to achieve by promoting the creation and perpetuation of a subclass of illiterates within our boundaries, surely adding to the problems and costs of unemployment, welfare, and crime. It is thus clear that whatever savings might be achieved by denying these children an education, they are wholly insubstantial in light of the costs involved to these children, the State, and the Nation. VI 39 If the State is to deny a discrete group of innocent children the free public education that it offers to other children residing within its borders, that denial must be justified by a showing that it furthers some substantial state interest. No such showing was made here. Accordingly, the judgment of the Court of Appeals in each of these cases is 40 Affirmed. 41 Justice MARSHALL, concurring. 42 While I join the Court's opinion, I do so without in any way retreating from my opinion in San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 70-133, 93 S.Ct. 1278, 1315-1348, 36 L.Ed.2d 16 (1973) (dissenting opinion). I continue to believe that an individual's interest in education is fundamental, and that this view is amply supported "by the unique status accorded public education by our society, and by the close relationship between education and some of our most basic constitutional values." Id., at 111, 93 S.Ct., at 1336. Furthermore, I believe that the facts of these cases demonstrate the wisdom of rejecting a rigidified approach to equal protection analysis, and of employing an approach that allows for varying levels of scrutiny depending upon "the constitutional and societal importance of the interest adversely affected and the recognized invidiousness of the basis upon which the particular classification is drawn." Id., at 99, 93 S.Ct., at 1330. See also Dandridge v. Williams, 397 U.S. 471, 519-521, 90 S.Ct. 1153, 1178-1180, 25 L.Ed.2d 491 (1970) (MARSHALL, J., dissenting). It continues to be my view that a class-based denial of public education is utterly incompatible with the Equal Protection Clause of the Fourteenth Amendment. 43 Justice BLACKMUN, concurring. 44 I join the opinion and judgment of the Court. 45 Like Justice POWELL, I believe that the children involved in this litigation "should not be left on the streets uneducated." Post, at 238. I write separately, however, because in my view the nature of the interest at stake is crucial to the proper resolution of these cases. 46 The "fundamental rights" aspect of the Court's equal protection analysis—the now-familiar concept that governmental classifications bearing on certain interests must be closely scrutinized—has been the subject of some controversy. Justice Harlan, for example, warned that "[v]irtually every state statute affects important rights. . . . [T]o extend the 'compelling interest' rule to all cases in which such rights are affected would go far toward making this Court a 'super-legislature.' " Shapiro v. Thompson, 394 U.S. 618, 661, 89 S.Ct. 1322, 1345, 22 L.Ed.2d 600 (1969) (dissenting opinion). Others have noted that strict scrutiny under the Equal Protection Clause is unnecessary when classifications infringing enumerated constitutional rights are involved, for "a state law that impinges upon a substantive right or liberty created or conferred by the Constitution is, of course, presumptively invalid, whether or not the law's purpose or effect is to create any classifications." San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 61, 93 S.Ct. 1278, 1311, 36 L.Ed.2d 16 (1973) (Stewart, J., concurring). See Shapiro v. Thompson, 394 U.S., at 659, 89 S.Ct., at 1344 (Harlan, J., dissenting). Still others have suggested that fundamental rights are not properly a part of equal protection analysis at all, because they are unrelated to any defined principle of equality.1 47 These considerations, combined with doubts about the judiciary's ability to make fine distinctions in assessing the effects of complex social policies, led the Court in Rodriguez to articulate a firm rule: fundamental rights are those that "explicitly or implicitly [are] guaranteed by the Constitution." 411 U.S., at 33-34, 93 S.Ct., 1296-1297. It therefore squarely rejected the notion that "an ad hoc determination as to the social or economic importance" of a given interest is relevant to the level of scrutiny accorded classifications involving that interest, id., at 32, 93 S.Ct., at 1296, and made clear that "[i]t is not the province of this Court to create substantive constitutional rights in the name of guaranteeing equal protection of the laws." Id., at 33, 93 S.Ct., at 1296. 48 I joined Justice POWELL's opinion for the Court in Rodriguez, and I continue to believe that it provides the appropriate model for resolving most equal protection disputes. Classifications infringing substantive constitutional rights necessarily will be invalid, if not by force of the Equal Protection Clause, then through operation of other provisions of the Constitution. Conversely, classifications bearing on nonconstitutional interests even those involving "the most basic economic needs of impoverished human beings," Dandridge v. Williams, 397 U.S. 471, 485, 90 S.Ct. 1153, 1161, 25 L.Ed.2d 491 (1970)—generally are not subject to special treatment under the Equal Protection Clause, because they are not distinguishable in any relevant way from other regulations in "the area of economics and social welfare." Ibid. 49 With all this said, however, I believe the Court's experience has demonstrated that the Rodriguez formulation does not settle every issue of "fundamental rights" arising under the Equal Protection Clause. Only a pedant would insist that there are no meaningful distinctions among the multitude of social and political interests regulated by the States, and Rodriguez does not stand for quite so absolute a proposition. To the contrary, Rodriguez implicitly acknowledged that certain interests, though not constitutionally guaranteed, must be accorded a special place in equal protection analysis. Thus, the Court's decisions long have accorded strict scrutiny to classifications bearing on the right to vote in state elections, and Rodriguez confirmed the "constitutional underpinnings of the right to equal treatment in the voting process." 411 U.S., at 34, n. 74, 93 S.Ct., at 1297, n. 74. Yet "the right to vote, per se, is not a constitutionally protected right," id., at 35, n. 78, 93 S.Ct., at 1319, n. 78. See Harper v. Virginia Board of Elections, 383 U.S. 663, 665, 86 S.Ct. 1079, 1080, 16 L.Ed.2d 169 (1966); Rodriguez, 411 U.S., at 59, n. 2, 93 S.Ct., at 1310, n. 2 (Stewart, J., concurring). Instead, regulation of the electoral process receives unusual scrutiny because "the right to exercise the franchise in a free and unimpaired manner is preservative of other basic civil and political rights." Reynolds v. Sims, 377 U.S. 533, 562, 84 S.Ct. 1362, 1381, 12 L.Ed.2d 506 (1964). See Dunn v. Blumstein, 405 U.S. 330, 336, 92 S.Ct. 995, 999, 31 L.Ed.2d 274 (1972). In other words, the right to vote is accorded extraordinary treatment because it is, in equal protection terms, an extraordinary right: a citizen2 cannot hope to achieve any meaningful degree of individual political equality if granted an inferior right of participation in the political process. Those denied the vote are relegated, by state fiat, in a most basic way to second-class status. 50 It is arguable, of course, that the Court never should have applied fundamental rights doctrine in the fashion outlined above. Justice Harlan, for one, maintained that strict equal protection scrutiny was appropriate only when racial or analogous classifications were at issue. Shapiro v. Thompson, 394 U.S., at 658-663, 89 S.Ct., at 1344-1346 (dissenting opinion). See Reynolds v. Sims, 377 U.S., at 590-591, 84 S.Ct., at 1396 (Harlan, J., dissenting). But it is too late to debate that point, and I believe that accepting the principle of the voting cases—the idea that state classifications bearing on certain interests pose the risk of allocating rights in a fashion inherently contrary to any notion of "equality"—dictates the outcome here. As both Justice POWELL and THE CHIEF JUSTICE observe, the Texas scheme inevitably will create "a subclass of illiterate persons," post, at 241 (POWELL, J., concurring); see post, at 242, 254 (BURGER, C.J., dissenting); where I differ with THE CHIEF JUSTICE is in my conclusion that this makes the statutory scheme unconstitutional as well as unwise. 51 In my view, when the State provides an education to some and denies it to others, it immediately and inevitably creates class distinctions of a type fundamentally inconsistent with those purposes, mentioned above, of the Equal Protection Clause. Children denied an education are placed at a permanent and insurmountable competitive disadvantage, for an uneducated child is denied even the opportunity to achieve. And when those children are members of an identifiable group, that group—through the State's action—will have been converted into a discrete underclass. Other benefits provided by the State, such as housing and public assistance, are of course important; to an individual in immediate need, they may be more desirable than the right to be educated. But classifications involving the complete denial of education are in a sense unique, for they strike at the heart of equal protection values by involving the State in the creation of permanent class distinctions. Cf. Rodriguez, 411 U.S., at 115, n. 74, 93 S.Ct., at 1338, n. 74 (MARSHALL, J., dissenting). In a sense, then, denial of an education is the analogue of denial of the right to vote: the former relegates the individual to second-class social status; the latter places him at a permanent political disadvantage. 52 This conclusion is fully consistent with Rodriguez. The Court there reserved judgment on the constitutionality of a state system that "occasioned an absolute denial of educational opportunities to any of its children," noting that "no charge fairly could be made that the system [at issue in Rodriguez] fails to provide each child with an opportunity to acquire . . . basic minimal skills." Id., at 37, 93 S.Ct., at 1299. And it cautioned that in a case "involv[ing] the most persistent and difficult questions of educational policy, . . . [the] Court's lack of specialized knowledge and experience counsels against premature interference with the informed judgments made at the state and local levels." Id., at 42, 93 S.Ct., at 1301. Thus Rodriguez held, and the Court now reaffirms, that "a State need not justify by compelling necessity every variation in the manner in which education is provided to its population." Ante, at 223. Similarly, it is undeniable that education is not a "fundamental right" in the sense that it is constitutionally guaranteed. Here, however, the State has undertaken to provide an education to most of the children residing within its borders. And, in contrast to the situation in Rodriguez, it does not take an advanced degree to predict the effects of a complete denial of education upon those children targeted by the State's classification. In such circumstances, the voting decisions suggest that the State must offer something more than a rational basis for its classification.3 53 Concededly, it would seem ironic to discuss the social necessity of an education in a case that concerned only undocumented aliens "whose very presence in the state and this country is illegal." Post, at 250 (BURGER, C.J., dissenting). But because of the nature of the federal immigration laws and the pre-eminent role of the Federal Government in regulating immigration, the class of children here is not a monolithic one. Thus, the District Court in the Alien Children Education case found as a factual matter that a significant number of illegal aliens will remain in this country permanently, 501 F.Supp. 544, 558-559 (SD Tex.1980); that some of the children involved in this litigation are "documentable," id., at 573; and that "[m]any of the undocumented children are not deportable. None of the named plaintiffs is under an order of deportation." Id., at 583, n. 103. As the Court's alienage cases demonstrate, these children may not be denied rights that are granted to citizens, excepting only those rights bearing on political interests. See Nyquist v. Mauclet, 432 U.S. 1, 97 S.Ct. 2120, 53 L.Ed.2d 63 (1977). And, as Justice POWELL notes, the structure of the immigration statutes makes it impossible for the State to determine which aliens are entitled to residence, and which eventually will be deported. Post, at 240-241, n. 6. Indeed, any attempt to do so would involve the State in the administration of the immigration laws. Whatever the State's power to classify deportable aliens, then—and whatever the Federal Government's ability to draw more precise and more acceptable alienage classifications—the statute at issue here sweeps within it a substantial number of children who will in fact, and who may well be entitled to, remain in the United States. Given the extraordinary nature of the interest involved, this makes the classification here fatally imprecise. And, as the Court demonstrates, the Texas legislation is not otherwise supported by any substantial interests. 54 Because I believe that the Court's carefully worded analysis recognizes the importance of the equal protection and preemption interests I consider crucial, I join its opinion as well as its judgment. 55 Justice POWELL, concurring. 56 I join the opinion of the Court, and write separately to emphasize the unique character of the cases before us. 57 The classification in question severely disadvantages children who are the victims of a combination of circumstances. Access from Mexico into this country, across our 2,000-mile border, is readily available and virtually uncontrollable. Illegal aliens are attracted by our employment opportunities, and perhaps by other benefits as well. This is a problem of serious national proportions, as the Attorney General recently has recognized. See ante, at 218-219, n. 17. Perhaps because of the intractability of the problem, Congress—vested by the Constitution with the responsibility of protecting our borders and legislating with respect to aliens—has not provided effective leadership in dealing with this problem.1 It therefore is certain that illegal aliens will continue to enter the United States and, as the record makes clear, an unknown percentage of them will remain here. I agree with the Court that their children should not be left on the streets uneducated. 58 Although the analogy is not perfect, our holding today does find support in decisions of this Court with respect to the status of illegitimates. In Weber v. Aetna Casualty & Surety Co., 406 U.S. 164, 175, 92 S.Ct. 1400, 1406, 31 L.Ed.2d 768 (1972), we said: "[V]isiting . . . condemnation on the head of an infant" for the misdeeds of the parents is illogical, unjust, and "contrary to the basic concept of our system that legal burdens should bear some relationship to individual responsibility or wrongdoing." 59 In these cases, the State of Texas effectively denies to the school-age children of illegal aliens the opportunity to attend the free public schools that the State makes available to all residents. They are excluded only because of a status resulting from the violation by parents or guardians of our immigration laws and the fact that they remain in our country unlawfully. The appellee children are innocent in this respect. They can "affect neither their parents' conduct nor their own status." Trimble v. Gordon, 430 U.S. 762, 770, 97 S.Ct. 1459, 1465, 52 L.Ed.2d 31 (1977). 60 Our review in a case such as these is properly heightened.2 See id., at 767, 97 S.Ct., at 1463. Cf. Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976). The classification at issue deprives a group of children of the opportunity for education afforded all other children simply because they have been assigned a legal status due to a violation of law by their parents. These children thus have been singled out for a lifelong penalty and stigma. A legislative classification that threatens the creation of an underclass of future citizens and residents cannot be reconciled with one of the fundamental purposes of the Fourteenth Amendment. In these unique circumstances, the Court properly may require that the State's interests be substantial and that the means bear a "fair and substantial relation" to these interests.3 See Lalli v. Lalli, 439 U.S. 259, 265, 99 S.Ct. 518, 523, 58 L.Ed.2d 503 (1978) ("classifications based on illegitimacy . . . are invalid under the Fourteenth Amendment if they are not substantially related to permissible state interests"); id., at 271, 99 S.Ct., at 526 ("[a]s the State's interests are substantial, we now consider the means adopted"). 61 In my view, the State's denial of education to these children bears no substantial relation to any substantial state interest. Both of the District Courts found that an uncertain but significant percentage of illegal alien children will remain in Texas as residents and many eventually will become citizens. The discussion by the Court, ante, at Part V, of the State's purported interests demonstrates that they are poorly served by the educational exclusion. Indeed, the interests relied upon by the State would seem to be insubstantial in view of the consequences to the State itself of wholly uneducated persons living indefinitely within its borders. By contrast, access to the public schools is made available to the children of lawful residents without regard to the temporary nature of their residency in the particular Texas school district.4 The Court of Appeals and the District Courts that addressed these cases concluded that the classification could not satisfy even the bare requirements of rationality. One need not go so far to conclude that the exclusion of appellees' class5 of children from state-provided education is a type of punitive discrimination based on status that is impermissible under the Equal Protection Clause. 62 In reaching this conclusion, I am not unmindful of what must be the exasperation of responsible citizens and government authorities in Texas and other States similarly situated. Their responsibility, if any, for the influx of aliens is slight compared to that imposed by the Constitution on the Federal Government.6 So long as the ease of entry remains inviting, and the power to deport is exercised infrequently by the Federal Government, the additional expense of admitting these children to public schools might fairly be shared by the Federal and State Governments. But it hardly can be argued rationally that anyone benefits from the creation within our borders of a subclass of illiterate persons many of whom will remain in the State, adding to the problems and costs of both State and National Governments attendant upon unemployment, welfare, and crime. 63 Chief Justice BURGER, with whom Justice WHITE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 64 Were it our business to set the Nation's social policy, I would agree without hesitation that it is senseless for an enlightened society to deprive any children—including illegal aliens—of an elementary education. I fully agree that it would be folly—and wrong—to tolerate creation of a segment of society made up of illiterate persons, many having a limited or no command of our language.1 However, the Constitution does not constitute us as "Platonic Guardians" nor does it vest in this Court the authority to strike down laws because they do not meet our standards of desirable social policy, "wisdom," or "common sense." See TVA v. Hill, 437 U.S. 153, 194-195, 98 S.Ct. 2278, 2301-2302, 57 L.Ed.2d 312 (1978). We trespass on the assigned function of the political branches under our structure of limited and separated powers when we assume a policymaking role as the Court does today. 65 The Court makes no attempt to disguise that it is acting to make up for Congress' lack of "effective leadership" in dealing with the serious national problems caused by the influx of uncountable millions of illegal aliens across our borders.2 See ante, at 237-238 (POWELL, J., concurring). The failure of enforcement of the immigration laws over more than a decade and the inherent difficulty and expense of sealing our vast borders have combined to create a grave socioeconomic dilemma. It is a dilemma that has not yet even been fully assessed, let alone addressed. However, it is not the function of the Judiciary to provide "effective leadership" simply because the political branches of government fail to do so. 66 The Court's holding today manifests the justly criticized judicial tendency to attempt speedy and wholesale formulation of "remedies" for the failures—or simply the laggard pace—of the political processes of our system of government. The Court employs, and in my view abuses, the Fourteenth Amendment in an effort to become an omnipotent and omniscient problem solver. That the motives for doing so are noble and compassionate does not alter the fact that the Court distorts our constitutional function to make amends for the defaults of others. 67 * In a sense, the Court's opinion rests on such a unique confluence of theories and rationales that it will likely stand for little beyond the results in these particular cases. Yet the extent to which the Court departs from principled constitutional adjudication is nonetheless disturbing. 68 I have no quarrel with the conclusion that the Equal Protection Clause of the Fourteenth Amendment applies to aliens who, after their illegal entry into this country, are indeed physically "within the jurisdiction" of a state. However, as the Court concedes, this "only begins the inquiry." Ante, at 215. The Equal Protection Clause does not mandate identical treatment of different categories of persons. Jefferson v. Hackney, 406 U.S. 535, 549, 92 S.Ct. 1724, 1732, 32 L.Ed.2d 285 (1972); Reed v. Reed, 404 U.S. 71, 75, 92 S.Ct. 251, 253, 30 L.Ed.2d 225 (1971); Tigner v. Texas, 310 U.S. 141, 147-148, 60 S.Ct. 879, 882, 84 L.Ed. 1124 (1940). 69 The dispositive issue in these cases, simply put, is whether, for purposes of allocating its finite resources, a state has a legitimate reason to differentiate between persons who are lawfully within the state and those who are unlawfully there. The distinction the State of Texas has drawn—based not only upon its own legitimate interests but on classifications established by the Federal Government in its immigration laws and policies—is not unconstitutional. A. 70 The Court acknowledges that, except in those cases when state classifications disadvantage a "suspect class" or impinge upon a "fundamental right," the Equal Protection Clause permits a state "substantial latitude" in distinguishing between different groups of persons. Ante, at 216-217. Moreover, the Court expressly—and correctly—rejects any suggestion that illegal aliens are a suspect class, ante, at 219, n. 19, or that education is a fundamental right, ante, at 221,223. Yet by patching together bits and pieces of what might be termed quasi-suspect-class and quasi-fundamental-rights analysis, the Court spins out a theory custom-tailored to the facts of these cases. 71 In the end, we are told little more than that the level of scrutiny employed to strike down the Texas law applies only when illegal alien children are deprived of a public education, see ante, at 223-224.3 If ever a court was guilty of an unabashedly result-oriented approach, this case is a prime example. 72 (1) 73 The Court first suggests that these illegal alien children, although not a suspect class, are entitled to special solicitude under the Equal Protection Clause because they lack "control" over or "responsibility" for their unlawful entry into this country. Ante, at 220, 223-224. Similarly, the Court appears to take the position that § 21.031 is presumptively "irrational" because it has the effect of imposing "penalties" on "innocent" children. Ibid. See also ante, at 2406 (POWELL, J., concurring).4 However, the Equal Protection Clause does not preclude legislators from classifying among persons on the basis of factors and characteristics over which individuals may be said to lack "control." Indeed, in some circumstances persons generally, and children in particular, may have little control over or responsibility for such things as their ill health, need for public assistance, or place of residence. Yet a state legislature is not barred from considering, for example, relevant differences between the mentally healthy and the mentally ill, or between the residents of different counties,5 simply because these may be factors unrelated to individual choice or to any "wrongdoing." The Equal Protection Clause protects against arbitrary and irrational classifications, and against invidious discrimination stemming from prejudice and hostility; it is not an all-encompassing "equalizer" designed to eradicate every distinction for which persons are not "responsible." The Court does not presume to suggest that appellees' purported lack of culpability for their illegal status prevents them from being deported or otherwise "penalized" under federal law. Yet would deportation be any less a "penalty" than denial of privileges provided to legal residents?6 Illegality of presence in the United States does not—and need not—depend on some amorphous concept of "guilt" or "innocence" concerning an alien's entry. Similarly, a state's use of federal immigration status as a basis for legislative classification is not necessarily rendered suspect for its failure to take such factors into account. 74 The Court's analogy to cases involving discrimination against illegitimate children—see ante, at 220; ante, at 238-239 (POWELL, J., concurring)—is grossly misleading. The State has not thrust any disabilities upon appellees due to their "status of birth." Cf. Weber v. Aetna Casualty & Surety Co., 406 U.S. 164, 176, 92 S.Ct. 1400, 1407, 31 L.Ed.2d 768 (1972). Rather, appellees' status is predicated upon the circumstances of their concededly illegal presence in this country, and is a direct result of Congress' obviously valid exercise of its "broad constitutional powers" in the field of immigration and naturalization. U.S. Const., Art. I, § 8, cl. 4; see Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 419, 68 S.Ct. 1138, 1142, 92 L.Ed. 1478 (1948). This Court has recognized that in allocating governmental benefits to a given class of aliens, one "may take into account the character of the relationship between the alien and this country." Mathews v. Diaz, 426 U.S. 67, 80, 96 S.Ct. 1883, 1891, 48 L.Ed.2d 478 (1976). When that "relationship" is a federally prohibited one, there can, of course, be no presumption that a state has a constitutional duty to include illegal aliens among the recipients of its governmental benefits.7 (2) 75 The second strand of the Court's analysis rests on the premise that, although public education is not a constitutionally guaranteed right, "neither is it merely some governmental 'benefit' indistinguishable from other forms of social welfare legislation." Ante, at 221. Whatever meaning or relevance this opaque observation might have in some other context,8 it simply has no bearing on the issues at hand. Indeed, it is never made clear what the Court's opinion means on this score. 76 The importance of education is beyond dispute. Yet we have held repeatedly that the importance of a governmental service does not elevate it to the status of a "fundamental right" for purposes of equal protection analysis. San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 30-31, 93 S.Ct. 1278, 1295, 36 L.Ed.2d 16 (1973); Lindsey v. Normet, 405 U.S. 56, 73-74, 92 S.Ct. 862, 874, 31 L.Ed.2d 36 (1972). In San Antonio Independent School Dist., supra, Justice POWELL, speaking for the Court, expressly rejected the proposition that state laws dealing with public education are subject to special scrutiny under the Equal Protection Clause. Moreover, the Court points to no meaningful way to distinguish between education and other governmental benefits in this context. Is the Court suggesting that education is more "fundamental" than food, shelter, or medical care? 77 The Equal Protection Clause guarantees similar treatment of similarly situated persons, but it does not mandate a constitutional hierarchy of governmental services. Justice POWELL, speaking for the Court in San Antonio Independent School Dist., supra, 411 U.S., at 31, 93 S.Ct., at 1295, put it well in stating that to the extent this Court raises or lowers the degree of "judicial scrutiny" in equal protection cases according to a transient Court majority's view of the societal importance of the interest affected, we "assum[e] a legislative role and one for which the Court lacks both authority and competence." Yet that is precisely what the Court does today. See also Shapiro v. Thompson, 394 U.S. 618, 655-661, 89 S.Ct. 1322, 1342-1345, 22 L.Ed.2d 600 (1969) (Harlan, J., dissenting). 78 The central question in these cases, as in every equal protection case not involving truly fundamental rights "explicitly or implicitly guaranteed by the Constitution," San Antonio Independent School Dist., supra, 411 U.S., at 33-34, 93 S.Ct., at 1296-1297, is whether there is some legitimate basis for a legislative distinction between different classes of persons. The fact that the distinction is drawn in legislation affecting access to public education—as opposed to legislation allocating other important governmental benefits, such as public assistance, health care, or housing—cannot make a difference in the level of scrutiny applied. B 79 Once it is conceded—as the Court does—that illegal aliens are not a suspect class, and that education is not a fundamental right, our inquiry should focus on and be limited to whether the legislative classification at issue bears a rational relationship to a legitimate state purpose. Vance v. Bradley, 440 U.S. 93, 97, 99 S.Ct. 939, 942, 59 L.Ed.2d 171 (1979); Dandridge v. Williams, 397 U.S. 471, 485-487, 90 S.Ct. 1153, 1161-1162, 25 L.Ed.2d 491 (1970); see ante, at 216.9 80 The State contends primarily that § 21.031 serves to prevent undue depletion of its limited revenues available for education, and to preserve the fiscal integrity of the State's school-financing system against an ever-increasing flood of illegal aliens—aliens over whose entry or continued presence it has no control. Of course such fiscal concerns alone could not justify discrimination against a suspect class or an arbitrary and irrational denial of benefits to a particular group of persons. Yet I assume no Member of this Court would argue that prudent conservation of finite state revenues is per se an illegitimate goal. Indeed, the numerous classifications this Court has sustained in social welfare legislation were invariably related to the limited amount of revenues available to spend on any given program or set of programs. See, e.g., Jefferson v. Hackney, 406 U.S., at 549-551, 92 S.Ct., at 1733-1734; Dandridge v. Williams, supra, at 487, 90 S.Ct., at 1162. The significant question here is whether the requirement of tuition from illegal aliens who attend the public schools—as well as from residents of other states, for example—is a rational and reasonable means of furthering the State's legitimate fiscal ends.10 81 Without laboring what will undoubtedly seem obvious to many, it simply is not "irrational" for a state to conclude that it does not have the same responsibility to provide benefits for persons whose very presence in the state and this country is illegal as it does to provide for persons lawfully present. By definition, illegal aliens have no right whatever to be here, and the state may reasonably, and constitutionally, elect not to provide them with governmental services at the expense of those who are lawfully in the state.11 In DeCanas v. Bica, 424 U.S. 351, 357, 96 S.Ct. 933, 937, 47 L.Ed.2d 43 (1976), we held that a State may protect its "fiscal interests and lawfully resident labor force from the deleterious effects on its economy resulting from the employment of illegal aliens." And only recently this Court made clear that a State has a legitimate interest in protecting and preserving the quality of its schools and "the right of its own bona fide residents to attend such institutions on a preferential tuition basis." Vlandis v. Kline, 412 U.S. 441, 453, 93 S.Ct. 2230, 2237, 37 L.Ed.2d 63 (1973) (emphasis added). See also Elkins v. Moreno, 435 U.S. 647, 663-668, 98 S.Ct. 1338, 1348-1350, 55 L.Ed.2d 614 (1978). The Court has failed to offer even a plausible explanation why illegality of residence in this country is not a factor that may legitimately bear upon the bona fides of state residence and entitlement to the benefits of lawful residence.12 82 It is significant that the Federal Government has seen fit to exclude illegal aliens from numerous social welfare programs, such as the food stamp program, 7 U.S.C. § 2015(f) (1976 ed. and Supp.IV) and 7 CFR § 273.4 (1981), the old-age assistance, aid to families with dependent children, aid to the blind, aid to the permanently and totally disabled, and supplemental security income programs, 45 CFR § 233.50 (1981), the Medicare hospital insurance benefits program, 42 U.S.C. § 1395i-2 and 42 CFR § 405.205(b) (1981), and the Medicaid hospital insurance benefits for the aged and disabled program, 42 U.S.C. § 1395o and 42 CFR § 405.103(a)(4) (1981). Although these exclusions do not conclusively demonstrate the constitutionality of the State's use of the same classification for comparable purposes, at the very least they tend to support the rationality of excluding illegal alien residents of a state from such programs so as to preserve the state's finite revenues for the benefit of lawful residents. See Mathews v. Diaz, 426 U.S., at 80, 96 S.Ct., at 1891; see also n. 7, supra. 83 The Court maintains—as if this were the issue—that "barring undocumented children from local schools would not necessarily improve the quality of education provided in those schools." Ante, at 229. See 458 F.Supp. 569, 577 (ED Tex.1978).13 However, the legitimacy of barring illegal aliens from programs such as Medicare or Medicaid does not depend on a showing that the barrier would "improve the quality" of medical care given to persons lawfully entitled to participate in such programs. Modern education, like medical care, is enormously expensive, and there can be no doubt that very large added costs will fall on the State or its local school districts as a result of the inclusion of illegal aliens in the tuition-free public schools. The State may, in its discretion, use any savings resulting from its tuition requirement to "improve the quality of education" in the public school system, or to enhance the funds available for other social programs, or to reduce the tax burden placed on its residents; each of these ends is "legitimate." The State need not show, as the Court implies, that the incremental cost of educating illegal aliens will send it into bankruptcy, or have a " 'grave impact on the quality of education,' " ante, at 229; that is not dispositive under a "rational basis" scrutiny. In the absence of a constitutional imperative to provide for the education of illegal aliens, the State may "rationally" choose to take advantage of whatever savings will accrue from limiting access to the tuition-free public schools to its own lawful residents, excluding even citizens of neighboring States.14 84 Denying a free education to illegal alien children is not a choice I would make were I a legislator. Apart from compassionate considerations, the long-range costs of excluding any children from the public schools may well outweigh the costs of educating them. But that is not the issue; the fact that there are sound policy arguments against the Texas Legislature's choice does not render that choice an unconstitutional one. II 85 The Constitution does not provide a cure for every social ill, nor does it vest judges with a mandate to try to remedy every social problem. Lindsey v. Normet, 405 U.S., at 74, 92 S.Ct., at 874. See Reynolds v. Sims, 377 U.S. 533, 624-625, 84 S.Ct. 1362, 1414, 12 L.Ed.2d 506 (1964) (Harlan, J., dissenting). Moreover, when this Court rushes in to remedy what it perceives to be the failings of the political processes, it deprives those processes of an opportunity to function. When the political institutions are not forced to exercise constitutionally allocated powers and responsibilities, those powers, like muscles not used, tend to atrophy. Today's cases, I regret to say, present yet another example of unwarranted judicial action which in the long run tends to contribute to the weakening of our political processes.15 86 Congress, "vested by the Constitution with the responsibility of protecting our borders and legislating with respect to aliens," ante, at 237 (POWELL, J., concurring), bears primary responsibility for addressing the problems occasioned by the millions of illegal aliens flooding across our southern border. Similarly, it is for Congress, and not this Court, to assess the "social costs borne by our Nation when select groups are denied the means to absorb the values and skills upon which our social order rests." Ante, at 221; see ante, at 223-224. While the "specter of a permanent caste" of illegal Mexican residents of the United States is indeed a disturbing one, see ante, at 218-219, it is but one segment of a larger problem, which is for the political branches to solve. I find it difficult to believe that Congress would long tolerate such a self-destructive result—that it would fail to deport these illegal alien families or to provide for the education of their children. Yet instead of allowing the political processes to run their course—albeit with some delay—the Court seeks to do Congress' job for it, compensating for congressional inaction. It is not unreasonable to think that this encourages the political branches to pass their problems to the Judiciary. 87 The solution to this seemingly intractable problem is to defer to the political processes, unpalatable as that may be to some. 1 That section provides, in pertinent part: "(a) All children who are citizens of the United States or legally admitted aliens and who are over the age of five years and under the age of 21 years on the first day of September of any scholastic year shall be entitled to the benefits of the Available School Fund for that year. "(b) Every child in this state who is a citizen of the United States or a legally admitted alien and who is over the age of five years and not over the age of 21 years on the first day of September of the year in which admission is sought shall be permitted to attend the public free schools of the district in which he resides or in which his parent, guardian, or the person having lawful control of him resides at the time he applies for admission. "(c) The board of trustees of any public free school district of this state shall admit into the public free schools of the district free of tuition all persons who are either citizens of the United States or legally admitted aliens and who are over five and not over 21 years of age at the beginning of the scholastic year if such person or his parent, guardian or person having lawful control resides within the school district." 2 Despite the enactment of § 21.031 in 1975, the School District had continued to enroll undocumented children free of charge until the 1977-1978 school year. In July 1977, it adopted a policy requiring undocumented children to pay a "full tuition fee" in order to enroll. Section 21.031 had not provided a definition of "a legally admitted alien." Tyler offered the following clarification: "A legally admitted alien is one who has documentation that he or she is legally in the United States, or a person who is in the process of securing documentation from the United States Immigration Service, and the Service will state that the person is being processed and will be admitted with proper documentation." App. to Juris. Statement in No. 80-1538, p. A-38. 3 The court contrasted this group with those illegal aliens who entered the country alone in order to earn money to send to their dependents in Mexico, and who in many instances remained in this country for only a short period of time. 458 F.Supp., at 578. 4 Plaintiffs' expert, Dr. Gilbert Cardenas, testified that "fifty to sixty per cent . . . of current legal alien workers were formerly illegal aliens." Id., at 577. A defense witness, Rolan Heston, District Director of the Houston District of the Immigration and Naturalization Service, testified that "undocumented children can and do live in the United States for years, and adjust their status through marriage to a citizen or permanent resident." Ibid. The court also took notice of congressional proposals to "legalize" the status of many unlawful entrants. Id., at 577-578. See also n. 17, infra. 5 The court found § 21.031 inconsistent with the scheme of national regulation under the Immigration and Nationality Act, and with federal laws pertaining to funding and discrimination in education. The court distinguished DeCanas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976), by emphasizing that the state bar on employment of illegal aliens involved in that case mirrored precisely the federal policy, of protecting the domestic labor market, underlying the immigration laws. The court discerned no express federal policy to bar illegal immigrants from education. 458 F.Supp., at 590-592. 6 The Court of Appeals noted that DeCanas v. Bica, supra, had not foreclosed all state regulation with respect to illegal aliens, and found no express or implied congressional policy favoring the education of illegal aliens. The court therefore concluded that there was no pre-emptive conflict between state and federal law. 628 F.2d, at 451-454. 7 The court concluded that § 21.031 was not pre-empted by federal laws or international agreements. 501 F.Supp., at 584-596. 8 Appellees in both cases continue to press the argument that § 21.031 is pre-empted by federal law and policy. In light of our disposition of the Fourteenth Amendment issue, we have no occasion to reach this claim. 9 It would be incongruous to hold that the United States, to which the Constitution assigns a broad authority over both naturalization and foreign affairs, is barred from invidious discrimination with respect to unlawful aliens, while exempting the States from a similar limitation. See 426 U.S., at 84-86, 96 S.Ct., at 1893-1894. 10 Although we have not previously focused on the intended meaning of this phrase, we have had occasion to examine the first sentence of the Fourteenth Amendment, which provides that "[a]ll persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States. . . ." (Emphasis added.) Justice Gray, writing for the Court in United States v. Wong Kim Ark, 169 U.S. 649, 18 S.Ct. 456, 42 L.Ed. 890 (1898), detailed at some length the history of the Citizenship Clause, and the predominantly geographic sense in which the term "jurisdiction" was used. He further noted that it was "impossible to construe the words 'subject to the jurisdiction thereof,' in the opening sentence [of the Fourteenth Amendment], as less comprehensive than the words 'within its jurisdiction,' in the concluding sentence of the same section; or to hold that persons 'within the jurisdiction' of one of the States of the Union are not 'subject to the jurisdiction of the United States.' " Id., at 687, 18 S.Ct., at 471. Justice Gray concluded that "[e]very citizen or subject of another country, while domiciled here, is within the allegiance and the protection, and consequently subject to the jurisdiction, of the United States." Id., at 693, 18 S.Ct., at 473. As one early commentator noted, given the historical emphasis on geographic territoriality, bounded only, if at all, by principles of sovereignty and allegiance, no plausible distinction with respect to Fourteenth Amendment "jurisdiction" can be drawn between resident aliens whose entry into the United States was lawful, and resident aliens whose entry was unlawful. See C. Bouve, Exclusion and Expulsion of Aliens in the United States 425-427 (1912). 11 In his separate opinion, Justice Field addressed the relationship between the Fifth and Fourteenth Amendments: "The term 'person,' used in the Fifth Amendment, is broad enough to include any and every human being within the jurisdiction of the republic. A resident, alien born, is entitled to the same protection under the laws that a citizen is entitled to. He owes obedience to the laws of the country in which he is domiciled, and, as a consequence, he is entitled to the equal protection of those laws. . . . The contention that persons within the territorial jurisdiction of this republic might be beyond the protection of the law was heard with pain on the argument at the bar—in face of the great constitutional amendment which declares that no State shall deny to any person within its jurisdiction the equal protection of the laws." Wong Wing v. United States, 163 U.S., at 242-243, 16 S.Ct., at 982-983 (concurring in part and dissenting in part). 12 Leng May Ma v. Barber, 357 U.S. 185, 78 S.Ct. 1072, 2 L.Ed.2d 1246 (1958), relied on by appellants, is not to the contrary. In that case the Court held, as a matter of statutory construction, that an alien paroled into the United States pursuant to § 212(d)(5) of the Immigration and Nationality Act, 8 U.S.C. § 1182(d)(5) (1952 ed.), was not "within the United States" for the purpose of availing herself of § 243(h), which authorized the withholding of deportation in certain circumstances. The conclusion reflected the longstanding distinction between exclusion proceedings, involving the determination of admissibility, and deportation proceedings. The undocumented children who are appellees here, unlike the parolee in Leng May Ma, supra, could apparently be removed from the country only pursuant to deportation proceedings. 8 U.S.C. § 1251(a)(2). See 1A C. Gordon & H. Rosenfield, Immigration Law and Procedure § 3.16b, p. 3-161 (1981). 13 Representative Bingham's views are also reflected in his comments on the Civil Rights Bill of 1866. He repeatedly referred to the need to provide protection, not only to the freedmen, but to "the alien and stranger," and to "refugees . . . and all men." Cong. Globe, 39th Cong., 1st Sess., 1292 (1866). 14 Several formulations might explain our treatment of certain classifications as "suspect." Some classifications are more likely than others to reflect deep-seated prejudice rather than legislative rationality in pursuit of some legitimate objective. Legislation predicated on such prejudice is easily recognized as incompatible with the constitutional understanding that each person is to be judged individually and is entitled to equal justice under the law. Classifications treated as suspect tend to be irrelevant to any proper legislative goal. See McLaughlin v. Florida, 379 U.S. 184, 192, 85 S.Ct. 283, 288, 13 L.Ed.2d 222 (1964); Hirabayashi v. United States, 320 U.S. 81, 100, 63 S.Ct. 1375, 1385, 87 L.Ed. 1774 (1943). Finally, certain groups, indeed largely the same groups, have historically been "relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process." San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 28, 93 S.Ct. 1278, 1293, 36 L.Ed.2d 16 (1973); Graham v. Richardson, 403 U.S. 365, 372, 91 S.Ct. 1848, 1852, 29 L.Ed.2d 538 (1971); see United States v. Carolene Products Co., 304 U.S. 144, 152-153, n. 4, 58 S.Ct. 778, 783-784, n. 4, 82 L.Ed. 1234 (1938). The experience of our Nation has shown that prejudice may manifest itself in the treatment of some groups. Our response to that experience is reflected in the Equal Protection Clause of the Fourteenth Amendment. Legislation imposing special disabilities upon groups disfavored by virtue of circumstances beyond their control suggests the kind of "class or caste" treatment that the Fourteenth Amendment was designed to abolish. 15 In determining whether a class-based denial of a particular right is deserving of strict scrutiny under the Equal Protection Clause, we look to the Constitution to see if the right infringed has its source, explicitly or implicitly, therein. But we have also recognized the fundamentality of participation in state "elections on an equal basis with other citizens in the jurisdiction," Dunn v. Blumstein, 405 U.S. 330, 336, 92 S.Ct. 995, 999, 31 L.Ed.2d 274 (1972), even though "the right to vote, per se, is not a constitutionally protected right." San Antonio Independent School Dist., supra, at 35, n. 78, 93 S.Ct., at 1298, n. 78. With respect to suffrage, we have explained the need for strict scrutiny as arising from the significance of the franchise as the guardian of all other rights. See Harper v. Virginia Bd. of Elections, 383 U.S. 663, 667, 86 S.Ct. 1079, 1081, 16 L.Ed.2d 169 (1966); Reynolds v. Sims, 377 U.S. 533, 562, 84 S.Ct. 1362, 1381, 12 L.Ed.2d 506 (1964); Yick Wo v. Hopkins, 118 U.S. 356, 370, 6 S.Ct. 1064, 1071, 30 L.Ed. 220 (1886). 16 See Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976); Lalli v. Lalli, 439 U.S. 259, 99 S.Ct. 518, 58 L.Ed.2d 503 (1978). This technique of "intermediate" scrutiny permits us to evaluate the rationality of the legislative judgment with reference to well-settled constitutional principles. "In expounding the Constitution, the Court's role is to discern 'principles sufficiently absolute to give them roots throughout the community and continuity over significant periods of time, and to lift them above the level of the pragmatic political judgments of a particular time and place.' " University of California Regents v. Bakke, 438 U.S. 265, 299, 98 S.Ct. 2733, 2752, 57 L.Ed.2d 750 (1978) (opinion of POWELL, J.), quoting A. Cox, The Role of the Supreme Court in American Government 114 (1976). Only when concerns sufficiently absolute and enduring can be clearly ascertained from the Constitution and our cases do we employ this standard to aid us in determining the rationality of the legislative choice. 17 The Attorney General recently estimated the number of illegal aliens within the United States at between 3 and 6 million. In presenting to both the Senate and House of Representatives several Presidential proposals for reform of the immigration laws—including one to "legalize" many of the illegal entrants currently residing in the United States by creating for them a special status under the immigration laws—the Attorney General noted that this subclass is largely composed of persons with a permanent attachment to the Nation, and that they are unlikely to be displaced from our territory: "We have neither the resources, the capability, nor the motivation to uproot and deport millions of illegal aliens, many of whom have become, in effect, members of the community. By granting limited legal status to the productive and law-abiding members of this shadow population, we will recognize reality and devote our enforcement resources to deterring future illegal arrivals." Joint Hearing before the Subcommittee on Immigration, Refugees, and International Law of the House Committee on the Judiciary and the Subcommittee on Immigration and Refugee Policy of the Senate Committee on the Judiciary, 97th Cong., 1st Sess., 9 (1981) (testimony of William French Smith, Attorney General). 18 As the District Court observed in No. 80-1538, the confluence of Government policies has resulted in "the existence of a large number of employed illegal aliens, such as the parents of plaintiffs in this case, whose presence is tolerated, whose employment is perhaps even welcomed, but who are virtually defenseless against any abuse, exploitation, or callous neglect to which the state or the state's natural citizens and business organizations may wish to subject them." 458 F.Supp., at 585. 19 We reject the claim that "illegal aliens" are a "suspect class." No case in which we have attempted to define a suspect class, see, e.g., n. 14, supra, has addressed the status of persons unlawfully in our country. Unlike most of the classifications that we have recognized as suspect, entry into this class, by virtue of entry into this country, is the product of voluntary action. Indeed, entry into the class is itself a crime. In addition, it could hardly be suggested that undocumented status is a "constitutional irrelevancy." With respect to the actions of the Federal Government, alienage classifications may be intimately related to the conduct of foreign policy, to the federal prerogative to control access to the United States, and to the plenary federal power to determine who has sufficiently manifested his allegiance to become a citizen of the Nation. No State may independently exercise a like power. But if the Federal Government has by uniform rule prescribed what it believes to be appropriate standards for the treatment of an alien subclass, the States may, of course, follow the federal direction. See DeCanas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976). 20 Because the State does not afford noncitizens the right to vote, and may bar noncitizens from participating in activities at the heart of its political community, appellants argue that denial of a basic education to these children is of less significance than the denial to some other group. Whatever the current status of these children, the courts below concluded that many will remain here permanently and that some indeterminate number will eventually become citizens. The fact that many will not is not decisive, even with respect to the importance of education to participation in core political institutions. "[T]he benefits of education are not reserved to those whose productive utilization of them is a certainty . . . ." 458 F.Supp., at 581, n. 14. In addition, although a noncitizen "may be barred from full involvement in the political arena, he may play a role perhaps even a leadership role—in other areas of import to the community." Nyquist v. Mauclet, 432 U.S. 1, 12, 97 S.Ct. 2120, 2126, 53 L.Ed.2d 63 (1977). Moreover, the significance of education to our society is not limited to its political and cultural fruits. The public schools are an important socializing institution, imparting those shared values through which social order and stability are maintained. 21 If the constitutional guarantee of equal protection was available only to those upon whom Congress affirmatively granted its benefit, the State's argument would be virtually unanswerable. But the Equal Protection Clause operates of its own force to protect anyone "within [the State's] jurisdiction" from the State's arbitrary action. See Part II, supra. The question we examine in text is whether the federal disapproval of the presence of these children assists the State in overcoming the presumption that denial of education to innocent children is not a rational response to legitimate state concerns. 22 Appellant School District sought at oral argument to characterize the alienage classification contained in § 21.031 as simply a test of residence. We are unable to uphold § 21.031 on that basis. Appellants conceded that if, for example, a Virginian or a legally admitted Mexican citizen entered Tyler with his school-age children, intending to remain only six months, those children would be viewed as residents entitled to attend Tyler schools. Tr. of Oral Arg. 31-32. It is thus clear that Tyler's residence argument amounts to nothing more than the assertion that illegal entry, without more, prevents a person from becoming a resident for purposes of enrolling his children in the public schools. A State may not, however, accomplish what would otherwise be prohibited by the Equal Protection Clause, merely by defining a disfavored group as nonresident. And illegal entry into the country would not, under traditional criteria, bar a person from obtaining domicile within a State. C. Bouve, Exclusion and Expulsion of Aliens in the United States 340 (1912). Appellants have not shown that the families of undocumented children do not comply with the established standards by which the State historically tests residence. Apart from the alienage limitation, § 21.031(b) requires a school district to provide education only to resident children. The school districts of the State are as free to apply to undocumented children established criteria for determining residence as they are to apply those criteria to any other child who seeks admission. 23 Although the State has no direct interest in controlling entry into this country, that interest being one reserved by the Constitution to the Federal Government, unchecked unlawful migration might impair the State's economy generally, or the State's ability to provide some important service. Despite the exclusive federal control of this Nation's borders, we cannot conclude that the States are without any power to deter the influx of persons entering the United States against federal law, and whose numbers might have a discernible impact on traditional state concerns. See DeCanas v. Bica, 424 U.S., at 354-356, 96 S.Ct., at 935-936. 24 The courts below noted the ineffectiveness of the Texas provision as a means of controlling the influx of illegal entrants into the State. See 628 F.2d, at 460-461; 458 F.Supp., at 585; 501 F.Supp., at 578 ("The evidence demonstrates that undocumented persons do not immigrate in search for a free public education. Virtually all of the undocumented persons who come into this country seek employment opportunities and not educational benefits. . . . There was overwhelming evidence . . . of the unimportance of public education as a stimulus for immigration") (footnote omitted). 25 Nor does the record support the claim that the educational resources of the State are so direly limited that some form of "educational triage " might be deemed a reasonable (assuming that it were a permissible) response to the State's problems. Id., at 579-581. 1 See, e.g., Perry, Modern Equal Protection: A Conceptualization and Appraisal, 79 Colum.L.Rev. 1023, 1075-1983 (1979). 2 I use the term "citizen" advisedly. The right to vote, of course, is a political interest of concern to citizens. The right to an education, in contrast, is a social benefit of relevance to a substantial number of those affected by Texas' statutory scheme, as is discussed below. 3 The Court concludes that the provision at issue must be invalidated "unless it furthers some substantial goal of the State." Ante, at 224. Since the statute fails to survive this level of scrutiny, as the Court demonstrates, there is no need to determine whether a more probing level of review would be appropriate. 1 Article I, § 8, cl. 4, of the Constitution provides: "The Congress shall have Power . . . To establish an uniform Rule of Naturalization." The Federal Government has "broad constitutional powers in determining what aliens shall be admitted to the United States, the period they may remain, regulation of their conduct before naturalization, and the terms and conditions of their naturalization." Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 419, 68 S.Ct. 1138, 1142, 92 L.Ed. 1478 (1948). See Graham v. Richardson, 403 U.S. 365, 378, 91 S.Ct. 1848, 1855, 29 L.Ed.2d 534 (1971) (regulation of aliens is "constitutionally entrusted to the Federal Government"). The Court has traditionally shown great deference to federal authority over immigration and to federal classifications based upon alienage. See, e.g., Fiallo v. Bell, 430 U.S. 787, 792, 97 S.Ct. 1473, 1477, 52 L.Ed.2d 50 (1977) ("it is important to underscore the limited scope of judicial inquiry into immigration legislation"); Harisiades v. Shaughnessy, 342 U.S. 580, 588-589, 72 S.Ct. 512, 518-519, 96 L.Ed. 586 (1952) ("It is pertinent to observe that any policy toward aliens is vitally and intricately interwoven with contemporaneous policies in regard to the conduct of foreign relations, the war power, and the maintenance of a republican form of government. Such matters are so exclusively entrusted to the political branches of government as to be largely immune from judicial inquiry or interference"). Indeed, even equal protection analysis in this area is based to a large extent on an underlying theme of pre-emption and exclusive federal power over immigration. See Takahashi v. Fish & Game Comm'n, supra, at 420, 68 S.Ct., at 1143 (the Federal Government has admitted resident aliens to the country "on an equality of legal privileges with all citizens under non-discriminatory laws" and the States may not alter the terms of this admission). Compare Graham v. Richardson, supra, and Sugarman v. Dougall, 413 U.S. 634, 93 S.Ct. 2842, 37 L.Ed.2d 853 (1973), with Mathews v. Diaz, 426 U.S. 67, 96 S.Ct. 1883, 48 L.Ed.2d 478 (1976), and Hampton v. Mow Sun Wong, 426 U.S. 88, 96 S.Ct. 1895, 48 L.Ed.2d 495 (1976). Given that the States' power to regulate in this area is so limited, and that this is an area of such peculiarly strong federal authority, the necessity of federal leadership seems evident. 2 I emphasize the Court's conclusion that strict scrutiny is not appropriately applied to this classification. This exacting standard of review has been reserved for instances in which a "fundamental" constitutional right or a "suspect" classification is present. Neither is present in these cases, as the Court holds. 3 THE CHIEF JUSTICE argues in his dissenting opinion that this heightened standard of review is inconsistent with the Court's decision in San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973). But in Rodriguez no group of children was singled out by the State and then penalized because of their parents' status. Rather, funding for education varied across the State because of the tradition of local control. Nor, in that case, was any group of children totally deprived of all education as in these cases. If the resident children of illegal aliens were denied welfare assistance, made available by government to all other children who qualify, this also—in my opinion—would be an impermissible penalizing of children because of their parents' status. 4 The State provides free public education to all lawful residents whether they intend to reside permanently in the State or only reside in the State temporarily. See ante, at 227, n. 22. Of course a school district may require that illegal alien children, like any other children, actually reside in the school district before admitting them to the schools. A requirement of de facto residency, uniformly applied, would not violate any principle of equal protection. 5 The classes certified in these cases included all undocumented school-age children of Mexican origin residing in the school district, see ante, at 206, or the State. See In re Alien Children Education Litigation, 501 F.Supp. 544, 553 (SD Tex.1980). Even so, it is clear that neither class was thought to include mature Mexican minors who were solely responsible for violating the immigration laws. In 458 F.Supp. 569 (ED Tex.1978), the court characterized plaintiffs as "entire families who have migrated illegally." Id., at 578. Each of the plaintiff children in that case was represented by a parent or guardian. Similarly the court in In re Alien Children Education Litigation found that "[u]ndocumented children do not enter the United States unaccompanied by their parents." 501 F.Supp., at 573. A different case would be presented in the unlikely event that a minor, old enough to be responsible for illegal entry and yet still of school age, entered this country illegally on his own volition. 6 In addition, the States' ability to respond on their own to the problems caused by this migration may be limited by the principles of pre-emption that apply in this area. See, e.g., Hines v. Davidowitz, 312 U.S. 52, 61 S.Ct. 399, 85 L.Ed. 581 (1941). In DeCanas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976), the Court found that a state law making it a criminal offense to employ illegal aliens was not pre-empted by federal authority over aliens and immigration. The Court found evidence that Congress intended state regulation in this area. Id., at 361, 96 S.Ct., at 939 ("there is evidence . . . that Congress intends that States may, to the extent consistent with federal law, regulate the employment of illegal aliens"). Moreover, under federal immigration law, only immigrant aliens and nonimmigrant aliens with special permission are entitled to work. See 1 C. Gordon & H. Rosenfield, Immigration Law and Procedure, §§ 1.34a, 1.36, 2.6b (1981). Because federal law clearly indicates that only certain specified aliens may lawfully work in the country and because these aliens have documentation establishing this right, the State in DeCanas was able to identify with certainty which aliens had a federal permission to work in this country. The State did not need to concern itself with an alien's current or future deportability. By contrast, there is no comparable federal guidance in the area of education. No federal law invites state regulation; no federal regulations identify those aliens who have a right to attend public schools. In addition, the Texas educational exclusion requires the State to make predictions as to whether individual aliens eventually will be found to be deportable. But it is impossible for a State to determine which aliens the Federal Government will eventually deport, which the Federal Government will permit to stay, and which the Federal Government will ultimately naturalize. Until an undocumented alien is ordered deported by the Federal Government, no State can be assured that the alien will not be found to have a federal permission to reside in the country, perhaps even as a citizen. Indeed, even the Immigration and Naturalization Service cannot predict with certainty whether any individual alien has a right to reside in the country until deportation proceedings have run their course. See, e.g., 8 U.S.C. §§ 1252, 1253(h), 1254 (1976 ed. and Supp.IV). 1 It does not follow, however, that a state should bear the costs of educating children whose illegal presence in this country results from the default of the political branches of the Federal Government. A state has no power to prevent unlawful immigration, and no power to deport illegal aliens; those powers are reserved exclusively to Congress and the Executive. If the Federal Government, properly chargeable with deporting illegal aliens, fails to do so, it should bear the burdens of their presence here. Surely if illegal alien children can be identified for purposes of this litigation, their parents can be identified for purposes of prompt deportation. 2 The Department of Justice recently estimated the number of illegal aliens within the United States at between 3 and 6 million. Joint Hearing before the Subcommittee on Immigration, Refugees, and International Law of the House Committee on the Judiciary and the Subcommittee on Immigration and Refugee Policy of the Senate Committee on the Judiciary, 97th Cong., 1st Sess., 7 (1981) (testimony of Attorney General Smith). Other estimates run as high as 12 million. See Strout, Closing the Door on Immigration, Christian Science Monitor, May 21, 1982, p. 22, col. 4. 3 The Court implies, for example, that the Fourteenth Amendment would not require a state to provide welfare benefits to illegal aliens. 4 Both the opinion of the Court and Justice POWELL's concurrence imply that appellees are being "penalized" because their parents are illegal entrants. Ante, at 220; ante, at 238-239,and 239. 3 (POWELL, J., concurring). However, Texas has classified appellees on the basis of their own illegal status, not that of their parents. Children born in this country to illegal alien parents, including some of appellees' siblings, are not excluded from the Texas schools. Nor does Texas discriminate against appellees because of their Mexican origin or citizenship. Texas provides a free public education to countless thousands of Mexican immigrants who are lawfully in this country. 5 Appellees "lack control" over their illegal residence in this country in the same sense as lawfully resident children lack control over the school district in which their parents reside. Yet in San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973), we declined to review under "heightened scrutiny" a claim that a State discriminated against residents of less wealthy school districts in its provision of educational benefits. There was no suggestion in that case that a child's "lack of responsibility" for his residence in a particular school district had any relevance to the proper standard of review of his claims. The result was that children lawfully here but residing in different counties received different treatment. 6 Indeed, even children of illegal alien parents born in the United States can be said to be "penalized" when their parents are deported. 7 It is true that the Constitution imposes lesser constraints on the Federal Government than on the states with regard to discrimination against lawfully admitted aliens. E.g., Mathews v. Diaz, 426 U.S. 67, 96 S.Ct. 1883, 48 L.Ed.2d 478 (1976); Hampton v. Mow Sun Wong, 426 U.S. 88, 96 S.Ct. 1895, 48 L.Ed.2d 495 (1976). This is because "Congress and the President have broad power over immigration and naturalization which the States do not possess," Hampton, supra, at 95, 96 S.Ct., at 1901, and because state discrimination against legally resident aliens conflicts with and alters "the conditions lawfully imposed by Congress upon admission, naturalization and residence of aliens in the United States or the several states." Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 419, 68 S.Ct. 1138, 1142, 92 L.Ed. 1478 (1948). However, the same cannot be said when Congress has decreed that certain aliens should not be admitted to the United States at all. 8 In support of this conclusion, the Court's opinion strings together quotations drawn from cases addressing such diverse matters as the right of individuals under the Due Process Clause to learn a foreign language, Meyer v. Nebraska, 262 U.S. 390, 43 S.Ct. 625, 67 L.Ed. 1042 (1923); the First Amendment prohibition against state-mandated religious exercises in the public schools, Abington School District v. Schempp, 374 U.S. 203, 83 S.Ct. 1560, 10 L.Ed.2d 844 (1963); and state impingements upon the free exercise of religion, Wisconsin v. Yoder, 406 U.S. 205, 92 S.Ct. 1526, 32 L.Ed.2d 15 (1972). However, not every isolated utterance of this Court retains force when wrested from the context in which it was made. 9 This "rational basis standard" was applied by the Court of Appeals. 628 F.2d 448, 458-461 (1980). 10 The Texas law might also be justified as a means of deterring unlawful immigration. While regulation of immigration is an exclusively federal function, a state may take steps, consistent with federal immigration policy, to protect its economy and ability to provide governmental services from the "deleterious effects" of a massive influx of illegal immigrants. DeCanas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976); ante, at 228, n. 23. The Court maintains that denying illegal aliens a free public education is an "ineffectual" means of deterring unlawful immigration, at least when compared to a prohibition against the employment of illegal aliens. Ante, at 228-229. Perhaps that is correct, but it is not dispositive; the Equal Protection Clause does not mandate that a state choose either the most effective and all-encompassing means of addressing a problem or none at all. Dandridge v. Williams, 397 U.S. 471, 486-487, 90 S.Ct. 1153, 1162, 25 L.Ed.2d 491 (1970). Texas might rationally conclude that more significant "demographic or economic problem[s]," ante, at 228, are engendered by the illegal entry into the State of entire families of aliens for indefinite periods than by the periodic sojourns of single adults who intend to leave the State after short-term or seasonal employment. It blinks reality to maintain that the availability of governmental services such as education plays no role in an alien family's decision to enter, or remain in, this country; certainly, the availability of a free bilingual public education might well influence an alien to bring his children rather than travel alone for better job opportunities. 11 The Court suggests that the State's classification is improper because "[a]n illegal entrant might be granted federal permission to continue to reside in this country, or even to become a citizen." Ante, at 226. However, once an illegal alien is given federal permission to remain, he is no longer subject to exclusion from the tuition-free public schools under § 21.031. The Court acknowledges that the Tyler Independent School District provides a free public education to any alien who has obtained, or is in the process of obtaining, documentation from the United States Immigration and Naturalization Service. See ante, at 206, n. 2. Thus, Texas has not taken it upon itself to determine which aliens are or are not entitled to United States residence. Justice BLACKMUN's assertion that the Texas statute will be applied to aliens "who may well be entitled to . . . remain in the United States," ante, at 236 (concurring opinion), is wholly without foundation. 12 The Court's opinion is disingenuous when it suggests that the State has merely picked a "disfavored group" and arbitrarily defined its members as nonresidents. Ante, at 227, n. 22. Appellees' "disfavored status" stems from the very fact that federal law explicitly prohibits them from being in this country. Moreover, the analogies to Virginians or legally admitted Mexican citizens entering Texas, ibid., are spurious. A Virginian's right to migrate to Texas, without penalty, is protected by the Constitution, see, e.g., Shapiro v. Thompson, 394 U.S. 618, 89 S.Ct. 1322, 22 L.Ed.2d 600 (1969); and a lawfully admitted alien's right to enter the State is likewise protected by federal law. See Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 68 S.Ct. 1138, 92 L.Ed. 1478 (1948). Cf. Zobel v. Williams, 457 U.S. 55, 102 S.Ct. 2309, 72 L.Ed.2d 672. 13 The District Court so concluded primarily because the State would decrease its funding to local school districts in proportion to the exclusion of illegal alien children. 458 F.Supp., at 577. 14 I assume no Member of the Court would challenge Texas' right to charge tuition to students residing across the border in Louisiana who seek to attend the nearest school in Texas. 15 Professor Bickel noted that judicial review can have a "tendency over time seriously to weaken the democratic process." A. Bickel, The Least Dangerous Branch 21 (1962). He reiterated James Bradley Thayer's observation that " 'the exercise of [the power of judicial review], even when unavoidable, is always attended with a serious evil, namely, that the correction of legislative mistakes comes from the outside, and the people thus lose the political experience, and the moral education and stimulus that comes from fighting the question out in the ordinary way, and correcting their own errors. The tendency of a common and easy resort to this great function, now lamentably too common, is to dwarf the political capacity of the people, and to deaden its sense of moral responsibility.' " Id., at 22 (quoting J. Thayer, John Marshall 106-107 (1901)).

12

457 U.S. 393 102 S.Ct. 2498 73 L.Ed.2d 93 CALIFORNIA, et al., Appellants,v.GRACE BRETHREN CHURCH et al. UNITED STATES, et al., Appellants, v. GRACE BRETHREN CHURCH et al. GRACE BRETHREN CHURCH, et al., Appellants, v. UNITED STATES et al. Nos. 81-31, 81-228, 81-455. Argued March 30, 1982. Decided June 18, 1982. Syllabus The Federal Unemployment Tax Act established a cooperative federal-state scheme to provide benefits to unemployed workers. The Act requires employers to pay an excise tax on wages paid to employees in "covered" employment, but entitles them to a credit on the federal tax for contributions paid into federally approved state unemployment compensation programs. The Act, in 26 U.S.C. § 3309(b), exempts from mandatory state coverage employees of, inter alia, "an organization which is operated primarily for religious purposes and which is operated, supervised, controlled, or principally supported by a church or convention or association of churches." A number of California churches and religious schools, including religious schools unaffiliated with any church, brought suit in Federal District Court to enjoin the Secretary of Labor from conditioning his approval of the California unemployment insurance program on its coverage of plaintiffs' employees, and to enjoin the State from collecting both tax information and the state unemployment compensation tax. The District Court conducted various proceedings and issued several opinions and orders extending over almost a year and a half, in one of which proceedings it rejected the Federal Government's argument that the court was barred from granting injunctive relief by the Tax Injunction Act, which provides that district courts "shall not enjoin, suspend or restrain" the assessment or collection of any state tax where "a plain, speedy and efficient remedy" may be had in the courts of such State. Ultimately, as pertinent here, on the ground that the benefit entitlement decisions for employees of the religious schools unaffiliated with churches risked excessive entanglement with religion in violation of the Establishment Clause of the First Amendment, the court permanently enjoined the state defendants from collecting unemployment taxes from such schools but did not issue an injunction against the federal defendants as to the schools because it had no information as to what response the Secretary of Labor would make to the court's conclusion that the state defendants could not constitutionally impose state unemployment taxes on the employees of such schools. The court said that if the Secretary instituted decertification proceedings against California for failing to collect the taxes on behalf of such employees, the parties could apply to the court for further relief. Held : 1. This Court has jurisdiction to hear these appeals under 28 U.S.C. § 1252, which permits appeals to this Court from a federal-court judgment holding an Act of Congress unconstitutional in any civil action to which the United States or any of its agencies, or any officer or employee thereof, is a party. While the District Court did not expressly hold § 3309(b) unconstitutional as applied to religious schools unaffiliated with churches, the effect of its several opinions and orders was to make "the United States or its officers . . . bound by a holding of unconstitutionality." McLucas v. DeChamplain, 421 U.S. 21, 31, 95 S.Ct. 1365, 1371, 43 L.Ed.2d 699. Pp. 404-407. 2. The Tax Injunction Act deprived the District Court of jurisdiction to issue declaratory and injunctive relief. Pp. 407-419. (a) That Act prohibits declaratory as well as injunctive relief. Because the declaratory judgment procedure "may in every practical sense operate to suspend collection of the state taxes until the litigation is ended," Great Lakes Dredge & Dock Co. v. Huffman, 319 U.S. 293, 299, 63 S.Ct. 1070, 1073, 87 L.Ed. 1407, the very language of the Act—"suspend or restrain" the assessment or collection of state taxes—suggests that a district court is prohibited from issuing declaratory relief in state tax cases. Moreover, because there is little practical difference between injunctive and declaratory relief, it is unlikely that Congress intended to prohibit taxpayers from seeking one form of relief, while permitting them to seek another, thereby defeating the principal purpose of the Tax Injunction Act "to limit drastically" federal-court interference with the assessment and collection of state taxes. Pp. 407-411 (b) A state-court remedy is "plain, speedy and efficient" within the meaning of the Tax Injunction Act only if it "provides the taxpayer with a 'full hearing and judicial determination' at which she may raise any and all constitutional objections to the tax." Rosewell v. LaSalle National Bank, 450 U.S. 503, 514, 101 S.Ct. 1221, 1229, 67 L.Ed.2d 464. Pp. 411-413. (c) Here, because the taxpayers in question could seek a refund of their state unemployment insurance taxes through state administrative and judicial procedures, and thereby obtain state judicial review of their constitutional claims, their remedy under state law was "plain, speedy and efficient" within the meaning of the Tax Injunction Act. There is no merit to the taxpayers' argument that the California refund procedures did not constitute a "plain, speedy and efficient remedy" because their First Amendment claims could be effectively remedied only by injunctive relief and that such relief was unavailable in California. First, under California procedures, the taxpayers should be able to challenge the constitutionality of the unemployment tax in state court before extensive entanglement occurs, and state tax collection agencies can be expected to abide by resulting state-court rulings. Second, to the extent that any entanglement occurs before state review of the constitutional questions, that entanglement would not be reduced by seeking relief instead in the federal courts. Moreover, to carve out a special exception for taxpayers who raise First Amendment claims would undermine the Tax Injunction Act's primary purpose. Pp. 413-417. (d) Where the District Court was without jurisdiction, this Court will not consider the merits of the taxpayers' First Amendment claims. McLucas v. DeChamplain, supra, and Weinberger v. Salfi, 422 U.S. 749, 95 S.Ct. 2457, 45 L.Ed.2d 522, distinguished. Pp. 418-419 Vacated and remanded. Harriet S. Shapiro, Washington, D. C., for the United States, et al. Jeffrey M. Vesely, Los Angeles, Cal., for the State of California, et al. William B. Ball, Harrisburg, Pa., for Grace Brethren Church, et al. Justice O'CONNOR delivered the opinion of the Court. 1 The principal question presented by the parties to these appeals is whether certain state and federal statutes violate the Establishment and Free Exercise Clauses of the First Amendment1 by requiring religious schools unaffiliated with any church to pay unemployment insurance taxes. We do not reach this substantive question, however, holding instead that the Tax Injunction Act, 28 U.S.C. § 1341,2 deprived the District Court of jurisdiction to hear these challenges. Accordingly, we vacate the judgment below. 2 * Last Term, in St. Martin Evangelical Lutheran Church v. South Dakota, 451 U.S. 772, 101 S.Ct. 2142, 68 L.Ed.2d 612 (1981), this Court considered statutory and constitutional challenges to provisions of the Federal Unemployment Tax Act (FUTA), 26 U.S.C. §§ 3301-3311 (1976 ed. and Supp.IV). Because the present claims involve the same provisions that we interpreted in St. Martin, we recount only briefly the substance and legislative history of the relevant statutes before turning to the facts in the present cases. A. 3 In FUTA,3 Congress has authorized a cooperative federal-state scheme to provide benefits to unemployed workers. The Act requires employers to pay an excise tax on wages paid to employees in "covered" employment,4 but entitles them to a credit of up to 90% of the federal tax for contributions they have paid into federally approved state unemployment compensation programs.5 One of the requirements for federal approval is that state programs "cover" certain broad categories of employment. 4 Until 1970, 26 U.S.C. § 3306(c)(8) excluded from the definition of covered employment "service performed in the employ of a religious, charitable, educational, or other [tax exempt] organization." Pub.L. 86-778, § 533, 74 Stat. 984. As a consequence, such organizations were not required to pay either federal excise taxes or state unemployment compensation taxes. In 1970, Congress amended FUTA to require state plans to cover employees of nonprofit organizations, state hospitals, and state institutions of higher education, thus eliminating the broad exemption available to nonprofit organizations.6 See § 3309(a)(1). At the same time, Congress enacted § 3309(b) to exempt from mandatory state coverage a narrow class of religious and educational employees, i.e., Congress exempted services performed. 5 "(1) in the employ of (A) a church or convention or association of churches, or (B) an organization which is operated primarily for religious purposes and which is operated, supervised, controlled, or principally supported by a church or convention or association of churches; 6 "(2) by a duly ordained, commissioned, or licensed minister of a church in the exercise of his ministry or by a member of a religious order in the exercise of duties required by such order; 7 "(3) in the employ of a school which is not an institution of higher education." Pub.L. 91-373, § 104(b)(1), 84 Stat. 698. 8 In 1976, Congress again amended FUTA, this time eliminating the substance of § 3309(b)(3), thereby removing the blanket exemption for school employees. See Unemployment Compensation Amendments of 1976, Pub.L. 94-566, § 115(b)(1), 90 Stat. 2670.7 In order to maintain compliance with FUTA, the States promptly amended their corresponding state programs. See, e.g., Cal.Un.Ins.Code Ann. §§ 634.5(a), (b) (West Supp.1982). B 9 The plaintiffs in these cases, a number of California churches and religious schools, sought to enjoin the Secretary of Labor from conditioning his approval of the California unemployment insurance program on its coverage of the plaintiffs' employees, and to enjoin the State from collecting both tax information and the state tax.8 For the purposes of evaluating their statutory and constitutional claims, the District Court divided the plaintiffs into three classes of employers: Category I represents those schools that are part of the corporate structure of a church or association of churches; Category II includes schools that are separate corporations formed by a church or association of churches; and Category III includes schools that are "operated primarily for religious purposes, but which [are] not operated, supervised, controlled or principally supported by a church or convention or association of churches, i.e., an independent, non-church affiliated religious school." Supplemental Opinion, reprinted in App. to Juris. Statement in No. 81-31, p. 71 (J.S.App.).9 10 On September 21, 1979, the District Court granted a preliminary injunction against the State, restraining it from collecting the state unemployment tax from the Category I plaintiffs. See id., at 51. The basis for the court's order was its conclusion that the plaintiffs were exempt from mandatory state coverage under § 3309(b)(1), and alternatively, that if they were not exempt under the terms of FUTA, collection of the tax from the plaintiffs would involve excessive governmental entanglement with religion, in violation of the Establishment Clause of the First Amendment. See J.S.App. 58-65. 11 In the same opinion, the District Court rejected the Federal Government's argument that, because the state remedy was "plain, speedy and efficient," the Tax Injunction Act, 28 U.S.C. § 1341, barred the court from granting injunctive relief. Considering first the availability of injunctive relief from the state courts, the court concluded that state statutory and constitutional provisions10 made such relief "at best, uncertain." J.S.App. 66. The court then concluded that a state suit for a refund was an inadequate remedy because the plaintiffs claimed not only that their property had been taken unlawfully, but also that the "very process of determining whether any tax is due at all results in a violation of their First Amendment rights." Id., at 67. Because this First Amendment injury was "irreparable" once the taxes had been collected, only an injunction against collection of the tax could remedy the plaintiffs' claims. Accordingly, because there existed no "plain, speedy and efficient" remedy in the state courts, the District Court concluded that it had jurisdiction to grant injunctive and declaratory relief. 12 In a supplemental opinion filed June 2, 1980, the court clarified its earlier opinion, stating expressly that the preliminary injunction covered only Category I plaintiffs. See id., at 71. For the same reasons that it had granted the initial preliminary injunction, however, the court extended the preliminary injunction to Category II plaintiffs. The court continued to deny relief to the Category III plaintiffs after concluding that they were not covered by the statutory exemptions in § 3309(b) and that the risk of excessive governmental entanglement with religion was too small to violate the Establishment Clause. J.S.App. 77-79.11 13 Finally, on April 3, 1981, the court filed a second supplemental opinion ruling on all of the plaintiffs' motions for permanent injunctions enjoining the State from collecting unemployment compensation taxes and the Federal Government from conditioning approval of the state unemployment compensation programs on their inclusion of the plaintiffs' employees. See id., at 1. Considering first the statutory claims, the court concluded that Category I and Category II schools, but not Category III schools, are exempt from coverage under 26 U.S.C. § 3309(b) and the corresponding state provision, Cal.Un.Ins.Code Ann. § 634.5(a) (West Supp. 1982). J.S.App. 3-15.12 The court also found that the benefit entitlement decisions for employees of Category III schools risk excessive governmental entanglement with religion in violation of the Establishment Clause of the First Amendment. Id., at 25-33.13 Consequently, the court held that "constitutional considerations bar the application of the scheme" to the Category III plaintiffs. Id., at 33. 14 Based on these findings, the court issued orders permanently enjoining the federal defendants from requiring state unemployment insurance programs to cover Category I and Category II schools as a precondition for federal approval of the state programs, id., at 47, 51, and permanently enjoining the state defendants from "collecting, or attempting to collect, unemployment compensation . . . taxes" from the Category I, II, or III schools. Id., at 47, 50. The court expressly held Cal.Un.Ins.Code Ann. § 634.5(a) (West Supp. 1982) unconstitutional. See J.S.App. 45, 46. The court did not issue an injunction against the federal defendants as to Category III schools because it 15 "has no information indicating what response, if any, the Secretary will make to the Court's conclusion that the state defendants may not constitutionally impose the state unemployment compensation tax scheme on the Category 3 employees of non-church affiliated schools. . . . If the Secretary, in response to failure by the state defendants to collect unemployment compensation taxes on behalf of Category 3 employees, institutes decertification proceedings against the State of California, the parties may apply to this Court for further relief." Second Supplemental Opinion reprinted in J.S.App. 44, n. 39. 16 Following issuance of the court's injunction, this Court decided St. Martin Evangelical Lutheran Church v. South Dakota, holding that § 3309(b)(1)(A) exempts Category I schools from mandatory coverage under the state unemployment insurance programs. Although no Category II schools were before the Court in St. Martin, the Court noted in a footnote that 17 "[t]o establish exemption from FUTA, a separately incorporated church school (or other organization) must satisfy the requirements of § 3309(b)(1)(B): (1) that the organization 'is operated primarily for religious purposes,' and (2) that it is 'operated, supervised, controlled, or principally supported by a church or convention or association of churches.' " 451 U.S., at 782-783, n. 12, 101 S.Ct., at 2148, n. 12. 18 As a result of this opinion, the Secretary of Labor reconsidered his position and decided that both Category I and Category II schools are statutorily exempt from mandatory coverage under FUTA. Consequently, the federal defendants, as well as the state defendants, have not appealed the District Court's injunction involving Category I and Category II schools, but only that part of the District Court order involving the Category III schools.14 II 19 An initial matter requiring our attention is whether this Court has jurisdiction to hear these appeals.15 Congress has provided that 20 "[a]ny party may appeal to the Supreme Court from an interlocutory or final judgment, decree or order of any court of the United States . . . holding an Act of Congress unconstitutional in any civil action, suit, or proceeding to which the United States or any of its agencies, or any officer or employee thereof, as such officer or employee, is a party." 28 U.S.C. § 1252. 21 The only possible doubt regarding our appellate jurisdiction under this provision is the requirement that the District Court hold "an Act of Congress unconstitutional." 22 In McLucas v. DeChamplain, 421 U.S. 21, 95 S.Ct. 1365, 43 L.Ed.2d 699 (1975), we stated that § 1252 was an unambiguous exception to the policy of minimizing the mandatory docket of this Court. Indeed, the "language of the statute sufficiently demonstrates its purpose: to afford immediate review in this Court in civil actions to which the United States or its officers are parties and thus will be bound by a holding of unconstitutionality." Id., at 31, 95 S.Ct., at 1371. Moreover, this Court has appellate jurisdiction under § 1252 "when the ruling of unconstitutionality is made in the application of the statute to a particular circumstance, . . . rather than upon the challenged statute as a whole." Fleming v. Rhodes, 331 U.S. 100, 102-103, 67 S.Ct. 1140, 1141-1142, 91 L.Ed. 1368 (1947) (discussing the predecessor to § 1252, Act of Aug. 24, 1937, 50 Stat. 751). See United States v. Christian Echoes National Ministry, Inc., 404 U.S. 561, 563, 92 S.Ct. 663, 664, 30 L.Ed.2d 716 (1972) (per curiam ); United States v. Darusmont, 449 U.S. 292, 293, 101 S.Ct. 549, 550, 66 L.Ed.2d 513 (1981). Finally, § 1252 provides jurisdiction even though the lower court did not expressly declare a federal statute unconstitutional, so long as a determination that a statutory provision was unconstitutional "was a necessary predicate to the relief" that the lower court granted. United States v. Clark, 445 U.S. 23, 26, n. 2, 100 S.Ct. 895, 899, n. 2, 63 L.Ed.2d 171 (1980).16 23 In the present case, the District Court did not expressly hold § 3309(b) of FUTA unconstitutional as applied to the Category III appellees,17 but the effect of its several opinions and orders was to make "the United States or its officers . . . bound by a holding of unconstitutionality." McLucas v. DeChamplain, supra, at 31, 95 S.Ct., at 1371. For example, while discussing the Establishment Clause claim of the Category III schools, the District Court held: 24 "Since such entanglement [involving the resolution of questions of faith and doctrine by secular tribunals] is inevitable during the benefit eligibility determination process if religious schools are brought within the scope of the unemployment compensation tax scheme, constitutional considerations bar the application of the scheme to them." Second Supplemental Opinion, reprinted in J.S.App. 33 (emphasis added). 25 Examination of other portions of the court's opinion makes clear that the court's use of the word "scheme" refers to the combined federal and state provisions. See, e.g., id., at 26 (expressly referring to both federal and state statutory provisions in discussing the "unemployment compensation scheme"); id., at 25 (referring to the intent of Congress and the California Legislature in discussing the "unemployment compensation tax scheme"). Moreover, the District Court's analysis leading to its order holding the California provision unconstitutional is based solely on its understanding of the operation and effect of FUTA, which of course prompted the passage of the corresponding state statute in the first place.18 Cf. St. Martin Evangelical Lutheran Church v. South Dakota, 451 U.S., at 780, n. 9, 101 S.Ct., at 2147, n. 9 (holding that the Court could review the South Dakota Supreme Court's interpretation of its unemployment compensation tax statute because its "analysis depended entirely on its understanding of the meaning of FUTA and the First Amendment"). Finally, in its second supplemental opinion, the court made clear that if the Secretary "institutes decertification proceedings against the State of California" for failing to collect unemployment compensation taxes on behalf of Category III employees, "the parties may apply to this Court for further relief," which can only mean injunctive relief against the Secretary. J.S.App. 44, n. 39. Under these circumstances, it is clear that the Secretary is "bound by a holding of unconstitutionality," and that this Court has jurisdiction under § 1252 to hear this appeal. III 26 As we noted above, the District Court declared Cal.Un.Ins.Code Ann. § 634.5(a) (West Supp. 1982) unconstitutional and enjoined the state defendants from collecting state unemployment compensation taxes from the Category III schools.19 In the course of granting this declaratory and injunctive relief, the court expressly rejected the Federal Government's argument that the Tax Injunction Act, 28 U.S.C. § 1341, deprived the court of jurisdiction. See J.S.App. 65-69. Consequently, before reaching the merits of the appellees' claim, we must decide whether the District Court correctly ruled that it had jurisdiction under the Tax Injunction Act to issue declaratory and injunctive relief. 27 The Tax Injunction Act states simply that the district courts "shall not enjoin, suspend or restrain the . . . collection of any tax under State law where a plain, speedy and efficient remedy may be had in the courts of such State." It is plain from this language that the Tax Injunction Act prohibits a federal district court, in most circumstances, from issuing an injunction enjoining the collection of state taxes. Although this Court once reserved the question,20 we now conclude that the Act also prohibits a district court from issuing a declaratory judgment holding state tax laws unconstitutional. 28 Initially, we observe that the Act divests the district court not only of jurisdiction to issue an injunction enjoining state officials, but also of jurisdiction to take actions that "suspend or restrain" the assessment and collection of state taxes. Because the declaratory judgment "procedure may in every practical sense operate to suspend collection of the state taxes until the litigation is ended," Great Lakes Dredge & Dock Co. v. Huffman, 319 U.S. 293, 299, 63 S.Ct. 1070, 1073, 87 L.Ed. 1407 (1943), the very language of the Act suggests that a federal court is prohibited from issuing declaratory relief in state tax cases.21 Additionally, because there is little practical difference between injunctive and declaratory relief, we would be hard pressed to conclude that Congress intended to prohibit taxpayers from seeking one form of anticipatory relief against state tax officials in federal court, while permitting them to seek another, thereby defeating the principal purpose of the Tax Injunction Act: "to limit drastically federal district court jurisdiction to interfere with so important a local concern as the collection of taxes." Rosewell v. LaSalle National Bank, 450 U.S. 503, 522, 101 S.Ct. 1221, 1233, 67 L.Ed.2d 464 (1981).22 As Justice BRENNAN stated in his opinion concurring in part and dissenting in part in Perez v. Ledesma, 401 U.S. 82, 128, n. 17, 91 S.Ct. 674, 698, n. 17, 27 L.Ed.2d 701 (1971): 29 "If federal declaratory relief were available to test state tax assessments, state tax administration might be thrown into disarray, and taxpayers might escape the ordinary procedural requirements imposed by state law. During the pendency of the federal suit the collection of revenue under the challenged law might be obstructed, with consequent damage to the State's budget, and perhaps a shift to the State of the risk of taxpayer insolvency. Moreover, federal constitutional issues are likely to turn on questions of state tax law, which, like issues of state regulatory law, are more properly heard in the state courts." 30 See Fair Assessment in Real Estate Assn., Inc. v. McNary, 454 U.S. 100, 108-109, n. 6, 102 S.Ct. 177, 182, n. 6, 70 L.Ed.2d 271 (1981).23 31 Consequently, because Congress' intent in enacting the Tax Injunction Act was to prevent federal-court interference with the assessment and collection of state taxes, we hold that the Act prohibits declaratory as well as injunctive relief. Accordingly, the District Court in these cases was without jurisdiction to declare the California tax provision unconstitutional or to issue its injunction against state authorities unless the appellees had no "plain, speedy and efficient remedy" in the state courts. 32 Last Term, in Rosewell v. LaSalle National Bank, this Court had occasion to consider the meaning of the "plain, speedy and efficient" exception in the Tax Injunction Act. After reviewing previous decisions24 and the legislative history of the Act,25 the Court concluded that the "plain, speedy and efficient" exception requires the "state-court remedy [to meet] certain minimal procedural criteria." 450 U.S., at 512, 101 S.Ct., at 1228 (emphasis in original). In particular, a state-court remedy is "plain, speedy and efficient" only if it "provides the taxpayer with a 'full hearing and judicial determination' at which she may raise any and all constitutional objections to the tax." Id., at 514, 101 S.Ct., at 1229 (quoting LaSalle National Bank v. County of Cook, 57 Ill.2d 318, 324, 312 N.E.2d 252, 255-256 (1974)).26 Applying these considerations, the Rosewell Court held that an Illinois tax scheme, requiring the taxpayer to pay an allegedly unconstitutional tax27 and seek a refund through state administrative and judicial procedures, was a "plain, speedy and efficient remedy" within the meaning of the Tax Injunction Act. In reaching this holding, the Court specifically relied on legislative Reports demonstrating congressional awareness that refunds were the exclusive remedy in many state tax systems.28 33 The holding in Rosewell reflects not only Congress' express command in the Tax Injunction Act, but also the historical reluctance of the federal courts to interfere with the operation of state tax systems if the taxpayer had available an adequate remedy in the state courts. As this Court stated in Dows v. City of Chicago, 11 Wall. 108, 110, 78 U.S. 108, 110, 20 L.Ed. 65 (1871), long before enactment of the Tax Injunction Act: 34 "No court of equity will . . . allow its injunction to issue to restrain [state officers collecting state taxes], except where it may be necessary to protect the rights of the citizen whose property is taxed, and he has no adequate remedy by the ordinary processes of the law. It must appear that the enforcement of the tax would lead to a multiplicity of suits, or produce irreparable injury, . . . before the aid of a court of equity can be invoked."29 35 In order to accommodate these concerns and be faithful to the congressional intent "to limit drastically" federal-court interference with state tax systems, we must construe narrowly the "plain, speedy and efficient" exception to the Tax Injunction Act. 36 With these cases and principles in mind, we turn to the California provisions to determine whether there exists a "plain, speedy and efficient" state remedy for the appellees' claim. B 37 There is no dispute that appellees in the present cases can seek a refund of the California unemployment tax through state administrative and judicial procedures. Once a taxpayer has sought from, and been denied a refund by, the appropriate state agency, see Cal.Un.Ins.Code Ann. §§ 1176-1185 (West 1972 and Supp.1982),30 he may file an action in Superior Court for a refund of the taxes paid, raising all arguments against the validity of the tax. Cal.Un.Ins.Code Ann. § 1241 (West Supp.1982). If the taxpayer is unsuccessful at trial, he may appeal the decision to higher state courts and ultimately seek review in this Court. Nothing in this scheme prevents the taxpayer from "rais[ing] any and all constitutional objections to the tax" in the state courts. Rosewell v. LaSalle National Bank, 450 U.S., at 514, 101 S.Ct., at 1229. As the Court in Rosewell noted, the "Act contemplates nothing more." Id., at 516, n. 19, 101 S.Ct., at 1230, n. 19.31 Moreover, assuming that the appellees' constitutional claims are meritorious, an issue on which we express no view, there is every reason to believe that once a state appellate court has declared the tax unconstitutional the appropriate state agencies will respect that declaration. See Pacific Motor Transport Co. v. State Bd. of Equalization, 28 Cal.App.3d 230, 236, 104 Cal.Rptr. 558, 562 (1972) (noting that while the "relief afforded may not 'prevent or enjoin' or otherwise hamper present or future tax assessment or collection effort . . . [i]t will be presumed that the governmental agency will respect a judicial declaration concerning a regulation's validity"). Accordingly, it appears that Rosewell is directly applicable to the present cases, and that the District Court had no jurisdiction to hear the appellees' claims. 38 The appellees contend, however, that the California refund procedures do not constitute a "plain, speedy and efficient remedy" because their claims can be remedied only by injunctive relief, and that such relief is unavailable in California courts to restrain the collection of state taxes. See n. 10, supra. Injunctive relief is necessary, the appellees claim, because prior to state judicial review, the employer must meet certain recordkeeping, registration, and reporting requirements, see Cal.Un.Ins.Code Ann. §§ 1085, 1086, 1088, 1092 (West 1972 and Supp. 1982), and potentially is subject to administrative benefit eligibility hearings32 in violation of the appellees' First Amendment rights. The appellees thus fear that their constitutional rights will be violated before they have an opportunity to challenge the constitutionality of the unemployment tax scheme in state court. 39 This argument is unpersuasive. First, nothing in the California scheme precludes the appellees from challenging the unemployment tax before a benefit eligibility hearing is held for one of their former employees. As soon as an employer makes its first payment to the state unemployment insurance fund, it may file for a refund and, after exhausting state administrative remedies, seek a judicial determination of the constitutionality of the tax.33 If the employer ultimately prevails on his constitutional argument, the state taxing authorities can be expected to respect that court's holding in future administrative proceedings. See Pacific Motor Transport Co. v. State Board of Equalization, supra, at 236, 104 Cal.Rptr., at 562. Thus, before any entanglement from the benefit eligibility hearings occurs, the appellees should be able to challenge the constitutionality of the state unemployment insurance taxes. 40 Second, while an employer may be subject to some recordkeeping and reporting requirements, or even a benefit eligibility hearing, pending the resolution of its constitutional claims in state court, it will be subject to the same burdens even if it seeks relief from the federal courts. Thus, whatever harm the appellees may suffer pending resolution of their constitutional claims, that harm is not reduced by seeking relief in federal court. Stated differently, there are no apparent advantages to federal-court relief that make state-court remedies less than "plain, speedy and efficient."34 41 Finally, we must keep in mind that at the time that it passed the Tax Injunction Act, Congress was well aware that refund procedures were the sole remedy in many States for unlawfully collected taxes. See S.Rep.No.1035, 75th Cong., 1st Sess., 1 (1937); H.R.Rep.No.1503, 75th Cong., 1st Sess., 2 (1937).35 Carving out a special exception for taxpayers raising First Amendment claims would undermine significantly Congress' primary purpose "to limit drastically federal district court jurisdiction to interfere with so important a local concern as the collection of taxes." Rosewell v. LaSalle National Bank, 450 U.S., at 522, 101 S.Ct., at 1233.36 Because we do not believe that Congress intended federal injunctions and declaratory judgments to disrupt state tax administration when state refund procedures are available, we decline to find an exception in the Tax Injunction Act for the appellees' claims.37 Accordingly, because the appellees could seek a refund of their state unemployment insurance taxes, and thereby obtain state judicial review of their constitutional claims, we hold that their remedy under state law was "plain, speedy and efficient" within the meaning of the Tax Injunction Act, and consequently, that the District Court had no jurisdiction to issue injunctive or declaratory relief.38 C 42 Despite the absence of jurisdiction in the District Court, the federal defendants urge us to consider the merits of the appellees' First Amendment claims because of the "public interest in, and the Secretary's need for, a definitive interpretation of 26 U.S.C. § 3309(b)." Brief for United States 21. The Government bases this argument on our decision in McLucas v. DeChamplain, 421 U.S., at 32, 95 S.Ct., at 1372, in which we held that "whether the District Court did or did not have jurisdiction to act, this case is properly here under § 1252." See also Weinberger v. Salfi, 422 U.S. 749, 763, n. 8, 95 S.Ct. 2457, 2466, n. 8, 45 L.Ed.2d 522 (1975). 43 The Government's argument is unavailing, however, for in McLucas and Salfi, some federal trial court had jurisdiction,39 whereas in the present cases, no federal district court had jurisdiction. If this Court were nonetheless to reach the First Amendment issues presented in these appeals, the litigants would have sidestepped neatly Congress' intent and our longstanding policy "to limit drastically" federal interference in the administration of state taxes when a "plain, speedy and efficient" state remedy is available.40 Accordingly, we do not reach the appellees' First Amendment claims. 44 The judgment of the District Court is vacated, and the cases are remanded for further proceedings consistent with this opinion. 45 So ordered. 46 Justice STEVENS, with whom Justice BLACKMUN joins, dissenting. 47 Appellee Grace Brethren Church filed suit against the United States Secretary of Labor and other defendants in the United States District Court for the Central District of California claiming that the Federal Unemployment Tax Act violates the First Amendment to the Federal Constitution. The District Court held the Act unconstitutional. Pursuant to a federal statute providing for expedited review in cases of this kind,1 the defendants appealed directly to this Court, byassing the Court of Appeals precisely as Congress intended. Recognizing the need for prompt review of a constitutional question affecting the nationwide operation of a federal statute, the Court holds that this special jurisdiction was properly invoked. It then reaches the curious conclusion that, because some of the defendants are California taxing authorities that administer this cooperative federal-state program in that State, Congress intended that only state courts could pass on the constitutional validity of this federal statute.2 Neither the language nor the legislative history of the Tax Injunction Act requires such a strange result. 48 The Tax Injunction Act provides that federal district courts "shall not enjoin, suspend, or restrain" the activities of state taxing authorities. The preclusion of federal injunctive relief was a response to a specific problem that concerned Congress in 1937. In the States in which taxpayers were required to challenge a tax assessment in a refund suit, only taxpayers that could sue state taxing authorities in federal court could obtain injunctive relief. The privileged taxpayers were primarily the foreign corporations that could invoke federal diversity jurisdiction. These federal suits were objectionable not only because of this discrimination but also because state treasuries often were deprived of tax revenues while the federal suits were adjudicated and because the federal suits involved only state-law questions that were more appropriate for state-court resolution. See S.Rep.No.1035, 75th Cong., 1st Sess., 1-2 (1937); 81 Cong.Rec. 1416-1417 (1937) (Sen. Bone); see also Rosewell v. LaSalle National Bank, 450 U.S. 503, 533, 101 S.Ct. 1221, 1239, 67 L.Ed.2d 464 (STEVENS, J., dissenting). 49 A literal reading of the Tax Injunction Act manifestly does not preclude the declaratory judgment entered in this litigation. Nor do the concerns that gave rise to its enactment require such a bar. Appellees' challenge is based on the Federal Constitution and is directed at a federal-state program administered according to federal requirements. Only federal questions are involved. 50 In Great Lakes Dredge & Dock Co. v. Huffman, 319 U.S. 293, 63 S.Ct. 1070, 87 L.Ed. 1407, the Court recognized that equitable considerations can in some cases provide adequate justification for federal courts to withhold declaratory relief when the Tax Injunction Act precluded injunctive relief. In the intervening 40 years, this equitable doctrine has been sufficient to protect state tax laws from unnecessary federal interference. Today, however, the Court confronts a situation in which the challenge to a state tax does not implicate these concerns.3 Ironically, the absence in these unusual cases of the traditional justification for a ban on declaratory relief seems to spur the Court to revise the Tax Injunction Act to preclude declaratory as well as injunctive relief. To accomplish this revision, the Court must ignore the plain meaning of the statute and the limited concerns that gave rise to its enactment. The Court instead relies upon the legislative history of the Johnson Act, ch. 283, 48 Stat. 775, see ante, at 409-410, n. 22, even though that statute was enacted before the Declaratory Judgment Act, ch. 512, 48 Stat. 955. Even if that suspect analysis could be overlooked, the fact remains that, after the Declaratory Judgment Act was on the books for three years, Congress did not see fit to bar declaratory relief when it expressly precluded injunctive relief in the Tax Injunction Act. The avoidance of a decision on the merits in this litigation hardly seems worth the Court's nimble exercise in lawmaking.4 51 The Court has both the power and the duty to decide the merits. I therefore respectfully dissent. 1 The First Amendment provides in pertinent part that "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof." The Free Exercise and Establishment Clauses apply to the States through the Due Process Clause of the Fourteenth Amendment. See Cantwell v. Connecticut, 310 U.S. 296, 303, 60 S.Ct. 900, 903, 84 L.Ed. 1213 (1940); Everson v. Board of Education, 330 U.S. 1, 15, 67 S.Ct. 504, 511, 91 L.Ed. 711 (1947). 2 The Act provides: "The district courts shall not enjoin, suspend or restrain the assessment, levy or collection of any tax under State law where a plain, speedy and efficient remedy may be had in the courts of such State." 3 FUTA was enacted originally as Title IX of the Social Security Act of 1935, ch. 531, 49 Stat. 639. 4 See 26 U.S.C. § 3301. 5 See 26 U.S.C. § 3302 (1976 ed. and Supp.IV). Each state program receives annual approval after the Secretary of Labor finds that it complies with federal statutory standards. See 26 U.S.C. §§ 3304(a), (c) (1976 ed. and Supp.IV). The federal standards for the state programs are contained in §§ 3304 and 3309. If a state plan complies with federal standards, the State is authorized to receive a federal grant to administer the state plan. See 29 U.S.C. § 49d(b); 42 U.S.C. § 501. 6 See Employment Security Amendments of 1970, Pub.L. 91-373, § 104(b)(1), 84 Stat. 697. Under §§ 3309(a)(2) and 3304(a)(6)(B), such nonprofit organizations were given the option of either making the same contribution to the state unemployment compensation fund required of other employers, or reimbursing the fund for unemployment compensation payments actually made to the nonprofit organizations' former employees. Although nonprofit organizations were covered by federally approved state unemployment compensation laws, they continued to be exempt from the federal excise tax on wages because the definition of "employment" in § 3306(c)(8), excluding services performed for such organizations, remained unchanged. 7 In its place, Congress substituted an unrelated provision. 8 This litigation grew out of two suits, one filed in the District Court by Grace Brethren Church et al. (Case No. CV 79-93 MRP), and the other filed in state court by the Lutheran Church Missouri Synod. The Secretary of Labor successfully removed the Lutheran Church case (Case No. CV 79-162 MRP) to the District Court, which consolidated the cases for trial. 9 Category I and II schools comprise schools from the Lutheran Church case, see Order (filed Apr. 3, 1981), reprinted in J.S.App. 49, as well as some of the schools from the Grace Brethren case. See Order (filed Apr. 3, 1981), reprinted in J.S.App. 45. Category III schools include only schools from the Grace Brethren case. See J.S.App. 46. 10 California Un.Ins.Code Ann. § 1851 (West 1972) provides: "No injunction or writ of mandate or other legal or equitable process shall issue in any suit, action or proceeding, in any court against this State or against any officer thereof to prevent or enjoin the collection of any contribution sought to be collected under this division." California Const., Art. XIII, § 32, provides: "No legal or equitable process shall issue in any proceeding in any court against this State or any officer thereof to prevent or enjoin the collection of any tax. After payment of a tax claimed to be illegal, an action may be maintained to recover the tax paid, with interest, in such manner as may be provided by the Legislature." Despite the apparently unambiguous language of these provisions, the District Court considered the availability of injunctive relief only "uncertain" because of state decisions indicating that injunctive relief may be available when the plaintiff challenges the state tax law as being unconstitutional. See Las Animas & San Joaquin Land Co. v. Preciado, 167 Cal. 580, 587, 140 P. 239, 242 (1914) (injunction available to restrain a school district from assessing property taxes on land over which it has no authority); Bueneman v. City of Santa Barbara, 8 Cal.2d 405, 407, 65 P.2d 884, 886 (1937) (statutory provision precluding courts from enjoining execution of public laws for public benefit does not apply to claims that a taxing statute is unconstitutional). More recent decisions, however, have held injunctive relief to be precluded. See Modern Barber Colleges, Inc. v. California Employment Stabilization Comm'n, 31 Cal.2d 720, 723, 192 P.2d 916, 918 (1948) (holding that a provision in the Unemployment Insurance Act, similar to § 1851, prohibited injunctive relief, leaving the taxpayer only with the option to pay the tax and seek a refund); Aronoff v. Franchise Tax Board, 60 Cal.2d 177, 180, 32 Cal.Rptr. 1, 383 P.2d 409, 411 (1963) (holding that Cal.Const., Art. XIII, § 15, and Cal.Rev. & Tax Code Ann. § 19081 (West 1970) preclude issuance of an injunction to prevent collection of additional income taxes). Relying on Aronoff, a District Court of Appeal held that Cal.Const., Art. XIII, § 32 (which, in 1974, became the successor to § 15), and the corresponding statutory provision, Cal.Un.Ins.Code Ann. § 1851 (West 1972), prohibit the courts from enjoining the collection of unemployment insurance taxes. Lorco Properties, Inc. v. Department of Benefit Payments, 57 Cal.App.3d 809, 815, 129 Cal.Rptr. 312, 315 (1976). Recently, in Pacific Gas & Electric Co. v. State Board of Equalization, 27 Cal.3d 277, 279, 165 Cal.Rptr. 122, 123, 611 P.2d 463, 464 (1980), the California Supreme Court held that under Cal.Const., Art. XIII, § 32, a taxpayer was barred from seeking relief compelling the state tax board to adjust the taxpayer's real property assessments. The court expressly held that there were no equitable exceptions to this rule, id., at 282, 165 Cal.Rptr., at 125, 611 P.2d, at 466, and reaffirmed the importance of the state policy to permit the uninterrupted collection of taxes. Cf. Pacific Motor Transport Co. v. State Board of Equalization, 28 Cal.App.3d 230, 236, 104 Cal.Rptr. 558, 562 (1972) (noted without approval in Pacific Gas & Electric Co. v. State Board of Equalization, supra, and holding that a taxpayer could seek declaratory relief to challenge the validity of a tax regulation, but that such relief could not " 'prevent or enjoin' or otherwise hamper present or future tax assessment or collection effort"). 11 The court also rejected the arguments offered by the Category III plaintiffs that imposition of the tax violates the Free Exercise Clause, and that the unique statutory treatment of Category III plaintiffs violates equal protection. J.S.App. 78. 12 The court held alternatively that if the Secretary of Labor's interpretation of § 3309(b) were correct (i.e., Category I and II schools were not exempt from coverage), then that provision violated the First Amendment because it caused excessive governmental entanglement with religion by requiring "[i]ntrusive monitoring of the activities of employees of religious schools in order to determine whether or not those employees are exempt from unemployment insurance . . . taxes" and by requiring "[i]nvolvement of state officials in the resolution of questions of religious doctrine in the course of determining the benefit eligibility of discharged employees of religious schools." Order (filed Apr. 3, 1981), reprinted in J.S.App. 45, 46; Order (filed Apr. 3, 1981), reprinted in id., at 49, 50. 13 The court again rejected the plaintiffs' argument that statutory coverage of Category III schools violates the Free Exercise Clause of the First Amendment, id., at 16-25, and found it unnecessary to reach the Category III plaintiffs' equal protection claim. Id., at 35. 14 See J.S.App. 11-12; Juris. Statement in No. 81-228, pp. 4, n. 2, 6, n. 5. The Category III schools are parties only in the Grace Brethren case, the suit originally filed in federal court. See n. 8, supra. The Grace Brethren appellees filed a cross-appeal (No. 81-455) claiming that the District Court erred in holding that FUTA and the corresponding California statutory provisions do not violate the Free Exercise Clause of the First Amendment. The cross-appeal, however, is unnecessary to preserve this argument since under this Court's Rule 10.5 "an appellee, without filing a cross-appeal, [may] defend a judgment on any ground that the law and record permit and that would not expand the relief he has been granted." The plaintiffs in the Lutheran Church case have filed a brief in support of the judgment below. Because, however, neither the State nor the Federal Government appealed from that part of the judgment involving the Lutheran Church plaintiffs, we do not address their claims. 15 In our order setting these cases for oral argument, we postponed the question of jurisdiction until consideration of the merits. See 454 U.S. 961, 102 S.Ct. 499, 70 L.Ed.2d 376 (1981). 16 In Clark, the Court of Claims simply ordered relief based on its earlier decision in another case. In that earlier decision, the court had declared the challenged statutory provision unconstitutional. See Gentry v. United States, 212 Ct.Cl. 1, 546 F.2d 343 (1976), rehearing denied, 212 Ct.Cl. 27, 551 F.2d 852 (1977). 17 See Order (filed Apr. 3, 1981), reprinted in J.S.App. 45, 46 (holding Cal.Un.Ins.Code Ann. § 634.5(a) (West Supp. 1982) unconstitutional, but making no direct reference to § 3309(b)). 18 The court's analysis of Category I and II schools also demonstrates that it believed FUTA, as applied to Category III schools, to be unconstitutional. In its discussion of Category I and II schools, the court held that if it were to follow the Secretary's interpretation of § 3309, i.e., if no exemption existed, then FUTA would be unconstitutional as applied to those schools in part because of the excessive governmental entanglement in the benefit eligibility hearing. See n. 12, supra. Since the court also found an entanglement problem with respect to benefit eligibility hearings for Category III schools, and since there is no statutory exemption for those schools, it follows that the District Court must have believed that FUTA was unconstitutional as applied to the Category III plaintiffs. 19 No federal tax is involved in this case, for the services performed for Category III schools are exempted by § 3306(c)(8) from the definition of employment for which the federal excise tax must be paid. 20 See Great Lakes Dredge & Dock Co. v. Huffman, 319 U.S. 293, 299, 63 S.Ct. 1070, 1073, 87 L.Ed. 1407 (1943). 21 In enacting the Declaratory Judgment Act, Congress recognized the substantial effect declaratory relief would have on legal disputes. Thus, while Congress perceived declaratory judgments as a device to reduce federal-court abuses associated with injunctions, Congress also recognized that declaratory relief would "settle controversies," S.Rep.No. 1005, 73d Cong., 2d Sess., 2 (1934), and permit the federal courts "the power to exercise in some instances preventive relief." H.R.Rep.No.1264, 73d Cong., 2d Sess., 2 (1934). 22 To be sure, in enacting the Tax Injunction Act, Congress considered primarily injunctions against state officials because that form of anticipatory relief was the principal weapon used by businesses to delay or avoid paying state taxes. See, e.g., S.Rep.No.1035, 75th Cong., 1st Sess., 1-2 (1937); 81 Cong.Rec. 1416 (1937) (remarks of Sen. Bone). Moreover, it is arguable that Congress' failure to mention the Declaratory Judgment Act, enacted only three years earlier, indicates that Congress intended the Tax Injunction Act to prohibit only federal injunctive relief. Nevertheless, the legislative history of the Tax Injunction Act demonstrates that Congress worried not so much about the form of relief available in the federal courts, as about divesting the federal courts of jurisdiction to interfere with state tax administration. Both the Senate and House Reports, as well as the congressional debates of the Act, expressly rely on the congressional purpose underlying the Johnson Act of 1934, 28 U.S.C. § 1342, which divests the district courts of jurisdiction of any suit to "enjoin, suspend, or restrain the operation" of any public utility commission order. See S.Rep.No.1035, supra, at 2; H.R.Rep.No.1503, 75th Cong., 1st Sess., 2 (1937); 81 Cong.Rec. 1415-1417 (1937) (remarks of Sen. Bone). The legislative history of the Johnson Act, in turn, makes clear that its purpose was to prevent public utilities from going to federal district court to challenge state administrative orders or avoid state administrative and judicial proceedings. See, e.g., S.Rep.No.125, 73d Cong., 1st Sess., 3 (1933) (in support of the Johnson bill, declaring that a utility "will be required, in all cases where a State has set up a public utility commission, to proceed before that commission if it has any complaint. It can appeal from this State board to the State courts and, if it is dissatisfied with the final judgment of the supreme court of the State, it can take an appeal to the Supreme Court of the United States"); id., at 33 ("It is the jurisdiction which Congress has given to Federal courts to pass on matters of State regulation which holds up the laws of the States, prevents the officials of the States from doing their duty, and robs the people of the benefit which would accrue to them, if the commissions which they have set up by law in the various States were permitted to perform their duty"); H.R.Rep.No.1194, 73d Cong., 2d Sess., 2 (1934) (in opposition to the Johnson bill, declaring that it "seeks to withdraw completely from the district courts of the United States all jurisdiction in suits relating to orders of State administrative boards or commissions affecting rates chargeable by public utilities"); 78 Cong.Rec. 1916 (1934) (remarks of Sen. Johnson); id., at 1918 ("the object is to make [the utilities] subject to the jurisdiction of the laws of our States; to give them their rights in every instance to the trial of the question of fact first before the public-utility commission, to give them every legal right they have, and if any right that is guaranteed by the Constitution is infringed upon at all, then, of course, the legal right of appeal ultimately from the highest tribunal in the State to the United States Supreme Court"); id., at 8324 (remarks of Rep. Mapes) ("It is simply a question as to whether or not States are going to be allowed to perform their proper functions in the supervision and fixing of rates, without interference of Federal law. It is a question as to whether or not Congress is going to continue to permit the utilities in important cases to thwart the will of the States and the State authorities. . . . This bill will only deprive the lower Federal courts of the jurisdiction they now have over rate cases"); id., at 8328 (remarks of Rep. Lewis) ("The Johnson bill absolutely abolishes the jurisdiction of the United States courts in rate cases"); id., at 8338 (remarks of Rep. Tarver) ("The Johnson bill contains but one substantive proposition, and that is to divest the district courts of the United States of jurisdiction in public-utility rate cases"); id., at 8419 (remarks of Rep. Hancock) ("the Johnson bill seeks to [save time and money] by divesting the Federal courts of all jurisdiction in public-utility cases except the right of appeal to the Supreme Court of the United States after the final decision of the State court of last resort"). 23 This Court has long recognized the dangers inherent in disrupting the administration of state tax systems. See, e.g., Dows v. City of Chicago, 11 Wall. 108, 110, 78 S.Ct. 108, 110, 20 L.Ed. 65 (1871) ("It is upon taxation that the several States chiefly rely to obtain the means to carry on their respective governments, and it is of the utmost importance to all of them that the modes adopted to enforce the taxes levied should be interfered with as little as possible. Any delay in the proceedings of the officers, upon whom the duty is devolved of collecting the taxes, may derange the operations of government, and thereby cause serious detriment to the public"). 24 See Tully v. Griffin, Inc., 429 U.S. 68, 74, 97 S.Ct. 219, 223, 50 L.Ed.2d 227 (1976); Hillsborough v. Cromwell, 326 U.S. 620, 625, 66 S.Ct. 445, 449, 90 L.Ed. 358 (1946); Great Lakes Dredge & Dock Co. v. Huffman, 319 U.S., at 300-301, 63 S.Ct., at 1074. 25 See 81 Cong.Rec. 1416 (1937) (remarks of Sen. Bone); S.Rep.No.1035, 75th Cong., 1st Sess., 2 (1937). The Court also relied on the legislative history of the Johnson Act of 1934, 28 U.S.C. § 1342 (prohibiting federal-court interference with orders issued by state administrative agencies to public utilities), on which the Tax Injunction Act was modeled. 26 See also 450 U.S., at 515, and n. 19, 517, 101 S.Ct. 1230, and n. 19, 1231 (making clear that some opportunity to raise constitutional objections is the most important consideration); S.Rep.No.1035, supra, at 2 (under the Tax Injunction Act, a "full hearing and judicial determination of the controversy is assured. An appeal to the Supreme Court of the United States is available as in other cases"). 27 The plaintiff in Rosewell had claimed that requiring payment of the county property tax violated her equal protection and due process rights. 28 See S.Rep.1035, supra, at 1 (state "statutes generally provide that taxpayers may contest their taxes only in refund actions after payment under protest"); H.R.Rep.No.1503, 75th Cong., 1st Sess., 2 (1937). 29 See also Boise Artesian Hot and Cold Water Co. v. Boise City, 213 U.S. 276, 282, 29 S.Ct. 426, 428, 53 L.Ed. 796 (1909) (holding that "the illegality or unconstitutionality of a state or municipal tax or imposition is not of itself a ground for equitable relief in the courts of the United States. In such a case the aggrieved party is left to his remedy at law, when that remedy is as complete, practicable and efficient as the remedy in equity"); Singer Sewing Machine Co. v. Benedict, 229 U.S. 481, 488, 33 S.Ct. 942, 944, 57 L.Ed. 1288 (1913) (holding that federal courts will not enjoin the collection of unconstitutional state taxes where the taxpayer "ha[s] a plain, adequate and complete remedy" at law); Great Lakes Dredge & Dock Co. v. Huffman, supra, at 299, 63 S.Ct., at 1073 (holding that the same "considerations which have led federal courts of equity to refuse to enjoin the collection of state taxes, save in exceptional cases, require a like restraint in the use of the declaratory judgment procedure"); Fair Assessment in Real Estate Assn., Inc. v. McNary, 454 U.S. 100, 103, 102 S.Ct. 177, 179, 70 L.Ed.2d 271 (1981) (noting that the Tax Injunction Act, "and the decisions of this Court which preceded it, reflect the fundamental principle of comity between federal courts and state governments that is essential to 'Our Federalism,' particularly in the area of state taxation"). 30 Apparently, California taxpayers cannot raise their constitutional challenges in the administrative tax refund proceeding unless an appellate court already has sustained such a challenge. See Cal.Const., Art. III, § 3.5, which provides in part that "[a]n administrative agency . . . has no power: "(a) To declare a statute unenforceable, or refuse to enforce a statute, on the basis of it being unconstitutional unless an appellate court has made a determination that such statute is unconstitutional; "(b) To declare a statute unconstitutional." 31 Significantly, the California administrative and judicial scheme for challenging a tax assessment is remarkably similar to the Illinois scheme that we upheld in Rosewell as "plain, speedy and efficient." See 450 U.S., at 508-509, and nn. 6, 7, 101 S.Ct., at 1226-1227, and nn. 6, 7. In fact, the California tax scheme is more favorable to the taxpayer than the Illinois scheme in that it requires the State to pay interest on improperly collected taxes. See Cal.Un.Ins.Code Ann. § 1242 (West Supp.1982). This Court has not hesitated to declare a state refund provision inadequate to bar federal relief if the taxpayer's opportunity to raise his constitutional claims in the state proceedings is uncertain. In Hillsborough v. Cromwell, 326 U.S. 620, 66 S.Ct. 445, 90 L.Ed. 358 (1946), the taxpayer could not raise his constitutional challenge in the administrative proceedings, and appeal to the state courts was discretionary with those courts. Consequently, because "there [was] such uncertainty concerning the New Jersey remedy as to make it speculative," id., at 625, 66 S.Ct., at 449, the Court held that the taxpayer could seek declaratory relief in federal court. 32 Under Cal.Un.Ins.Code Ann. § 1256 (West Supp. 1982), a former employee can collect unemployment benefits only if he has not been dismissed for "misconduct" or has not "left his most recent work voluntarily without good cause." 33 Part of the appellees' argument for the necessity of injunctive relief rests on the premise that payments to the state fund are made only after a benefit eligibility hearing has been held. Under 26 U.S.C. §§ 3309(a)(2) and 3304(a)(6)(B), however, the States are required to give nonprofit organizations, including the appellees, the option either of making regular contributions to the state unemployment insurance fund or of reimbursing the fund for payments actually made to the employers' former employees. The nonprofit organizations are not required to choose the reimbursement method, however, and can make regular payments to the fund in advance of any employee being discharged. 34 Our conclusion that the state-court remedy is plain, speedy and efficient is reenforced by our observation that it took the appellees in these cases over two years to obtain injunctive relief in federal court. 35 The dissent errs when it states, without authority, that the Tax Injunction Act is not applicable to these cases because of the "layers of review that must be exhausted in the California system." Post, at 422, n. 4. Certainly, nothing in the legislative history of the Act suggests that requiring a taxpayer to seek a refund first through administrative procedures makes the state remedy less than "plain, speedy and efficient." Moreover, the legislative history of the Johnson Act, after which the Tax Injunction Act was modeled, see n. 22, supra, makes clear congressional intent that a state remedy is "plain, speedy and efficient" even though a utility must proceed first through administrative and then judicial proceedings in order to challenge the constitutionality of utility rates. See S.Rep.No.125, 73d Cong., 1st Sess., 2-3 (1933). 36 In addition, there seems to be no principled basis for limiting the appellees' argument to First Amendment claims. Any employer required to pay state taxes in a manner allegedly violating the Equal Protection Clause, for example, might argue that the absence of state injunctive relief permitted the infliction of an irreparable injury that could be remedied only by a federal injunction. 37 We also reject the appellees' argument to the extent that it assumes that the state courts will not protect their constitutional rights. As we stated in another context: "[W]e are unwilling to assume that there now exists a general lack of appropriate sensitivity to constitutional rights in the trial and appellate courts of the several States. State courts, like federal courts, have a constitutional obligation to safeguard personal liberties and to uphold federal law." Stone v. Powell, 428 U.S. 465, 494, n. 35, 96 S.Ct. 3037, 3052, n. 35, 49 L.Ed.2d 1067 (1976). 38 The state defendants also argue that because the Federal Government is an indispensable party to this action, and could not be compelled to submit to state-court jurisdiction, the state courts could not afford the appellees complete relief. Consequently, the state defendants reason, the Tax Injunction Act does not deprive the District Court of jurisdiction. See Brief for Appellants State of California et al. 35. The error in this argument is its premise; as St. Martin Evangelical Lutheran Church v. South Dakota, 451 U.S. 772, 101 S.Ct. 2142, 68 L.Ed.2d 612 (1981), demonstrates, the Federal Government need not be a party in order for the appellees to litigate their statutory and constitutional claims. Finally, none of the parties suggests that we avoid the jurisdictional bar of the Tax Injunction Act by restricting our review to the appellees' challenge to 26 U.S.C. § 3309(b), and disregarding their challenge to the corresponding state provisions, §§ 634.5(a), (b). Such a suggestion would be untenable since, after all, the state provisions were enacted in order to comply with federal statutory requirements, and consequently are identically worded to the federal provisions. Thus, a challenge to FUTA would be a direct effort to "enjoin, suspend or restrain" state tax officials from collecting unemployment taxes from the appellees. Alternatively, if the challenge to FUTA would not affect the actions of state officials, there would be serious doubts whether the appellees were injured by FUTA's provisions. Accordingly, we vacate not only the District Court's judgment with respect to the appellees' state claims, but also its judgment regarding the constitutionality of FUTA. 39 In both of those cases, the question was whether a single district judge or a three-judge district court had jurisdiction. In the present cases, by contrast, the issue is whether the federal courts or the state courts have jurisdiction. 40 Similarly, the state defendants' reliance on Williams v. Zbaraz, 448 U.S. 358, 100 S.Ct. 2694, 65 L.Ed.2d 831 (1980), is misplaced. In that case, the District Court had held unconstitutional a federal statute that the parties had not challenged. We held that because there was no case or controversy on that issue, the District Court had exceeded its jurisdiction for that issue. Id., at 367, 100 S.Ct., at 2700. Nevertheless, because of the holding of unconstitutionality we concluded that we had jurisdiction under § 1252 to "review the 'whole case.' " Id., at 368, 100 S.Ct., at 2700. That review, however, was restricted to those issues over which the District Court had had jurisdiction, and we vacated that portion of the judgment holding the federal statute unconstitutional. Ibid. 1 Title 28 U.S.C. § 1252 provides: "Any party may appeal to the Supreme Court from an interlocutory or final judgment, decree or order of any court of the United States . . . holding an Act of Congress unconstitutional in any civil action, suit, or proceeding to which the United States or any of its agencies, or any officer or employee thereof, as such officer or employee, is a party." 2 A further irony is that the Secretary of Labor, who is certainly the principal defendant even if not an indispensable party, could remove such an action if it were filed in state court. Indeed, with respect to one of the actions consolidated in the District Court, the Secretary did just that. 3 Indeed, to the extent that equitable considerations are implicated they favor the procedure followed in these cases whereby expedited review in this Court is available. 4 There is an independent reason why the Tax Injunction Act does not preclude federal declaratory relief in this litigation. When one compares the layers of review that must be exhausted in the California system with the direct appeal to this Court provided by 28 U.S.C. § 1252, one surely cannot conclude that the state system provides the "plain, speedy and efficient" remedy that Congress intended for the resolution of the federal questions these cases present.

89

457 U.S. 368 102 S.Ct. 2485 73 L.Ed.2d 74 UNITED STATES, Petitionerv.Learley Reed GOODWIN. No. 80-2195. Argued April 21, 1982. Decided June 18, 1982. Syllabus After initially expressing an interest in plea bargaining on misdemeanor charges, respondent decided not to plead guilty and requested a trial by jury. While the misdemeanor charges were still pending, he was indicted and convicted in Federal District Court on a felony charge arising out of the same incident as the misdemeanor charges. Respondent moved to set aside the verdict on the ground of prosecutorial vindictiveness, contending that the felony indictment gave rise to an impermissible appearance of retaliation. The District Court denied the motion. The Court of Appeals reversed, holding that, although the prosecutor did not act with actual vindictiveness in seeking a felony indictment, the Due Process Clause prohibits the Government from bringing more serious charges against the defendant after he has invoked his right to a jury trial, unless the prosecutor comes forward with objective evidence that the increased charges could not have been brought before the defendant exercised his right. Believing that the circumstances surrounding the felony indictment gave rise to a genuine risk of retaliation, the court adopted a legal presumption of prosecutorial vindictiveness. Held : A presumption of prosecutorial vindictiveness was not warranted in this case, and absent such a presumption no due process violation was established. Pp. 372-384. (a) In cases in which action detrimental to a defendant has been taken after the exercise of a legal right, the presumption of an improper vindictive motive has been applied only where a reasonable likelihood of vindictiveness existed. North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656; Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628. Cf. Bordenkircher v. Hayes, 434 U.S. 357, 98 S.Ct. 663, 54 L.Ed.2d 604. Pp. 372-380. (b) A change in the prosecutor's charging decision made after an initial trial is completed is much more likely to be improperly motivated than is a pretrial decision. It is unrealistic to assume that a prosecutor's probable response to such pretrial motions as to be tried by a jury is to seek to penalize and to deter. Here, the timing of the prosecutor's action suggests that a presumption of vindictiveness was not warranted. A prosecutor should remain free before trial to exercise his discretion to determine the extent of the societal interest in the prosecution. The initial charges filed by a prosecutor may not reflect the extent to which an individual is legitimately subject to prosecution. Bordenkircher, supra. Pp. 380-382. (c) The nature of the right asserted by respondent confirms that a presumption of vindictiveness was not warranted in this case. The mere fact that a defendant refuses to plead guilty and forces the government to prove its case is insufficient to warrant a presumption that subsequent changes in the charging decision are unwarranted. Bordenkircher, supra. Pp. 382-383. (d) The fact that respondent, as opposed to having a bench trial, requested a jury trial does not compel a special presumption of prosecutorial vindictiveness whenever additional charges are thereafter brought. While there may have been an opportunity for vindictiveness here, a mere opportunity for vindictiveness is insufficient to justify the imposition of a prophylactic rule. The possibility that a prosecutor would respond to a defendant's pretrial demand for a jury trial by bringing charges not in the public interest that could be explained only as a penalty imposed on the defendant is sounlikely that a presumption of vindictiveness is certainly not warranted. Pp.383-384 4th Cir., 637 F.2d 250, reversed and remanded. Andrew L. Frey, Washington, D.C., for petitioner. Paul W. Spence, Baltimore, Md., for respondent. Justice STEVENS delivered the opinion of the Court. 1 This case involves presumptions. The question presented is whether a presumption that has been used to evaluate a judicial or prosecutorial response to a criminal defendant's exercise of a right to be retried after he has been convicted should also be applied to evaluate a prosecutor's pretrial response to a defendant's demand for a jury trial. 2 After the respondent requested a trial by jury on pending misdemeanor charges, he was indicted and convicted on a felony charge. Believing that the sequence of events gave rise to an impermissible appearance of prosecutorial retaliation against the defendant's exercise of his right to be tried by jury, the United States Court of Appeals for the Fourth Circuit reversed the felony conviction. 637 F.2d 250. Because this case presents an important question concerning the scope of our holdings in North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656, and Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628, we granted the Government's petition for certiorari. 454 U.S. 1079, 102 S.Ct. 632, 70 L.Ed.2d 613. 3 * Respondent Goodwin was stopped for speeding by a United States Park Policeman on the Baltimore-Washington Parkway. Goodwin emerged from his car to talk to the policeman. After a brief discussion, the officer noticed a clear plastic bag underneath the armrest next to the driver's seat of Goodwin's car. The officer asked Goodwin to return to his car and to raise the armrest. Respondent did so, but as he raised the armrest he placed the car into gear and accelerated rapidly. The car struck the officer, knocking him first onto the back of the car and then onto the highway. The policeman returned to his car, but Goodwin eluded him in a high-speed chase. 4 The following day, the officer filed a complaint in the District Court charging respondent with several misdemeanor and petty offenses, including assault. Goodwin was arrested and arraigned before a United States Magistrate. The Magistrate set a date for trial, but respondent fled the jurisdiction. Three years later Goodwin was found in custody in Virginia and was returned to Maryland. 5 Upon his return, respondent's case was assigned to an attorney from the Department of Justice, who was detailed temporarily to try petty crime and misdemeanor cases before the Magistrate. The attorney did not have authority to try felony cases or to seek indictments from the grand jury. Respondent initiated plea negotiations with the prosecutor, but later advised the Government that he did not wish to plead guilty and desired a trial by jury in the District Court.1 6 The case was transferred to the District Court and responsibility for the prosecution was assumed by an Assistant United States Attorney. Approximately six weeks later, after reviewing the case and discussing it with several parties, the prosecutor obtained a four-count indictment charging respondent with one felony count of forcibly assaulting a federal officer and three related counts arising from the same incident.2 A jury convicted respondent on the felony count and on one misdemeanor count. 7 Respondent moved to set aside the verdict on the ground of prosecutorial vindictiveness, contending that the indictment on the felony charge gave rise to an impermissible appearance of retaliation. The District Court denied the motion, finding that "the prosecutor in this case has adequately dispelled any appearance of retaliatory intent."3 8 Although the Court of Appeals readily concluded that "the prosecutor did not act with actual vindictiveness in seeking a felony indictment," 637 F.2d, at 252, it nevertheless reversed. Relying on our decisions in North Carolina v. Pearce, supra, and Blackledge v. Perry, supra, the court held that the Due Process Clause of the Fifth Amendment prohibits the Government from bringing more serious charges against a defendant after he has invoked his right to a jury trial, unless the prosecutor comes forward with objective evidence to show that the increased charges could not have been brought before the defendant exercised his rights. Because the court believed that the circumstances surrounding the felony indictment gave rise to a genuine risk of retaliation, it adopted a legal presumption designed to spare courts the "unseemly task" of probing the actual motives of the prosecutor. 637 F.2d, at 255. II 9 To punish a person because he has done what the law plainly allows him to do is a due process violation "of the most basic sort." Bordenkircher v. Hayes, 434 U.S. 357, 363, 98 S.Ct. 663, 668, 54 L.Ed.2d 604. In a series of cases beginning with North Carolina v. Pearce and culminating in Bordenkircher v. Hayes, the Court has recognized this basic—and itself uncontroversial principle. For while an individual certainly may be penalized for violating the law, he just as certainly may not be punished for exercising a protected statutory or constitutional right.4 10 The imposition of punishment is the very purpose of virtually all criminal proceedings. The presence of a punitive motivation, therefore, does not provide an adequate basis for distinguishing governmental action that is fully justified as a legitimate response to perceived criminal conduct from governmental action that is an impermissible response to noncriminal, protected activity. Motives are complex and difficult to prove. As a result, in certain cases in which action detrimental to the defendant has been taken after the exercise of a legal right, the Court has found it necessary to "presume" an improper vindictive motive. Given the severity of such a presumption, however—which may operate in the absence of any proof of an improper motive and thus may block a legitimate response to criminal conduct—the Court has done so only in cases in which a reasonable likelihood of vindictiveness exists. 11 In North Carolina v. Pearce, the Court held that neither the Double Jeopardy Clause nor the Equal Protection Clause prohibits a trial judge from imposing a harsher sentence on retrial after a criminal defendant successfully attacks an initial conviction on appeal. The Court stated, however, that "[i]t can hardly be doubted that it would be a flagrant violation [of the Due Process Clause] of the Fourteenth Amendment for a state trial court to follow an announced practice of imposing a heavier sentence upon every reconvicted defendant for the explicit purpose of punishing the defendant for his having succeeded in getting his original conviction set aside." 395 U.S., at 723-724, 89 S.Ct., at 2080. The Court continued: 12 "Due process of law, then, requires that vindictiveness against a defendant for having successfully attacked his first conviction must play no part in the sentence he receives after a new trial. And since the fear of such vindictiveness may unconstitutionally deter a defendant's exercise of the right to appeal or collaterally attack his first conviction, due process also requires that a defendant be freed of apprehension of such a retaliatory motivation on the part of the sentencing judge." Id., at 725, 89 S.Ct., at 2080. 13 In order to assure the absence of such a motivation, the Court concluded: 14 "[W]henever a judge imposes a more severe sentence upon a defendant after a new trial, the reasons for his doing so must affirmatively appear. Those reasons must be based upon objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original sentencing proceeding. And the factual data upon which the increased sentence is based must be made part of the record, so that the constitutional legitimacy of the increased sentence may be fully reviewed on appeal." Id., at 726, 89 S.Ct., at 2081. 15 In sum, the Court applied a presumption of vindictiveness, which may be overcome only by objective information in the record justifying the increased sentence.5 16 In Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628, the Court confronted the problem of increased punishment upon retrial after appeal in a setting different from that considered in Pearce. Perry was convicted of assault in an inferior court having exclusive jurisdiction for the trial of misdemeanors. The court imposed a 6-month sentence. Under North Carolina law, Perry had an absolute right to a trial de novo in the Superior Court, which possessed felony jurisdiction. After Perry filed his notice of appeal, the prosecutor obtained a felony indictment charging him with assault with a deadly weapon. Perry pleaded guilty to the felony and was sentenced to a term of five to seven years in prison. 17 In reviewing Perry's felony conviction and increased sentence,6 this Court first stated the essence of the holdings in Pearce and the cases that had followed it: 18 "The lesson that emerges from Pearce, Colten, and Chaffin is that the Due Process Clause is not offended by all possibilities of increased punishment upon retrial after appeal, but only by those that pose a realistic likelihood of 'vindictiveness.' " 417 U.S., at 27, 94 S.Ct., at 2102. 19 The Court held that the opportunities for vindictiveness in the situation before it were such "as to impel the conclusion that due process of law requires a rule analogous to that of the Pearce case." Ibid. It explained: "A prosecutor clearly has a considerable stake in discouraging convicted misdemeanants from appealing and thus obtaining a trial de novo in the Superior Court, since such an appeal will clearly require increased expenditures of prosecutorial resources before the defendant's conviction becomes final, and may even result in a formerly convicted defendant's going free. And, if the prosecutor has the means readily at hand to discourage such appeals—by 'upping the ante' through a felony indictment whenever a convicted misdemeanant pursues his statutory appellate remedy—the State can insure that only the most hardy defendants will brave the hazards of a de novo trial." Id., at 27-28, 94 S.Ct., at 2102. 20 The Court emphasized in Blackledge that it did not matter that no evidence was present that the prosecutor had acted in bad faith or with malice in seeking the felony indictment.7 As in Pearce, the Court held that the likelihood of vindictiveness justified a presumption that would free defendants of apprehension of such a retaliatory motivation on the part of the prosecutor.8 21 Both Pearce and Blackledge involved the defendant's exercise of a procedural right that caused a complete retrial after he had been once tried and convicted. The decisions in these cases reflect a recognition by the Court of the institutional bias inherent in the judicial system against the retrial of issues that have already been decided. The doctrines of stare decisis, res judicata, the law of the case, and double jeopardy all are based, at least in part, on that deep-seated bias. While none of these doctrines barred the retrials in Pearce and Blackledge, the same institutional pressure that supports them might also subconsciously motivate a vindictive prosecutorial or judicial response to a defendant's exercise of his right to obtain a retrial of a decided question. 22 In Bordenkircher v. Hayes, 434 U.S. 357, 98 S.Ct. 663, 54 L.Ed.2d 604, the Court for the first time considered an allegation of vindictiveness that arose in a pretrial setting. In that case the Court held that the Due Process Clause of the Fourteenth Amendment did not prohibit a prosecutor from carrying out a threat, made during plea negotiations, to bring additional charges against an accused who refused to plead guilty to the offense with which he was originally charged. The prosecutor in that case had explicitly told the defendant that if he did not plead guilty and "save the court the inconvenience and necessity of a trial" he would return to the grand jury to obtain an additional charge that would significantly increase the defendant's potential punishment.9 The defendant refused to plead guilty and the prosecutor obtained the indictment. It was not disputed that the additional charge was justified by the evidence, that the prosecutor was in possession of this evidence at the time the original indictment was obtained, and that the prosecutor sought the additional charge because of the accused's refusal to plead guilty to the original charge. 23 In finding no due process violation, the Court in Bordenkircher considered the decisions in Pearce and Blackledge, and stated: 24 "In those cases the Court was dealing with the State's unilateral imposition of a penalty upon a defendant who had chosen to exercise a legal right to attack his original conviction—a situation 'very different from the give-andtake negotiation common in plea bargaining between the prosecution and defense, which arguably possess relatively equal bargaining power.' Parker v. North Carolina, 397 U.S. 790, 809 [90 S.Ct. 1458, 1479-1480, 25 L.Ed.2d 785] (opinion of BRENNAN, J.)." 434 U.S., at 362, 98 S.Ct., at 667. 25 The Court stated that the due process violation in Pearce and Blackledge "lay not in the possibility that a defendant might be deterred from the exercise of a legal right . . . but rather in the danger that the State might be retaliating against the accused for lawfully attacking his conviction." 434 U.S., at 363, 98 S.Ct., at 667-668. 26 The Court held, however, that there was no such element of punishment in the "give-and-take" of plea negotiation, so long as the accused "is free to accept or reject the prosecution's offer." Ibid. The Court noted that, by tolerating and encouraging the negotiation of pleas, this Court had accepted as constitutionally legitimate the simple reality that the prosecutor's interest at the bargaining table is to persuade the defendant to forgo his constitutional right to stand trial. The Court concluded: 27 "We hold only that the course of conduct engaged in by the prosecutor in this case, which no more than openly presented the defendant with the unpleasant alternatives of forgoing trial or facing charges on which he was plainly subject to prosecution, did not violate the Due Process Clause of the Fourteenth Amendment." Id., at 365, 98 S.Ct., at 669. 28 The outcome in Bordenkircher was mandated by this Court's acceptance of plea negotiation as a legitimate process.10 In declining to apply a presumption of vindictiveness, the Court recognized that "additional" charges obtained by a prosecutor could not necessarily be characterized as an impermissible "penalty." Since charges brought in an original indictment may be abandoned by the prosecutor in the course of plea negotiation—in often what is clearly a "benefit" to the defendant—changes in the charging decision that occur in the context of plea negotiation are an inaccurate measure of improper prosecutorial "vindictiveness."11 An initial indictment—from which the prosecutor embarks on a course of plea negotiation—does not necessarily define the extent of the legitimate interest in prosecution. For just as a prosecutor may forgo legitimate charges already brought in an effort to save the time and expense of trial, a prosecutor may file additional charges if an initial expectation that a defendant would plead guilty to lesser charges proves unfounded.12 III 29 This case, like Bordenkircher, arises from a pretrial decision to modify the charges against the defendant. Unlike Bordenkircher, however, there is no evidence in this case that could give rise to a claim of actual vindictiveness; the prosecutor never suggested that the charge was brought to influence the respondent's conduct.13 The conviction in this case may be reversed only if a presumption of vindictiveness—applicable in all cases—is warranted. 30 There is good reason to be cautious before adopting an inflexible presumption of prosecutorial vindictiveness in a pretrial setting. In the course of preparing a case for trial, the prosecutor may uncover additional information that suggests a basis for further prosecution or he simply may come to realize that information possessed by the State has a broader significance. At this stage of the proceedings, the prosecutor's assessment of the proper extent of prosecution may not have crystallized. In contrast, once a trial begins—and certainly by the time a conviction has been obtained—it is much more likely that the State has discovered and assessed all of the information against an accused and has made a determination, on the basis of that information, of the extent to which he should be prosecuted. Thus, a change in the charging decision made after an initial trial is completed is much more likely to be improperly motivated than is a pretrial decision. 31 In addition, a defendant before trial is expected to invoke procedural rights that inevitably impose some "burden" on the prosecutor. Defense counsel routinely file pretrial motions to suppress evidence; to challenge the sufficiency and form of an indictment; to plead an affirmative defense; to request psychiatric services; to obtain access to government files; to be tried by jury. It is unrealistic to assume that a prosecutor's probable response to such motions is to seek to penalize and to deter. The invocation of procedural rights is an integral part of the adversary process in which our criminal justice system operates. 32 Thus, the timing of the prosecutor's action in this case suggests that a presumption of vindictiveness is not warranted. A prosecutor should remain free before trial to exercise the broad discretion entrusted to him to determine the extent of the societal interest in prosecution. An initial decision should not freeze future conduct.14 As we made clear in Bordenkircher, the initial charges filed by a prosecutor may not reflect the extent to which an individual is legitimately subject to prosecution.15 33 The nature of the right asserted by the respondent confirms that a presumption of vindictiveness is not warranted in this case. After initially expressing an interest in plea negotiation, respondent decided not to plead guilty and requested a trial by jury in District Court. In doing so, he forced the Government to bear the burdens and uncertainty of a trial. This Court in Bordenkircher made clear that the mere fact that a defendant refuses to plead guilty and forces the government to prove its case is insufficient to warrant a presumption that subsequent changes in the charging decision are unjustified. Respondent argues that such a presumption is warranted in this case, however, because he not only requested a trial—he requested a trial by jury. 34 We cannot agree. The distinction between a bench trial and a jury trial does not compel a special presumption of prosecutorial vindictiveness whenever additional charges are brought after a jury is demanded. To be sure, a jury trial is more burdensome than a bench trial. The defendant may challenge the selection of the venire; the jury itself must be impaneled; witnesses and arguments must be prepared more carefully to avoid the danger of a mistrial. These matters are much less significant, however, than the facts that before either a jury or a judge the State must present its full case against the accused and the defendant is entitled to offer a full defense. As compared to the complete trial de novo at issue in Blackledge, a jury trial—as opposed to a bench trial—does not require duplicative expenditures of prosecutorial resources before a final judgment may be obtained. Moreover, unlike the trial judge in Pearce, no party is asked "to do over what it thought it had already done correctly."16 A prosecutor has no "personal stake" in a bench trial and thus no reason to engage in "self-vindication" upon a defendant's request for a jury trial.17 Perhaps most importantly, the institutional bias against the retrial of a decided question that supported the decisions inPearce and Blackledge simply has no counterpart in this case.18 35 There is an opportunity for vindictiveness, as there was in Colten and Chaffin. Those cases demonstrate, however, that a mere opportunity for vindictiveness is insufficient to justify the imposition of a prophylactic rule. As Blackledge makes clear, "the Due Process Clause is not offended by all possibilities of increased punishment . . . but only by those that pose a realistic likelihood of 'vindictiveness.' " 417 U.S., at 27, 94 S.Ct., at 2102. The possibility that a prosecutor would respond to a defendant's pretrial demand for a jury trial by bringing charges not in the public interest that could be explained only as a penalty imposed on the defendant is so unlikely that a presumption of vindictiveness certainly is not warranted. IV 36 In declining to apply a presumption of vindictiveness, we of course do not foreclose the possibility that a defendant in an appropriate case might prove objectively that the prosecutor's charging decision was motivated by a desire to punish him for doing something that the law plainly allowed him to do.19 In this case, however, the Court of Appeals stated: "On this record we readily conclude that the prosecutor did not act with actual vindictiveness in seeking a felony indictment." 637 F.2d, at 252. Respondent does not challenge that finding. Absent a presumption of vindictiveness, no due process violation has been established. 37 The judgment of the Court of Appeals is reversed. The case is remanded for further proceedings consistent with this opinion. 38 It is so ordered. 39 Justice BLACKMUN, concurring in the judgment. 40 Like Justice BRENNAN, I believe that our precedents mandate the conclusion that "a realistic likelihood of 'vindictiveness' " arises in this context. Blackledge v. Perry, 417 U.S. 21, 27, 94 S.Ct. 2098, 2102, 40 L.Ed.2d 628 (1974). The Assistant United States Attorney responsible for increasing the charges against respondent was aware of the initial charging decision; he had the means available to discourage respondent from electing a jury trial in District Court; he had a substantial stake in dissuading respondent from exercising that option; and he was familiar with, and sensitive to, the institutional interests that favored a trial before the Magistrate. 41 Moreover, I find no support in our prior cases for any distinction between pretrial and post-trial vindictiveness. As I have said before: "Prosecutorial vindictiveness in any context is still prosecutorial vindictiveness. The Due Process Clause should protect an accused against it, however it asserts itself." Bordenkircher v. Hayes, 434 U.S. 357, 368, 98 S.Ct. 663, 670, 54 L.Ed.2d 604 (1978) (dissenting opinion). And, as Justice BRENNAN points out, Bordenkircher does not dictate the result here. In fact, in Bordenkircher the Court expressly distinguished and left unresolved cases such as this one, "where the prosecutor without notice brought an additional and more serious charge after plea negotiations relating only to the original [charges] had ended with the defendant's insistence on pleading not guilty." Id., at 360, 98 S.Ct., at 666. 42 The Court's ruling in Bordenkircher did not depend on a distinction between the pretrial and post-trial settings: rather, the Court declined to apply its prior opinions in Blackledge and North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969), because those cases involved "the State's unilateral imposition of a penalty," rather than " 'the give-and-take negotiation common in plea bargaining.' " 434 U.S., at 362, 98 S.Ct., at 667, quoting Parker v. North Carolina, 397 U.S. 790, 809, 90 S.Ct. 1458, 1479, 25 L.Ed.2d 785 (1970) (opinion of BRENNAN, J.). Here, as in Pearce and Blackledge, the prosecutor unilaterally imposed a penalty in response to respondent's exercise of a legal right. 43 Adopting the prophylactic rule of Pearce and Blackledge in this case will not, as the Court would insist, undercut "the broad discretion entrusted to [the prosecutor] to determine the extent of the societal interest in prosecution." Ante, at 382. "[T]he prosecutor initially 'makes a discretionary determination that the interests of the state are served by not seeking more serious charges.' " Bordenkircher v. Hayes, 434 U.S., at 367, 98 S.Ct., at 670 (dissenting opinion), quoting Hayes v. Cowan, 547 F.2d 42, 44 (CA6 1976). Moreover, the Due Process Clause does not deprive a prosecutor of the flexibility to add charges after a defendant has decided not to plead guilty and has elected a jury trial in District Court—so long as the adjustment is based on "objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original" charging decision. North Carolina v. Pearce, 395 U.S., at 726, 89 S.Ct., at 2081. In addition, I believe that the prosecutor adequately explains an increased charge by pointing to objective information that he could not reasonably have been aware of at the time charges were initially filed. Cf. ante, at 381. 44 Because I find that the Assistant United States Attorney's explanation for seeking a felony indictment satisfies these standards, see ante, at 371, n.2, I conclude that the Government has dispelled the appearance of vindictiveness and, therefore, that the imposition of additional charges did not violate respondent's due process rights. Accordingly, I concur in the judgment. 45 Justice BRENNAN, with whom Justice MARSHALL joins, dissenting. 46 In Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974), this Court held that the Due Process Clause prohibits a prosecutor from responding to the defendant's invocation of his statutory right to a trial de novo by bringing more serious charges against him that arise out of the same conduct. In the case before us, the prosecutor responded to the defendant's invocation of his statutory and constitutional right to a trial by jury by raising petty offenses to felony charges based on the same conduct. Yet the Court holds, in the teeth of Blackledge, that here there is no denial of due process. In my view, Blackledge requires affirmance of the Court of Appeals, and the Court's attempt to distinguish that case from the present one is completely unpersuasive. 47 The salient facts of this case are quite simple. Respondent was originally charged with several petty offenses and misdemeanors—speeding, reckless driving, failing to give aid at the scene of an accident, fleeing from a police officer, and assault by striking a police officer—arising from his conduct on the Baltimore-Washington Parkway. Assuming that respondent had been convicted on every count charged in this original complaint, the maximum punishment to which he conceivably could have been exposed was fines of $3,500 and 28 months in prison.1 Because all of the charges against respondent were petty offenses or misdemeanors, they were scheduled for trial before a magistrate, see 28 U.S.C. § 636(a)(3); 18 U.S.C. § 3401(a), who was not authorized to conduct jury trials, see ante, at 371, n.1. In addition, the case was assigned to a prosecutor who, owing to inexperience, was not even authorized to try felony cases. Thus the Government recognized that respondent's alleged crimes were relatively minor, and attempted to dispose of them in an expedited manner. But respondent frustrated this attempt at summary justice by demanding a jury trial in Federal District Court. This was his right, of course, not only under the applicable statute, 18 U.S.C. § 3401(b), but also under the Constitution.2 48 Respondent's demand required that the case be transferred from the Magistrate's Court in Hyattsville to the District Court in Baltimore, and that the prosecution be reassigned to an Assistant United States Attorney, who was authorized to prosecute cases in the District Court. The new prosecutor sought and obtained a second, four-count indictment, in which the same conduct originally charged as petty-offense and misdemeanor counts was now charged as a misdemeanor and two felonies: assaulting, resisting, or impeding a federal officer with a deadly weapon, and assault with a dangerous weapon. If we assume (as before) that respondent was convicted on all of these charges, his maximum exposure to punishment had now become fines of $11,500 and 15 years in prison.3 Respondent's claim below was that such an elevation of the charges against him from petty offenses to felonies, following his exercise of his statutory and constitutional right to a jury trial, reflected prosecutorial vindictiveness that denied him due process of law. 49 The Court attempts to denigrate respondent's claim by asserting that this case "involves presumptions," ante, at 369, and by arguing that "there is no evidence in this case that could give rise to a claim of actual vindictiveness," ante, at 380 (emphasis in original). By casting respondent's claim in terms of a "mere" legal presumption, the Court hopes to make that claim appear to be unreal or technical. But such an approach is contrary to the letter and spirit of Blackledge. There we focused upon the accused's "apprehension of . . . retaliatory motivation," 417 U.S., at 28, 94 S.Ct., at 2102, and we held that the Due Process Clause is violated when situations involving increased punishment "pose a realistic likelihood of 'vindictiveness,' " id., at 27, 94 S.Ct., at 2102. In such situations, the criminal defendant's apprehension of retaliatory motivation does not amount to an unreal or technical violation of his constitutional rights. On the contrary, as we recognized in North Carolina v. Pearce, 395 U.S. 711, 725, 89 S.Ct. 2072, 2080, 23 L.Ed.2d 656 (1969), "the fear of such vindictiveness may unconstitutionally deter a defendant's exercise" of his rights. 50 The Court does not contend that Blackledge is inapplicable to instances of pretrial as well as post-trial vindictiveness. But after examining the record before us for objective indications of such vindictiveness, the Court concludes, ante, at 382, that "a presumption of vindictiveness is not warranted in this case." With all respect, I disagree both with the Court's conclusion and with its reasoning. In my view, the question here is not one of "presumptions." Rather, I would analyze respondent's claim in the terms employed by our precedents. Did the elevation of the charges against respondent "pose a realistic likelihood of 'vindictiveness?' " See Blackledge v. Perry, 417 U.S., at 27, 94 S.Ct., at 2102. Is it possible that "the fear of such vindictiveness may unconstitutionally deter" a person in respondent's position from exercising his statutory and constitutional right to a jury trial? See North Carolina v. Pearce, supra, 395 U.S., at 725, 89 S.Ct., at 2080. The answer to these questions is plainly "Yes." 51 The Court suggests, ante, at 383, that the distinction between a bench trial and a jury trial is unimportant in this context. Such a suggestion is demonstrably fallacious. Experienced criminal practitioners, for both prosecution and defense, know that a jury trial entails far more prosecutorial work than a bench trial. Defense challenges to the potential-juror array, voir dire examination of potential jurors, and suppression hearings all take up a prosecutor's time before a jury trial, adding to his scheduling difficulties and caseload. More care in the preparation of his requested instructions, of his witnesses, and of his own remarks is necessary in order to avoid mistrial or reversible error. And there is always the specter of the "irrational" acquittal by a jury that is unreviewable on appeal. Thus it is simply inconceivable that a criminal defendant's election to be tried by jury would be a matter of indifference to his prosecutor. On the contrary, the prosecutor would almost always prefer that the defendant waive such a "troublesome" right. And if the defendant refuses to do so, the prosecutor's subsequent elevation of the charges against the defendant manifestly poses a realistic likelihood of vindictiveness. 52 The truth of my conclusion, and the patent fallacy of the Court's, is particularly evident on the record before us. The practical effect of respondent's demand for a jury trial was that the Government had to transfer the case from a trial before a Magistrate in Hyattsville to a trial before a District Judge and jury in Baltimore, and had to substitute one prosecutor for another. The Government thus suffered not only administrative inconvenience: It also lost the value of the preparation and services of the first prosecutor, and was forced to commit a second prosecutor to prepare the case from scratch. Thus, just as in Blackledge, respondent's election had the effect of "clearly requir[ing] increased expenditures of prosecutorial resources before the defendant's conviction" could finally be achieved. 417 U.S., at 27, 94 S.Ct., at 2102. And, to paraphrase Blackledge, 53 "if the prosecutor has the means readily at hand to discourage such [elections]—by 'upping the ante' through a felony indictment . . . —the State can insure that only the most hardy defendants will brave the hazards of a [jury] trial." Cf. id., at 27-28, 94 S.Ct., at 2102. 54 I conclude that the facts of this case easily support the inference of "a realistic likelihood of vindictiveness." 55 The Court discusses Bordenkircher v. Hayes, 434 U.S. 357, 98 S.Ct. 663, 54 L.Ed.2d 604 (1978), ante, at 377-380, and suggests some analogy between that case and the present one, ante, at 380. In my view, such an analogy is quite inapt. Bordenkircher dealt only with the context of plea bargaining and with the narrow situation in which the prosecutor "openly presented the defendant with the unpleasant alternatives of forgoing trial or facing [increased] charges." 434 U.S., at 365, 98 S.Ct., at 669. Bordenkircher did not remotely suggest that a pretrial increase in charges, made as a response to a demand for jury trial, would not present a realistic likelihood of vindictiveness when the demand put the prosecution to an added burden such as that imposed in this case. Indeed, Bordenkircher expressly distinguished its facts from those in Blackledge and Pearce : "In those cases the Court was dealing with the State's unilateral imposition of a penalty upon a defendant who had chosen to exercise a legal right . . . —a situation 'very different from the give-and-take negotiation common in plea bargaining . . . .' " 434 U.S., at 362, 98 S.Ct., at 667, quoting Parker v. North Carolina, 397 U.S. 790, 809, 90 S.Ct. 1458, 1479, 25 L.Ed.2d 785 (1970). The facts in this case plainly fit within the pattern of Pearce and Blackledge, not of Bordenkircher. There was no ongoing "give-and-take negotiation" between respondent and the Government, and there was the "unilateral imposition of a penalty" in response to respondent's choice "to exercise a legal right." 56 Because it seems clear to me that Blackledge requires it, I would affirm the judgment of the Court of Appeals. 1 At that time, there was no statutory provision allowing a trial by jury before a magistrate. 2 By affidavit, the Assistant United States Attorney later set forth his reasons for this action: (1) he considered respondent's conduct on the date in question to be a serious violation of law, (2) respondent had a lengthy history of violent crime, (3) the prosecutor considered respondent's conduct to be related to major narcotics transactions, (4) the prosecutor believed that respondent had committed perjury at his preliminary hearing, and (5) respondent had failed to appear for trial as originally scheduled. The Government attorney stated that his decision to seek a felony indictment was not motivated in any way by Goodwin's request for a jury trial in District Court. 3 App. to Pet. for Cert. 22a; cf. n.2, supra. The District Court considered the merits of respondent's motion even though it was not timely filed in accordance with Rule 12(b)(1) of the Federal Rules of Criminal Procedure. The District Court found sufficient "cause" for respondent's procedural default pursuant to Federal Rule of Criminal Procedure 12(f). The Court of Appeals did not consider the propriety of the District Court's ruling in this regard and neither do we. 4 "[F]or an agent of the State to pursue a course of action whose objective is to penalize a person's reliance on his legal rights is 'patently unconstitutional.' " Bordenkircher v. Hayes, 434 U.S. 357, 363, 98 S.Ct. 663, 668, 54 L.Ed.2d 604 (quoting Chaffin v. Stynchcombe, 412 U.S. 17, 32-33, n. 20, 93 S.Ct. 1977, 1986, n. 20, 36 L.Ed.2d 714). 5 Two subsequent cases developed the principles set forth in Pearce. In Colten v. Kentucky, 407 U.S. 104, 92 S.Ct. 1953, 32 L.Ed.2d 584, the Court refused to apply the prophylactic rule of Pearce to an allegation of vindictiveness that arose in a case involving Kentucky's two-tier system for adjudicating less serious criminal charges. In that system, a defendant who is convicted and sentenced in an inferior court is entitled to a trial de novo in a court of general jurisdiction. The defendant in Colten exercised that right and received a more severe sentence from the court of general jurisdiction. This Court found that "[t]he possibility of vindictiveness, found to exist in Pearce, is not inherent in the Kentucky two-tier system." 407 U.S., at 116, 92 S.Ct., at 1960. The Court emphasized that the second trial was conducted, and the final sentence was imposed, by a different court that was not asked "to do over what it thought it had already done correctly." Id., at 117, 92 S.Ct., at 1960. The Court noted: "It may often be that the superior court will impose a punishment more severe than that received from an inferior court. But it no more follows that such a sentence is a vindictive penalty for seeking a superior court trial than that the inferior court imposed a lenient penalty." Ibid. Ultimately, the Court described the sentence received from the inferior tribunal as "in effect . . . no more than an offer in settlement." Id., at 119, 92 S.Ct., at 1961. In Chaffin v. Stynchcombe, 412 U.S. 17, 93 S.Ct. 1977, 36 L.Ed.2d 714, the Court held that the prophylactic rule of Pearce does not apply when the second sentence is imposed on retrial by a jury. The Court emphasized that the decision in Pearce "was premised on the apparent need to guard against vindictiveness in the resentencing process." 412 U.S., at 25, 93 S.Ct., at 1982 (emphasis in original). The Court found that the possibility of vindictiveness was de minimis when resentencing was by jury in a properly controlled retrial. The Court noted that (1) the jury typically will not be aware of the prior sentence, (2) the jury, unlike a judge who has been reversed, will have no personal stake in the prior conviction and no motivation to engage in self-vindication, and (3) the jury will not likely be sensitive to the institutional interests that might occasion higher sentences by a judge desirous of discouraging what he regards as meritless appeals. Id., at 26-27, 93 S.Ct., at 1982-1983. 6 The Court held that in pleading guilty Perry had not waived the right "not to be haled into court at all upon the felony charge." 417 U.S., at 30, 94 S.Ct., at 2104. 7 There is, of course, no evidence that the prosecutor in this case acted in bad faith or maliciously in seeking a felony indictment against Perry." Id., at 28, 94 S.Ct., at 2102. 8 The presumption again could be overcome by objective evidence justifying the prosecutor's action. The Court noted: "This would clearly be a different case if the State had shown that it was impossible to proceed on the more serious charge at the outset, as in Diaz v. United States, 223 U.S. 442 [32 S.Ct. 250, 56 L.Ed. 500]." Id., at 29, n.7, 94 S.Ct., at 2103, n.7. 9 The prosecutor advised the defendant that he would obtain an indictment under the Kentucky Habitual Criminal Act, which would subject the accused to a mandatory sentence of life imprisonment by reason of his two prior felony convictions. Absent the additional indictment, the defendant was subject to a punishment of 2 to 10 years in prison. 10 Cf. 434 U.S., at 364-365, 98 S.Ct., at 669 ("To hold that the prosecutor's desire to induce a guilty plea . . . may play no part in his charging decision, would contradict the very premises that underlie the concept of plea bargaining itself"). If a prosecutor could not threaten to bring additional charges during plea negotiation, and then obtain those charges when plea negotiation failed, an equally compelling argument could be made that a prosecutor's initial charging decision could never be influenced by what he hoped to gain in the course of plea negotiation. Whether "additional" charges were brought originally and dismissed, or merely threatened during plea negotiations, the prosecutor could be accused of using those charges to induce a defendant to forgo his right to stand trial. If such use of "additional" charges were presumptively invalid, the institution of plea negotiation could not survive. Thus, to preserve the plea negotiation process, with its correspondent advantages for both the defendant and the State, the Court in Bordenkircher held that "additional" charges may be used to induce a defendant to plead guilty. Once that conclusion was accepted, it necessarily followed that it did not matter whether the "additional" charges were obtained in the original indictment or merely threatened in plea negotiations and obtained once those negotiations broke down. In the former situation, the prosecutor could be said simply to have "anticipated" that the defendant might refuse to plead guilty and, as a result, to have placed his "threat" in the original indictment. Cf. id., at 360-361, 98 S.Ct., at 666 ("As a practical matter, in short, this case would be no different if the grand jury had indicted Hayes as a recidivist from the outset, and the prosecutor had offered to drop that charge as part of the plea bargain"). The decision in Bordenkircher also was influenced by the fact that, had the Court recognized a distinction of constitutional dimension between the dismissal of charges brought in an original indictment and the addition of charges after plea negotiation, the aggressive prosecutor would merely be prompted "to bring the greater charge initially in every case, and only thereafter to bargain." Id., at 368, 98 S.Ct., at 670 (BLACKMUN, J., dissenting). The consequences of such a decision often would be prejudicial to defendants, for an accused "would bargain against a greater charge, face the likelihood of increased bail, and run the risk that the court would be less inclined to accept a bargained plea." Ibid. Moreover, in those cases in which a defendant accepted the prosecution's offer, his reputation would be spared the unnecessary damage that would result from the placement of the additional charge on the public record. 11 The Court in Bordenkircher stated that the validity of a pretrial charging decision must be measured against the broad discretion held by the prosecutor to select the charges against an accused. "Within the limits set by the legislature's constitutionally valid definition of chargeable offenses, 'the conscious exercise of some selectivity in enforcement is not itself a federal constitutional violation' so long as 'the selection was [not] deliberately based upon an unjustifiable standard such as race, religion, or other arbitrary classification.' " Id., at 364, 98 S.Ct., at 668-69 (quoting Oyler v. Boles, 368 U.S. 448, 456, 82 S.Ct. 501, 506, 7 L.Ed.2d 446). A charging decision does not levy an improper "penalty" unless it results solely from the defendant's exercise of a protected legal right, rather than the prosecutor's normal assessment of the societal interest in prosecution. See Westen & Westin, A Constitutional Law of Remedies for Broken Plea Bargains, 66 Calif.L.Rev. 471, 486 (1978). 12 In rejecting a presumption of vindictiveness, the Court in Bordenkircher did not foreclose the possibility that a defendant might prove through objective evidence an improper prosecutorial motive. In the case before it, however, the Court did not find such proof in the fact that the prosecutor had stated explicitly that additional charges were brought to persuade the defendant to plead guilty. The fact that the prosecutor threatened the defendant did not prove that the action threatened was not permissible; the prosecutor's conduct did not establish that the additional charges were brought solely to "penalize" the defendant and could not be justified as a proper exercise of prosecutorial discretion. 13 See n. 12, supra. 14 We recognize that prosecutors may be trained to bring all legitimate charges against an individual at the outset. Certainly, a prosecutor should not file any charge until he has investigated fully all of the circumstances surrounding a case. To presume that every case is complete at the time an initial charge is filed, however, is to presume that every prosecutor is infallible—an assumption that would ignore the practical restraints imposed by often limited prosecutorial resources. Moreover, there are certain advantages in avoiding a rule that would compel prosecutors to attempt to place every conceivable charge against an individual on the public record from the outset. See n. 10, supra. 15 Respondent argues that the Court's refusal to presume vindictiveness in Bordenkircher is not controlling in this case because he had refused to plead guilty and the plea negotiation process was over. Respondent's argument is not strengthened, however, by the fact that the additional charge in this case was brought outside the context of plea negotiation. The fact that the increased charge in Bordenkircher was brought after a "warning" made during plea negotiation was the principal basis for the defendant's claim that the charge was an unjustified response to his legal right to stand trial. But cf. n. 12, supra. Respondent's argument in this case has no such predicate; unlike the defendant in Bordenkircher, the only evidence respondent is able to marshal in support of his allegation of vindictiveness is that the additional charge was brought at a point in time after his exercise of a protected legal right. 16 Cf. Colten v. Kentucky, 407 U.S., at 117, 92 S.Ct., at 1960. 17 Cf. Chaffin v. Stynchcombe, 412 U.S., at 27, 93 S.Ct., at 1983. 18 Indeed, there is a strong tradition in this country in favor of jury trials, despite the additional burdens that they entail for all parties. In many cases—and for many reasons—both the judge and the prosecutor may prefer to have a case tried by jury. See, e.g., Vines v. Muncy, 553 F.2d 342 (CA4 1977); United States v. Morlang, 531 F.2d 183 (CA4 1975); United States v. Ceja, 451 F.2d 399 (CA1 1971); see also Fed.Rule Crim.Proc. 23(a). In Singer v. United States, 380 U.S. 24, 85 S.Ct. 783, 13 L.Ed.2d 630, this Court held that a criminal defendant does not have a constitutional right to waive a jury trial and to have his case tried before a judge alone. The Court stated: "Trial by jury has been established by the Constitution as the 'normal and . . . preferable mode of disposing of issues of fact in criminal cases.' Patton v. United States, 281 U.S. 276, 312 [50 S.Ct. 253, 263, 74 L.Ed. 854]." Id., at 35, 85 S.Ct., at 790. 19 As the Government states in its brief: "Accordingly, while the prosecutor's charging decision is presumptively lawful, and the prosecutor is not required to sustain any burden of justification for an increase in charges, the defendant is free to tender evidence to the court to support a claim that enhanced charges are a direct and unjustifiable penalty for the exercise of a procedural right. Of course, only in a rare case would a defendant be able to overcome the presumptive validity of the prosecutor's actions through such a demonstration." Brief for United States 28, n. 9. 1 Two counts of "speeding" and one count of "reckless driving," in violation of 36 CFR §§ 50.31, 50.32 (1981), are each punishable by fines of not more than $500, or imprisonment for not more than six months, or both, 36 CFR § 50.5(a) (1981). One count of "failing to give aid at the scene of an accident," in violation of 18 U.S.C. §§ 7, 13, Md.Transp.Code Ann. §§ 20-102, 20-104 (1977), is punishable by a fine of not more than $1,000, or imprisonment for not more than four months, or both, §§ 27-101(c)(12), (14). One count of "fleeing from a police officer," in violation of 18 U.S.C. §§ 7, 13, Md.Transp.Code Ann. § 21-904 (1977), is punishable by a fine of not more than $500, § 27-101(b). One count of "assault by striking" a police officer, in violation of 18 U.S.C. § 113(d), is punishable by a fine of not more than $500, or imprisonment for not more than six months, or both. 2 See District of Columbia v. Colts, 282 U.S. 63, 73-74, 51 S.Ct. 52, 53, 75 L.Ed. 177 (1930); United States v. Hamdan, 552 F.2d 276, 278-280 (CA9 1977); United States v. Sanchez-Meza, 547 F.2d 461, 464-465 (CA9 1976); United States v. Potvin, 481 F.2d 380, 381-383 (CA10 1973). 3 "Assaulting, resisting, or impeding" a federal officer with a deadly weapon, in violation of 18 U.S.C. § 111, is punishable by a fine of not more than $10,000, or imprisonment for not more than 10 years, or both. "Assault with a dangerous weapon," in violation of 18 U.S.C. § 113(c), is punishable by a fine of not more than $1,000, or imprisonment for not more than five years, or both. A third count in the new indictment was "fleeing from a police officer," in violation of 18 U.S.C. §§ 7, 13, Md.Transp.Code Ann. § 21-904 (1977), which is punishable by a fine of not more than $500, § 27-101(b). The fourth count of the indictment was "failure to appear," in violation of 18 U.S.C. § 3150.

01

457 U.S. 273 102 S.Ct. 2432 73 L.Ed.2d 1 State of CALIFORNIA, ex rel. STATE LANDS COMMISSION, Plaintiffv.UNITED STATES. No. 89, Original. Argued March 29, 1982. Decided June 18, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1131, 103 S.Ct. 14. Syllabus Held: The United States, not California, has title to oceanfront land created through accretion, resulting from construction of a jetty, to land owned by the United States on the coast of California. Pp. 278-288. (a) A dispute over accretions to oceanfront land where title rests with or was derived from the Federal Government is to be determined by federal law. Hughes v. Washington, 389 U.S. 290, 88 S.Ct. 438, 19 L.Ed.2d 530; Wilson v. Omaha Indian Tribe, 442 U.S. 653, 99 S.Ct. 2529, 61 L.Ed.2d 153. Under federal law, accretion, whatever its cause, belongs to the upland owner. Pp. 278-283. (b) This is not a case where, as a matter of choice of law, state law should be borrowed and applied as the federal rule for deciding the substantive legal issue. Congress addressed the issue of accretions to federal land in the Submerged Lands Act, which vested title in the States to the lands underlying the territorial sea and confirmed the title of the States to the tidelands up to the line of mean high tide, but which in § 5 withheld from the grant to the States all "accretions" to coastal lands acquired or reserved by the United States. In light of this latter provision, borrowing for federal-law purposes a state rule that would divest federal ownership is foreclosed. Moreover, this is not a case in which federal common law must be created, since it has long been settled under federal law that the right to future accretions is an inherent and essential attribute of the littoral or riparian owner. Pp. 283-285. (c) Only land underneath inland waters was included in the initial grant to the States under the equal-footing doctrine, United States v. California, 332 U.S. 19, 67 S.Ct. 1658, 91 L.Ed. 1889, and hence California cannot properly claim that title to the land in question here was vested in the State by that doctrine and confirmed by the Submerged Lands Act. The latter Act was a constitutional exercise of Congress' power to dispose of federal property and "did not impair the validity" of the United States v. California decision, United States v. Louisiana, 363 U.S. 1, 7, 20, 80 S.Ct. 961, 967, 973, 4 L.Ed.2d 1025. To accept California's argument would require rejecting not only Hughes, supra, but also the established federal rule that accretions belong to the upland owner. Pp. 285-286. (d) Section 2(a)(3) of the Submerged Lands Act, defining "lands beneath navigable waters" that fall within the Act's general grant to the States as including all "made" lands that formerly were lands beneath navigable water, does not apply to the gradual process by which sand accumulated along the shore, although caused by a jetty. To the extent that accretions are to be considered "made" land, they would fall within the reservation by the United States in the Act of "all lands filled in, built up, or otherwise reclaimed by the United States for its own use." In any event, § 5(a) of the Act expressly withholds from the grant to the States all "accretions" to lands reserved by the United States. Pp. 286-288. (e) Section 3(a) of the Submerged Lands Act, confirming the title of persons who, on June 5, 1950, were entitled to lands beneath navigable water "under the law of the respective states in which the land is located," means nothing more than that state law determines the proper beneficiary of the grant of land under the Act. Federal law determines the scope of the grant under the Act in the first instance. P. 2441. The United States' motion for judgment on the pleadings granted. Bruce S. Flushman, San Francisco, Cal., for plaintiff. Louis F. Claiborne, Washington, D. C., for defendant. Justice WHITE delivered the opinion of the Court. 1 The issue before the Court is the ownership of oceanfront land created through accretion to land owned by the United States on the coast of California. The decision turns on whether federal or state law governs the issue. 2 * From the time of California's admission to the Union in 1850, the United States owned the upland on the north side of the entrance channel to Humboldt Bay, Cal. In 1859 and 1871, the Secretary of the Interior ordered that certain of these lands, which fronted on the Pacific Ocean, the channel, and Humboldt Bay be reserved from public sale.1 Since that time the land has been continuously possessed by the United States and used as a Coast Guard Reservation. The Pacific shoreline along the Coast Guard site remained substantially unchanged until near the turn of the century when the United States began construction of two jetties at the entrance to Humboldt Bay.2 The jetty constructed on the north side of the entrance resulted in fairly rapid accretion on the ocean side of the Coast Guard Reservation, so that formerly submerged lands became uplands.3 One hundred and eightyfour acres of upland were created by the seaward movement of the ordinary high-water mark. This land, which remains barren save for a watchtower, is the subject of the dispute in this case. 3 The controversy arose in 1977 when the Coast Guard applied for permission from California to use this land to construct the watchtower.4 At this time it became evident that both California and the United States asserted ownership of the land. The United States eventually built the watchtower without obtaining California's permission.5 Invoking our original jurisdiction, California then filed this suit to quiet title to the subject land.6 We granted leave for California to file a bill of complaint. 454 U.S. 809, 102 S.Ct. 83, 70 L.Ed.2d 78 (1981). 4 California alleges that upon its admission to the Union on September 9, 1850, Act of Sept. 9, 1850, 9 Stat. 452, and by confirmation in the Submerged Lands Act, 67 Stat. 29, 43 U.S.C. § 1301 et seq., California became vested with absolute title to the tidelands and the submerged lands upon which, after construction of the jetties, alluvion was deposited, resulting in formation of the subject land. Because the accretion formed on sovereign state land, California maintains that its law should govern ownership. Under California law, a distinction is drawn between accretive changes to a boundary caused by natural forces and boundary changes caused by the construction of artificial objects. For natural accretive changes, the upland boundary moves seaward as the alluvion is deposited, resulting in a benefit to the upland owner. Los Angeles v. Anderson, 206 Cal. 662, 667, 275 P. 789, 791 (1929). When accretion is caused by construction of artificial works, however, the boundary does not move but becomes fixed at the ordinary high water mark at the time the artificial influence is introduced. Carpenter v. Santa Monica, 63 Cal.App.2d 772, 794, 147 P.2d 964, 975 (1944). It is not disputed that the newly formed land in controversy was created by the construction of the jetty. Therefore, if state law governs, California would prevail. 5 By its answer, and supporting memoranda, the United States contends that the formerly submerged lands were never owned by California before passage of the Submerged Lands Act in 1953, and that the disputed land was not granted to California by the Act. The United States also submits that the case is governed by federal rather than state law and that under long-established federal law, accretion, whatever its cause, belongs to the upland owner. Jones v. Johnston, 18 How. 150, 156, 15 L.Ed. 320 (1856); County of St. Clair v. Lovingston, 23 Wall. 46, 66, 21 L.Ed. 813 (1874); Jefferis v. East Omaha Land Co., 134 U.S. 178, 189-193, 10 S.Ct. 518, 520-22, 33 L.Ed. 872 (1890); Beaver v. United States, 350 F.2d 4, 10-11 (CA9 1965).7 If such federal law controls, title to the deposited land vested in the United States as the accretions formed. 6 Recognizing that the choice-of-law issue was clearly drawn, California moved for summary judgment and the United States moved for judgment on the pleadings. No essential facts being in dispute, a special master was not appointed and the case was briefed and argued. We conclude that federal law governs the decision in this case and that the land in dispute is owned by the United States. II 7 In Borax Consolidated, Ltd. v. Los Angeles, 296 U.S. 10, 56 S.Ct. 23, 80 L.Ed. 9 (1935), the city filed suit to quiet its title to land claimed to be tideland and to belong to the city by virtue of a grant from the State. The defendant claimed by virtue of a patent from the United States issued after California entered the Union. In an opinion by Chief Justice Hughes, and with a single dissent, the Court held that if the land in question was tideland, the title passed to California at the time of her admission to the Union in 1850; that it remained to be determined whether the land at issue was tideland; and that this issue was "necessarily a federal question" controlled by federal law. The Court said: 8 "Petitioners claim under a federal patent which, according to the plat, purported to convey land bordering on the Pacific Ocean. There is no question that the United States was free to convey the upland, and the patent affords no ground for holding that it did not convey all the title that the United States had in the premises. The question as to the extent of this federal grant, that is, as to the limit of the land conveyed, or the boundary between the upland and the tideland, is necessarily a federal question. It is a question which concerns the validity and effect of an act done by the United States; it involves the ascertainment of the essential basis of a right asserted under federal law. Packer v. Bird, 137 U.S. 661, 669, 670 [11 S.Ct. 210, 211, 212, 34 L.Ed. 819]; Brewer-Elliott Oil Co. v. United States, 260 U.S. 77, 87 [43 S.Ct. 60, 64, 67 L.Ed. 140]; United States v. Holt Bank, 270 U.S. 49, 55, 56 [46 S.Ct. 197, 199, 70 L.Ed. 465]; United States v. Utah, 283 U.S. 64, 75 [51 S.Ct. 438, 440, 75 L.Ed. 844]. Rights and interests in the tideland, which is subject to the sovereignty of the State, are matters of local law. Barney v. Keokuk [4 Otto 324, 342], 94 U.S. 324, 338 [24 L.Ed. 224]; Shively v. Bowlby, [152 U.S. 1,] 40 [, 14 S.Ct. 548, 562, 38 L.Ed. 331]; Hardin v. Jordan, 140 U.S. 371, 382 [11 S.Ct. 808, 812, 35 L.Ed. 428]; Port of Seattle v. Oregon & Washington R. Co., 255 U.S. 56, 63 [41 S.Ct. 237, 239, 65 L.Ed. 500]". Borax Consolidated, Ltd. v. Los Angeles, supra, 296 U.S., at 22, 56 S.Ct., at 29. 9 The Court went on to hold that tidelands extend to the mean high-water line, which the Court then defined as a matter of federal law. 10 There was no question of accretions to the shoreline of the property involved in Borax. But some 30 years later, Mrs. Stella Hughes, the successor in interest to the owner of oceanfront property patented by the United States prior to the entry of the State of Washington into the Union, sued the State seeking to quiet her title to accretions that had become attached to her land and that had caused a seaward movement of the shoreline. Under Washington law, the accretions belonged to the State, the owner of the tidelands, and Mrs. Hughes would no longer own property fronting on the ocean. Under federal law accretions are the property of the upland owner. The trial court found that federal law applied. The Washington Supreme Court reversed, holding that Washington law applied and that the State owned any land that accreted after statehood. Hughes v. State, 67 Wash.2d 799, 410 P.2d 20 (1966). 11 We in turn reversed, reaffirming the decision in Borax that federal law determined the boundary between state-owned tidelands and property granted under a federal patent and holding that the same law applied to determine the boundary between state-owned tidelands and oceanfront property where accretions had extended the shoreline seaward. Hughes v. Washington, 389 U.S. 290, 88 S.Ct. 438, 19 L.Ed.2d 530 (1967).8 The justification for employing federal law was the special nature of the coastal boundary question: "The rule deals with waters that lap both the lands of the State and the boundaries of the international sea. This relationship, at this particular point of the marginal sea, is too close to the vital interest of the Nation in its own boundaries to allow it to be governed by any law but the 'supreme Law of the Land.' " Id., at 293, 88 S.Ct., at 440. We went on to decide that under federal law, the federal grantee of the uplands had the right to the accumulated accretions. 12 Except for the fact that in the present case the upland to which the accretions attached has always been owned by the United States, this case and Hughes are similarly situated. Unless Hughes is to be overruled, judgment must be entered for the United States. 13 California urges that for all intents and purposes Hughes has already been eviscerated by Oregon ex rel. State Land Board v. Corvallis Sand & Gravel Co., 429 U.S. 363, 97 S.Ct. 582, 50 L.Ed.2d 550 (1977). Corvallis involved a dispute between the State of Oregon and an Oregon corporation over the ownership of land that became part of a riverbed because of avulsive changes in the river's course. The Oregon Court of Appeals affirmed the trial court's award of the land to the corporation because that was the result dictated by federal common law, which, under Bonelli Cattle Co. v. Arizona, 414 U.S. 313, 94 S.Ct. 517, 38 L.Ed.2d 526 (1973), was the proper source of law. A majority of this Court reversed, overruling Bonelli and holding that the disputed ownership of the riverbed should be decided solely as a matter of Oregon law. Bonelli's error was said to have been reliance on the equal-footing doctrine as a source of federal common law.9 Once the equal-footing doctrine had vested title to the riverbed in Arizona, "it did not operate after that date to determine what effect on titles the movement of the river might have." 429 U.S., at 371, 94 S.Ct., at 587. State, rather than federal law, should have been applied. 14 California urges that in rejecting Bonelli and holding that disputes about the title to lands granted by the United States are to be settled by state law, the Court also rejected Hughes since that case involved land that had been patented by the United States to private owners. We do not agree. Corvallis itself recognized that federal law would continue to apply if "there were present some other principle of federal law requiring state law to be displaced." 429 U.S., at 371, 94 S.Ct., at 587. For example, the effects of accretive and avulsive changes in the course of a navigable stream forming an interstate boundary is determined by federal law. Id., at 375, 94 S.Ct., at 589. The Corvallis opinion also recognized that Bonelli did not rest upon Hughes and that the Hughes Court considered oceanfront property "sufficiently different . . . so as to justify a 'federal common law' rule of riparian proprietorship." 429 U.S., at 377, n.6, 94 S.Ct., at 590, n.6. The Corvallis decision did not purport to disturb Hughes. 15 Wilson v. Omaha Indian Tribe, 442 U.S. 653, 99 S.Ct. 2529, 61 L.Ed.2d 153 (1979), made clear that Corvallis also does not apply "where the [United States] Government has never parted with title and its interest in the property continues." 442 U.S., at 670, 99 S.Ct., at 2539.10 The dispute inCorvallis was between the State and a private owner of land previously in federal possession. In contrast, the riparian owner in Wilson was the United States, holding reservation land in trust for the Omaha Indian Tribe. The issue was the effect of accretive or avulsive changes in the course of a navigable stream. State boundaries were not involved. What we said in Wilson is at least equally applicable here where the United States has held title to, occupied, and utilized the littoral land for over 100 years: "[T]he general rule recognized by Corvallis does not oust federal law in this case. Here, we are not dealing with land titles merely derived from a federal grant, but with land with respect to which the United States has never yielded title or terminated its interest." 442 U.S., at 670, 99 S.Ct., at 2539. 16 We conclude, based on Hughes v. Washington and Wilson v. Omaha Indian Tribe, that a dispute over accretions to oceanfront land where title rests with or was derived from the Federal Government is to be determined by federal law. III 17 Controversies governed by federal law do not inevitably require resort to uniform federal rules. Wilson v. Omaha Indian Tribe, supra, at 672, 99 S.Ct., at 2540. It may be determined as a matter of choice of law that, although federal law should govern a given question, state law should be borrowed and applied as the federal rule for deciding the substantive legal issue at hand. Board of Commissioners of Jackson County v. United States, 308 U.S. 343, 60 S.Ct. 285, 84 L.Ed. 313 (1939); Royal Indemnity Co. v. United States, 313 U.S. 289, 61 S.Ct. 995, 85 L.Ed. 1361 (1941). This is not such a case. First, and dispositive in itself, is the fact that Congress has addressed the issue of accretions to federal land. The Submerged Lands Act, 43 U.S.C. § 1301 et seq., vested title in the States to the lands underlying the territorial sea, which, in California's case, extended three miles seaward from the ordinary low-water line. The Act also confirmed the title of the States to the tidelands up to the line of mean high tide. Section 5(a) of the Act, however withheld from the grant to the States all "accretions" to coastal lands acquired or reserved by the United States.11 43 U.S.C. § 1313(a). In light of this provision, borrowing for federal-law purposes a state rule that would divest federal ownership is foreclosed. In Wilson, where we did adopt state law as the federal rule, no special federal concerns, let alone a statutory directive, required a federal common-law rule. 18 Moreover, this is not a case in which federal common law must be created. For over 100 years it has been settled under federal law that the right to future accretions is an inherent and essential attribute of the littoral or riparian owner. New Orleans v. United States, 10 Pet. 662, 717, 9 L.Ed. 661 (1836); County of St. Clair v. Lovingston, 23 Wall., at 68. " 'Almost all jurists and legislators, . . . both ancient and modern, have agreed that the owner of the land thus bounded is entitled to these additions.' " Jefferis v. East Omaha Land Co., 134 U.S., at 189, 10 S.Ct., at 520, quoting Banks v. Ogden, 2 Wall. 57, 67, 17 L.Ed. 818 (1865). We rejected the invitation to rely on state law in Hughes, which California readily admits is a case "in which the facts and issues are essentially identical," Statement in Support of Motion for Leave to File Complaint 16, and we see no reason at this juncture to adopt California's minority rule on artificial accretions,12 even if we were free to do so. 19 Applying the federal rule that accretions, regardless of cause, accrue to the upland owner, we conclude that title to the entire disputed land in issue is vested in the United States. IV 20 Despite Hughes and Wilson, California claims ownership of the disputed lands because all of the accretions were deposited on tidelands and submerged lands, title to which, California submits, was vested in the State by the equal-footing doctrine and confirmed by the Submerged Lands Act. But California's claim to the land underlying the territorial sea was firmly rejected in United States v. California, 332 U.S. 19, 67 S.Ct. 1658, 91 L.Ed. 1889 (1947), which held that only land underneath inland waters was included in the initial grant to the States under the equal-footing doctrine. Furthermore, the Submerged Lands Act was a constitutional exercise of Congress' power to dispose of federal property, Alabama v. Texas, 347 U.S. 272, 273-274, 74 S.Ct. 481, 482, 98 L.Ed. 689 (1954), and "did not impair the validity" of the California decision, United States v. Louisiana, 363 U.S. 1, 7, 20, 80 S.Ct. 961, 967, 973, 4 L.Ed.2d 1025 (1960).13 In any event, whatever the ownership of the submerged lands, this approach, based as it is on the equal-footing doctrine and the federal statute, is not a claim that state law should govern but a claim that the historic rule that accretions belong to the upland owner is wrong and should be replaced with a rule awarding title to the owner of the land on which the accretions took place. To accept this submission, however, would require rejecting not only Hughes, but also the long-established federal rule that accretions belong to the upland owner—a doctrine consistent with the majority rule prevailing in the States. See Part III, supra. Indeed, the proposed rule is also inconsistent with California's own law that accretions attributable to natural causes belong to the upland owner. For all these reasons, we refuse the invitation to depart from the long-settled rule.14 21 Independent of the above analysis, California claims that the United States expressly surrendered title to the disputed land through the Submerged Lands Act. California argues the subject land falls within the general grant to the States of "lands beneath naviagable waters." Section 2(a)(3) of the Act defines "lands beneath navigable waters" to include "all filled in, made, or reclaimed lands which formerly were lands beneath navigable waters." 43 U.S.C. § 1301(a)(3). Because the jetty construction caused fairly rapid accretion, and, but for the construction of the jetties, the subject land would have remained submerged, California submits the accretion-formed land is "made" land, whose title rests in California by virtue of the Submerged Lands Act. 22 We do not read this provision of the Act as applying to the gradual process by which sand accumulated along the shore, although caused by a jetty affecting the action of the sea.15 Moreover, to the extent that the accretions are to be considered "made" land, they would fall within the reservation by the United States of "all lands filled in, built up, or otherwise reclaimed by the United States for its own use." This follows from the congressional object to assure each sovereign the continuing benefit of landfill and like work performed by each.16 In any event, § 5(a) of the Act expressly withholds from the grant to the States all "accretions" to lands reserved by the United States, and both California and the United States agree that the exposure of the formerly submerged lands in dispute constitutes "accretion." This reading of the Act adheres to the principle that federal grants are to be construed strictly in favor of the United States. United States v. Grand River Dam Authority, 363 U.S. 229, 235, 80 S.Ct. 1134, 1138, 4 L.Ed.2d 1186 (1960); United States v. Union Pacific R. Co., 353 U.S. 112, 116, 77 S.Ct. 685, 687, 1 L.Ed.2d 693 (1957). 23 Finally, California submits that the Act granted title to the State by confirming the title of persons who, on June 5, 1950, were entitled to such lands "under the law of the respective States in which the land is located . . . ." 43 U.S.C. § 1311(a). This provision means nothing more than that state law determines the proper beneficiary of the grant of land under the Act; it is clear that federal law determines the scope of the grant under the Act in the first instance. V 24 We reaffirm today that federal law determines the boundary of oceanfront lands owned or patented by the United States. Applying the federal rule that accretions of whatever cause belong to the upland owner, we find that title to the disputed parcel rests with the United States. Accordingly, California's motion for summary judgment is denied, and the United States' motion for judgment on the pleadings is granted. The parties, or either of them, may, before September 27, 1982, submit a proposed decree to carry this opinion into effect, failing which the Court will prepare and enter an appropriate decree at the next Term of Court. 25 It is so ordered. 26 Justice REHNQUIST, with whom Justice STEVENS and Justice O'CONNOR join, concurring in the judgment. 27 I concur in the judgment. I believe that our decision in Wilson v. Omaha Indian Tribe, 442 U.S. 653, 99 S.Ct. 2529, 61 L.Ed.2d 153 (1979), requires the application of federal common law to resolve this title dispute between the United States and California, and that § 5(a) of the Submerged Lands Act indicates the source of that law. 28 The dispute in this case concerns the ownership of artificially caused accretions on oceanfront property belonging to the United States. The dispute centers on the legal effect of the movement of the "mean high-water mark." That mark separates the fastlands continuously owned by the United States from the "tidelands"—the area of partially submerged lands between the mean high- and low-water marks. California's claim of title to the tidelands is based upon the equal-footing doctrine. Because the tidelands belong to it and because the accretions formed on the tidelands, California contends that state law applies to resolve this title dispute between it and the United States. The rule adopted by the California courts regarding artificially caused accretions holds that title to accreted land vests with the State rather than the riparian or littoral owner. The United States contends that federal common law applies and argues that the federal common-law rule holds that title to land formed by accretion vests in the owner of the riparian land. 29 The dispute in this case is similar to that in Wilson v. Omaha Indian Tribe. We held in Wilson that federal common law and not state law governs title disputes resulting from changes in the course of a navigable stream where an instrumentality of the Federal Government is the riparian owner. 442 U.S., at 669-671, 99 S.Ct., at 2538-2539. The rule of Oregon ex rel. State Land Board v. Corvallis Sand & Gravel Co., 429 U.S. 363, 97 S.Ct. 582, 50 L.Ed.2d 550 (1977), was distinguished. The Corvallis rule—that state law governs—applies where the dispute over the legal effect of a shifting riverbed does not involve claims of title by a federal instrumentality. 30 I agree with the Court that the Wilson rule applies to oceanfront property as well as riverfront property where the Federal Government is the littoral owner. Wilson should apply to the movement of the high-water mark along the ocean in a fashion similar to the way it applies to changes in the bed of a navigable stream. In the instant case, as in Wilson, it is irrelevant that the accretion, as a geographical "fact," formed on land within the State's dominion, be it a river bottom or the ocean tidelands. The fact is that bothWilson and the instant case concern title disputes over changes in the shoreline where the Federal Government owns land along the shoreline. 31 In Wilson, we held that state law supplied the applicable rule of decision even though federal common law applied to resolve the title dispute. We found no need for a uniform national rule and no reason why federal interests should not be treated under the same rules of property that would apply to private persons. In contrast to Wilson, however, I agree with the Court that Congress in § 5(a) of the Submerged Lands Act has supplied the rule of decision. Section 5(a) withholds from the grant to the States all accretions to coastal lands acquired or reserved by the United States. I also agree with the Court that California did not acquire the disputed lands pursuant to the "made lands" provisions in § 2(a)(3). 32 Consequently, the Court's discussion regarding the continuing vitality of Hughes v. Washington, 389 U.S. 290, 88 S.Ct. 438, 19 L.Ed.2d 530 (1967), is dicta. Hughes is unnecessary to the resolution of choice-of-law issues in title disputes between the Federal Government and a State or private person. Reliance on Hughes would be necessary only if we were to hold that federal common law, rather than state law, applied in a title dispute between a federal patentee and a State or private persons as to lands fronting an ocean. The instant case does not present that issue. It is difficult to reconcile Hughes with Corvallis and we should postpone that endeavor until required to undertake it. 33 In summary, I think this case can be easily resolved as a title dispute between the United States and California concerning the legal effect of movement of the Pacific Ocean's high-water mark. Wilson and the Submerged Lands Act resolve the dispute. The continuing vitality of Hughes should be left to another day. 1 Secretarial Order, December 27, 1859; Secretarial Order, August 19, 1871. See Exhibit C to Exhibits in Support of California's Motion for Leave to File Complaint. 2 Construction of the jetties commenced on the South Spit in 1889 and on the North Spit in 1890. U. S. Army Corps of Engineers, San Francisco District, Survey Report on Humboldt Bay, California, App. I, Shoreline Changes 2-3, 8-9 (Feb. 10, 1950), Exhibit D (hereafter cited as Corps Report). The north jetty was a massive work, having a total length of 7,500 feet. 3 The United States and California agree that the seaward shift of the shoreline was caused by the construction of the jetties. A study by the Army Corps of Engineers found: "With the inauguration of jetty construction in 1890, there began a series of interruptions in normal littoral transport [of sand]. With each increment in length of the jetties the [Humboldt] bar was pushed seaward. Consequent decrease in offshore depths caused the shore to advance on each side of the inlet." Id., at 8, ¶ 25. After jetty construction, ". . . the Humboldt bar . . . shifted and reformed seaward of its 1870 position, and the ocean high-water shore line along the north spit . . . shifted seaward. The seaward advance of the north spit shore line was most pronounced upon reconstruction of the north jetty in 1917." Id., at 9, ¶ 25. 4 California does not contend that, having applied for a state permit, the United States is estopped from asserting its claim to ownership of the disputed land. Tr. of Oral Arg. at 5-6. Such an argument is foreclosed by United States v. California, 332 U.S. 19, 39-40, 67 S.Ct. 1658, 1668-69, 91 L.Ed. 1889 (1947) (footnote omitted): "[O]fficers who have no authority at all to dispose of Government property cannot by their conduct cause the Government to lose its valuable rights by their acquiescence, laches, or failure to act." See also United States v. City and County of San Francisco, 310 U.S. 16, 31-32, 60 S.Ct. 749, 757, 84 L.Ed. 1050 (1940); Utah v. United States, 284 U.S. 534, 545-546, 52 S.Ct. 232, 235, 76 L.Ed. 469 (1932). 5 In May 1978, California transmitted a proposed permit to the United States to allow construction of the watchtower. See Corps Report, Exhibit F. A few days later, the Bureau of Land Management of the Department of the Interior formally advised the Coast Guard and the California Commission that the United States claimed the disputed acreage as accretion. Letter of June 5, 1978, attached to Corps Report, Exhibit G. The proposed permit was never executed. 6 Disputes between a State and the United States over ownership of property are fully within our original jurisdiction over cases in "which a State shall be Party," Art. III, § 2, cl. 2. Although our jurisdiction over this matter is concurrent with that of the district courts, California v. Arizona, 440 U.S. 59, 65, 99 S.Ct. 919, 923, 59 L.Ed.2d 144 (1979); 28 U.S.C. § 1251(b)(2), we have previously indicated that coastal boundary disputes are appropriately brought as original actions in this Court. United States v. Alaska, 422 U.S. 184, 186, n.2, 95 S.Ct. 2240, 2244, n.2, 45 L.Ed.2d 109 (1975). The United States has waived its immunity to suit in actions brought against it to quiet title to land. 28 U.S.C. § 1346(f). See California v. Arizona, supra, 440 U.S., at 65-68, 99 S.Ct., at 923-924. 7 California's claim that Wilson v. Omaha Indian Tribe, 442 U.S. 653, 672, 99 S.Ct. 2529, 2540, 61 L.Ed.2d 153 (1979), determined that there was no "federal common law" of accretion and avulsion, is a misunderstanding of that decision. We said only that "[t]he federal law applied in boundary cases . . . does not necessarily furnish the appropriate rules to govern" a case not involving a boundary dispute. Too much is also read into dictum in Oregon ex rel. State Land Board v. Corvallis Sand & Gravel Co., 429 U.S. 363, 380-381, n. 8, 97 S.Ct. 582, 591-592, n. 8, 50 L.Ed.2d 550 (1977), taking issue with the dissent's meaning of the term "federal common law." 8 All participating Justices joined except Justice Stewart, who concurred on grounds that the State's claim to the property constituted a taking without compensation. He rejected the majority's application of federal law to the question. Justice MARSHALL took no part in the case. 9 The equal-footing principle holds that all States admitted to the Union possess the same rights and sovereignty as the original 13 States. Pollard's Lessee v. Hagan, 3 How. 212, 229, 11 L.Ed. 565 (1845); Shively v. Bowlby, 152 U.S. 1, 26, 30, 14 S.Ct. 548, 557, 559, 38 L.Ed. 331 (1894). 10 The majority opinion in Corvallis appears to recognize that its rule does not extend to land remaining in federal hands: " 'We hold the true principle to be this, that whenever the question in any Court, state or federal, is, whether a title to land which had once been property of the United States has passed, that question must be resolved by the laws of the United States; but that whenever, according to these laws, the title shall have passed, then that property, like all other property in the state, is subject to state legislation; so far as that legislation is consistent with the admission that the title passed and vested according to the laws of the United States.' " 429 U.S., at 377, 94 S.Ct., at 590 (quoting Wilcox v. Jackson, 13 Pet. 498, 517, 10 L.Ed. 264 (1839); emphasis added by Corvallis Court). 11 In relevant part, § 5(a) of the Act, 62 Stat. 32, 43 U.S.C. § 1313(a), excepts from the grant to the States "all tracts or parcels of land together with all accretions thereto, . . . title to which has been lawfully and expressly acquired by the United States . . . and . . . all lands expressly retained by or ceded to the United States when the State entered the Union. . . ." Although "accretions" are expressly mentioned only in connection with federal "acquired lands," accretions to retained lands should be similarly excepted from the grant to the States. Former Solicitor General Cox, in an opinion approved by the Attorney General, explained: "There can be no doubt that Congress intended each of the various categories of lands excepted by section 5(a) to include accretions. The terms of section 5(a) make this clear. The customary rights of landowners are set forth in full in the first of the several exceptions listed in section 5(a). Thus, it speaks of 'all tracts or parcels of land together with all accretions thereto, resources therein, or improvements thereon. . . .' Each of the other exceptions speaks simply of 'all lands.' Obviously, the more comprehensive word 'lands' was used instead of 'tracts or parcels of land' and the explicit reference to accretions, resources and improvements was omitted in order to avoid repetition. There is no reasonable basis for any other conclusion. Congress would not have limited its exceptions of 'all accretions thereto, resources therein, or improvements thereon' to lands 'lawfully and expressly acquired by the United States' from any State or its grantees and then denied them where the lands were 'expressly retained' or 'acquired by the United States by eminent domain proceedings, purchase, cession, gift, or otherwise in a proprietary capacity. . . .' " 42 Op.Atty.Gen. 241, 264 (1963). 12 In United States v. California, O. T. 1951, No. 6, Orig., California argued that the "Court should adopt the federal rule that accretions formed by gradual and imperceptible degrees even though induced by artificial structures accrue to the owner of the adjoining land." Brief in Relation to Report of Special Master 90. California suggested "ample reasons why [the] exceptional California view should not be extended and applied in determining the boundaries of the marginal sea off California." Id., at 91. Those reasons included the fact that the California rule is contrary to that adopted by courts of most other States, that the application of state law would lead to varying results in different States, and that the California rule was devised for wholly inapplicable reasons. 13 See also Alabama v. Texas, 347 U.S., at 273-274, 74 S.Ct., at 482; United States v. California, 381 U.S. 139, 145-148, 85 S.Ct. 1401, 1405-06, 14 L.Ed.2d 296 (1965); United States v. Louisiana, 389 U.S. 155, 156-157, 88 S.Ct. 367, 368, 19 L.Ed.2d 383 (1967); Texas Boundary Case, 394 U.S. 1, 2, 89 S.Ct. 768, 769, 22 L.Ed.2d 36 (1969); United States v. Maine, 420 U.S. 515, 524-526, 95 S.Ct. 1155, 1160-61, 43 L.Ed.2d 363 (1975); United States v. Louisiana, 446 U.S. 253, 256, 268, 100 S.Ct. 1618, 1621, 1627, 64 L.Ed.2d 196 (1980). 14 For the same reasons, we reject California's alternative theory that the equal-footing doctrine vests title in the State to all lands that ever were tidelands. California argues that as deposition occurred on submerged land, these areas went to a tideland phase—vesting title in the State—before eventually emerging as uplands. Federal law governs the scope of title initially vested by the equal-footing doctrine; at most, this argument suggests a different federal rule should apply to former tidelands. The suggestion has little to recommend it. Even leaving aside the concerns expressed in text, we see no reason for an exceptional rule to apply to land that once was, but no longer is, tideland. Moreover, implementation of the rule would require plotting the high- and low-water lines at all intervening times between statehood and the present. 15 The word "made" was inserted into the provision in a bill introduced by Congressman Walter. H.R. 8137, 81st Cong., 2d Sess., § 2(a)(2) (1950). The Report on that measure describes it as "in substance, the same" as earlier proposals omitting the term. H.R.Rep.No. 2078, 81st Cong., 2d Sess., 3 (1950). Throughout Congress' consideration of the bill there was no comment on the "made" land provision. No Member of either House ever suggested that § 1301(a)(3) covered accretions that were attributable to artificial works. Against this background, we find no significance in the two casual references by Robert Moses and Senator Daniel to naturally formed accretions as "made." Hearings on S.J.Res. 13 et al. before the Senate Committee on Interior and Insular Affairs, 83d Cong., 1st Sess., 158 (1953) (remarks of Robert Moses); id., at 193-194 (remarks of Sen. Daniel). 16 The interpretive opinion rendered by former Solicitor General Cox, while including naturally formed islands within the "made" language of § 2(a)(3), rejects the suggestion that accretion to the mainland, whether or not directly attributable to artificial causes, is included in the Submerged Lands Act grant to the States. 42 Op.Atty.Gen., at 259-265, 266-267. We express no opinion on the Act's treatment of naturally formed islands in the marginal sea.

910

73 L.Ed.2d 48 102 S.Ct. 2466 457 U.S. 332 ARIZONA, Petitionerv.MARICOPA COUNTY MEDICAL SOCIETY et al. No. 80-419. Argued Nov. 4, 1981. Decided June 18, 1982. Syllabus Respondent foundations for medical care were organized by respondent Maricopa County Medical Society and another medical society to promote fee-for-service medicine and to provide the community with a competitive alternative to existing health insurance plans. The foundations, by agreement of their member doctors, established the maximum fees the doctors may claim in full payment for health services provided to policyholders of specified insurance plans. Petitioner State of Arizona filed a complaint against respondents in Federal District Court, alleging that they were engaged in an illegal price-fixing conspiracy in violation of § 1 of the Sherman Act. The District Court denied the State's motion for partial summary judgment, but certified for interlocutory appeal the question whether the maximum-fee agreements were illegal per se under § 1 of the Sherman Act. The Court of Appeals affirmed the denial of the motion for partial summary judgment and held that the certified question could not be answered without evaluating the purpose and effect of the agreements at a full trial. Held: The maximum-fee agreements, as price-fixing agreements, are per se unlawful under § 1 of the Sherman Act. Pp. 342-357. (a) The agreements do not escape condemnation under the per se rule against price-fixing agreements because they are horizontal and fix maximum prices. Horizontal agreements to fix maximum prices are on the same legal—even if not economic—footing as agreements to fix minimum or uniform prices. Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219; Albrecht v. Herald Co., 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998. The per se rule is violated here by a price restraint that tends to provide the same economic rewards to all practitioners regardless of their skill, experience, training, or willingness to employ innovative and difficult procedures in individual cases. Such a restraint may also discourage entry into the market and may deter experimentation and new developments by individual entrepreneurs. P. 348. (b) Nor does the fact that doctors rather than nonprofessionals are the parties to the price-fixing agreements preclude application of the per se rule. Respondents do not claim that the quality of the professional services their members provide is enhanced by the price restraint, Goldfarb v. Virginia State Bar, 421 U.S. 773, 95 S.Ct. 2004, 44 L.Ed.2d 572, and National Society of Professional Engineers v. United States, 435 U.S. 679, 98 S.Ct. 1355, 55 L.Ed.2d 637, distinguished, and their claim that the price restraint will make it easier for customers to pay does not distinguish the medical profession from any other provider of goods or services. Pp. 348-349. (c) That the judiciary has had little antitrust experience in the health care industry is insufficient reason for not applying the per se rule here. "[T]he Sherman Act, so far as price-fixing agreements are concerned, establishes one uniform rule applicable to all industries alike." United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 222, 60 S.Ct. 811, 843, 84 L.Ed. 1129. Pp. 349-351. (d) The per se rule is not rendered inapplicable in this case for the alleged reason that the agreements in issue have procompetitive justification. The anticompetitive potential in all price-fixing agreements justifies their facial invalidation even if procompetitive justifications are offered for some. Even when respondents are given every benefit of doubt, the record in this case is not inconsistent with the presumption that respondents' agreements will not significantly enhance competition. The most that can be said for having doctors fix the maximum prices is that doctors may be able to do it more efficiently than insurers, but there is no reason to believe any savings that might accrue from this arrangement would be sufficiently great to affect the competitiveness of these kinds of insurance plans. Pp. 351-354. (e) Respondents' maximum-fee schedules do not involve price-fixing in only a literal sense. Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1, distinguished. As agreements among independent competing entrepreneurs, they fit squarely into the horizontal price-fixing mold. Pp. 355-357 9th Cir., 643 F.2d 553, reversed. Kenneth R. Reed, Phoenix, Ariz., for petitioner. Stephen M. Shapiro, Washington, D. C., for the United States, as amicus curiae, by special leave of Court. Philip P. Berelson, Phoenix, Ariz., for respondents. [Amicus Curiae Information from pages 334-336 intentionally omitted] Justice STEVENS delivered the opinion of the Court. 1 The question presented is whether § 1 of the Sherman Act, 26 Stat. 209, as amended, 15 U.S.C. § 1, has been violated by agreements among competing physicians setting, by majority vote, the maximum fees that they may claim in full payment for health services provided to policyholders of specified insurance plans. The United States Court of Appeals for the Ninth Circuit held that the question could not be answered without evaluating the actual purpose and effect of the agreements at a full trial. 643 F.2d 553 (1980). Because the undisputed facts disclose a violation of the statute, we granted certiorari, 450 U.S. 979, 101 S.Ct. 1512, 67 L.Ed.2d 813 (1981), and now reverse. 2 * In October 1978 the State of Arizona filed a civil complaint against two county medical societies and two "foundations for medical care" that the medical societies had organized. The complaint alleged that the defendants were engaged in illegal price-fixing conspiracies.1 After the defendants filed answers, one of the medical societies was dismissed by consent, the parties conducted a limited amount of pretrial discovery, and the State moved for partial summary judgment on the issue of liability. The District Court denied the motion,2 but entered an order pursuant to 28 U.S.C. § 1292(b), certifying for interlocutory appeal the question "whether the FMC membership agreements, which contain the promise to abide by maximum fee schedules, are illegal per se under section 1 of the Sherman Act."3 3 The Court of Appeals, by a divided vote, affirmed the District Court's order refusing to enter partial summary judgment, but each of the three judges on the panel had a different view of the case. Judge Sneed was persuaded that "the challenged practice is not a per se violation." 643 F.2d, at 560.4 Judge Kennedy, although concurring, cautioned that he had not found "these reimbursement schedules to be per se proper, [or] that an examination of these practices under the rule of reason at trial will not reveal the proscribed adverse effect on competition, or that this court is foreclosed at some later date, when it has more evidence, from concluding that such schedules do constitute per se violations." Ibid.5 Judge Larson dissented, expressing the view that a per se rule should apply and, alternatively, that a rule-of-reason analysis should condemn the arrangement even if a per se approach was not warranted. Id., at 563-569.6 4 Because the ultimate question presented by the certiorari petition is whether a partial summary judgment should have been entered by the District Court, we must assume that the respondents' version of any disputed issue of fact is correct. We therefore first review the relevant undisputed facts and then identify the factual basis for the respondents' contention that their agreements on fee schedules are not unlawful. II 5 The Maricopa Foundation for Medical Care is a nonprofit Arizona corporation composed of licensed doctors of medicine, osteopathy, and podiatry engaged in private practice. Approximately 1,750 doctors, representing about 70% of the practitioners in Maricopa County, are members. 6 The Maricopa Foundation was organized in 1969 for the purpose of promoting fee-for-service medicine and to provide the community with a competitive alternative to existing health insurance plans.7 The foundation performs three primary activities. It establishes the schedule of maximum fees that participating doctors agree to accept as payment in full for services performed for patients insured under plans approved by the foundation. It reviews the medical necessity and appropriateness of treatment provided by its members to such insured persons. It is authorized to draw checks on insurance company accounts to pay doctors for services performed for covered patients. In performing these functions, the foundation is considered an "insurance administrator" by the Director of the Arizona Department of Insurance. Its participating doctors, however, have no financial interest in the operation of the foundation. 7 The Pima Foundation for Medical Care, which includes about 400 member doctors,8 performs similar functions. For the purposes of this litigation, the parties seem to regard the activities of the two foundations as essentially the same. No challenge is made to their peer review or claim administration functions. Nor do the foundations allege that these two activities make it necessary for them to engage in the practice of establishing maximum-fee schedules. 8 At the time this lawsuit was filed,9 each foundation made use of "relative values" and "conversion factors" in compiling its fee schedule. The conversion factor is the dollar amount used to determine fees for a particular medical specialty. Thus, for example, the conversion factors for "medicine" and "laboratory" were $8 and $5.50, respectively, in 1972, and $10 and $6.50 in 1974. The relative value schedule provides a numerical weight for each different medical service—thus, an office consultation has a lesser value than a home visit. The relative value was multiplied by the conversion factor to determine the maximum fee. The fee schedule has been revised periodically. The foundation board of trustees would solicit advice from various medical societies about the need for change in either relative values or conversion factors in their respective specialties. The board would then formulate the new fee schedule and submit it to the vote of the entire membership.10 9 The fee schedules limit the amount that the member doctors may recover for services performed for patients insured under plans approved by the foundations. To obtain this approval the insurers—including self-insured employers as well as insurance companies11—agree to pay the doctors' charges up to the scheduled amounts, and in exchange the doctors agree to accept those amounts as payment in full for their services. The doctors are free to charge higher fees to uninsured patients, and they also may charge any patient less than the scheduled maxima. A patient who is insured by a foundation-endorsed plan is guaranteed complete coverage for the full amount of his medical bills only if he is treated by a foundation member. He is free to go to a nonmember physician and is still covered for charges that do not exceed the maximum-fee schedule, but he must pay any excess that the nonmember physician may charge. 10 The impact of the foundation fee schedules on medical fees and on insurance premiums is a matter of dispute. The State of Arizona contends that the periodic upward revisions of the maximum-fee schedules have the effect of stabilizing and enhancing the level of actual charges by physicians, and that the increasing level of their fees in turn increases insurance premiums. The foundations, on the other hand, argue that the schedules impose a meaningful limit on physicians' charges, and that the advance agreement by the doctors to accept the maxima enables the insurance carriers to limit and to calculate more efficiently the risks they underwrite and therefore serves as an effective cost-containment mechanism that has saved patients and insurers millions of dollars. Although the Attorneys General of 40 different States, as well as the Solicitor General of the United States and certain organizations representing consumers of medical services, have filed amicus curiae briefs supporting the State of Arizona's position on the merits, we must assume that the respondents' view of the genuine issues of fact is correct. 11 This assumption presents, but does not answer, the question whether the Sherman Act prohibits the competing doctors from adopting, revising, and agreeing to use a maximum-fee schedule in implementation of the insurance plans. III 12 The respondents recognize that our decisions establish that price-fixing agreements are unlawful on their face. But they argue that the per se rule does not govern this case because the agreements at issue are horizontal and fix maximum prices, are among members of a profession, are in an industry with which the judiciary has little antitrust experience, and are alleged to have procompetitive justifications. Before we examine each of these arguments, we pause to consider the history and the meaning of the per se rule against price-fixing agreements. A. 13 Section 1 of the Sherman Act of 1890 literally prohibits every agreement "in restraint of trade."12 In United States v. Joint Traffic Assn., 171 U.S. 505, 19 S.Ct. 25, 43 L.Ed. 259 (1898), we recognized that Congress could not have intended a literal interpretation of the word "every"; since Standard Oil Co. of New Jersey v. United States, 221 U.S. 1, 31 S.Ct. 502, 55 L.Ed. 619 (1911), we have analyzed most restraints under the so-called "rule of reason." As its name suggests, the rule of reason requires the factfinder to decide whether under all the circumstances of the case the restrictive practice imposes an unreasonable restraint on competition.13 14 The elaborate inquiry into the reasonableness of a challenged business practice entails significant costs. Litigation of the effect or purpose of a practice often is extensive and complex. Northern Pacific R. Co. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514, 518, 2 L.Ed.2d 545 (1958). Judges often lack the expert understanding of industrial market structures and behavior to determine with any confidence a practice's effect on competition. United States v. Topco Associates, Inc., 405 U.S. 596, 609-610, 92 S.Ct. 1126, 1134, 31 L.Ed.2d 515 (1972). And the result of the process in any given case may provide little certainty or guidance about the legality of a practice in another context. Id., at 609, n. 10, 92 S.Ct., at 1134, n.10; Northern Pacific R. Co. v. United States, supra, 356 U.S., at 5, 78 S.Ct., at 518. 15 The costs of judging business practices under the rule of reason, however, have been reduced by the recognition of per se rules.14 Once experience with a particular kind of restraint enables the Court to predict with confidence that the rule of reason will condemn it, it has applied a conclusive presumption that the restraint is unreasonable.15 As in every rule of general application, the match between the presumed and the actual is imperfect. For the sake of business certainty and litigation efficiency, we have tolerated the invalidation of some agreements that a fullblown inquiry might have proved to be reasonable.16 16 Thus the Court in Standard Oil recognized that inquiry under its rule of reason ended once a price-fixing agreement was proved, for there was "a conclusive presumption which brought [such agreements] within the statute." 221 U.S., at 65, 31 S.Ct., at 517. By 1927, the Court was able to state that "it has . . . often been decided and always assumed that uniform price-fixing by those controlling in any substantial manner a trade or business in interstate commerce is prohibited by the Sherman Law." United States v. Trenton Potteries Co., 273 U.S. 392, 398, 47 S.Ct. 377, 379, 71 L.Ed. 700. 17 "The aim and result of every price-fixing agreement, if effective, is the elimination of one form of competition. The power to fix prices, whether reasonably exercised or not, involves power to control the market and to fix arbitrary and unreasonable prices. The reasonable price fixed today may through economic and business changes become the unreasonable price of tomorrow. Once established, it may be maintained unchanged because of the absence of competition secured by the agreement for a price reasonable when fixed. Agreements which create such potential power may well be held to be in themselves unreasonable or unlawful restraints, without the necessity of minute inquiry whether a particular price is reasonable or unreasonable as fixed and without placing on the government in enforcing the Sherman Law the burden of ascertaining from day to day whether it has become unreasonable through the mere variation of economic conditions." Id., at 397-398, 47 S.Ct., at 379. 18 Thirteen years later, the Court could report that "for over forty years this Court has consistently and without deviation adhered to the principle that price-fixing agreements are unlawful per se under the Sherman Act and that no showing of so-called competitive abuses or evils which those agreements were designed to eliminate or alleviate may be interposed as a defense." United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 218, 60 S.Ct. 811, 841, 84 L.Ed. 1129 (1940). In that case a glut in the spot market for gasoline had prompted the major oil refiners to engage in a concerted effort to purchase and store surplus gasoline in order to maintain stable prices. Absent the agreement, the companies argued, competition was cutthroat and self-defeating. The argument did not carry the day: 19 "Any combination which tampers with price structures is engaged in an unlawful activity. Even though the members of the price-fixing group were in no position to control the market, to the extent that they raised, lowered, or stabilized prices they would be directly interfering with the free play of market forces. The Act places all such schemes beyond the pale and protects that vital part of our economy against any degree of interference. Congress has not left with us the determination of whether or not particular price-fixing schemes are wise or unwise, healthy or destructive. It has not permitted the age-old cry of ruinous competition and competitive evils to be a defense to price-fixing conspiracies. It has no more allowed genuine or fancied competitive abuses as a legal justification for such schemes than it has the good intentions of the members of the combination. If such a shift is to be made, it must be done by the Congress. Certainly Congress has not left us with any such choice. Nor has the Act created or authorized the creation of any special exception in favor of the oil industry. Whatever may be its peculiar problems and characteristics, the Sherman Act, so far as price-fixing agreements are concerned, establishes one uniform rule applicable to all industries alike." Id., at 221-222, 60 S.Ct., at 843. 20 The application of the per se rule to maximum-price-fixing agreements in Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219 (1951), followed ineluctably from Socony-Vacuum: 21 "For such agreements, no less than those to fix minimum prices, cripple the freedom of traders and thereby restrain their ability to sell in accordance with their own judgment. We reaffirm what we said in United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 223 [60 S.Ct. 811, 844, 84 L.Ed. 1129]: 'Under the Sherman Act a combination formed for the purpose and with the effect of raising, depressing, fixing, pegging, or stabilizing the price of a commodity in interstate or foreign commerce is illegal per se.' " 340 U.S., at 213, 60 S.Ct., at 839. 22 Over the objection that maximum-price-fixing agreements were not the "economic equivalent" of minimum-price-fixing agreements,17 Kiefer-Stewart was reaffirmed in Albrecht v. Herald Co., 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998 (1968): 23 "Maximum and minimum price fixing may have different consequences in many situations. But schemes to fix maximum prices, by substituting the perhaps erroneous judgment of a seller for the forces of the competitive market, may severely intrude upon the ability of buyers to compete and survive in that market. Competition, even in a single product, is not cast in a single mold. Maximum prices may be fixed too low for the dealer to furnish services essential to the value which goods have for the consumer or to furnish services and conveniences which consumers desire and for which they are willing to pay. Maximum price fixing may channel distribution through a few large or specifically advantaged dealers who otherwise would be subject to significant nonprice competition. Moreover, if the actual price charged under a maximum price scheme is nearly always the fixed maximum price, which is increasingly likely as the maximum price approaches the actual cost of the dealer, the scheme tends to acquire all the attributes of an arrangement fixing minimum prices." Id., at 152-153, 88 S.Ct., at 872-873 (footnote omitted). 24 We have not wavered in our enforcement of the per se rule against price fixing. Indeed, in our most recent price-fixing case we summarily reversed the decision of another Ninth Circuit panel that a horizontal agreement among competitors to fix credit terms does not necessarily contravene the antitrust laws. Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 100 S.Ct. 1925, 64 L.Ed.2d 580 (1980). B 25 Our decisions foreclose the argument that the agreements at issue escape per se condemnation because they are horizontal and fix maximum prices. Kiefer-Stewart and Albrecht place horizontal agreements to fix maximum prices on the same legal—even if not economic—footing as agreements to fix minimum or uniform prices.18 The per se rule "is grounded on faith in price competition as a market force [and not] on a policy of low selling prices at the price of eliminating competition." Rahl, Price Competition and the Price Fixing Rule—Preface and Perspective, 57 Nw. U.L.Rev. 137, 142 (1962). In this case the rule is violated by a price restraint that tends to provide the same economic rewards to all practitioners regardless of their skill, their experience, their training, or their willingness to employ innovative and difficult procedures in individual cases. Such a restraint also may discourage entry into the market and may deter experimentation and new developments by individual entrepreneurs. It may be a masquerade for an agreement to fix uniform prices, or it may in the future take on that character. 26 Nor does the fact that doctors—rather than nonprofessionals are the parties to the price-fixing agreements support the respondents' position. In Goldfarb v. Virginia State Bar, 421 U.S. 773, 788, n. 17, 95 S.Ct. 2004, 2013, n.17, 44 L.Ed.2d 572 (1975), we stated that the "public service aspect, and other features of the professions, may require that a particular practice, which could properly be viewed as a violation of the Sherman Act in another context, be treated differently." See National Society of Professional Engineers v. United States, 435 U.S. 679, 696, 98 S.Ct. 1355, 1367, 55 L.Ed.2d 637 (1978). The price-fixing agreements in this case, however, are not premised on public service or ethical norms. The respondents do not argue, as did the defendants in Goldfarb and Professional Engineers, that the quality of the professional service that their members provide is enhanced by the price restraint. The respondents' claim for relief from the per se rule is simply that the doctors' agreement not to charge certain insureds more than a fixed price facilitates the successful marketing of an attractive insurance plan. But the claim that the price restraint will make it easier for customers to pay does not distinguish the medical profession from any other provider of goods or services. 27 We are equally unpersuaded by the argument that we should not apply the per se rule in this case because the judiciary has little antitrust experience in the health care industry.19 The argument quite obviously is inconsistent withSocony-Vacuum. In unequivocal terms, we stated that, "[w]hatever may be its peculiar problems and characteristics, the Sherman Act, so far as price-fixing agreements are concerned, establishes one uniform rule applicable to all industries alike." 310 U.S., at 222, 60 S.Ct., at 843. We also stated that "[t]he elimination of so-called competitive evils [in an industry] is no legal justification" for price-fixing agreements, id., at 220, 60 S.Ct., at 843, yet the Court of Appeals refused to apply the per se rule in this case in part because the health care industry was so far removed from the competitive model.20 Consistent with our prediction in Socony-Vacuum, 310 U.S., at 221, 60 S.Ct., at 843, the result of this reasoning was the adoption by the Court of Appeals of a legal standard based on the reasonableness of the fixed prices,21 an inquiry we have so often condemned.22 Finally, the argument that the per se rule must be rejustified for every industry that has not been subject to significant antitrust litigation ignores the rationale for per se rules, which in part is to avoid "the necessity for an incredibly complicated and prolonged economic investigation into the entire history of the industry involved, as well as related industries, in an effort to determine at large whether a particular restraint has been unreasonable—an inquiry so often wholly fruitless when undertaken." Northern Pacific R. Co. v. United States, 356 U.S., at 5, 78 S.Ct., at 518. 28 The respondents' principal argument is that the per se rule is inapplicable because their agreements are alleged to have procompetitive justifications. The argument indicates a misunderstanding of the per se concept. The anticompetitive potential inherent in all price-fixing agreements justifies their facial invalidation even if procompetitive justifications are offered for some.23 Those claims of enhanced competition are so unlikely to prove significant in any particular case that we adhere to the rule of law that is justified in its general application. Even when the respondents are given every benefit of the doubt, the limited record in this case is not inconsistent with the presumption that the respondents' agreements will not significantly enhance competition. 29 The respondents contend that their fee schedules are procompetitive because they make it possible to provide consumers of health care with a uniquely desirable form of insurance coverage that could not otherwise exist. The features of the foundation-endorsed insurance plans that they stress are a choice of doctors, complete insurance coverage, and lower premiums. The first two characteristics, however, are hardly unique to these plans. Since only about 70% of the doctors in the relevant market are members of either foundation, the guarantee of complete coverage only applies when an insured chooses a physician in that 70%. If he elects to go to a nonfoundation doctor, he may be required to pay a portion of the doctor's fee. It is fair to presume, however, that at least 70% of the doctors in other markets charge no more than the "usual, customary, and reasonable" fee that typical insurers are willing to reimburse in full.24 Thus, in Maricopa and Pima Counties as well as in most parts of the country, if an insured asks his doctor if the insurance coverage is complete, presumably in about 70% of the cases the doctor will say "Yes" and in about 30% of the cases he will say "No." 30 It is true that a binding assurance of complete insurance coverage—as well as most of the respondents' potential for lower insurance premiums25—can be obtained only if the insurer and the doctor agree in advance on the maximum fee that the doctor will accept as full payment for a particular service. Even if a fee schedule is therefore desirable, it is not necessary that the doctors do the price fixing.26 The record indicates that the Arizona Comprehensive Medical/Dental Program for Foster Children is administered by the Maricopa Foundation pursuant to a contract under which the maximum-fee schedule is prescribed by a state agency rather than by the doctors.27 This program and the Blue Shield plan challenged in Group Life & Health Insurance Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261 (1979), indicate that insurers are capable not only of fixing maximum reimbursable prices but also of obtaining binding agreements with providers guaranteeing the insured full reimbursement of a participating provider's fee. In light of these examples, it is not surprising that nothing in the record even arguably supports the conclusion that this type of insurance program could not function if the fee schedules were set in a different way. 31 The most that can be said for having doctors fix the maximum prices is that doctors may be able to do it more efficiently than insurers. The validity of that assumption is far from obvious,28 but in any event there is no reason to believe that any savings that might accrue from this arrangement would be sufficiently great to affect the competitiveness of these kinds of insurance plans. It is entirely possible that the potential or actual power of the foundations to dictate the terms of such insurance plans may more than offset the theoretical efficiencies upon which the respondents' defense ultimately rests.29 C 32 Our adherence to the per se rule is grounded not only on economic prediction, judicial convenience, and business certainty, but also on a recognition of the respective roles of the Judiciary and the Congress in regulating the economy. United States v. Topco Associates, Inc., 405 U.S., at 611-612, 92 S.Ct., at 1135. Given its generality, our enforcement of the Sherman Act has required the Court to provide much of its substantive content. By articulating the rules of law with some clarity and by adhering to rules that are justified in their general application, however, we enhance the legislative prerogative to amend the law. The respondents' arguments against application of the per se rule in this case therefore are better directed to the Legislature. Congress may consider the exception that we are not free to read into the statute.30 IV 33 Having declined the respondents' invitation to cut back on the per se rule against price fixing, we are left with the respondents' argument that their fee schedules involve price fixing in only a literal sense. For this argument, the respondents rely upon Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1 (1979). 34 In Broadcast Music we were confronted with an antitrust challenge to the marketing of the right to use copyrighted compositions derived from the entire membership of the American Society of Composers, Authors and Publishers (ASCAP). The so-called "blanket license" was entirely different from the product that any one composer was able to sell by himself.31 Although there was little competition among individual composers for their separate compositions, the blanket-license arrangement did not place any restraint on the right of any individual copyright owner to sell his own compositions separately to any buyer at any price.32 But a "necessary consequence" of the creation of the blanket license was that its price had to be established. Id., at 21, 99 S.Ct., at 1563. We held that the delegation by the composers to ASCAP of the power to fix the price for the blanket license was not a species of the price-fixing agreements categorically forbidden by the Sherman Act. The record disclosed price fixing only in a "literal sense." Id., at 8, 99 S.Ct., at 1556. 35 This case is fundamentally different. Each of the foundations is composed of individual practitioners who compete with one another for patients. Neither the foundations nor the doctors sell insurance, and they derive no profits from the sale of health insurance policies. The members of the foundations sell medical services. Their combination in the form of the foundation does not permit them to sell any different product.33 Their combination has merely permitted them to sell their services to certain customers at fixed prices and arguably to affect the prevailing market price of medical care. 36 The foundations are not analogous to partnerships or other joint arrangements in which persons who would otherwise be competitors pool their capital and share the risks of loss as well as the opportunities for profit. In such joint ventures, the partnership is regarded as a single firm competing with other sellers in the market. The agreement under attack is an agreement among hundreds of competing doctors concerning the price at which each will offer his own services to a substantial number of consumers. It is true that some are surgeons, some anesthesiologists, and some psychiatrists, but the doctors do not sell a package of three kinds of services. If a clinic offered complete medical coverage for a flat fee, the cooperating doctors would have the type of partnership arrangement in which a price-fixing agreement among the doctors would be perfectly proper. But the fee agreements disclosed by the record in this case are among independent competing entrepreneurs. They fit squarely into the horizontal price-fixing mold. 37 The judgment of the Court of Appeals is reversed. 38 It is so ordered. 39 Justice BLACKMUN and Justice O'CONNOR took no part in the consideration or decision of this case. 40 Justice POWELL, with whom THE CHIEF JUSTICE and Justice REHNQUIST join, dissenting. 41 The medical care plan condemned by the Court today is a comparatively new method of providing insured medical services at predetermined maximum costs. It involves no coercion. Medical insurance companies, physicians, and patients alike are free to participate or not as they choose. On its face, the plan seems to be in the public interest. 42 The State of Arizona challenged the plan on a per se antitrust theory. The District Court denied the State's summary judgment motion, and—because of the novelty of the issue—certified the question of per se liability for an interlocutory appeal. On summary judgment, the record and all inferences therefrom must be viewed in the light most favorable to the respondents. Nevertheless, rather than identifying clearly the controlling principles and remanding for decision on a completed record, this Court makes its own per se judgment of invalidity. The respondents' contention that the "consumers" of medical services are benefited substantially by the plan is given short shrift. The Court concedes that "the parties conducted [only] a limited amount of pretrial discovery," ante, at 336, leaving undeveloped facts critical to an informed decision of this case. I do not think today's decision on an incomplete record is consistent with proper judicial resolution of an issue of this complexity, novelty, and importance to the public. I therefore dissent. 43 * The Maricopa and Pima Foundations for Medical Care are professional associations of physicians organized by the medical societies in their respective counties.1 The foundations were established to make available a type of prepaid medical insurance plan, aspects of which are the target of this litigation. Under the plan, the foundations insure no risks themselves. Rather, their key function is to secure agreement among their member physicians to a maximum-price schedule for specific medical services. Once a fee schedule has been agreed upon following a process of consultation and balloting, the foundations invite private insurance companies to participate by offering medical insurance policies based upon the maximum-fee schedule.2 The insurers agree to offer complete reimbursement to their insureds for the full amount of their medical bills—so long as these bills do not exceed the maximum-fee schedule. 44 An insured under a foundation-sponsored plan is free to go to any physician. The physician then bills the foundation directly for services performed.3 If the insured has chosen a physician who is not a foundation member and the bill exceeds the foundation maximum-fee schedule, the insured is liable for the excess. If the billing physician is a foundation member, the foundation disallows the excess pursuant to the agreement each physician executed upon joining the foundation.4 Thus, the plan offers complete coverage of medical expenses but still permits an insured to choose any physician. II 45 This case comes to us on a plaintiff's motion for summary judgment after only limited discovery. Therefore, as noted above, the inferences to be drawn from the record must be viewed in the light most favorable to the respondents. United States v. Diebold, Inc., 369 U.S. 654, 655, 82 S.Ct. 993, 994, 8 L.Ed.2d 176 (1962). This requires, as the Court acknowledges, that we consider the foundation arrangement as one that "impose[s] a meaningful limit on physicians' charges," that "enables the insurance carriers to limit and to calculate more efficiently the risks they underwrite," and that "therefore serves as an effective cost-containment mechanism that has saved patients and insurers millions of dollars." Ante, at 342. The question is whether we should condemn this arrangement forthwith under the Sherman Act, a law designed to benefit consumers. 46 Several other aspects of the record are of key significance but are not stressed by the Court. First, the foundation arrangement forecloses no competition. Unlike the classic cartel agreement, the foundation plan does not instruct potential competitors: "Deal with consumers on the following terms and no others." Rather, physicians who participate in the foundation plan are free both to associate with other medical insurance plans at any fee level, high or low—and directly to serve uninsured patients—at any fee level, high or low. Similarly, insurers that participate in the foundation plan also remain at liberty to do business outside the plan with any physician—foundation member or not—at any fee level. Nor are physicians locked into a plan for more than one year's membership. See n. 1, supra. Thus freedom to compete, as well as freedom to withdraw, is preserved. The Court cites no case in which a remotely comparable plan or agreement is condemned on a per se basis. 47 Second, on this record we must find that insurers represent consumer interests. Normally consumers search for high quality at low prices. But once a consumer is insured5—i.e., has chosen a medical insurance plan—he is largely indifferent to the amount that his physician charges if the coverage is full, as under the foundation-sponsored plan. 48 The insurer, however, is not indifferent. To keep insurance premiums at a competitive level and to remain profitable, insurers including those who have contracts with the foundations—step into the consumer's shoes with his incentive to contain medical costs. Indeed, insurers may be the only parties who have the effective power to restrain medical costs, given the difficulty that patients experience in comparing price and quality for a professional service such as medical care. 49 On the record before us, there is no evidence of opposition to the foundation plan by insurance companies—or, for that matter, by members of the public. Rather seven insurers willingly have chosen to contract out to the foundations the task of developing maximum-fee schedules.6 Again, on the record before us, we must infer that the foundation plan—open as it is to insurers, physicians, and the public—has in fact benefited consumers by "enabl[ing] the insurance carriers to limit and to calculate more efficiently the risks they underwrite." Ante, at 342. Nevertheless, even though the case is here on an incomplete summary judgment record, the Court conclusively draws contrary inferences to support its per se judgment. III 50 It is settled law that once an arrangement has been labeled as "price fixing" it is to be condemned per se. But it is equally well settled that this characterization is not to be applied as a talisman to every arrangement that involves a literal fixing of prices. Many lawful contracts, mergers, and partnerships fix prices. But our cases require a more discerning approach. The inquiry in an antitrust case is not simply one of "determining whether two or more potential competitors have literally 'fixed' a 'price.' . . . [Rather], it is necessary to characterize the challenged conduct as falling within or without that category of behavior to which we apply the label 'per se price fixing.' That will often, but not always, be a simple matter." Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1, 9, 99 S.Ct. 1551, 1557, 60 L.Ed.2d 1 (1979). 51 Before characterizing an arrangement as a per se price-fixing agreement meriting condemnation, a court should determine whether it is a " 'naked restrain[t] of trade with no purpose except stifling of competition.' " United States v. Topco Associates, Inc., 405 U.S. 596, 608, 92 S.Ct. 1126, 1133, 31 L.Ed.2d 515 (1972), quoting White Motor Co. v. United States, 372 U.S. 253, 263, 83 S.Ct. 696, 702, 9 L.Ed.2d 738 (1963). See also Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 49-50, 97 S.Ct. 2549, 2557, 53 L.Ed.2d 568 (1977). Such a determination is necessary because "departure from the rule-of-reason standard must be based upon demonstrable economic effect rather than . . . upon formalistic line drawing." Id., at 58-59, 97 S.Ct., at 2561-2562. As part of this inquiry, a court must determine whether the procompetitive economies that the arrangement purportedly makes possible are substantial and realizable in the absence of such an agreement. 52 For example, in National Society of Professional Engineers v. United States, 435 U.S. 679, 98 S.Ct. 1355, 55 L.Ed.2d 637 (1978), we held unlawful as a per se violation an engineering association's canon of ethics that prohibited competitive bidding by its members. After the parties had "compiled a voluminous discovery and trial record," id., at 685, 98 S.Ct., at 1362, we carefully considered—rather than rejected out of hand—the engineers' "affirmative defense" of their agreement: that competitive bidding would tempt engineers to do inferior work that would threaten public health and safety. Id., at 693, 98 S.Ct., at 1366. We refused to accept this defense because its merits "confirm[ed] rather than refut[ed] the anticompetitive purpose and effect of [the] agreement." Ibid. The analysis incident to the "price fixing" characterization found no substantial procompetitive efficiencies. See also Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 646, n. 8, and 649-650, 100 S.Ct. 1925, 1927, n. 8, and 1928-1929, 64 L.Ed.2d 580 (1980) (challenged arrangement condemned because it lacked "a procompetitive justification" and had "no apparent potentially redeeming value"). 53 In Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., supra, there was minimum price fixing in the most "literal sense." Id., at 8, 99 S.Ct., at 1556. We nevertheless agreed, unanimously,7 that an arrangement by which copyright clearinghouses sold performance rights to their entire libraries on a blanket rather than individual basis did not warrant condemnation on a per se basis. Individual licensing would have allowed competition between copyright owners. But we reasoned that licensing on a blanket basis yielded substantial efficiencies that otherwise could not be realized. See id., at 20-21, 99 S.Ct., at 1562-1563. Indeed, the blanket license was itself "to some extent, a different product." Id., at 22, 99 S.Ct., at 1563.8 54 In sum, the fact that a foundation-sponsored health insurance plan literally involves the setting of ceiling prices among competing physicians does not, of itself, justify condemning the plan as per se illegal. Only if it is clear from the record that the agreement among physicians is "so plainly anticompetitive that no elaborate study of [its effects] is needed to establish [its] illegality" may a court properly make a per se judgment. National Society of Professional Engineers v. United States, supra, at 692, 98 S.Ct., at 1365. And, as our cases demonstrate, the per se label should not be assigned without carefully considering substantial benefits and procompetitive justifications. This is especially true when the agreement under attack is novel, as in this case. See Broadcast Music, supra, at 9-10, 99 S.Ct., at 1556-1557; United States v. Topco Associates, Inc., supra, at 607-608, 92 S.Ct., at 1133 ("It is only after considerable experience with certain business relationships that courts classify them as per se violations"). IV 55 The Court acknowledges that the per se ban against price fixing is not to be invoked every time potential competitors literally fix prices. Ante, at 355-357. One also would have expected it to acknowledge that per se characterization is inappropriate if the challenged agreement or plan achieves for the public procompetitive benefits that otherwise are not attainable. The Court does not do this. And neither does it provide alternative criteria by which the per se characterization is to be determined. It is content simply to brand this type of plan as "price fixing" and describe the agreement in Broadcast Music—which also literally involved the fixing of prices—as "fundamentally different." Ante, at 356. 56 In fact, however, the two agreements are similar in important respects. Each involved competitors and resulted in cooperative pricing.9 Each arrangement also was prompted by the need for better service to the consumers.10 And each arrangement apparently makes possible a new product by reaping otherwise unattainable efficiencies.11 The Court's effort to distinguish Broadcast Music thus is unconvincing.12 57 The Court, in defending its holding, also suggests that "respondents' arguments against application of the per se rule . . . are better directed to the Legislature." Ante, at 354-355. This is curious advice. The Sherman Act does not mention per se rules. And it was not Congress that decided Broadcast Music and the other relevant cases. Since the enactment of the Sherman Act in 1890, it has been the duty of courts to interpret and apply its general mandate—and to do so for the benefit of consumers. 58 As in Broadcast Music, the plaintiff here has not yet discharged its burden of proving that respondents have entered a plainly anticompetitive combination without a substantial and procompetitive efficiency justification. In my view, the District Court therefore correctly refused to grant the State's motion for summary judgment.13 This critical and disputed issue of fact remains unresolved. See Fed.Rule Civ.Proc. 56(c). V 59 I believe the Court's action today loses sight of the basic purposes of the Sherman Act. As we have noted, the antitrust laws are a "consumer welfare prescription." Reiter v. Sonotone Corp., 442 U.S. 330, 343, 99 S.Ct. 2326, 2333, 60 L.Ed.2d 931 (1979). In its rush to condemn a novel plan about which it knows very little, the Court suggests that this end is achieved only by invalidating activities that may have some potential for harm. But the little that the record does show about the effect of the plan suggests that it is a means of providing medical services that in fact benefits rather than injures persons who need them. 60 In a complex economy, complex economic arrangements are commonplace. It is unwise for the Court, in a case as novel and important as this one, to make a final judgment in the absence of a complete record and where mandatory inferences create critical issues of fact. 1 The complaint alleged a violation of § 1 of the Sherman Act as well as of the Arizona antitrust statute. The state statute is interpreted in conformity with the federal statute. 643 F.2d 553, 554, n. 1 (CA9 1980). The State of Arizona prayed for an injunction but did not ask for damages. 2 The District Court offered three reasons for its decision. First, citing Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977), the court stated that "a recent antitrust trend appears to be emerging where the Rule of Reason is the preferred method of determining whether a particular practice is in violation of the antitrust law." App. to Pet. for Cert. 43. Second, "the two Supreme Court cases invalidating maximum price-fixing, [Kiefer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219 (1951), and Albrecht v. Herald Co., 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998 (1968) ], need not be read as establishing a per se rule." Id., at 44. Third, "a profession is involved here." Id., at 45. Under the rule-of-reason approach, the plaintiff's motion for partial summary judgment on the issue of liability could not be granted "because there is insufficient evidence as to the [purpose and effect of the allegedly unlawful practices and the power of the defendants.]" Id., at 47. The District Court also denied the defendants' motion to dismiss based on the ground that they were engaged in the business of insurance within the meaning of the McCarran-Ferguson Act, 15 U.S.C. § 1011 et seq. See App. to Pet. for Cert. 39-41. The defendants did not appeal that portion of the District Court order. 643 F.2d, at 559, and n. 7. 3 The quoted language is the Court of Appeals' phrasing of the question. Id., at 554. The District Court had entered an order on June 5, 1979, providing, in relevant part: "The plaintiff's motion for partial summary judgment on the issue of liability is denied with leave to file a similar motion based on additional evidence if appropriate." App. to Pet. for Cert. 48. On August 8, 1979, the District Court entered a further order providing: "The Order of this Court entered June 5, 1979 is amended by addition of the following: This Court's determination that the Rule of Reason approach should be used in analyzing the challenged conduct in the instant case to determine whether a violation of Section 1 of the Sherman Act has occurred involves a question of law as to which there is substantial ground for difference of opinion and an immediate appeal from the Order denying plaintiff's motion for partial summary judgment on the issue of liability may materially advance the ultimate determination of the litigation. Therefore, the foregoing Order and determination of the Court is certified for interlocutory appeal pursuant to 28 U.S.C. § 1292(b)." Id., at 50-51. 4 Judge Sneed explained his reluctance to apply the per se rule substantially as follows: The record did not indicate the actual purpose of the maximum-fee arrangements or their effect on competition in the health care industry. It was not clear whether the assumptions made about typical price restraints could be carried over to that industry. Only recently had this Court applied the antitrust laws to the professions. Moreover, there already were such significant obstacles to pure competition in the industry that a court must compare the prices that obtain under the maximum-fee arrangements with those that would otherwise prevail rather than with those that would prevail under ideal competitive conditions. Furthermore, the Ninth Circuit had not applied Keifer-Stewart Co. v. Joseph E. Seagram & Sons, Inc., 340 U.S. 211, 71 S.Ct. 259, 95 L.Ed. 219 (1951), and Albrecht v. Herald Co., 390 U.S. 145, 88 S.Ct. 869, 19 L.Ed.2d 998 (1968), to horizontal agreements that establish maximum prices; some of the economic assumptions underlying the rule against maximum price fixing were not sound. 5 Judge Kennedy's concurring opinion concluded as follows: "There does not now appear to be a controlling or definitive analysis of the market impact caused by the arrangements under scrutiny in this case, but trial may reveal that the arrangements are, at least in their essentials, not peculiar to the medical industry and that they should be condemned." 643 F.2d, at 560. 6 Judge Larson stated, in part: "Defendants formulated and dispersed relative value guides and conversion factor lists which together were used to set an upper limit on fees received from third-party payors. It is clear that these activities constituted maximum price-fixing by competitors. Disregarding any 'special industry' facts, this conduct is per se illegal. Precedent alone would mandate application of the per se standard. "I find nothing in the nature of either the medical profession or the health care industry that would warrant their exemption from per se rules for price-fixing." Id., at 563-564 (citations omitted). 7 Most health insurance plans are of the fee-for-service type. Under the typical insurance plan, the insurer agrees with the insured to reimburse the insured for "usual, customary, and reasonable" medical charges. The third-party insurer, and the insured to the extent of any excess charges, bears the economic risk that the insured will require medical treatment. An alternative to the fee-for-service type of insurance plan is illustrated by the health maintenance organizations authorized under the Health Maintenance Organization Act of 1973, 42 U.S.C. § 300e et seq. Under this form of prepaid health plan, the consumer pays a fixed periodic fee to a functionally integrated group of doctors in exchange for the group's agreement to provide any medical treatment that the subscriber might need. The economic risk is thus borne by the doctors. 8 The record contains divergent figures on the percentage of Pima County doctors that belong to the foundation. A 1975 publication of the foundation reported 80%; a 1978 affidavit by the executive director of the foundation reported 30%. 9 In 1980, after the District Court and the Court of Appeals had rendered judgment, both foundations apparently discontinued the use of relative values and conversion factors in formulating the fee schedules. Moreover, the Maricopa Foundation that year amended its bylaws to provide that the fee schedule would be adopted by majority vote of its board of trustees and not by vote of its members. The challenge to the foundation activities as we have described them in the text, however, is not mooted by these changes. See United States v. W. T. Grant Co., 345 U.S. 629, 73 S.Ct. 894, 97 L.Ed. 1303 (1953). 10 The parties disagree over whether the increases in the fee schedules are the cause or the result of the increases in the prevailing rate for medical services in the relevant markets. There appears to be agreement, however, that 85-95% of physicians in Maricopa County bill at or above the maximum reimbursement levels set by the Maricopa Foundation. 11 Seven different insurance companies underwrite health insurance plans that have been approved by the Maricopa Foundation, and three companies underwrite the plans approved by the Pima Foundation. The record contains no firm data on the portion of the health care market that is covered by these plans. The State relies upon a 1974 analysis indicating that insurance plans endorsed by the Maricopa Foundation had about 63% of the prepaid health care market, but the respondents contest the accuracy of this analysis. 12 "Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal. . . ." 15 U.S.C. § 1. 13 Justice Brandeis provided the classic statement of the rule of reason in Chicago Bd. of Trade v. United States, 246 U.S. 231, 238, 38 S.Ct. 242, 62 L.Ed. 683 (1918): "The true test of legality is whether the restraint imposed is such as merely regulates and perhaps thereby promotes competition or whether it is such as may suppress or even destroy competition. To determine that question the court must ordinarily consider the facts peculiar to the business to which the restraint is applied; its condition before and after the restraint was imposed; the nature of the restraint and its effect, actual or probable. The history of the restraint, the evil believed to exist, the reason for adopting the particular remedy, the purpose or end sought to be attained, are all relevant facts. This is not because a good intention will save an otherwise objectionable regulation or the reverse; but because knowledge of intent may help the court to interpret facts and to predict consequences." 14 For a thoughtful and brief discussion of the costs and benefits of rule-of-reason versus per se rule analysis of price-fixing agreements, see F. Scherer, Industrial Market Structure and Economic Performance 438-443 (1970). Professor Scherer's "opinion, shared by a majority of American economists concerned with antitrust policy, is that in the present legal framework the costs of implementing a rule of reason would exceed the benefits derived from considering each restrictive agreement on its merits and prohibiting only those which appear unreasonable." Id., at 440. 15 "Among the practices which the courts have heretofore deemed to be unlawful in and of themselves are price fixing, division of markets, group boycotts, and tying arrangements." Northern Pacific R. Co. v. United States, 356 U.S., at 5, 78 S.Ct., at 518 (citations omitted). See United States v. Columbia Steel Co., 334 U.S. 495, 522-523, 68 S.Ct. 1107, 1121-1122, 92 L.Ed. 1533 (1948). 16 Thus, in applying the per se rule to invalidate the restrictive practice in United States v. Topco Associates, Inc., 405 U.S. 596, 92 S.Ct. 1126, 31 L.Ed.2d 515 (1972), we stated that "[w]hether or not we would decide this case the same way under the rule of reason used by the District Court is irrelevant to the issue before us." Id., at 609, 92 S.Ct., at 1134. The Court made the same point in Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S., at 50, n. 16, 97 S.Ct., at 2557, n. 16: "Per se rules thus require the Court to make broad generalizations about the social utility of particular commercial practices. The probability that anticompetitive consequences will result from a practice and the severity of those consequences must be balanced against its procompetitive consequences. Cases that do not fit the generalization may arise, but a per se rule reflects the judgment that such cases are not sufficiently common or important to justify the time and expense necessary to identify them." 17 Albrecht v. Herald Co., 390 U.S., at 156, 88 S.Ct., at 875 (Harlan, J., dissenting). 18 It is true that in Keifer-Stewart, as in Albrecht, the agreement involved a vertical arrangement in which maximum resale prices were fixed. But the case also involved an agreement among competitors to impose the resale price restraint. In any event, horizontal restraints are generally less defensible than vertical restraints. See Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977); Easterbrook, Maximum Price Fixing, 48 U.Chi.L.Rev. 886, 890, n. 20 (1981). 19 The argument should not be confused with the established position that a new per se rule is not justified until the judiciary obtains considerable rule-of-reason experience with the particular type of restraint challenged. See White Motor Co. v. United States, 372 U.S. 253, 83 S.Ct. 696, 9 L.Ed.2d 738 (1963). Nor is our unwillingness to examine the economic justification of this particular application of the per se rule against price fixing inconsistent with our reexamination of the general validity of the per se rule rejected in Continental T.V., Inc. v. GTE Sylvania Inc., supra. 20 "The health care industry, moreover, presents a particularly difficult area. The first step to understanding is to recognize that not only is access to the medical profession very time consuming and expensive both for the applicant and society generally, but also that numerous government subventions of the costs of medical care have created both a demand and supply function for medical services that is artificially high. The present supply and demand functions of medical services in no way approximate those which would exist in a purely private competitive order. An accurate description of those functions moreover is not available. Thus, we lack baselines by which could be measured the distance between the present supply and demand functions and those which would exist under ideal competitive conditions." 643 F.2d, at 556. 21 "Perforce we must take industry as it exists, absent the challenged feature, as our baseline for measuring anticompetitive impact. The relevant inquiry becomes whether fees paid to doctors under that system would be less than those payable under the FMC maximum fee agreement. Put differently, confronted with an industry widely deviant from a reasonably free competitive model, such as agriculture, the proper inquiry is whether the practice enhances the prices charged for the services. In simplified economic terms, the issue is whether the maximum fee arrangement better permits the attainment of the monopolist's goal, viz., the matching of marginal cost to marginal revenue, or in fact obstructs that end." Ibid. 22 In the first price-fixing case arising under the Sherman Act, the Court was required to pass on the sufficiency of the defendants' plea that they had established rates that were actually beneficial to consumers. Assuming the factual validity of the plea, the Court rejected the defense as a matter of law. United States v. Trans-Missouri Freight Assn., 166 U.S. 290, 17 S.Ct. 540, 41 L.Ed. 1007 (1897). In National Society of Professional Engineers v. United States, 435 U.S. 679, 689, 98 S.Ct. 1355, 1364, 55 L.Ed.2d 637 (1978), we referred to Judge Taft's "classic rejection of the argument that competitors may lawfully agree to sell their goods at the same price as long as the agreed-upon price is reasonable." See United States v. Addyston Pipe & Steel Co., 85 F. 271 (CA6 1898), aff'd, 175 U.S. 211, 20 S.Ct. 96, 44 L.Ed. 136 (1899). In our latest price-fixing case, we reiterated the point: "It is no excuse that the prices fixed are themselves reasonable." Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 647, 100 S.Ct. 1925, 1927, 64 L.Ed.2d 580 (1980). 23 "Whatever economic justification particular price-fixing agreements may be thought to have, the law does not permit an inquiry into their reasonableness. They are all banned because of their actual or potential threat to the central nervous system of the economy." United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 226, n. 59, 60 S.Ct. 811, 844, n. 59, 84 L.Ed. 1129 (1940). 24 According to the respondents' figures, this presumption is well-founded. See Brief for Respondents 42, n. 120. 25 We do not perceive the respondents' claim of procompetitive justification for their fee schedules to rest on the premise that the fee schedules actually reduce medical fees and accordingly reduce insurance premiums, thereby enhancing competition in the health insurance industry. Such an argument would merely restate the long-rejected position that fixed prices are reasonable if they are lower than free competition would yield. It is arguable, however, that the existence of a fee schedule, whether fixed by the doctors or by the insurers, makes it easier—and to that extent less expensive—for insurers to calculate the risks that they underwrite and to arrive at the appropriate reimbursement on insured claims. 26 According to a Federal Trade Commission staff report: "Until the mid-1960's, most Blue Shield plans determined in advance how much to pay for particular procedures and prepared fee schedules reflecting their determinations. Fee schedules are still used in approximately 25 percent of Blue Shield contracts." Bureau of Competition, Federal Trade Commission, Medical Participation in Control of Blue Shield and Certain Other Open-Panel Medical Prepayment Plans 128 (1979). We do not suggest that Blue Shield plans are not actually controlled by doctors. Indeed, as the same report discusses at length, the belief that they are has given rise to considerable antitrust litigation. See also D. Kass & P. Pautler, Bureau of Economics, Federal Trade Commission, Staff Report on Physician Control of Blue Shield Plans (1979). Nor does this case present the question whether an insurer may, consistent with the Sherman Act, fix the fee schedule and enter into bilateral contracts with individual doctors. That question was not reached in Group Life & Health Insurance Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261 (1979). See id., at 210, n. 5, 99 S.Ct., at 1072, n. 5. In an amicus curiae brief, the United States expressed its opinion that such an arrangement would be legal unless the plaintiffs could establish that a conspiracy among providers was at work. Brief for United States as Amicus Curiae, O.T.1978, No. 77-952, pp. 10-11. Our point is simply that the record provides no factual basis for the respondents' claim that the doctors must fix the fee schedule. 27 In that program the foundation performs the peer review function as well as the administrative function of paying the doctors' claims. 28 In order to create an insurance plan under which the doctor would agree to accept as full payment a fee prescribed in a fixed schedule, someone must canvass the doctors to determine what maximum prices would be high enough to attract sufficient numbers of individual doctors to sign up but low enough to make the insurance plan competitive. In this case that canvassing function is performed by the foundation; the foundation then deals with the insurer. It would seem that an insurer could simply bypass the foundation by performing the canvassing function and dealing with the doctors itself. Under the foundation plan, each doctor must look at the maximum-fee schedule fixed by his competitors and vote for or against approval of the plan (and, if the plan is approved by majority vote, he must continue or revoke his foundation membership). A similar, if to some extent more protracted, process would occur if it were each insurer that offered the maximum-fee schedule to each doctor. 29 In this case it appears that the fees are set by a group with substantial power in the market for medical services, and that there is competition among insurance companies in the sale of medical insurance. Under these circumstances the insurance companies are not likely to have significantly greater bargaining power against a monopoly of doctors than would individual consumers of medical services. 30 "[Congress] can, of course, make per se rules inapplicable in some or all cases, and leave courts free to ramble through the wilds of economic theory in order to maintain a flexible approach." United States v. Topco Associates, Inc., 405 U.S., at 610, n. 10, 92 S.Ct., at 1134, n. 10. Indeed, it has exempted certain industries from the full reach of the Sherman Act. See, e.g., 7 U.S.C. §§ 291, 292 (Capper-Volstead Act, agricultural cooperatives); 15 U.S.C. §§ 1011-1013 (McCarran-Ferguson Act, insurance); 49 U.S.C. § 5b (Reed-Bulwinkle Act, rail and motor carrier rate-fixing bureaus); 15 U.S.C. § 1801 (newspaper joint operating agreements). 31 "Thus, to the extent the blanket license is a different product, ASCAP is not really a joint sales agency offering the individual goods of many sellers, but is a separate seller offering its blanket license, of which the individual compositions are raw material." 441 U.S., at 22, 99 S.Ct., at 1563 (footnote omitted). 32 "Here, the blanket-license fee is not set by competition among individual copyright owners, and it is a fee for the use of any of the compositions covered by the license. But the blanket license cannot be wholly equated with a simple horizontal arrangement among competitors. ASCAP does set the price for its blanket license, but that license is quite different from anything any individual owner could issue. The individual composers and authors have neither agreed not to sell individually in any other market nor use the blanket license to mask price fixing in such other markets." Id., at 23-24, 99 S.Ct., at 1564 (footnote omitted). 33 It may be true that by becoming a member of the foundation the individual practitioner obtains a competitive advantage in the market for medical services that he could not unilaterally obtain. That competitive advantage is the ability to attract as customers people who value both the guarantee of full health coverage and a choice of doctors. But, as we have indicated, the setting of the price by doctors is not a "necessary consequence" of an arrangement with an insurer in which the doctor agrees not to charge certain insured customers more than a fixed price. 1 The Pima Foundation is open to any Pima County area physician licensed in Arizona. It has a renewable 5-year membership term. A voluntary resignation provision permits earlier exit on the January 1 following announcement of an intent to resign. The Maricopa Foundation admits physicians who are members of their county medical society. The Maricopa Foundation has a renewable 1-year term of membership. Initial membership may be for a term of less than a year so that a uniform annual termination date for all members can be maintained. The medical societies are professional associations of physicians practicing in the particular county. The Pima County Medical Society, but not the Pima Foundation, has been dismissed from the case pursuant to a consent decree. 2 Three private carriers underwrite various Pima Foundation-sponsored plans: Arizona Blue Cross-Blue Shield, Pacific Mutual Life Insurance Co., and Connecticut General Life Insurance Co. The latter two companies also underwrite plans for the Maricopa Foundation, as do five other private insurance companies. Apparently large employers, such as the State of Arizona and Motorola, also act as foundation-approved insurers with respect to their employees' insurance plans. 3 The foundations act as the insurance companies' claims agents on a contract basis. They administer the claims and, to some extent, review the medical necessity and propriety of the treatment for which a claim is entered. The foundations charge insurers a fee for their various services. In recent years, this fee has been set at 4% of the insurers' premiums. 4 This agreement provides in part that the physician agrees "to be bound . . . with respect to maximum fees . . . by any fee determination by the [f]oundation consistent with the schedule adopted by the [foundation physician] membership . . . ." App. 31-32. The agreement also provides that foundation members "understand and agree that participating membership in the [f]oundation shall not affect the method of computation or amount of fees billed by me with respect to any medical care for any patient." Ibid. 5 At least seven insurance companies are competing in the relevant market. See n. 2, supra. At this stage of the case we must infer that they are competing vigorously and successfully. The term "consumer"—commonly used in antitrust cases and literature—is used herein to mean persons who need or may need medical services from a physician. 6 The State introduced no evidence on its summary judgment motion supporting its apparent view that insurers effectively can perform this function themselves, without physician participation. It is clear, however, that price and quality of professional services—unlike commercial products—are difficult to compare. Cf. Bates v. State Bar of Arizona, 433 U.S. 350, 391-395, 97 S.Ct. 2691, 2712-2715, 53 L.Ed.2d 810 (1977) (opinion of POWELL, J.). This is particularly true of medical service. Presumably this is a reason participating insurers wish to utilize the foundations' services. 7 See Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S., at 25, 99 S.Ct., at 1565 (Stevens, J., dissenting in part) ("The Court holds that ASCAP's blanket license is not a species of price fixing categorically forbidden by the Sherman Act. I agree with that holding"). 8 Cf. Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 54, 97 S.Ct. 2549, 2559, 53 L.Ed.2d 568 (1977) (identifying achievement of efficiencies as "redeeming virtue" in decision sustaining an agreement against per se challenge); L. Sullivan, Law of Antitrust § 74, p. 200 (1977) (per se characterization inappropriate if price agreement achieves great economies of scale and thereby improves economic performance); id., § 66, p. 180 (higher burden might reasonably be placed on plaintiff where agreement may involve efficiencies). 9 In this case the physicians in effect vote on foundation maximum-fee schedules. In Broadcast Music, the copyright owners aggregated their copyrights into a group package, sold rights to the package at a group price, and distributed the proceeds among themselves according to an agreed-upon formula. See Columbia Broadcasting System, Inc. v. American Society of Composers, Authors and Publishers, 562 F.2d 130, 135-136 (CA2 1977). 10 In this case, the foundations' maximum-fee schedules attempt to rectify the inflationary consequence of patients' indifference to the size of physicians' bills and insurers' commitment to reimburse whatever "usual, customary, and reasonable" charges physicians may submit. In Broadcast Music, the market defect inhered in the fact that "those who performed copyrighted music for profit were so numerous and widespread, and most performances so fleeting, that as a practical matter it was impossible for the many individual copyright owners to negotiate with and license the users and to detect unauthorized uses." 441 U.S., at 4-5, 99 S.Ct., at 1554. 11 In this case, the record before us indicates that insurers—those best situated to decide and best motivated to inspire trust in their judgment—believe that the foundations are the most efficient providers of the maximum-fee scheduling service. In Broadcast Music, we found that the blanket copyright clearinghouse system "reduce[d] costs absolutely . . . ." Id., at 21, 99 S.Ct., at 1563. 12 The Court states that in Broadcast Music "there was little competition among individual composers for their separate compositions." Ante, at 355. This is an irrational ground for distinction. Competition could have existed, 441 U.S., at 6, 99 S.Ct., at 1555; see also 562 F.2d, at 134-135, 138, but did not because of the cooperative agreement. That competition yet persists among physicians is not a sensible reason to invalidate their agreement while refusing similarly to condemn the Broadcast Music agreements that were completely effective in eliminating competition. The Court also offers as a distinction that the foundations do not permit the creation of "any different product." Ante, at 356. But the foundations provide a "different product" to precisely the same extent as did Broadcast Music's clearinghouses. The clearinghouses provided only what copyright holders offered as individual sellers—the rights to use individual compositions. The clearinghouses were able to obtain these same rights more efficiently, however, because they eliminated the need to engage in individual bargaining with each individual copyright owner. See 441 U.S., at 21-22, 99 S.Ct., at 1563. In the same manner, the foundations set up an innovative means to deliver a basic service—insured medical care from a wide range of physicians of one's choice—in a more economical manner. The foundations' maximum-fee schedules replace the weak cost containment incentives in typical "usual, customary, and reasonable" insurance agreements with a stronger cost control mechanism: an absolute ceiling on maximum fees that can be charged. The conduct of the insurers in this case indicates that they believe that the foundation plan as it presently exists is the most efficient means of developing and administering such schedules. At this stage in the litigation, therefore, we must agree that the foundation plan permits the more economical delivery of the basic insurance service—"to some extent, a different product." Broadcast Music, 441 U.S., at 22, 99 S.Ct., at 1563. 13 Medical services differ from the typical service or commercial product at issue in an antitrust case. The services of physicians, rendered on a patient-by-patient basis, rarely can be compared by the recipient. A person requiring medical service or advice has no ready way of comparing physicians or of "shopping" for quality medical service at a lesser price. Primarily for this reason, the foundations—operating the plan at issue—perform a function that neither physicians nor prospective patients can perform individually. On a collective—and average—basis, the physicians themselves express a willingness to render certain identifiable services for not more than specified fees, leaving patients free to choose the physician. We thus have a case in which we derive little guidance from the conventional "perfect market" analysis of antitrust law. I would give greater weight than the Court to the uniqueness of medical services, and certainly would not invalidate on a per se basis a plan that may in fact perform a uniquely useful service. Affirmance of the District Court's holding would not have immunized the medical service plan at issue. Nor would it have foreclosed an eventual conclusion on remand that the arrangement should be deemed per se invalid. And if the District Court had found that petitioner had failed to establish a per se violation of the Sherman Act, the question would have remained whether the plan comports with the rule of reason. See, e.g., United States v. United States Gypsum Co., 438 U.S. 422, 441, n. 16, 98 S.Ct. 2864, 2875, n. 16, 57 L.Ed.2d 854 (1978).

78

457 U.S. 291 102 S.Ct. 2442 73 L.Ed.2d 16 Mark J. MILLS et al., Petitionersv.Rubie ROGERS et al. No. 80-1417. Argued Jan. 13, 1982. Decided June 18, 1982. Syllabus Respondents, present or former mental patients at a Massachusetts state hospital, instituted a class action against petitioner officials and staff of the hospital in Federal District Court, alleging that forcible administration of antipsychotic drugs to patients violated rights protected by the Federal Constitution. The court held that mental patients enjoy constitutionally protected liberty and privacy interests in deciding for themselves whether to submit to drug therapy; that under state law an involuntary commitment provides no basis for an inference of legal "incompetency" to make such decision; and that without consent either by the patient or the guardian of a patient who has been adjudicated incompetent, the patient's liberty interests may be overridden only in an emergency. The Court of Appeals affirmed in part and reversed in part. It agreed with the District Court's first two holdings above, but reached different conclusions as to the circumstances under which state interests might override the patient's liberty interests. The Court of Appeals reserved to the District Court, on remand, the task of developing mechanisms to ensure adequate procedural protection of the patient's interests. This Court granted certiorari to determine whether an involuntarily committed mental patient has a constitutional right to refuse treatment with antipsychotic drugs. Shortly thereafter the Massachusetts Supreme Judicial Court ruled on the rights—under both Massachusetts common law and the Federal Constitution—of a noninstitutionalized incompetent mental patient as to involuntary treatment with antipsychotic drugs. Held: The Court of Appeals' judgment is vacated, and the case is remanded for that court's consideration, in the first instance, of whether the correct disposition of this case is affected by the Massachusetts Supreme Judicial Court's intervening decision. Pp. 298-306. (a) Assuming (as the parties agree) that the Constitution recognizes a liberty interest in avoiding the unwanted administration of antipsychotic drugs, a substantive issue remains as to the definition of that protected constitutional interest, as well as identification of the conditions under which competing state interests might outweigh it. There is also a procedural issue concerning the minimum procedures required by the Constitution for determining that an individual's liberty interest actually is outweighed in a particular instance. As a practical matter both issues are intertwined with questions of state law, which may create liberty interests and procedural protections broader than those protected by the Federal Constitution. If so, the minimal requirements of the Federal Constitution would not be controlling, and would not need to be identified in order to determine the legal rights and duties of persons within the State. Pp. 298-300. (b) While the record is unclear as to respondents' position in the District Court concerning the effect of state law on their asserted federal rights, in their brief in this Court they clearly assert state-law arguments as alternative grounds for affirming both the "substantive" and "procedural" decisions of the Court of Appeals. In applying the policy of avoiding unnecessary decisions of constitutional issues, it is not clear which, if any, constitutional issues now must be decided to resolve the controversy between the parties. Because of its greater familiarity both with the record and with Massachusetts law, the Court of Appeals is better situated than this Court to determine how the intervening state-court decision may have changed the law of Massachusetts and how any changes may affect this case. Pp. 304-306. 634 F.2d 650, 1 Cir., vacated and remanded. Stephen Schultz, Bridgeport, Conn., for petitioners. Richard Wayne Cole, Boston, Mass., for respondents. Justice POWELL delivered the opinion of the Court. 1 The Court granted certiorari in this case to determine whether involuntarily committed mental patients have a constitutional right to refuse treatment with antipsychotic drugs. 2 * This litigation began on April 27, 1975, when respondent Rubie Rogers and six other persons filed suit against various officials and staff of the May and Austin Units of the Boston State Hospital. The plaintiffs all were present or former mental patients at the institution. During their period of institutionalization all had been forced to accept unwanted treatment with antipsychotic drugs.1 Alleging that forcible administration of these drugs violated rights protected by the Constitution of the United States, the plaintiffs—respondents here sought compensatory and punitive damages and injunctive relief.2 3 The District Court certified the case as a class action. See Rogers v. Okin, 478 F.Supp. 1342, 1352, n.1 (Mass.1979). Although denying relief in damages, the court held that mental patients enjoy constitutionally protected liberty and privacy interests in deciding for themselves whether to submit to drug therapy.3 The District Court found that an involuntary "commitment" provides no basis for an inference of legal "incompetency" to make this decision under Massachusetts law. Id., at 1361-1362.4 Until a judicial finding of incompetency has been made, the court concluded, the wishes of the patients generally must be respected. Id., at 1365-1368. Even when a state court has rendered a determination of incompetency, the District Court found that the patient's right to make treatment decisions is not forfeited, but must be exercised on his behalf by a court-appointed guardian. Id., at 1364. Without consent either by the patient or his guardian, the court held, the patient's liberty interests may be overridden only in an emergency.5 4 The Court of Appeals for the First Circuit affirmed in part and reversed in part. Rogers v. Okin, 634 F.2d 650 (1980). It agreed that mental patients have a constitutionally protected interest in deciding for themselves whether to undergo treatment with antipsychotic drugs. Id., at 653.6 It also accepted the trial court's conclusion that Massachusetts law recognizes involuntarily committed persons as presumptively competent to assert this interest on their own behalf. See id., at 657-659. The Court of Appeals reached different conclusions, however, as to the circumstances under which state interests might override the liberty interests of the patient. 5 The Court of Appeals found that the State has two interests that must be weighed against the liberty interests asserted by the patient: a police power interest in maintaining order within the institution and in preventing violence, see id., at 655, and a parens patriae interest in alleviating the sufferings of mental illness and in providing effective treatment, see id., at 657. The court held that the State, under its police powers, may administer medication forcibly only upon a determination that "the need to prevent violence in a particular situation outweighs the possibility of harm to the medicated individual" and that "reasonable alternatives to the administration of antipsychotics [have been] ruled out." Id., at 656. Criticizing the District Court for imposing what it regarded as a more rigid standard, the Court of Appeals held that a hospital's professional staff must have substantial discretion in deciding when an impending emergency requires involuntary medication.7 The Court of Appeals reserved to the District Court, on remand, the task of developing mechanisms to ensure that staff decisions under the "police power" standard accord adequate procedural protection to "the interests of the patients."8 6 With respect to the State's parens patriae powers, the Court of Appeals accepted the District Court's state-law distinction between patients who have and patients who have not been adjudicated incompetent. Where a patient has not been found judicially to be "incompetent" to make treatment decisions under Massachusetts law,9 the court ruled that the parens patriae interest will justify involuntary medication only when necessary to prevent further deterioration in the patient's mental health. See id., at 660. The Court of Appeals reversed the District Court's conclusion that a guardian must be appointed to make nonemergency treatment decisions on behalf of incompetent patients. Even for incompetent patients, however, it ruled that the State's parens patriae interest would justify prescription only of such treatment as would be accepted voluntarily by "the individual himself . . . were he competent" to decide. Id., at 661.10 The Court of Appeals held that the patient's interest in avoiding undesired drug treatment generally must be protected procedurally by a judicial determination of "incompetency."11 If such a determination were made, further on-the-scene procedures still would be required before antipsychotic drugs could be administered forcibly in a particular instance. Ibid.12 7 Because the judgment of the Court of Appeals involved constitutional issues of potentially broad significance,13 we granted certiorari. Okin v. Rogers, 451 U.S. 906, 101 S.Ct. 1972, 68 L.Ed.2d 293 (1981). II A. 8 The principal question on which we granted certiorari is whether an involuntarily committed mental patient has a constitutional right to refuse treatment with antipsychotic drugs.14 This question has both substantive and procedural aspects. See 634 F.2d, at 656, 661; Rennie v. Klein, 653 F.2d 836, 841 (CA3 1981). The parties agree that the Constitution recognizes a liberty interest in avoiding the unwanted administration of antipsychotic drugs.15 Assuming that they are correct in this respect, the substantive issue involves a definition of that protected constitutional interest, as well as identification of the conditions under which competing state interests might outweigh it. See Youngberg v. Romeo, 457 U.S. 307, 319-320, 102 S.Ct. 2452, 2460, 73 L.Ed.2d 28; Bell v. Wolfish, 441 U.S. 520, 560, 99 S.Ct. 1861, 1885, 60 L.Ed.2d 447 (1979); Roe v. Wade, 410 U.S. 113, 147-154, 93 S.Ct. 705, 724, 727, 35 L.Ed.2d 147 (1973); Jacobson v. Massachusetts, 197 U.S. 11, 25-27 (1905). The procedural issue concerns the minimum procedures required by the Constitution for determining that the individual's liberty interest actually is outweighed in a particular instance. See Parham v. J.R., 442 U.S. 584, 606, 99 S.Ct. 2493, 2506, 61 L.Ed.2d 101 (1979); Mathews v. Eldridge, 424 U.S. 319, 335, 96 S.Ct. 893, 903, 47 L.Ed.2d 18 (1976). 9 As a practical matter both the substantive and procedural issues are intertwined with questions of state law. In theory a court might be able to define the scope of a patient's federally protected liberty interest without reference to state law.16 Having done so, it then might proceed to adjudicate the procedural protection required by the Due Process Clause for the federal interest alone. Cf. Vitek v. Jones, 445 U.S. 480, 491-494, 100 S.Ct. 1254, 1262-1264, 63 L.Ed.2d 552 (1980). For purposes of determining actual rights and obligations, however, questions of state law cannot be avoided. Within our federal system the substantive rights provided by the Federal Constitution define only a minimum. State law may recognize liberty interests more extensive than those independently protected by the Federal Constitution. See Greenholtz v. Nebraska Penal Inmates, 442 U.S. 1, 7, 12, 99 S.Ct. 2100, 2106, 60 L.Ed.2d 668 (1979); Oregon v. Hass, 420 U.S. 714, 719, 95 S.Ct. 1215, 1219, 43 L.Ed.2d 570 (1975); see also Brennan, State Constitutions and the Protection of Individual Rights, 90 Harv.L.Rev. 489 (1977). If so, the broader state protections would define the actual substantive rights possessed by a person living within that State. 10 Where a State creates liberty interests broader than those protected directly by the Federal Constitution, the procedures mandated to protect the federal substantive interests also might fail to determine the actual procedural rights and duties of persons within the State. Because state-created liberty interests are entitled to the protection of the federal Due Process Clause, see, e.g., Vitek v. Jones, supra, 445 U.S. at 488, 100 S.Ct. at 1261; Greenholtz v. Nebraska Penal Inmates, supra, 442 U.S. at 7, 99 S.Ct., at 2103, the full scope of a patient's due process rights may depend in part on the substantive liberty interests created by state as well as federal law. Moreover, a State may confer procedural protections of liberty interests that extend beyond those minimally required by the Constitution of the United States. If a State does so, the minimal requirements of the Federal Constitution would not be controlling, and would not need to be identified in order to determine the legal rights and duties of persons within that State. B 11 Roughly five months after the Court of Appeals decided this case, and shortly after this Court granted certiorari, the Supreme Judicial Court of Massachusetts announced its decision in Guardianship of Roe, 383 Mass. 415, 421 N.E.2d 40 (1981) (Roe ). Roe involved the right of a noninstitutionalized but mentally incompetent person to refuse treatment with antipsychotic drugs. Expressly resting its decision on the common law of Massachusetts as well as on the Federal Constitution,17 Massachusetts' highest court held in Roe that a person has a protected liberty interest in " 'decid[ing] for himself whether to submit to the serious and potentially harmful medical treatment that is represented by the administration of antipsychotic drugs.' " Id., at 433, n. 9, 421 N.E.2d, at 51, n. 9.18 The court found again apparently on the basis of the common law of Massachusetts as well as the Constitution of the United States—that this interest of the individual is of such importance that it can be overcome only by "an overwhelming State interest." Id., at 434, 421 N.E.2d, at 51. Roe further held that a person does not forfeit his protected liberty interest by virtue of becoming incompetent, but rather remains entitled to have his "substituted judgment" exercised on his behalf. Ibid. Defining this "substituted judgment" as one for which "[n]o medical expertise is required," id., at 435, 421 N.E.2d, at 52, the Massachusetts Supreme Judicial Court required a judicial determination of substituted judgment before drugs could be administered in a particular instance,19 except possibly in cases of medical emergency.20 C 12 The Massachusetts Supreme Court stated that its decision was limited to cases involving noninstitutionalized mental patients. See id., at 417, 441-442, 452-455, 421 N.E.2d, at 42, 55, 61-62.21 Nonetheless, respondents have argued in this Court that Roe may influence the correct disposition of the case at hand.22 We agree. 13 Especially in the wake of Roe, it is distinctly possible that Massachusetts recognizes liberty interests of persons adjudged incompetent that are broader than those protected directly by the Constitution of the United States. Compare Roe, supra, at 434, 421 N.E.2d, at 51 (protected liberty interest in avoiding unwanted treatment continues even when a person becomes incompetent and creates a right of incompetents to have their "substituted judgment" determined), with Addington v. Texas, 441 U.S. 418, 429-430, 99 S.Ct. 1804, 1811, 60 L.Ed.2d 323 (1979) (because a person "who is suffering from a debilitating mental illness" is not "wholly at liberty," and because the complexities of psychiatric diagnosis "render certainties virtually beyond reach," "practical considerations" may require "a compromise between what it is possible to prove and what protects the rights of the individual"). If the state interest is broader, the substantive protection that the Constitution affords against the involuntary administration of antipsychotic drugs would not determine the actual substantive rights and duties of persons in the State of Massachusetts. 14 Procedurally, it also is quite possible that a Massachusetts court, as a matter of state law, would require greater protection of relevant liberty interests than the minimum adequate to survive scrutiny under the Due Process Clause. Compare Roe, supra, 483 Mass., at 434, 421 N.E.2d, at 51 ("We have . . . stated our preference for judicial resolution of certain legal issues arising from proposed extraordinary medical treatment. . ."), with Youngberg v. Romeo, 457 U.S., at 322-323, 102 S.Ct., at 2462 ("[T]here certainly is no reason to think judges or juries are better qualified than appropriate professionals in making [treatment] decisions"), and with Parham v. J.R., 442 U.S., at 608, n. 16, 99 S.Ct., at 2507, n. 16 (Courts must not "unduly burde[n] the legitimate efforts of the states to deal with difficult social problems. The judicial model for factfinding for all constitutionally protected interests, regardless of their nature, can turn rational decisionmaking into an unmanageable enterprise").23 Again on this hypothesis state law would be dispositive of the procedural rights and duties of the parties to this case. 15 Finally, even if state procedural law itself remains unchanged by Roe, the federally mandated procedures will depend on the nature and weight of the state interests, as well as the individual interests, that are asserted. To identify the nature and scope of state interests that are to be balanced against an individual's liberty interests, this Court may look to state law. See, e.g., Roe v. Wade, 410 U.S., at 148 and n. 42, 151, and nn. 48-50, 93 S.Ct., at 724 and n. 42, 726, and nn. 48-50; Ingraham v. Wright, 430 U.S. 651, 661-663, 97 S.Ct. 1401, 1407-1408, 51 L.Ed.2d 711 (1977). Here we view the underlying state-law predicate for weighing asserted state interests as being put into doubt, if not altered, by Roe.24 D 16 It is unclear on the record presented whether respondents, in the District Court, did or did not argue the existence of "substantive" state-law liberty interests as a basis for their claim to procedural protection under the federal Due Process Clause, or whether they may have claimed state-law procedural protections for substantive federal interests.25 In their brief in this Court, however, respondents clearly assert state-law arguments as alternative grounds for affirming both the "substantive" and "procedural" decisions of the Court of Appeals. See Brief for Respondents, especially at 61, 71-72, 92-95. 17 Until certain questions have been answered, we think it would be inappropriate for us to attempt to weigh or even to identify relevant liberty interests that might be derived directly from the Constitution, independently of state law. It is this Court's settled policy to avoid unnecessary decisions of constitutional issues. See, e.g., City of Mesquite v. Aladdin's Castle, Inc., 455 U.S. 283, 294, 102 S.Ct. 1070, 1077, 71 L.Ed.2d 152 (1982); New York Transit Authority v. Beazer, 440 U.S. 568, 582-583, n. 22, 99 S.Ct. 1355, 1364, n. 22, 59 L.Ed.2d 587 (1979); Poe v. Ullman, 367 U.S. 497, 502-509, 81 S.Ct. 1752, 1755-1759, 6 L.Ed.2d 989 (1961); Ashwander v. TVA, 297 U.S. 288, 341, 347-348, 56 S.Ct. 466, 483, 80 L.Ed. 688 (1936) (Brandeis, J., concurring). This policy is supported, although not always required, by the prohibition against advisory opinions. Cf. United States v. Hastings, 296 U.S. 188, 193, 56 S.Ct. 218, 220, 80 L.Ed. 148 (1935) (review of one basis for a decision supported by another basis not subject to examination would represent "an expression of an abstract opinion"). 18 In applying this policy of restraint, we are uncertain here which if any constitutional issues now must be decided to resolve the controversy between the parties. In the wake of Roe, we cannot say with confidence that adjudication based solely on identification of federal constitutional interests would determine the actual rights and duties of the parties before us. And, as an additional cause for hesitation, our reading of the opinion of the Court of Appeals has left us in doubt as to the extent to which state issues were argued below and the degree to which the court's holdings may rest on subsequently altered state-law foundations. 19 Because of its greater familiarity both with the record and with Massachusetts law, the Court of Appeals is better situated than we to determine how Roe may have changed the law of Massachusetts and how any changes may affect this case. Accordingly, we think it appropriate for the Court of Appeals to determine in the first instance whether Roe requires revision of its holdings or whether it may call for the certification of potentially dispositive state-law questions to the Supreme Judicial Court of Massachusetts, see Bellotti v. Baird, 428 U.S. 132, 150-151, 96 S.Ct. 2857, 2867-2868, 49 L.Ed.2d 844 (1976).26 The Court of Appeals also may consider whether this is a case in which abstention now is appropriate. See generally Colorado River Water Conservation Dist. v. United States, 424 U.S. 800, 813-819, 96 S.Ct. 1236, 1244-1247, 47 L.Ed.2d 483 (1976). 20 The judgment of the Court of Appeals is therefore vacated, and the case is remanded for further proceedings consistent with this opinion. 21 So ordered. 1 As used in this litigation, the term "antipsychotic drugs" refers to medications such as Thorazine, Mellaril, Prolixin, and Haldol that are used in treating psychoses, especially schizophrenia. See Rogers v. Okin, 478 F.Supp. 1342, 1359-1360 (Mass.1979), aff'd in part and rev'd in part, 634 F.2d 650, 653 (CA1 1980). Sometimes called "major tranquilizers," these compounds were introduced into psychiatry in the early 1950's. See Cole & Davis, Antipsychotic Drugs, in 2 A. Freedman, H. Kaplan & B. Sadock, Comprehensive Textbook of Psychiatry II, pp. 1921-1922 (2d ed. 1975). It is not disputed that such drugs are "mind-altering." Their effectiveness resides in their capacity to achieve such effects. Citing authorities, petitioners assert that such drugs are essential not only to the treatment of individual disorders, but also to the preservation of institutional order generally needed for effective therapy. See Brief for Petitioners 17-41, 54-100. Respondents dispute this claim, also with support from medical authorities. Respondents also emphasize that antipsychotic drugs carry a significant risk of adverse side effects. These include such neurological syndromes as parkinsonisms, characterized by a mask-like face, retarded volitional movements, and tremors; akathisia, a clinical term for restlessness; dystonic reactions, including grimacing and muscle spasms; and tardive dyskinesia, a disease characterized in its mild form by involuntary muscle movements, especially around the mouth. Tardive dyskinesia can be even more disabling in its most severe forms. See Rogers v. Okin, 478 F.Supp., at 1360; Byck, Drugs and the Treatment of Psychiatric Disorders, in L. Goodman & A. Gilman, The Pharmalogical Basis of Therapeutics 152, 169 (5th ed. 1975). 2 The respondents also presented constitutional and statutory challenges to a hospital policy of secluding patients against their will. 478 F.Supp., at 1352. Their complaint additionally asserted claims for damages under state tort law. Id., at 1352, 1383. The District Court held that state law prevented seclusion except where necessary to prevent violence. See id., at 1371, 1374. Neither this decision, nor the denial of relief on the damages claims, is in issue before this Court. 3 The District Court characterized liberty to make "the intimate decision as to whether to accept or refuse [antipsychotic] medication" as "basic to any right of privacy" and therefore protected by the Constitution. See id., at 1366. The court did not derive this right from any particular constitutional provision, although it did observe that the "concept of a right of privacy . . . embodies First Amendment concerns." Ibid. In relying on the First Amendment the court reasoned that "the power to produce ideas is fundamental to our cherished right to communicate and is entitled to comparable constitutional protection." Id., at 1367. 4 Under the common law of torts, the right to refuse any medical treatment emerged from the doctrines of trespass and battery, which were applied to unauthorized touchings by a physician. See, e.g., Superintendent of Belchertown State School v. Saikewicz, 373 Mass. 728, 738-739, 370 N.E.2d 417, 424 (1977); W. Prosser, Law of Torts § 18 (4th ed. 1971). In this case the petitioners had argued—as they continue to argue—that the judicial commitment proceedings conducted under Massachusetts law, Mass.Gen.Laws Ann., ch. 123 (West Supp. 1982-1983), provide a determination of incompetency sufficient to warrant the State in providing treatment over the objections of the patient. In rejecting this argument as a matter of state law, the District Court relied principally on the language of the relevant Massachusetts statutes and on the regulations of the Department of Mental Health. See 478 F.Supp., at 1359, 1361 (citing Department of Mental Health Regulation § 221.02 ("No person shall be deprived of the right to manage his affairs . . . solely by reason of his admission or commitment to a facility except where there has been an adjudication that such person is incompetent"), and Mass.Gen.Laws Ann., ch. 123, § 25 (West Supp. 1982-1983) ("No person shall be deemed to be incompetent to manage his affairs . . . solely by reason of his admission or commitment in any capacity. . .")). The court also appears to have engaged in independent factfinding leading to the same conclusion: "The weight of the evidence persuades this court that, although committed mental patients do suffer at least some impairment of their relationship to reality, most are able to appreciate the benefits, risks, and discomfort that may reasonably be expected from receiving psychotropic medication." 478 F.Supp., at 1361. 5 The District Court defined an emergency as a situation in which failure to medicate "would result in a substantial likelihood of physical harm to th[e] patient, other patients, or to staff members of the institution." Id., at 1365. 6 The Court of Appeals termed it "intuitively obvious" that "a person has a constitutionally protected interest in being left free by the state to decide for himself whether to submit to the serious and potentially harmful medical treatment that is represented by the administration of antipsychotic drugs." 634 F.2d, at 653. Although the Court of Appeals found that the "precise textual source in the Constitution for the protection of this interest is unclear," ibid., it concluded that "a source in the Due Process Clause of the Fourteenth Amendment for the protection of this interest exists, most likely as part of the penumbral right to privacy, bodily integrity, or personal security." Ibid. The Court of Appeals found it unnecessary to examine the conclusion of the District Court that First Amendment interests also were implicated. 7 The Court of Appeals held that the District Court had erred in requiring what it construed as an overly simplistic mathematical calculation of the "quantitative" likelihood of harm. See id., at 656. 8 It asserted, apparently as a minimum, that "the determination that medication is necessary must be made by a qualified physician as to each individual patient to be medicated." Ibid. 9 A number of other States also distinguish between the standards governing involuntary commitment and those applying to determinations of incompetency to make treatment decisions. For a survey as of December 1, 1977, see Plotkin, Limiting the Therapeutic Orgy: Mental Patients' Right to Refuse Treatment, 72 Nw.U.L.Rev. 461, 504-525 (1977). The Court of Appeals for the Second Circuit has held that civil commitment does not raise even a presumption of incompetence. See Winters v. Miller, 446 F.2d 65 (1971). 10 In imposing this "substituted judgment" standard the Court of Appeals appears to have viewed its holding as mandated by the Federal Constitution. See 634 F.2d, at 661 ("In so holding, we do not imply that the Constitution. . .). But it followed its ultimate substantive conclusion with a citation to a Massachusetts case: "Cf. Superintendent of Belchertown v. Saikewicz," 373 Mass. 728, 370 N.E.2d 417 (1977). Saikewicz held that a court must apply the "substituted judgment" standard in determining whether to approve painful medical treatment for a profoundly retarded man incapable of giving informed consent. In Saikewicz the Massachusetts Supreme Judicial Court appears to have relied on both the Federal Constitution and the law of Massachusetts to support its decision. See id., at 738-741, 370 N.E.2d, at 424-425. But the Massachusetts court characterized its analysis as having identified a "constitutional right of privacy," id., at 739, 370 N.E.2d, at 424, thus creating some doubt as to the extent that the decision had an independent state-law basis. 11 The Court of Appeals appears to have agreed with the District Court that this determination, under Massachusetts law, would require a decision by the probate court under Mass.Gen.Laws Ann., ch. 123 § 25 (West Supp. 1982-1983); see ch. 201, §§ 1, 6, 12 (West Supp. 1982-1983) (appointment and powers of guardians). It suggested, however, that nonjudicial procedures would satisfy the federal constitutional requirements of due process. See 634 F.2d, at 659-660. 12 The Court of Appeals again instructed the District Court to develop procedural safeguards adequate to protect the patient's substantive interests. See id., at 661. 13 Constitutional questions involving the rights of committed mental patients to refuse antipsychotic drugs have been presented in other recent cases, including Rennie v. Klein, 653 F.2d 836 (CA3 1981), and Davis v. Hubbard, 506 F.Supp. 915 (ND Ohio 1980). On the issues raised, see generally Plotkin, supra; Shapiro, Legislating the Control of Behavior Control: Autonomy and the Coercive Use of Organic Therapies, 47 S.Cal.L.Rev. 237 (1974). 14 Pet. for Cert. 1. 15 In this Court petitioners appear to concede that involuntarily committed mental patients have a constitutional interest in freedom from bodily invasion, see Brief for Petitioners 43-47, but they deny that this interest is "fundamental." They also assert that it is outweighed in an appropriate balancing test by compelling state interests in administering antipsychotic drugs. Id., at 54-68. 16 As do the parties, we assume for purposes of this discussion that involuntarily committed mental patients do retain liberty interests protected directly by the Constitution, cf. O'Connor v. Donaldson, 422 U.S. 563, 95 S.Ct. 2486, 45 L.Ed.2d 396 (1975), and that these interests are implicated by the involuntary administration of antipsychotic drugs. Only "assuming" the existence of such interests, we of course intimate no view as to the weight of such interests in comparison with possible countervailing state interests. 17 See 383 Mass., at 417, and n. 1, 433, n. 9, 421 N.E.2d, at 42, and n. 1, 51, n. 9. 18 Although the Massachusetts court quoted this formulation from the decision of the Court of Appeals in Rogers v. Okin, 634 F.2d, at 653, the quotation is used to define the right, rather than to identify its legal source. Roe noted that Rogers v. Okin found the source of this right in the Due Process Clause of the Fourteenth Amendment. The court continued its discussion by stating its reliance on three bases, two of them not cited in Rogers v. Okin: the "inherent power of the court to prevent mistakes or abuses by guardians, whose authority comes from the Commonwealth," and the "common law" right of persons to decide what will be done with their bodies. 383 Mass., at 433, n. 9, 421 N.E.2d, at 51, n. 9. 19 See id., at 435, 421 N.E.2d, at 52: "The determination of what the incompetent individual would do if competent will probe the incompetent individual's values and preferences, and such an inquiry, in a case involving antipsychotic drugs [and a noninstitutionalized but incompetent patient], is best made in courts of competent jurisdiction." Having held that a "ward possesses but is incapable of exercising personally" the right to refuse antipsychotic drugs, the Massachusetts Supreme Court viewed the "primary dispute" as over "who ought to exercise this right on behalf of the ward." Id., at 434, 421 N.E.2d, at 51. The Supreme Judicial Court in Roe identified six "relevant" but "not exclusive" factors that should guide the decisions of the lower courts: "(1) the ward's expressed preferences regarding treatment; (2) his religious beliefs; (3) the impact upon the ward's family; (4) the probability of adverse side effects; (5) the consequences if treatment is refused; and (6) the prognosis with treatment." Id., at 444, 421 N.E.2d, at 57. It emphasized that the determination "must 'give the fullest possible expression to the character and circumstances' " of the individual patient and that "this is a subjective rather than an objective determination." Id., at 444, 421 N.E.2d, at 56 (citation and footnote omitted). 20 See id., at 440-441, 421 N.E.2d, at 54-55. 21 But cf. id., at 432, 421 N.E.2d, at 50 ("because of the likelihood of . . . the necessity of making similar determinations in other cases, we establish guidelines regarding the criteria to be used and the procedures to be followed in making a substituted judgment determination"), and id., at 453-454, 421 N.E.2d, at 62 ("We do not mean to imply that these [involuntarily committed] patients' rights are wholly unprotected or that their circumstances are entirely dissimilar to those we have discussed. We do suggest, however, that it would be imprudent to establish prematurely the relative importance of adverse interests. . ."). 22 Respondents first presented this argument in a motion to dismiss or in the alternative to certify certain questions to the Supreme Judicial Court of Massachusetts, filed in this Court on October 1, 1981. In their brief on the merits, respondents argue that Roe provides an alternative basis on which this Court could affirm the judgment of the Court of Appeals. 23 Even prior to Roe, the Court of Appeals concluded that Massachusetts state law, which it construed as requiring judicial determinations of incompetency separate from involuntary commitment proceedings, see 634 F.2d, at 658-659, "in many respects . . . goes well beyond the minimum requirements mandated by the Fourteenth Amendment," id., at 659 (footnote omitted). Roe now has taken the further step of requiring judicial procedure in every instance in which a guardian believes drug therapy necessary for a noninstitutionalized incompetent. 24 In Roe the Massachusetts court explicitly considered the implicated state interests, see 383 Mass., at 449, 421 N.E.2d, at 59, and concluded that the trial judge had erred in finding that the State had a "vital" parens patriae interest in "seeing that its residents function at the maximum level of their capacity," ibid. The Court of Appeals in this case had found and weighed a parens patriae interest. 634 F.2d, at 657-661. 25 Although relying primarily on federal constitutional grounds, the respondents' original complaint in the District Court could be construed as raising state-law guarantees either as alternative or as interrelated bases for relief. See Complaint in No. 75-1610-T (D.Mass.) (filed Apr. 27, 1975). In their briefs in the Court of Appeals, respondents relied unambiguously on state law in support of both the "substantive" and "procedural" rights that they now claim in this Court. See Brief for Plaintiff-Appellants in No. 79-1649, p. 44 ("Massachusetts law created a legal entitlement to be free from forced medications except in emergencies . . ."); Brief for Plaintiff-Appellees in No. 79-1648, p. 54 ("[T]he lower court's requirement that a guardian must decide whether an incompetent patient will receive psychotropic medication in a non-emergency was the correct application of state law and was not based upon constitutional authority") (emphasis omitted). 26 A certification procedure is provided by Mass. Rules of Court, Sup.Jud.Ct. Rule 1:03.

89

457 U.S. 307 102 S.Ct. 2452 73 L.Ed.2d 28 Duane YOUNGBERG, etc., et al., Petitioners,v.Nicholas ROMEO, an Incompetent, by his mother and next friend, Paula Romeo. No. 80-1429. Argued Jan. 11, 1982. Decided June 18, 1982. Syllabus Respondent, who is mentally retarded, was involuntarily committed to a Pennsylvania state institution. Subsequently, after becoming concerned about injuries which respondent had suffered at the institution, his mother filed an action as his next friend in Federal District Court for damages under 42 U.S.C. § 1983 against petitioner institution officials. She claimed that respondent had constitutional rights to safe conditions of confinement, freedom from bodily restraint, and training or "habilitation" and that petitioners knew, or should have known, about his injuries but failed to take appropriate preventive procedures, thus violating his rights under the Eighth and Fourteenth Amendments. In the ensuing jury trial, the District Court instructed the jury on the assumption that the Eighth Amendment was the proper standard of liability, and a verdict was returned for petitioners, on which judgment was entered. The Court of Appeals reversed and remanded for a new trial, holding that the Fourteenth, rather than the Eighth, Amendment provided the proper constitutional basis for the asserted rights. Held : Respondent has constitutionally protected liberty interests under the Due Process Clause of the Fourteenth Amendment to reasonably safe conditions of confinement, freedom from unreasonable bodily restraints, and such minimally adequate training as reasonably may be required by these interests. Whether respondent's constitutional rights have been violated must be determined by balancing these liberty interests against the relevant state interests. The proper standard for determining whether the State has adequately protected such rights is whether professional judgment in fact was exercised. And in determining what is "reasonable," courts must show deference to the judgment exercised by a qualified professional, whose decision is presumptively valid. Pp. 314-325. 3d Cir., 644 F.2d 147, vacated and remanded. David H. Allshouse, Harrisburg, Pa., for petitioners. Edmond A. Tiryak, Philadelphia, Pa., for respondents. [Amicus Curiae Information from page 308 intentionally omitted] Justice POWELL delivered the opinion of the Court. 1 The question presented is whether respondent, involuntarily committed to a state institution for the mentally retarded, has substantive rights under the Due Process Clause of the Fourteenth Amendment to (i) safe conditions of confinement; (ii) freedom from bodily restraints; and (iii) training or "habilitation."1 Respondent sued under 42 U.S.C. § 1983 three administrators of the institution, claiming damages for the alleged breach of his constitutional rights. 2 * Respondent Nicholas Romeo is profoundly retarded. Although 33 years old, he has the mental capacity of an 18-month-old child, with an I.Q. between 8 and 10. He cannot talk and lacks the most basic self-care skills. Until he was 26, respondent lived with his parents in Philadelphia. But after the death of his father in May 1974, his mother was unable to care for him. Within two weeks of the father's death, respondent's mother sought his temporary admission to a nearby Pennsylvania hospital. 3 Shortly thereafter, she asked the Philadelphia County Court of Common Pleas to admit Romeo to a state facility on a permanent basis. Her petition to the court explained that she was unable to care for Romeo or control his violence.2 As part of the commitment process, Romeo was examined by a physician and a psychologist. They both certified that respondent was severely retarded and unable to care for himself. App. 21a-22a and 28a-29a. On June 11, 1974, the Court of Common Pleas committed respondent to the Pennhurst State School and Hospital, pursuant to the applicable involuntary commitment provision of the Pennsylvania Mental Health and Mental Retardation Act, Pa.Stat.Ann., Tit. 50, § 4406(b) (Purdon 1969). 4 At Pennhurst, Romeo was injured on numerous occasions, both by his own violence and by the reactions of other residents to him. Respondent's mother became concerned about these injuries. After objecting to respondent's treatment several times, she filed this complaint on November 4, 1976, in the United States District Court for the Eastern District of Pennsylvania as his next friend. The complaint alleged that "[d]uring the period July, 1974 to the present, plaintiff has suffered injuries on at least sixty-three occasions." The complaint originally sought damages and injunctive relief from Pennhurst's director and two supervisors;3 it alleged that these officials knew, or should have known, that Romeo was suffering injuries and that they failed to institute appropriate preventive procedures, thus violating his rights under the Eighth and Fourteenth Amendments. 5 Thereafter, in late 1976, Romeo was transferred from his ward to the hospital for treatment of a broken arm. While in the infirmary, and by order of a doctor, he was physically restrained during portions of each day.4 These restraints were ordered by Dr. Gabroy, not a defendant here, to protect Romeo and others in the hospital, some of whom were in traction or were being treated intravenously. 7 Tr. 40, 49, 76-78. Although respondent normally would have returned to his ward when his arm healed, the parties to this litigation agreed that he should remain in the hospital due to the pending lawsuit. 5 id., at 248; 6 id., at 57-58 and 137. Nevertheless, in December 1977, a second amended complaint was filed alleging that the defendants were restraining respondent for prolonged periods on a routine basis. The second amended complaint also added a claim for damages to compensate Romeo for the defendants' failure to provide him with appropriate "treatment or programs for his mental retardation."5 All claims for injunctive relief were dropped prior to trial because respondent is a member of the class seeking such relief in another action.6 6 An 8-day jury trial was held in April 1978. Petitioners introduced evidence that respondent participated in several programs teaching basic self-care skills.7 A comprehensive behavior-modification program was designed by staff members to reduce Romeo's aggressive behavior,8 but that program was never implemented because of his mother's objections.9 Respondent introduced evidence of his injuries and of conditions in his unit.10 7 At the close of the trial, the court instructed the jury that "if any or all of the defendants were aware of and failed to take all reasonable steps to prevent repeated attacks upon Nicholas Romeo," such failure deprived him of constitutional rights. App. 73a. The jury also was instructed that if the defendants shackled Romeo or denied him treatment "as a punishment for filing this lawsuit," his constitutional rights were violated under the Eighth Amendment. Id., at 73a-75a. Finally, the jury was instructed that only if they found the defendants "deliberate[ly] indifferen[t] to the serious medical [and psychological] needs" of Romeo could they find that his Eighth and Fourteenth Amendment rights had been violated. Id., at 74a-75a.11 The jury returned a verdict for the defendants, on which judgment was entered. 8 The Court of Appeals for the Third Circuit, sitting en banc, reversed and remanded for a new trial. 644 F.2d 147 (1980). The court held that the Eighth Amendment, prohibiting cruel and unusual punishment of those convicted of crimes, was not an appropriate source for determining the rights of the involuntarily committed. Rather, the Fourteenth Amendment and the liberty interest protected by that Amendment provided the proper constitutional basis for these rights. In applying the Fourteenth Amendment, the court found that the involuntarily committed retain liberty interests in freedom of movement and in personal security. These were "fundamental liberties" that can be limited only by an "overriding, non-punitive" state interest. Id., at 157-158 (footnote omitted). It further found that the involuntarily committed have a liberty interest in habilitation designed to "treat" their mental retardation. Id., at 164-170.12 9 The en banc court did not, however, agree on the relevant standard to be used in determining whether Romeo's rights had been violated.13 Because physical restraint "raises a Presumption of a punitive sanction," The majority of the Court of Appeals concluded that it can be justified only by "compelling necessity." Id., at 159-160 (footnote omitted). A somewhat different standard was appropriate for the failure to provide for a resident's safety. The majority considered that such a failure must be justified by a showing of "substantial necessity." Id., at 164. Finally, the majority held that when treatment has been administered, those responsible are liable only if the treatment is not "acceptable in the light of present medical or other scientific knowledge." Id., at 166-167 and 173.14 10 Chief Judge Seitz, concurring in the judgment, considered the standards articulated by the majority as indistinguishable from those applicable to medical malpractice claims. In Chief Judge Seitz' view, the Constitution "only requires that the courts make certain that professional judgment in fact was exercised." Id., at 178. He concluded that the appropriate standard was whether the defendants' conduct was "such a substantial departure from accepted professional judgment, practice, or standards in the care and treatment of this plaintiff as to demonstrate that the defendants did not base their conduct on a professional judgment." Ibid.15 11 We granted the petition for certiorari because of the importance of the question presented to the administration of state institutions for the mentally retarded. 451 U.S. 982, 101 S.Ct. 2313, 68 L.Ed.2d 838 (1981). II 12 We consider here for the first time the substantive rights of involuntarily committed mentally retarded persons under the Fourteenth Amendment to the Constitution.16 In this case, respondent has been committed under the laws of Pennsylvania, and he does not challenge the commitment. Rather, he argues that he has a constitutionally protected liberty interest in safety, freedom of movement, and training within the institution; and that petitioners infringed these rights by failing to provide constitutionally required conditions of confinement. 13 The mere fact that Romeo has been committed under proper procedures does not deprive him of all substantive liberty interests under the Fourteenth Amendment. See, e.g., Vitek v. Jones, 445 U.S. 480, 491-494, 100 S.Ct. 1254, 1262-1264, 63 L.Ed.2d 552 (1980). Indeed, the state concedes that respondent has a right to adequate food, shelter, clothing, and medical care.17 We must decide whether liberty interests also exist in safety, freedom of movement, and training. If such interests do exist, we must further decide whether they have been infringed in this case. 14 Respondent's first two claims involve liberty interests recognized by prior decisions of this Court, interests that involuntary commitment proceedings do not extinguish.18 The first is a claim to safe conditions. In the past, this Court has noted that the right to personal security constitutes a "historic liberty interest" protected substantively by the Due Process Clause. Ingraham v. Wright, 430 U.S. 651, 673, 97 S.Ct. 1401, 1413, 51 L.Ed.2d 711 (1977). And that right is not extinguished by lawful confinement, even for penal purposes. See Hutto v. Finney, 437 U.S. 678, 98 S.Ct. 2565, 57 L.Ed.2d 522 (1978). If it is cruel and unusual punishment to hold convicted criminals in unsafe conditions, it must be unconstitutional to confine the involuntarily committed—who may not be punished at all—in unsafe conditions. 15 Next, respondent claims a right to freedom from bodily restraint. In other contexts, the existence of such an interest is clear in the prior decisions of this Court. Indeed, "[l]iberty from bodily restraint always has been recognized as the core of the liberty protected by the Due Process Clause from arbitrary governmental action." Greenholtz v. Nebraska Penal Inmates, 442 U.S. 1, 18, 99 S.Ct. 2100, 2109, 60 L.Ed.2d 668 (1979) (POWELL, J., concurring in part and dissenting in part). This interest survives criminal conviction and incarceration. Similarly, it must also survive involuntary commitment. B 16 Respondent's remaining claim is more troubling. In his words, he asserts a "constitutional right to minimally adequate habilitation." Brief for Respondent 8, 23, 45. This is a substantive due process claim that is said to be grounded in the liberty component of the Due Process Clause of the Fourteenth Amendment.19 The term "habilitation," used in psychiatry, is not defined precisely or consistently in the opinions below or in the briefs of the parties or the amici.20 As noted previously in n. 1, supra, the term refers to "training and development of needed skills." Respondent emphasizes that the right he asserts is for "minimal" training, see Brief for Respondent 34, and he would leave the type and extent of training to be determined on a case-by-case basis "in light of present medical or other scientific knowledge," id., at 45. 17 In addressing the asserted right to training, we start from established principles. As a general matter, a State is under no constitutional duty to provide substantive services for those within its border. See Harris v. McRae, 448 U.S. 297, 318, 100 S.Ct. 2671, 2689, 65 L.Ed.2d 784 (1980) (publicly funded abortions); Maher v. Roe, 432 U.S. 464, 469, 97 S.Ct. 2376, 2380, 53 L.Ed.2d 484 (1977) (medical treatment). When a person is institutionalized—and wholly dependent on the State—it is conceded by petitioners that a duty to provide certain services and care does exist, although even then a State necessarily has considerable discretion in determining the nature and scope of its responsibilities. See Richardson v. Belcher, 404 U.S. 78, 83-84, 92 S.Ct. 254, 258-259, 30 L.Ed.2d 231 (1971); Dandridge v. Williams, 397 U.S. 471, 478, 90 S.Ct. 1153, 1158, 25 L.Ed.2d 491 (1970). Nor must a State "choose between attacking every aspect of a problem or not attacking the problem at all." Id., at 486-487, 90 S.Ct., at 1162-1163. 18 Respondent, in light of the severe character of his retardation, concedes that no amount of training will make possible his release. And he does not argue that if he were still at home, the State would have an obligation to provide training at its expense. See Tr. of Oral Arg. 33. The record reveals that respondent's primary needs are bodily safety and a minimum of physical restraint, and respondent clearly claims training related to these needs.21 As we have recognized that there is a constitutionally protected liberty interest in safety and freedom from restraint, supra, at 315-316, training may be necessary to avoid unconstitutional infringement of those rights. On the basis of the record before us, it is quite uncertain whether respondent seeks any "habilitation" or training unrelated to safety and freedom from bodily restraints. In his brief to this Court, Romeo indicates that even the self-care programs he seeks are needed to reduce his aggressive behavior. See Brief for Respondent 21-22, 50. And in his offer of proof to the trial court, respondent repeatedly indicated that, if allowed to testify, his experts would show that additional training programs, including self-care programs, were needed to reduce his aggressive behavior. App. to Pet. for Cert. 98a-104a.22 If, as seems the case, respondent seeks only training related to safety and freedom from restraints, this case does not present the difficult question whether a mentally retarded person, involuntarily committed to a state institution, has some general constitutional right to training per se, even when no type or amount of training would lead to freedom.23 19 Chief Judge Seitz, in language apparently adopted by respondent, observed: 20 "I believe that the plaintiff has a constitutional right to minimally adequate care and treatment. The existence of a constitutional right to care and treatment is no longer a novel legal proposition." 644 F.2d, at 176. 21 Chief Judge Seitz did not identify or otherwise define—beyond the right to reasonable safety and freedom from physical restraint the "minimally adequate care and treatment" that appropriately may be required for this respondent.24 In the circumstances presented by this case, and on the basis of the record developed to date, we agree with his view and conclude that respondent's liberty interests require the State to provide minimally adequate or reasonable training to ensure safety and freedom from undue restraint. In view of the kinds of treatment sought by respondent and the evidence of record, we need go no further in this case.25 III A. 22 We have established that Romeo retains liberty interests in safety and freedom from bodily restraint. Yet these interests are not absolute; indeed to some extent they are in conflict. In operating an institution such as Pennhurst, there are occasions in which it is necessary for the State to restrain the movement of residents—for example, to protect them as well as others from violence.26 Similar restraints may also be appropriate in a training program. And an institution cannot protect its residents from all danger of violence if it is to permit them to have any freedom of movement. The question then is not simply whether a liberty interest has been infringed but whether the extent or nature of the restraint or lack of absolute safety is such as to violate due process. 23 In determining whether a substantive right protected by the Due Process Clause has been violated, it is necessary to balance "the liberty of the individual" and "the demands of an organized society." Poe v. Ullman, 367 U.S. 497, 542, 81 S.Ct. 1752, 1776, 6 L.Ed.2d 989 (1961) (Harlan, J., dissenting). In seeking this balance in other cases, the Court has weighed the individual's interest in liberty against the State's asserted reasons for restraining individual liberty. In Bell v. Wolfish, 441 U.S. 520, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979), for example, we considered a challenge to pretrial detainees' confinement conditions. We agreed that the detainees, not yet convicted of the crime charged, could not be punished. But we upheld those restrictions on liberty that were reasonably related to legitimate government objectives and not tantamount to punishment.27 See id., at 539, 99 S.Ct., at 1874. We have taken a similar approach in deciding procedural due process challenges to civil commitment proceedings. In Parham v. J. R., 442 U.S. 584, 99 S.Ct. 2493, 61 L.Ed.2d 101 (1979), for example, we considered a challenge to state procedures for commitment of a minor with parental consent. In determining that procedural due process did not mandate an adversarial hearing, we weighed the liberty interest of the individual against the legitimate interests of the State, including the fiscal and administrative burdens additional procedures would entail.28 Id., at 599-600, 99 S.Ct., at 2502-2503. 24 Accordingly, whether respondent's constitutional rights have been violated must be determined by balancing his liberty interests against the relevant state interests. If there is to be any uniformity in protecting these interests, this balancing cannot be left to the unguided discretion of a judge or jury. We therefore turn to consider the proper standard for determining whether a State adequately has protected the rights of the involuntarily committed mentally retarded. B 25 We think the standard articulated by Chief Judge Seitz affords the necessary guidance and reflects the proper balance between the legitimate interests of the State and the rights of the involuntarily committed to reasonable conditions of safety and freedom from unreasonable restraints. He would have held that "the Constitution only requires that the courts make certain that professional judgment in fact was exercised. It is not appropriate for the courts to specify which of several professionally acceptable choices should have been made." 644 F.2d, at 178. Persons who have been involuntarily committed are entitled to more considerate treatment and conditions of confinement than criminals whose conditions of confinement are designed to punish. Cf. Estelle v. Gamble, 429 U.S. 97, 104, 97 S.Ct. 285, 291, 50 L.Ed.2d 251 (1976). At the same time, this standard is lower than the "compelling" or "substantial" necessity tests the Court of Appeals would require a State to meet to justify use of restraints or conditions of less than absolute safety. We think this requirement would place an undue burden on the administration of institutions such as Pennhurst and also would restrict unnecessarily the exercise of professional judgment as to the needs of residents. 26 Moreover, we agree that respondent is entitled to minimally adequate training. In this case, the minimally adequate training required by the Constitution is such training as may be reasonable in light of respondent's liberty interests in safety and freedom from unreasonable restraints. In determining what is "reasonable" in this and in any case presenting a claim for training by a State we emphasize that courts must show deference to the judgment exercised by a qualified professional. By so limiting judicial review of challenges to conditions in state institutions, interference by the federal judiciary with the internal operations of these institutions should be minimized.29 Moreover, there certainly is no reason to think judges or juries are better qualified than appropriate professionals in making such decisions. See Parham v. J. R., supra, at 607, 99 S.Ct., at 2506-2507; bell V. WOLFIsh, supRa, at 544, 99 s.ct., at 1877 (courts shoulD not " 'second-guess the expert administrators on matters on which they are better informed' "). For these reasons, the decision, if made by a professional,30 is presumptively valid; liability may be imposed only when the decision by the professional is such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the person responsible actually did not base the decision on such a judgment.31 In an action for damages against a professional in his individual capacity, however, the professional will not be liable if he was unable to satisfy his normal professional standards because of budgetary constraints; in such a situation, good-faith immunity would bar liability. See n. 13, supra. IV 27 In deciding this case, we have weighed those postcommitment interests cognizable as liberty interests under the Due Process Clause of the Fourteenth Amendment against legitimate state interests and in light of the constraints under which most state institutions necessarily operate. We repeat that the State concedes a duty to provide adequate food, shelter, clothing, and medical care. These are the essentials of the care that the State must provide. The State also has the unquestioned duty to provide reasonable safety for all residents and personnel within the institution. And it may not restrain residents except when and to the extent professional judgment deems this necessary to assure such safety or to provide needed training. In this case, therefore, the State is under a duty to provide respondent with such training as an appropriate professional would consider reasonable to ensure his safety and to facilitate his ability to function free from bodily restraints. It may well be unreasonable not to provide training when training could significantly reduce the need for restraints or the likelihood of violence. 28 Respondent thus enjoys constitutionally protected interests in conditions of reasonable care and safety, reasonably nonrestrictive confinement conditions, and such training as may be required by these interests. Such conditions of confinement would comport fully with the purpose of respondent's commitment. Cf. Jackson v. Indiana, 406 U.S. 715, 738, 92 S.Ct. 1845, 1858, 32 L.Ed.2d 435 (1972); see n. 27, supra. In determining whether the State has met its obligations in these respects, decisions made by the appropriate professional are entitled to a presumption of correctness. Such a presumption is necessary to enable institutions of this type—often, unfortunately, overcrowded and understaffed—to continue to function. A single professional may have to make decisions with respect to a number of residents with widely varying needs and problems in the course of a normal day. The administrators, and particularly professional personnel, should not be required to make each decision in the shadow of an action for damages. 29 In this case, we conclude that the jury was erroneously instructed on the assumption that the proper standard of liability was that of the Eighth Amendment. We vacate the decision of the Court of Appeals and remand for further proceedings consistent with this decision. 30 So ordered. 31 Justice BLACKMUN, with whom Justice BRENNAN and Justice O'CONNOR join, concurring. 32 I join the Court's opinion. I write separately, however, to make clear why I believe that opinion properly leaves unresolved two difficult and important issues. 33 The first is whether the Commonwealth of Pennsylvania could accept respondent for "care and treatment," as it did under the Pennsylvania Mental Health and Mental Retardation Act of 1966, Pa.Stat.Ann., Tit. 50, § 4406(b) (Purdon 1969), and then constitutionally refuse to provide him any "treatment," as that term is defined by state law. Were that question properly before us, in my view there would be a serious issue whether, as a matter of due process, the State could so refuse. I therefore do not find that issue to be a "frivolous" one, as THE CHIEF JUSTICE does, post, at 330, n.1 34 In Jackson v. Indiana, 406 U.S. 715, 92 S.Ct. 1845, 32 L.Ed.2d 435 (1972), this Court, by a unanimous vote of all participating Justices, suggested a constitutional standard for evaluating the conditions of a civilly committed person's confinement: "At the least, due process requires that the nature and duration of commitment bear some reasonable relation to the purpose for which the individual is committed." Id., at 738, 92 S.Ct., at 1858. Under this standard, a State could accept a person for "safekeeping," then constitutionally refuse to provide him treatment. In such a case, commitment without treatment would bear a reasonable relation to the goal for which the person was confined. 35 If a state court orders a mentally retarded person committed for "care and treatment," however, I believe that due process might well bind the State to ensure that the conditions of his commitment bear some reasonable relation to each of those goals. In such a case, commitment without any "treatment" whatsoever would not bear a reasonable relation to the purposes of the person's confinement. 36 In respondent's case, the majority and principal concurring opinions in the Court of Appeals agreed that "[b]y basing [respondent's] deprivation of liberty at least partially upon a promise of treatment, the state ineluctably has committed the community's resources to providing minimal treatment." 644 F.2d 147, 168 (CA3 1980).2 Neither opinion clarified, however, whether respondent in fact had been totally denied "treatment," as that term is defined under Pennsylvania law. To the extent that the majority addressed the question, it found that "the evidence in the record, although somewhat contradictory, suggests not so much a total failure to treat as an inadequacy of treatment." Ibid. 37 This Court's reading of the record, ante, at 311-312, and n.7, supports that conclusion. Moreover, the Court today finds that respondent's entitlement to "treatment" under Pennsylvania law was not properly raised below. See ante, at 316, n. 19. Given this uncertainty in the record, I am in accord with the Court's decision not to address the constitutionality of a State's total failure to provide "treatment" to an individual committed under state law for "care and treatment." 38 The second difficult question left open today is whether respondent has an independent constitutional claim, grounded in the Due Process Clause of the Fourteenth Amendment, to that "habilitation" or training necessary to preserve those basic self-care skills he possessed when he first entered Pennhurst—for example, the ability to dress himself and care for his personal hygiene. In my view, it would be consistent with the Court's reasoning today to include within the "minimally adequate training required by the Constitution," ante, at 322, such training as is reasonably necessary to prevent a person's pre-existing self-care skills from deteriorating because of his commitment. 39 The Court makes clear, ante, at 315-316 and 324 that even after a person is committed to a state institution, he is entitled to such training as is necessary to prevent unreasonable losses of additional liberty as a result of his confinement—for example, unreasonable bodily restraints or unsafe institutional conditions. If a person could demonstrate that he entered a state institution with minimal self-care skills, but lost those skills after commitment because of the State's unreasonable refusal to provide him training, then, it seems to me, he has alleged a loss of liberty quite distinct from—and as serious as—the loss of safety and freedom from unreasonable restraints. For many mentally retarded people, the difference between the capacity to do things for themselves within an institution and total dependence on the institution for all of their needs is as much liberty as they ever will know. 40 Although respondent asserts a claim of this kind, I agree with the Court that "[o]n the basis of the record before us, it is quite uncertain whether respondent [in fact] seeks any 'habilitation' or training unrelated to safety and freedom from bodily restraints."3 Ante, at 318. Since the Court finds respondent constitutionally entitled at least to "such training as may be reasonable in light of [his] liberty interests in safety and freedom from unreasonable restraints," ante, at 322, I accept its decision not to address respondent's additional claim. 41 If respondent actually seeks habilitation in self-care skills not merely to reduce his aggressive tendencies, but also to maintain those basic self-care skills necessary to his personal autonomy within Pennhurst, I believe he is free on remand to assert that claim. Like the Court, I would be willing to defer to the judgment of professionals as to whether or not, and to what extent, institutional training would preserve respondent's pre-existing skills. Cf. ante, at 321-323. As the Court properly notes, "[p]rofessionals in the habilitation of the mentally retarded disagree strongly on the question whether effective training of all severely or profoundly retarded individuals is even possible." Ante, at 316, n. 20. 42 If expert testimony reveals that respondent was so retarded when he entered the institution that he had no basic self-care skills to preserve, or that institutional training would not have preserved whatever skills he did have, then I would agree that he suffered no additional loss of liberty even if petitioners failed to provide him training. But if the testimony establishes that respondent possessed certain basic self-care skills when he entered the institution, and was sufficiently educable that he could have maintained those skills with a certain degree of training, then I would be prepared to listen seriously to an argument that petitioners were constitutionally required to provide that training, even if respondent's safety and mobility were not imminently threatened by their failure to do so. 43 The Court finds it premature to resolve this constitutional question on this less than fully developed record. Because I agree with that conclusion, I concur in the Court's opinion. 44 Chief Justice BURGER, concurring in the judgment. 45 I agree with much of the Court's opinion. However, I would hold flatly that respondent has no constitutional right to training, or "habilitation," per se. The parties, and the Court, acknowledge that respondent cannot function outside the state institution, even with the assistance of relatives. Indeed, even now neither respondent nor his family seeks his discharge from state care. Under these circumstances, the State's provision of food, shelter, medical care, and living conditions as safe as the inherent nature of the institutional environment reasonably allows, serves to justify the State's custody of respondent. The State did not seek custody of respondent; his family understandably sought the State's aid to meet a serious need. 46 I agree with the Court that some amount of self-care instruction may be necessary to avoid unreasonable infringement of a mentally retarded person's interests in safety and freedom from restraint; but it seems clear to me that the Constitution does not otherwise place an affirmative duty on the State to provide any particular kind of training or habilitation—even such as might be encompassed under the essentially standardless rubric "minimally adequate training," to which the Court refers. See ante, at 319, and n. 24. Cf. 644 F.2d 147, 176 (CA3 1980) (Seitz, C.J., concurring in judgment). Since respondent asserts a right to "minimally adequate" habilitation "[q]uite apart from its relationship to decent care," Brief for Respondent 23, unlike the Court I see no way to avoid the issue.* Cf. ante, at 318. 47 I also point out that, under the Court's own standards, it is largely irrelevant whether respondent's experts were of the opinion that "additional training programs, including self-care programs, were needed to reduce [respondent's] aggressive behavior," ibid.—a prescription far easier for "spectators" to give than for an institution to implement. The training program devised for respondent by petitioners and other professionals at Pennhurst was, according to the Court's opinion, "presumptively valid"; and "liability may be imposed only when the decision by the professional is such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the person responsible actually did not base the decision on such a judgment." Ante, at 323. Thus, even if respondent could demonstrate that the training programs at Pennhurst were inconsistent with generally accepted or prevailing professional practice—if indeed there be such—this would not avail him so long as his training regimen was actually prescribed by the institution's professional staff. 48 Finally, it is worth noting that the District Court's instructions in this case were on the whole consistent with the Court's opinion today; indeed, some instructions may have been overly generous to respondent. Although the District Court erred in giving an instruction incorporating an Eighth Amendment "deliberate indifference" standard, the court also instructed, for example, that petitioners could be held liable if they "were aware of and failed to take all reasonable steps to prevent repeated attacks upon" respondent. See ante, at 312. Certainly if petitioners took "all reasonable steps" to prevent attacks on respondent, they cannot be said to have deprived him either of reasonably safe conditions or of training necessary to achieve reasonable safety. 1 The American Psychiatric Association explains: "The word 'habilitation,' . . . is commonly used to refer to programs for the mentally-retarded because mental retardation is . . . a learning disability and training impairment rather than an illness. [T]he principal focus of habilitation is upon training and development of needed skills." Brief for American Psychiatric Association as Amicus Curiae 4, n. 1. 2 Mrs. Romeo's petition to the Court of Common Pleas stated: "Since my husband's death I am unable to handle him. He becomes violent—kicks, punches, breaks glass; He can't speak—wants to express himself but can't. He is [a] constant 24 hr. care. [W]ithout my husband I am unable to care for him." App. 18a. 3 Petitioner Duane Youngberg was the Superintendent of Pennhurst; he had supervisory authority over the entire facility. Petitioner Richard Matthews was the Director of Resident Life at Pennhurst. Petitioner Marguerite Conley was Unit Director for the unit in which respondent lived. According to respondent, petitioners are administrators, not medical doctors. See Brief for Respondent 2. Youngberg and Matthews are no longer at Pennhurst. 4 Although the Court of Appeals described these restraints as "shackles," "soft" restraints, for the arms only, were generally used. 7 Tr. 53-55. 5 Respondent uses "treatment" as synonymous with "habilitation" or "training." See Brief for Respondent 21-23. 6 Pennhurst State School and Hospital v. Halderman, 451 U.S. 1, 101 S.Ct. 1531, 67 L.Ed.2d 694 (1981) (remanded for further proceedings). 7 Prior to his transfer to Pennhurst's hospital ward, Romeo participated in programs dealing with feeding, showering, drying, dressing, self-control, and toilet training, as well as a program providing interaction with staff members. Defendants' Exhibit 10; 3 Tr. 69-70; 5 id., at 44-56, 242-250; 6 id., at 162-166; 7 id., at 41-48. Some programs continued while respondent was in the hospital, 5 id., at 227, 248, 256; 6 id., at 50, 162-166, 6 id., at 32, 34, 41-48, and they reduced respondent's aggressive behavior to some extent, 7 id., at 45. 8 2 id., at 7; 5 id., at 88-90; 6 id., at 88, 200-203; Defendants' Exhibit 1, p. 9. The program called for short periods of separation from other residents and for use of "muffs" on plaintiff's hands for short periods of time, i.e., five minutes, to prevent him from harming himself or others. 9 1 Tr. 53; 4 id., at 25; 6 id., at 204. 10 The District Judge refused to allow testimony by two of Romeo's witnesses—trained professionals—indicating that Romeo would have benefited from more or different training programs. The trial judge explained that evidence of the advantages of alternative forms of treatment might be relevant to a malpractice suit, but was not relevant to a constitutional claim under § 1983. App. to Pet. for Cert. 101. 11 The "deliberate indifference" standard was adopted by this Court in Estelle v. Gamble, 429 U.S. 97, 104, 97 S.Ct. 285, 291, 50 L.Ed.2d 251 (1976), a case dealing with prisoners' rights to punishment that is not "cruel and unusual" under the Eighth Amendment. Although the District Court did not refer to Estelle v. Gamble in charging the jury, it erroneously used the deliberate-indifference standard articulated in that case. See App. 45a, 75a. 12 The Court of Appeals used "habilitation" and "treatment" as synonymous, though it regarded "habilitation" as more accurate in describing treatment needed by the mentally retarded. See 644 F.2d, at 165, and n. 40. 13 The existence of a qualified immunity defense was not at issue on appeal. The defendants had received instructions on this defense, App. 76a, and it was not challenged by respondent. 644 F.2d, at 173, n. 1. After citing Pierson v. Ray, 386 U.S. 547, 87 S.Ct. 1213, 18 L.Ed.2d 288 (1967), and Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974), the majority of the Court of Appeals noted that such instructions should be given again on the remand. 644 F.2d, at 171-172. 14 Actually, the court divided the right-to-treatment claim into three categories and adopted three standards, but only the standard described in text is at issue before this Court. The Court of Appeals also stated that if a jury finds that no treatment has been administered, it may hold the institution's administrators liable unless they can provide a compelling explanation for the lack of treatment, id., at 165, 173, but respondent does not discuss this precise standard in his brief and it does not appear to be relevant to the facts of this case. In addition, the court considered "least intrusive" analysis appropriate to justify severe intrusions on individual dignity such as permanent physical alteration or surgical intervention, id., at 165-166 and 173, but respondent concedes that this issue is not present in this case. 15 Judge Aldisert joined Chief Judge Seitz' opinion, but wrote separately to emphasize the nature of the difference between the majority opinion and that of the Chief Judge. On a conceptual level, Judge Aldisert thought that the court erred in abandoning the common-law method of deciding the case at bar rather than articulating broad principles unconnected with the facts of the case and of uncertain meaning. Id., at 182-183. And, on a pragmatic level, Judge Aldisert warned that neither juries nor those administering state institutions would receive guidance from the "amorphous constitutional law tenets" articulated in the majority opinion. Id., at 184. See id., at 183-185. Judge Garth also joined Chief Judge Seitz' opinion, and wrote separately to criticize the majority for addressing issues not raised by the facts of this case. Id., at 186. 16 In pertinent part, that Amendment provides that a State cannot deprive "any person of life, liberty, or property, without due process of law. . . ." U.S.Const., Amdt. 14, § 1. Respondent no longer relies on the Eighth Amendment as a direct source of constitutional rights. See Brief for Respondent 13, n. 12. 17 Brief for Petitioners 8, 11, 12, and n. 10; Brief for Respondent 15-16. See also Brief for State of Connecticut et al. as Amici Curiae 8. Petitioners argue that they have fully protected these interests. 18 Petitioners do not appear to argue to the contrary. See Brief for Petitioners 27-31. 19 Respondent also argues that because he was committed for care and treatment under state law he has a state substantive right to habilitation, which is entitled to substantive, not procedural, protection under the Due Process Clause of the Fourteenth Amendment. But this argument is made for the first time in respondent's brief to this Court. It was not advanced in the courts below, and was not argued to the Court of Appeals as a ground for reversing the trial court. Given the uncertainty of Pennsylvania law and the lack of any guidance on this issue from the lower federal courts, we decline to consider it now. See Dothard v. Rawlinson, 433 U.S. 321, 323, n. 1, 97 S.Ct. 2720, 2724, n. 1, 53 L.Ed.2d 786 (1977); Duignan v. United States, 274 U.S. 195, 200, 47 S.Ct. 566, 568, 71 L.Ed. 996 (1927); Old Jordan Milling Co. v. Societe Anonyme des Mines, 164 U.S. 261, 264-265, 17 S.Ct. 113, 114-115, 41 L.Ed. 427 (1896). 20 Professionals in the habilitation of the mentally retarded disagree strongly on the question whether effective training of all severely or profoundly retarded individuals is even possible. See, e.g., Favell, Risley, Wolfe, Riddle, & Rasmussen, The Limits of Habilitation: How Can We Identify Them and How Can We Change Them?, 1 Analysis and Intervention in Developmental Disabilities 37 (1981); Bailey, Wanted: A Rational Search for the Limiting Conditions of Habilitation in the Retarded, 1 Analysis and Intervention in Developmental Disabilities 45 (1981); Kauffman & Krouse, The Cult of Educability: Searching for the Substance of Things Hoped For; The Evidence of Things Not Seen, 1 Analysis and Intervention in Developmental Disabilities 53 (1981). 21 See, e.g., description of complaint supra, at 310. 22 See also Brief for Appellant in No. 78-1982, pp. 11-14, 20-21, and 24 (CA3). 23 In the trial court, respondent asserted that "state officials at a state mental hospital have a duty to provide residents . . . with such treatment as will afford them a reasonable opportunity to acquire and maintain those life skills necessary to cope as effectively as their capacities permit." App. to Pet. for Cert. 94a-95a. But this claim to a sweeping per se right was dropped thereafter. In his brief to this Court, respondent does not repeat it and, at oral argument, respondent's counsel explicitly disavowed any claim that respondent is constitutionally entitled to such treatment as would enable him "to achieve his maximum potential." Tr. of Oral Arg. 46-48. 24 Chief Judge Seitz used the term "treatment" as synonymous with training or habilitation. See 644 F.2d, at 181. 25 It is not feasible, as is evident from the variety of language and formulations in the opinions below and the various briefs here, to define or identify the type of training that may be required in every case. A court properly may start with the generalization that there is a right to minimally adequate training. The basic requirement of adequacy, in terms more familiar to courts, may be stated as that training which is reasonable in light of identifiable liberty interests and the circumstances of the case. A federal court, of course, must identify a constitutional predicate for the imposition of any affirmative duty on a State. Because the facts in cases of confinement of mentally retarded patients vary widely, it is essential to focus on the facts and circumstances of the case before a court. Judge Aldisert, in his concurring opinion in the court below, was critical of the "majority's abandonment of incremental decision-making in favor of promulgation of broad standards . . . [that] lac[k] utility for the groups most affected by this decision." Id., at 183-184. Judge Garth agreed that reaching issues not presented by the case requires a court to articulate principles and rules of law in "the absence of an appropriate record . . . and without the benefit of analysis, argument, or briefing" on such issues. Id., at 186. 26 In Romeo's case, there can be no question that physical restraint was necessary at times. See n. 2, supra. 27 See also Jackson v. Indiana, 406 U.S. 715, 738, 192 S.Ct. 1845, 1858, 32 L.Ed.2d 435 (1972) (holding that an incompetent pretrial detainee cannot, after a competency hearing, be held indefinitely without either criminal process or civil commitment; due process requires, at a minimum, some rational relation between the nature and duration of commitment and its purpose). This case differs in critical respects from Jackson, a procedural due process case involving the validity of an involuntary commitment. Here, respondent was committed by a court on petition of his mother who averred that in view of his condition she could neither care for him nor control his violence. N. 2, supra. Thus, the purpose of respondent's commitment was to provide reasonable care and safety, conditions not available to him outside of an institution. 28 See also Addington v. Texas, 441 U.S. 418, 99 S.Ct. 1804, 60 L.Ed.2d 323 (1979). In that case, we held that the State must prove the need for commitment by "clear and convincing" evidence. See id., at 431-432, 99 S.Ct., at 1812-1813. We reached this decision by weighing the individual's liberty interest against the State's legitimate interests in confinement. 29 See Parham v. J. R., 442 U.S. 584, 608, n. 16, 99 S.Ct. 2493, 2507, n. 16, 61 L.Ed.2d 101 (In limiting judicial review of medical decisions made by professionals, "it is incumbent on courts to design procedures that protect the rights of the individual without unduly burdening the legitimate efforts of the states to deal with difficult social problems"). See also Rhodes v. Chapman, 452 U.S. 337, 352, 101 S.Ct. 2392, 2402, 69 L.Ed.2d 59 (1981) ("[C]ourts cannot assume that state legislatures and prison officials are insensitive to the requirements of the Constitution or to the perplexing sociological problems of how best to achieve the goals of the penal function in the criminal justice system . . ."); Bell v. Wolfish, 441 U.S. 520, 539, 99 S.Ct. 1861, 1874, 60 L.Ed.2d 447 (1979) (In the context of conditions of confinement of pretrial detainees, "[c]ourts must be mindful that these inquiries spring from constitutional requirements and that judicial answers to them must reflect that fact rather than a court's idea of how best to operate a detention facility"); Wolff v. McDonnell, 418 U.S. 539, 556, 94 S.Ct. 2963, 2974-2975, 41 L.Ed.2d 935 (1974) (In considering a procedural due process claim in the context of prison, "there must be mutual accommodation between institutional needs and objectives and the provisions of the Constitution that are of general application"). See also Townsend & Mattson, The Interaction of Law and Special Education: Observing the Emperor's New Clothes, 1 Analysis and Intervention in Developmental Disabilities 75 (1981) (judicial resolution of rights of the handicapped can have adverse as well as positive effects on social change). 30 By "professional" decisionmaker, we mean a person competent, whether by education, training or experience, to make the particular decision at issue. Long-term treatment decisions normally should be made by persons with degrees in medicine or nursing, or with appropriate training in areas such as psychology, physical therapy, or the care and training of the retarded. Of course, day-to-day decisions regarding care—including decisions that must be made without delay—necessarily will be made in many instances by employees without formal training but who are subject to the supervision of qualified persons. 31 All members of the Court of Appeals agreed that respondent's expert testimony should have been admitted. This issue was not included in the questions presented for certiorari, and we have no reason to disagree with the view that the evidence was admissible. It may be relevant to whether petitioners' decisions were a substantial departure from the requisite professional judgment. See supra, at this page. 1 See also Garvey, Freedom and Choice in Constitutional Law, 94 Harv.L.Rev. 1756, 1787-1791 (1981); Welsch v. Likins, 550 F.2d 1122, 1126, and n. 6 (CA8 1977); Wyatt v. Aderholt, 503 F.2d 1305 (CA5 1974), aff'g Wyatt v. Stickney, 325 F.Supp. 781, 785 (MD Ala.1971). 2 In the principal concurring opinion, Chief Judge Seitz, for himself and three other judges, stated: "The state does not contest that it has placed the [respondent] in Pennhurst to provide basic care and treatment. Indeed, he has a right to treatment under state law, . . . and the fact that Pennhurst has programs and staff to treat patients is indicative of such a purpose. I believe that when the purpose of confining a mentally retarded person is to provide care and treatment, as is undoubtedly the case here, it violates the due process clause to fail to fulfill that purpose." 644 F.2d, at 176. 3 At trial, respondent's attorney requested a jury instruction that "[u]nder the Eighth and Fourteenth Amendments, state officials at a state mental hospital have a duty to provide residents of such institutions with such treatment as will afford them a reasonable opportunity to acquire and maintain those life skills necessary to cope as effectively as their capacities permit." App. to Pet. for Cert. 94a-95a (emphasis added). In this Court, respondent again argued that "without minimal habilitative efforts—basic training in fundamental life skills—institutionalized retarded persons not only will fail to develop such skills independently but also will lose the skills they may have brought with them into the institution. . . . Indeed, putting aside increased risks of physical harm, if a retarded individual loses all of his previously acquired skills through prolonged institutional neglect, then the State has worked positive injury . . . . Once [retarded persons] have been confined they have no one but the State to turn to for help in gaining additional skills or, at least, preserving whatever skills and abilities they have." Brief for Respondent 22-23 (emphasis added). Respondent's description of the expert testimony to be offered on remand, however, suggests that he seeks training in self-care skills primarily to ensure his personal safety and the safety of others. See, e.g., App. to Pet. for Cert. 100a (respondent's offer of proof that "when mentally retarded individuals learn alternative behavior, such as toilet training and dressing and so forth, [their] aggression decreases"); Brief for Respondent 22 (training in self-care skills is necessary to prevent development of "a variety of inappropriate, aggressive and self-destructive behaviors"). * Indeed, in the trial court respondent asserted a broad claim to such "treatment as [would] afford [him] a reasonable opportunity to acquire and maintain those life skills necessary to cope as effectively as [his] capacities permit." App. to Pet. for Cert. 94a. Respondent also maintains that, because state law purportedly creates a right to "care and treatment," he has a federal substantive right under the Due Process Clause to enforcement of this state right. See ante, at 316, n. 19. This contention is obviously frivolous; were every substantive right created by state law enforceable under the Due Process Clause, the distinction between state and federal law would quickly be obliterated.

34

457 U.S. 594 102 S.Ct. 2612 73 L.Ed.2d 245 Johann SCHMIDT, and Leland Pollard, etc., Petitioners,v.OAKLAND UNIFIED SCHOOL DISTRICT et al. No. 81-1444. June 21, 1982. PER CURIAM. 1 California Ed.Code Ann. § 39640 (West Supp.1982) requires school districts to award any contracts for work involving more than $12,000 to the "lowest responsible bidder." For projects over $100,000, the Oakland School District requires that to be considered responsible, general contractors must use minority-owned businesses for at least 25 percent of the dollar amount of the total bid. Petitioners submitted the low bid for an advertised project but were disqualified under the School District plan as not being responsible. They brought this action claiming damages and asserting that the affirmative-action plan violated not only the Federal Constitution but also state law. The Court of Appeals affirmed a judgment of the District Court upholding the plan on constitutional grounds. 662 F.2d 550 (1981). Although the Court of Appeals acknowledged that under one of its prior decisions, the plan at issue might be invalid under state law, it declined to decide the state-law question since it was a sensitive matter and petitioners could present it to the state courts. 2 If the affirmative-action plan is invalid under state law, the Court of Appeals need not have reached the federal constitutional issue. Nevertheless, the Court of Appeals declined to resolve the pendent state-law claim. Under Hagans v. Lavine, 415 U.S. 528, 546, 94 S.Ct. 1372, 1383, 39 L.Ed.2d 577 (1974), and Mine Workers v. Gibbs, 383 U.S. 715, 86 S.Ct. 1130, 16 L.Ed.2d 218 (1966), this was an abuse of discretion in the circumstances of this case. 3 We accordingly grant the petition for certiorari, vacate the judgment of the Court of Appeals, and remand the case for further proceedings consistent with this opinion. 4 So ordered.

89

73 L.Ed.2d 149 102 S.Ct. 2540 457 U.S. 465 BLUE SHIELD OF VIRGINIA, et al., Petitionersv.Carol McCREADY. No. 81-225. Argued March 24, 1982. Decided June 21, 1982. Syllabus Respondent employee was provided coverage under a prepaid group health plan purchased by her employer from petitioner Blue Shield of Virginia (Blue Shield). The plan provided reimbursement for part of the cost incurred by subscribers for outpatient treatment for mental and nervous disorders, including psychotherapy. However, Blue Shield's practice was to reimburse subscribers for services provided by psychiatrists but not by psychologists unless the treatment was supervised by and billed through a physician. Respondent was treated by a clinical psychologist and submitted claims to Blue Shield for the costs of the treatment. After the claims were routinely denied because they had not been billed through a physician, respondent brought a class action in Federal District Court, alleging that Blue Shield and petitioner Neuropsychiatric Society of Virginia, Inc., had engaged in an unlawful conspiracy in violation of § 1 of the Sherman Act to exclude psychologists from receiving compensation under Blue Shield's plans. She further alleged that Blue Shield's failure to reimburse was in furtherance of the conspiracy and had caused injury to her business or property for which she was entitled to treble damages under § 4 of the Clayton Act, which provides for recovery of such damages by "[a]ny person" injured "by reason of anything" prohibited in the antitrust laws. The District Court granted petitioners' motion to dismiss, holding that respondent had no standing under § 4 to maintain her suit. The Court of Appeals reversed. Held : Respondent has standing to maintain the action under § 4 of the Clayton Act. Pp. 472-485. (a) The lack of restrictive language in § 4 reflects Congress' expansive remedial purpose of creating a private enforcement mechanism to deter violators and deprive them of the fruits of their illegal actions, and to provide ample compensation to victims of antitrust violations. In the absence of some articulable consideration of statutory policy suggesting a contrary conclusion in a particular factual setting, § 4 is to be applied in accordance with its plain language and its broad remedial and deterrent objectives. Pp.472-473 (b) Permitting respondent to proceed does not offer the slightest possibility of a duplicative exaction from petitioners, Hawaii v. Standard Oil Co., 405 U.S. 251, 92 S.Ct. 885, 31 L.Ed.2d 184, and Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707, distinguished, since she had paid her psychologist, who thus was not injured by Blue Shield's refusal to reimburse respondent. And whatever the adverse effect of Blue Shield's actions on respondent's employer, who purchased the plan, it is not the employer as purchaser, but its employees as subscribers, who are out of pocket as a consequence of the plan's failure to pay benefits. Pp.473-475. (c) In determining whether a particular injury is too remote from the alleged violation to warrant § 4 standing, consideration is to be given (1) to the physical and economic nexus between the alleged violation and the harm to the plaintiff, and (2), more particularly, to the relationship of the injury alleged with those forms of injury about which Congress was likely to have been concerned in making defendant's conduct unlawful and in providing a private remedy under § 4. Pp. 476-478. (d) Respondent's injury is not rendered "remote" merely because the alleged goal of petitioners was to halt encroachment by psychologists into a market that physicians and psychiatrists sought to preserve for themselves. Here, the § 4 remedy cannot reasonably be restricted to those competitors whom petitioners hoped to eliminate from the market. Denying reimbursement to subscribers for the cost of treatment was the very means by which it is alleged that Blue Shield sought to achieve its alleged illegal ends, and respondent's injury was precisely the type of loss that the claimed violations would be likely to cause. Nor is the § 4 remedy unavailable to respondent on the asserted ground that standing should be limited to participants in the restrained market in group health care plans—that is, to entities, such as respondent's employer, who were purchasers of group health plans. Respondent did not allege a restraint in the market for group health plans, but instead premised her claim on the concerted refusal to reimburse under a plan that would permit reimbursement for psychologists' services. As a consumer of psychotherapy services entitled to financial benefits under the Blue Shield plan, she was within that area of the economy endangered by the breakdown of competitive conditions resulting from Blue Shield's selective refusal to reimburse. Pp. 478-481. (e) Section 4 standing is not precluded on the asserted ground that respondent's injury does not reflect the "anticompetitive" effect of the alleged boycott. Her injury was of a type that Congress sought to redress in providing a private remedy for violations of the antitrust laws. Respondent did not yield to Blue Shield's coercive pressure to induce its subscribers into selecting psychiatrists over psychologists for the services they required, but instead bore Blue Shield's sanction in the form of an increase in the net cost of her psychologist's services. In light of the conspiracy here alleged, respondent's injury "flows from that which makes defendants' acts unlawful," Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489, 97 S.Ct. 690, 697, 50 L.Ed.2d 701 and falls squarely within the area of congressional concern. Pp. 481-484. 649 F.2d 228 (4th Cir.), affirmed. Griffin B. Bell, Atlanta, Ga., for petitioners. Warwick R. Furr, II, Washington, D. C., for respondent. Justice BRENNAN delivered the opinion of the Court. 1 The antitrust complaint at issue in this case alleges that a group health plan's practice of refusing to reimburse subscribers for psychotherapy performed by psychologists, while providing reimbursement for comparable treatment by psychiatrists, was in furtherance of an unlawful conspiracy to restrain competition in the psychotherapy market. The question presented is whether a subscriber who employed the services of a psychologist has standing to maintain an action under § 4 of the Clayton Act based upon the plan's failure to provide reimbursement for the costs of that treatment. 2 * From September 1975 until January 1978, respondent Carol McCready was an employee of Prince William County, Va. As part of her compensation, the county provided her with coverage under a prepaid group health plan purchased from petitioner Blue Shield of Virginia (Blue Shield).1 The plan specifically provided reimbursement for a portion of the cost incurred by subscribers with respect to outpatient treatment for mental and nervous disorders, including psychotherapy. Pursuant to this provision, Blue Shield reimbursed subscribers for psychotherapy provided by psychiatrists. But Blue Shield did not provide reimbursement for the services of psychologists unless the treatment was supervised by and billed through a physician.2 While a subscriber to the plan, McCready was treated by a clinical psychologist. She submitted claims to Blue Shield for the costs of that treatment, but those claims were routinely denied because they had not been billed through a physician.3 3 In 1978, McCready brought this class action in the United States District Court for the Eastern District of Virginia, on behalf of all Blue Shield subscribers who had incurred costs for psychological services since 1973 but who had not been reimbursed.4 The complaint alleged that Blue Shield and petitioner Neuropsychiatric Society of Virginia, Inc., had engaged in an unlawful conspiracy in violation of § 1 of the Sherman Act, 26 Stat. 209, as amended, 15 U.S.C. § 1,5 "to exclude and boycott clinical psychologists from receiving compensation under" the Blue Shield plans. App. 55. McCready further alleged that Blue Shield's failure to reimburse had been in furtherance of the alleged conspiracy, and had caused injury to her business or property for which she was entitled to treble damages and attorney's fees under § 4 of the Clayton Act, 38 Stat. 731, 15 U.S.C. § 15.6 4 The District Court granted petitioners' motion to dismiss, holding that McCready had no standing under § 4 to maintain her suit.7 In the District Court's view, McCready's standing to maintain a § 4 action turned on whether she had suffered injury "within the sector of the economy competitively endangered by the defendants' alleged violations of the antitrust laws." App. 17. Noting that the goal of the alleged boycott was to exclude clinical psychologists from a segment of the psychotherapy market, the court concluded that the "sector of the economy competitively endangered" by the charged violation extended "no further than that area occupied by the psychologists." Id., at 18 (emphasis in original). Thus, while McCready clearly had suffered an injury by being denied reimbursement, this injury was "to indirect and remote to be considered 'antitrust injury.' " Ibid. 5 A divided panel of the United States Court of Appeals for the Fourth Circuit reversed, holding that McCready had alleged an injury within the meaning of § 4 of the Clayton Act and had standing to maintain the suit. 649 F.2d 228 (4th Cir. 1981). The court recognized that the goal of the alleged conspiracy was the exclusion of clinical psychologists from some segment of the psychotherapy market. But it held that the § 4 remedy was available to any person "whose property loss is directly or proximately caused by" a violation of the antitrust laws, and that McCready's loss was not "too remote or indirect to be covered by the Act." Id., at 231.8 The court thus remanded the case to the District Court for further proceedings. We granted certiorari. 454 U.S. 962, 102 S.Ct. 500, 72 L.Ed.2d 376 (1981). II 6 Section 4 of the Clayton Act, 38 Stat. 731, provides a treble-damages remedy to "[a]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws," 15 U.S.C. § 15 (emphasis added). As we noted in Reiter v. Sonotone Corp., 442 U.S. 330, 337, 99 S.Ct. 2326, 2330, 60 L.Ed.2d 931 (1979), "[o]n its face, § 4 contains little in the way of restrictive language." And the lack of restrictive language reflects Congress' "expansive remedial purpose" in enacting § 4: Congress sought to create a private enforcement mechanism that would deter violators and deprive them of the fruits of their illegal actions, and would provide ample compensation to the victims of antitrust violations. Pfizer Inc. v. India, 434 U.S. 308, 313-314, 98 S.Ct. 584, 587-88, 54 L.Ed.2d 563 (1978). See Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 485-486, and n. 10, 97 S.Ct. 690, 695-96, and n. 10, 50 L.Ed.2d 701 (1977); Perma Mufflers, Inc. v. International Parts Corp., 392 U.S. 134, 139, 88 S.Ct. 1981, 1984, 20 L.Ed.2d 982 (1968); American Society of Mechanical Engineers v. Hydrolevel Corp., 456 U.S. 556, 572-573, and n. 10, 102 S.Ct. 1935, 1945-1946, and n. 10, 72 L.Ed.2d 330 (1982). As we have recognized, "[t]he statute does not confine its protection to consumers, or to purchasers, or to competitors, or to sellers. . . . The Act is comprehensive in its terms and coverage, protecting all who are made victims of the forbidden practices by whomever they may be perpetrated." Mandeville Island Farms, Inc. v. American Crystal Sugar Co., 334 U.S. 219, 236, 68 S.Ct. 996, 1006, 92 L.Ed. 1328 (1948). 7 Consistent with the congressional purpose, we have refused to engraft artificial limitations on the § 4 remedy.9 Two recent cases illustrate the point. Pfizer Inc. v. India, supra, afforded the statutory phrase "any person" its "naturally broad and inclusive meaning," id., 434 U.S. at 312, 98 S.Ct., at 587, and held that it extends even to an action brought by a foreign sovereign. Similarly, Reiter v. Sonotone Corp., supra, rejected the argument that the § 4 remedy is available only to redress injury to commercial interests. In that case we afforded the statutory term "property" its "naturally broad and inclusive meaning," and held that a consumer has standing to seek a § 4 remedy reflecting the increase in the purchase price of goods that was attributable to a price-fixing conspiracy. 442 U.S., at 338, 99 S.Ct., at 2330. In sum, in the absence of some articulable consideration of statutory policy suggesting a contrary conclusion in a particular factual setting, we have applied § 4 in accordance with its plain language and its broad remedial and deterrent objectives. But drawing on statutory policy, our cases have acknowledged two types of limitation on the availability of the § 4 remedy to particular classes of persons and for redress of particular forms of injury. We treat these limitations in turn.10 8 In Hawaii v. Standard Oil Co., 405 U.S. 251, 92 S.Ct. 885, 31 L.Ed.2d 184 (1972), we held that § 4 did not authorize a State to sue in its parens patriae capacity for damages to its "general economy." Noting that a "large and ultimately indeterminable part of the injury to the 'general economy' . . . is no more than a reflection of injuries to the 'business or property' of consumers, for which they may recover themselves under § 4," we concluded that "[e]ven the most lengthy and expensive trial could not . . . cope with the problems of double recovery inherent in allowing damages" for injury to the State's quasi-sovereign interests. Id., at 264, 92 S.Ct., at 892. See Reiter v. Sonotone Corp., supra, 442 U.S., at 342, 99 S.Ct., at 2332. 9 In Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977), similar concerns prevailed. Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 88 S.Ct. 2224, 20 L.Ed.2d 1231 (1968), had held that an antitrust defendant could not relieve itself of its obligation to pay damages resulting from overcharges to a direct-purchaser plaintiff by showing that the plaintiff had passed the amount of the overcharge on to its own customers. Illinois Brick was an action by an indirect purchaser claiming damages from the antitrust violator measured by the amount that had been passed on to it. Relying in part on Hawaii v. Standard Oil Co., supra, the Court found unacceptable the risk of duplicative recovery engendered by allowing both direct and indirect purchasers to claim damages resulting from a single overcharge by the antitrust defendant. Illinois Brick, supra, 431 U.S., at 730-731, 97 S.Ct., at 2066-2067. The Court found that the splintered recoveries and litigative burdens that would result from a rule requiring that the impact of an overcharge be apportioned between direct and indirect purchasers could undermine the active enforcement of the antitrust laws by private actions. 431 U.S., at 745-747, 97 S.Ct., at 2074-2075. The Court concluded that direct purchasers rather than indirect purchasers were the injured parties who as a group were most likely to press their claims with the vigor that the § 4 treble-damages remedy was intended to promote. Id., at 735, 97 S.Ct., at 2069. 10 The policies identified in Hawaii and Illinois Brick plainly offer no support for petitioners here. Both cases focused on the risk of duplicative recovery engendered by allowing every person along a chain of distribution to claim damages arising from a single transaction that violated the antitrust laws. But permitting respondent to proceed in the circumstances of this case offers not the slightest possibility of a duplicative exaction from petitioners. McCready has paid her psychologist's bills; her injury consists of Blue Shield's failure to pay her. Her psychologist can link no claim of injury to himself arising from his treatment of McCready; he has been fully paid for his service and has not been injured by Blue Shield's refusal to reimburse her for the cost of his services. And whatever the adverse effect of Blue Shield's actions on McCready's employer, who purchased the plan, it is not the employer as purchaser, but its employees as subscribers, who are out of pocket as a consequence of the plan's failure to pay benefits.11 B 11 Analytically distinct from the restrictions on the § 4 remedy recognized in Hawaii and Illinois Brick, there is the conceptually more difficult question "of which persons have sustained injuries too remote [from an antitrust violation] to give them standing to sue for damages under § 4." Illinois Brick Co. v. Illinois, 431 U.S., at 728, n. 7, 97 S.Ct., at 2065, n. 7 (emphasis added).12 An antitrust violation may be expected to cause ripples of harm to flow through the Nation's economy; but "despite the broad wording of § 4 there is a point beyond which the wrongdoer should not be held liable." Id., at 760, 97 S.Ct., at 2082 (BRENNAN, J., dissenting). It is reasonable to assume that Congress did not intend to allow every person tangentially affected by an antitrust violation to maintain an action to recover threefold damages for the injury to his business or property. Of course, neither the statutory language nor the legislative history of § 4 offers any focused guidance on the question of which injuries are too remote from the violation and the purposes of the antitrust laws to form the predicate for a suit under § 4; indeed, the unrestrictive language of the section, and the avowed breadth of the congressional purpose, cautions us not to cabin § 4 in ways that will defeat its broad remedial objective. But the potency of the remedy implies the need for some care in its application. In the absence of direct guidance from Congress, and faced with the claim that a particular injury is too remote from the alleged violation to warrant § 4 standing, the courts are thus forced to resort to an analysis no less elusive than that employed traditionally by courts at common law with respect to the matter of "proximate cause."13 See Perkins v. Standard Oil Co., 395 U.S. 642, 649, 89 S.Ct. 1871, 1875, 23 L.Ed.2d 599 (1969); Karseal Corp. v. Richfield Oil Corp., 221 F.2d 358, 363 (CA9 1955). In applying that elusive concept to this statutory action, we look (1) to the physical and economic nexus between the alleged violation and the harm to the plaintiff, and (2), more particularly, to the relationship of the injury alleged with those forms of injury about which Congress was likely to have been concerned in making defendant's conduct unlawful and in providing a private remedy under § 4. 12 (1) 13 It is petitioners' position that McCready's injury is too "fortuitous" and too "incidental" to and "remote" from the alleged violation to provide the basis for a § 4 action.14 At the outset, petitioners argue that because the alleged conspiracy was directed by its protagonists at psychologists, and not at subscribers to group health plans, only psychologists might maintain suit. This argument may be quickly disposed of. 14 We do not think that because the goal of the conspirators was to halt encroachment by psychologists into a market that physicians and psychiatrists sought to preserve for themselves, McCready's injury is rendered "remote." The availability of the § 4 remedy to some person who claims its benefit is not a question of the specific intent of the conspirators. Here the remedy cannot reasonably be restricted to those competitors whom the conspirators hoped to eliminate from the market.15 McCready claims that she has been the victim of a concerted refusal to pay on the part of Blue Shield, motivated by a desire to deprive psychologists of the patronage of Blue Shield subscribers. Denying reimbursement to subscribers for the cost of treatment was the very means by which it is alleged that Blue Shield sought to achieve its illegal ends. The harm to McCready and her class was clearly foreseeable; indeed, it was a necessary step in effecting the ends of the alleged illegal conspiracy. Where the injury alleged is so integral an aspect of the conspiracy alleged, there can be no question but that the loss was precisely " 'the type of loss that the claimed violations . . . would be likely to cause.' " Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S., at 489, 97 S.Ct., at 697, quoting Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 125, 89 S.Ct. 1562, 1577, 23 L.Ed.2d 129 (1969). 15 Petitioners next argue that even if the § 4 remedy might be available to persons other than the competitors of the conspirators, it is not available to McCready because she was not an economic actor in the market that had been restrained. In petitioners' view, the proximate range of the violation is limited to the sector of the economy in which a violation of the type alleged would have its most direct anticompetitive effects. Here, petitioners contend that that market, for purposes of the alleged conspiracy, is the market in group health care plans. Thus, in petitioners' view, standing to redress the violation alleged in this case is limited to participants in that market—that is, to entities, such as McCready's employer, who were purchasers of group health plans, but not to McCready as a beneficiary of the Blue Shield plan.16 16 Petitioners misconstrue McCready's complaint. McCready does not allege a restraint in the market for group health plans. Her claim of injury is premised on a concerted refusal to reimburse under a plan that was, in fact, purchased and retained by her employer for her benefit, and that as a matter of contract construction and state law permitted reimbursement for the services of psychologists without any significant variation in the structure of the contractual relationship between her employer and Blue Shield.17 See n. 2, supra. As a consumer of psychotherapy services entitled to financial benefits under the Blue Shield plan, we think it clear that McCready was "within that area of the economy . . . endangered by [that] breakdown of competitive conditions" resulting from Blue Shield's selective refusal to reimburse. In re Multidistrict Vehicle Air Pollution M.D.L. No. 31, 481 F.2d 122, 129 (CA9 1973). 17 (2) 18 We turn finally to the manner in which the injury alleged reflects Congress' core concerns in prohibiting the antitrust defendants' course of conduct. Petitioners phrase their argument on this point in a manner that concedes McCready's participation in the market for psychotherapy services and rests instead on the notion that McCready's injury does not reflect the "anticompetitive" effect of the alleged boycott. They stress that McCready did not visit a psychiatrist whose fees were artificially inflated as a result of the competitive advantage he gained by Blue Shield's refusal to reimburse for the services of psychologists; she did not pay additional sums for the services of a physician to supervise and bill for the psychotherapy provided by her psychologist; and that there is no "claim that her psychologists' bills are higher than they would have been had the conspiracy not existed."18 In promoting this argument, petitioners rely heavily on language in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., supra. 19 In Brunswick, respondents were three bowling centers who complained that petitioner's acquisition of several financially troubled bowling centers violated § 7 of the Clayton Act by lessening competition or tending to create a monopoly. In seeking damages, "respondents attempted to show that had petitioner allowed the [acquired] centers to close, respondents' profits would have increased." Id., at 481, 97 S.Ct., at 693-94. The Court of Appeals endorsed the legal theory upon which respondents' claim was based, id., at 483, holding that "any loss 'causally linked' to 'the mere presence of the violator in the market' " was compensable under § 4, id., at 487, 97 S.Ct., at 696-97. We reversed, holding that the injury alleged by respondents was not of " 'the type that the statute was intended to forestall.' " Id., at 487-488, 97 S.Ct., at 696-697, quoting Wyandotte Transportation Co. v. United States, 389 U.S. 191, 202, 88 S.Ct. 379, 386, 19 L.Ed.2d 407 (1967). Indeed, the Court noted that respondents sought in damages "the profits they would have realized had competition been reduced." 429 U.S., at 488, 97 S.Ct., at 697 (emphasis added). 20 We can agree with petitioners' view of Brunswick as embracing the general principle that treble-damages recoveries should be linked to the procompetition policy of the antitrust laws. But petitioners seek to take Brunswick one significant step farther. In a passage upon which petitioners place much reliance, we stated: 21 "[F]or plaintiffs to recover treble damages on account of § 7 violations, they must prove more than injury causally linked to an illegal presence in the market. Plaintiffs must prove antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful. The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation. It should, in short, be 'the type of loss that the claimed violations . . . would be likely to cause.' Zenith Radio Corp. v. Hazeltine Research, 395 U.S., at 125 [89 S.Ct., at 1577]." Id., 429 U.S. at 489, 97 S.Ct., at 697-98 (emphasis in original; footnote omitted). 22 Relying on this language, petitioners reason that McCready can maintain no action under § 4 because her injury "did not reflect the anticompetitive effect" of the alleged violation. 23 Brunswick is not so limiting. Indeed, as we made clear in a footnote to the relied-upon passage, a § 4 plaintiff need not "prove an actual lessening of competition in order to recover. [C]ompetitors may be able to prove antitrust injury before they actually are driven from the market and competition is thereby lessened." Id., at 489, n. 14, 97 S.Ct., at 698 n. 14. Thus while an increase in price resulting from a dampening of competitive market forces is assuredly one type of injury for which § 4 potentially offers redress, see Reiter v. Sonotone Corp., 442 U.S. 330, 99 S.Ct. 2326, 60 L.Ed.2d 931 (1979), that is not the only form of injury remediable under § 4. We think it plain that McCready's injury was of a type that Congress sought to redress in providing a private remedy for violations of the antitrust laws. 24 McCready charges Blue Shield with a purposefully anticompetitive scheme. She seeks to recover as damages the sums lost to her as the consequence of Blue Shield's attempt to pursue that scheme.19 She alleges that Blue Shield sought to induce its subscribers into selecting psychiatrists over psychologists for the psychotherapeutic services they required,20 and that the heart of its scheme was the offer of a Hobson's choice to its subscribers. Those subscribers were compelled to choose between visiting a psychologist and forfeiting reimbursement, or receiving reimbursement by forgoing treatment by the practitioner of their choice. In the latter case, the antitrust injury would have been borne in the first instance by the competitors of the conspirators, and inevitably—though indirectly—by the customers of the competitors in the form of suppressed competition in the psychotherapy market; in the former case, as it happened, the injury was borne directly by the customers of the competitors. McCready did not yield to Blue Shield's coercive pressure, and bore Blue Shield's sanction in the form of an increase in the net cost of her psychologist's services. Although McCready was not a competitor of the conspirators, the injury she suffered was inextricably intertwined with the injury the conspirators sought to inflict on psychologists and the psychotherapy market. In light of the conspiracy here alleged we think that McCready's injury "flows from that which makes defendants' acts unlawful" within the meaning of Brunswick, and falls squarely within the area of congressional concern.21 III 25 Section 4 of the Clayton Act provides a remedy to "[a]ny person" injured "by reason of" anything prohibited in the antitrust laws. We are asked in this case to infer a limitation on the rule of recovery suggested by the plain language of § 4. But having reviewed our precedents and, more importantly, the policies of the antitrust laws, we are unable to identify any persuasive rationale upon which McCready might be denied redress under § 4 for the injury she claims. The judgment of the Court of Appeals is 26 Affirmed. 27 Justice REHNQUIST, with whom THE CHIEF JUSTICE and Justice O'CONNOR join, dissenting. 28 Respondent's alleged "antitrust injury" in this case arises from a health insurance coverage dispute with her insurer, petitioner Blue Shield of Virginia. Respondent's complaint is that Blue Shield reimburses its subscribers for treatment by psychiatrists, but not by psychologists unless their services are supervised and billed by treating physicians. Respondent was treated by a clinical psychologist, but when she submitted claims to Blue Shield, she was denied reimbursement. 29 Respondent alleged in her complaint that Blue Shield's refusal to reimburse her for the costs she incurred in obtaining the services of a psychologist furthered a conspiracy by petitioners "to exclude and boycott clinical psychologists from receiving compensation under" Blue Shield's plan. App. 55. Blue Shield's refusal-to-reimburse policy is alleged to constitute a form of economic pressure on McCready and other Blue Shield subscribers to obtain the services of psychiatrists rather than psychologists. By employing this economic pressure on Blue Shield subscribers, petitioners are alleged to have placed clinical psychologists at a competitive disadvantage with regard to psychiatrists in the market for insurance-reimbursed psychological services. 30 The Court concludes that McCready's inability to obtain reimbursement for the psychological services she actually obtained permits her to maintain an action to enforce the antitrust laws pursuant to § 4 of the Clayton Act. According to the Court, one who suffers economic loss as a necessary step in effecting the end of a conspiracy has "standing" to sue pursuant to § 4. Ante, at 479, 483-484. I disagree. 31 Section 4 of the Clayton Act authorizes suits for treble damages by "[a]ny person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws." 15 U.S.C. § 15. It is not enough, however, for a plaintiff merely to allege that the defendant violated the antitrust laws and that he was injured. Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 486-489, 97 S.Ct. 690, 696-98, 50 L.Ed.2d 701 (1977). See Hawaii v. Standard Oil Co., 405 U.S. 251, 263, n. 14, 92 S.Ct. 885, 891, n. 14, 31 L.Ed.2d 184 (1972). The injury suffered by the plaintiff must be of the type the antitrust laws were intended to forestall. Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., supra, 429 U.S., at 487-488, 97 S.Ct., at 696-97. 32 "Plaintiffs must prove antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful. The injury should reflect the anticompetitive effect either of the violation or of anticompetitive acts made possible by the violation. It should, in short, be 'the type of loss that the claimed violations . . . would be likely to cause.' " 429 U.S., at 489, 97 S.Ct., at 697 (citation omitted). 33 Although McCready alleges that she would have been reimbursed had it not been for the conspiracy, I do not think that she has made a sufficient allegation of "antitrust injury" within the meaning of Brunswick. 34 Standing alone, a refusal by an insurer to reimburse its insured does not constitute a violation of the Sherman Act. At most, such an action on the part of an insurer may amount to a breach of a contract or a violation of relevant state law regulating the insurance industry.1 According to the Court, however, what distinguishes this case from the typical insurance coverage dispute is either the purpose behind or the effect of Blue Shield's refusal to reimburse. If Blue Shield violated the antitrust laws by its nonreimbursement policy, it was only because that policy was used as a means of putting psychologists at a competitive disadvantage in relation to psychiatrists. 35 Two conceivable grounds therefore may be divined from the Court's opinion to support its conclusion that McCready has suffered "antitrust injury" when Blue Shield refused to reimburse her costs in obtaining the services of a psychologist. The first theory is that McCready may recover simply because petitioners' nonreimbursement policy was intended to put clinical psychologists at a competitive disadvantage. According to the Court, this must be so even if Blue Shield's refusal to reimburse her would be entirely legal under the antitrust laws in the absence of such a purpose to competitively injure third parties. Blue Shield's intent or purpose renders the discriminatory reimbursement policy illegal. Under this theory, it would seem to be irrelevant for the Court's purposes whether McCready obtained the services of a psychologist or a psychiatrist so long as the illegal intent is present and she suffered economic loss as a result.2 36 The second conceivable rationale is a flat rule that recovery is permitted by those persons who suffer economic loss as a necessary step in effecting a conspiracy to place third parties at a competitive disadvantage.3 Under this theory, McCready may recover merely by demonstrating that she was a "tool" of petitioners' effort to disable psychologists from competing with psychiatrists in the market for insurance-reimbursed psychological services. She may recover because she did not yield to the economic pressure imposed on her.4 The theory is that McCready may recover because her loss is linked to petitioners' efforts to enforce a "boycott" of third parties. 37 I believe that such reasoning is foreclosed by the Court's decision in Brunswick. In order to recover, a plaintiff must demonstrate that the nature of the injury he suffered is of the type that makes the challenged practice illegal. In Brunswick, the merger may well have violated § 7 of the Clayton Act in the abstract or even as to competitors not before the Court. Yet, we held that the plaintiffs in Brunswick could not recover because they did not suffer from the anticompetitive effects of the merger. We rejected the contention that it was sufficient to show merely that the defendant's merger violated § 7 and that there existed a causal link between that merger and an economic loss. 429 U.S., at 486-489, 97 S.Ct., at 696-98. Instead, the required showing is that the type of harm suffered by the plaintiff is that which makes the challenged practice illegal. Id., at 489, 97 S.Ct., at 697-98. 38 Therefore, McCready may not recover merely by showing that she has suffered an economic loss resulting from a practice the legality of which depends upon its effect on a third party. McCready must show that the challenged practice is illegal with regard to its effect upon her. But petitioners' policy is alleged to be illegal not by virtue of its effect upon Blue Shield's subscribers but because of its effect upon psychologists. McCready alleges no anticompetitive effect upon herself. She does not allege that the conspiracy has affected the availability of the psychological services she sought and actually obtained. Nor does she allege that the conspiracy affected the price of the treatment she received.5 She does not allege that her injury was caused by any reduction in competition between psychologists and psychiatrists, nor that it was the result of any success6 Blue Shield achieved in its "boycott" of psychologists. She seeks recovery solely on the basis that Blue Shield's reimbursement policy failed to alter her conduct in a fashion necessary to foreclose psychologists from obtaining the patronage of Blue Shield's subscribers. 39 If the important consideration is whether the challenged practice is illegal with regard to its effect on the plaintiff, then it would be irrelevant for the plaintiff's purposes that the conspiracy might also adversely affect competition on another level of the market. For example, a group of retailers may threaten to refuse to do business with those distributors that continue to do business with a disfavored retailer. If the distributors agreed to cooperate with the conspiring retailers, then the disfavored retailer would have an action against the agreeing distributors and the conspiring retailers. See, e.g., United States v. General Motors Corp., 384 U.S. 127, 86 S.Ct. 1321, 16 L.Ed.2d 415 (1966); Klor's, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207, 79 S.Ct. 705, 3 L.Ed.2d 741 (1959). I would think that a distributor who refused to go along with the retailers' conspiracy and thereby lost the conspiring retailers' business would also have an action against those retailers. Such an action would be based upon the conspirators' concerted refusal to deal with the distributor which itself would be unlawful under the antitrust laws. Such an action, unlike the instant case, would not depend upon the anticompetitive effect of the challenged practice upon a third party. The distributor would have an action not on the ground that he was caught in the middle of an attempted boycott of participants on another level of the market, but because he was boycotted. The boycott of the distributor puts him at a competitive disadvantage to those distributors who are unaffected by the retailers' conspiracy and to those distributors who agree to participate.7 40 McCready, however, does not allege that petitioners engaged in a concerted refusal to deal with her. As the Court is aware, ante, at 468-470, McCready has alleged that petitioners violated the antitrust laws by conspiring to exclude clinical psychologists from the coverage of Blue Shield plans, and that this conspiracy foreseeably injured her. The Court apparently concludes, however, that McCready has also sufficiently alleged that petitioners have engaged in a concerted refusal to deal with her, and that this is the gravamen of her antitrust complaint: "McCready alleges that she has been the victim of a concerted refusal by psychiatrists to reimburse through the Blue Shield plan." Ante, at 484, n.21. It may be that the Court today is merely holding that a boycottee has "standing" to sue under § 4. Were this the issue presented by this case, I have little doubt that the Court merely would have denied certiorari. 41 But McCready simply does not, and could not, claim standing as the target of a concerted refusal to deal. Neither Blue Shield nor the psychiatrists threatened to cease doing business with McCready if she obtained the services of a psychologist rather than a psychiatrist. McCready alleges only that under the Blue Shield policy she could not obtain reimbursement for services rendered by psychologists. If such a claim is sufficient to make out a concerted refusal to deal, then any consumer who could not obtain a product or service on the precise terms he desires could claim to be the victim of a "boycott." Most importantly, McCready alleges that Blue Shield's policy violates the antitrust laws only by virtue of its anticompetitive effect on psychologists. She does not allege that Blue Shield's policy is illegal in any way because of its effect on subscribers. 42 The Court, however, dismisses such concerns by stating in conclusory terms that "the injury [McCready] suffered was inextricably intertwined with the injury the conspirators sought to inflict on psychologists and the psychotherapy market." Ante, at 484. I trust that the Court is not holding that a plaintiff may escape dismissal of the complaint merely by alleging that he suffered an economic loss "inextricably intertwined" with an injury the defendants intended, but failed, to inflict upon a third party.8 Although the Court may view itself as successfully deciding this case on its peculiar facts, it has wholly failed to provide any sort of reasoned basis for its decision. Especially in the area of antitrust law, labels do not suffice when analysis is necessary. 43 I would reverse the judgment of the Court of Appeals because McCready has not alleged that she has suffered antitrust injury, but at best injury attributable to a breach of contract on the part of Blue Shield. 44 Justice STEVENS, dissenting. 45 Respondent is a consumer of psychotherapeutic services. The question is whether she has been injured in her "business or property by reason of anything forbidden in the antitrust laws."1 The alleged antitrust violation is an agreement between petitioners Neuropsychiatric Society of Virginia and Blue Shield that Blue Shield would refuse to reimburse subscribers for payments made to clinical psychologists for charges that were not billed through a physician. The objective of the alleged conspiracy was to induce subscribers to patronize psychiatrists instead of psychologists. 46 For purposes of decision, I assume that the alleged agreement is unlawful. In analyzing the sufficiency of respondent's damage claim, it is helpful first to consider the situation in which the conspiracy would have its maximum impact on the relevant market. Given their objective, petitioners' conspiracy would be most effective if they made it perfectly clear to subscribers that they would not be reimbursed if they consulted psychologists instead of psychiatrists. For without this information, a subscriber's choice between a psychologist and a psychiatrist would not be affected by the conspiracy. Thus, I first assume that the Blue Shield insurance policy did not cover services performed by psychologists and that subscribers as a class were fully aware of this exclusion. 47 On this assumption, a Blue Shield subscriber who is a potential consumer in the relevant market has at least three options. He may: (1) forgo treatment entirely; (2) go to a psychiatrist; or (3) go to a psychologist.2 If he exercises his first option, his illness may worsen but he will not have suffered any economic injury cognizable under the antitrust laws.3 If he exercises his second option, his property will not be diminished because Blue Shield will reimburse him for his payment to the psychiatrist. If he exercises his third option, his property will be diminished to the extent of his unreimbursed payment to the psychologist, but he will have received in exchange psychotherapeutic services that presumably were worth the payment.4 The fact that he voluntarily elected to spend money for services not covered by his insurance policy would have no greater legal significance than a similar voluntary decision by a person who was not a Blue Shield subscriber.5 It thus seems clear to me that whatever option the fully informed subscriber exercises, he would suffer no injury to his property by reason of the restriction of insurance coverage to psychotherapeutic services performed by psychiatrists. 48 This conclusion is reinforced by the fact that Blue Shield subscribers have the additional option of going to a psychologist while retaining their rights to reimbursement under the policy. According to respondent's complaint, Blue Shield did not refuse to reimburse all payments made by subscribers to psychologists, but only those payments not billed through a physician. Even if a fully informed subscriber's preference for psychologists over psychiatrists were protected by the antitrust laws, that preference was not denied by the antitrust violation alleged in this case.6 The Hobson's choice described by the Court, ante, at 483, simply does not fit this case. 49 The availability of this fourth option would seem to indicate that respondent, in fact, was not fully aware of the scope of her policy's coverage. If her lack of understanding was caused by fraud or deception, she should be able to recover in a common-law action. If the misunderstanding was her own fault, that circumstance should not provide a basis for an antitrust recovery that would not be available if she had been fully informed. 50 Nor is the deficiency in respondent's complaint cured if the assumption about the insurance coverage is reversed. Although her antitrust claim would be more credible if Blue Shield excluded coverage of services performed by psychologists, respondent alleged in the second count of her complaint that the insurance policy, properly construed under applicable principles of Virginia law, provided coverage for services performed by psychologists, but that Blue Shield nevertheless refused to reimburse her for the payments she made to her psychologist. If a subscriber does not suffer antitrust injury when the insurance policy excludes coverage of services performed by psychologists, it would be anomalous to conclude that the availability of a breach-of-contract claim would in any way enhance his standing. The right to recover under the federal antitrust laws cannot be derived from a right to recover under state law. 51 Because respondent's complaint discloses no basis for concluding that she has suffered an injury to her property by reason of the alleged antitrust violation, I respectfully dissent. 1 With petitioner Blue Shield of Southwestern Virginia. 2 Petitioners contend that the contract between the county and Blue Shield must be read to bar payments for the services of nonphysicians. Respondent counters that between 1962 and 1972 Blue Shield routinely reimbursed subscribers for psychotherapy provided by psychologists, and that this practice was revised in 1972 as a result of the alleged conspiracy. In addition, respondent notes that in 1973 the Virginia Legislature passed a "freedom of choice" statute, Va.Code § 38.1-824 (1981), that required Blue Shield to pay for services rendered by licensed psychologists. See Virginia Academy of Clinical Psychologists v. Blue Shield of Virginia, 624 F.2d 476, 478 (CA4 1980). She argues that Blue Shield's obligations must be read consistently with that statute, at least until that statute was held invalid as applied in Blue Cross of Virginia v. Commonwealth, 221 Va. 349, 269 S.E.2d 827 (1980). This case arises on a motion to dismiss. We therefore assume, as McCready has alleged, that but for the alleged conspiracy to deny payment, she would have been reimbursed by Blue Shield for the cost of her psychologist's services. 3 Apparently Blue Shield inadvertently paid one of McCready's claims. After the error was discovered, Blue Shield sought to obtain a refund from McCready for the amount paid. 649 F.2d 228, 230, n. 4 (1981). 4 A similar complaint was filed by the Virginia Academy of Clinical Psychologists (VACP) and its president against the same defendants. The District Court addressed the motions to dismiss filed in each of the cases in a single opinion. The court dismissed McCready's case—thus giving rise to the appellate decision at issue in this Court—but permitted the VACP case to proceed to trial. Following trial, the District Court entered judgment for the defendants, Virginia Academy of Clinical Psychologists v. Blue Shield of Virginia, 469 F.Supp. 552 (1979), but the Court of Appeals reversed with respect to defendant Blue Shield, 624 F.2d 476 (CA4 1980). The opinion of the Court of Appeals for the Fourth Circuit in the instant case states that the opinion in VACP "should be read in connection with" its own opinion. 649 F.2d, at 230. A brief recitation of the decision in the VACP case is thus helpful in understanding the precise nature of McCready's claim. In VACP, the Court of Appeals rejected the District Court's treatment of Blue Shield as a distinct entity for purposes of determining whether a conspiracy or agreement had been shown. 624 F.2d, at 479. The court found that "the Blue Shield Plans are combinations of physicians, operating under the direction and control of their physician members." Ibid. "Blue Shield Plans are not insurance companies, though they are, to a degree, insurers. Rather, they are generally characterized as prepaid health care plans, quantity purchasers of health care services. [I]n a real and legal sense, the Blue Shield Plans are agents of their member physicians." Id., at 480 (citations and footnote omitted). With respect to the question whether the alleged Blue Shield combination was "in restraint of trade," the Court of Appeals agreed with the District Court that the rule of reason was applicable, but held that the District Court had erred in finding no liability. The Court of Appeals observed that psychologists and psychiatrists compete in the psychotherapy market, and that the decisions of Blue Shield "necessarily dictate, to some extent," who will be chosen to provide psychotherapy. Id., at 485. Finding that Blue Shield's policy of denying reimbursement for the psychotherapeutic services of psychologists unless billed through physicians, was not merely a cost-containment device or simply "good medical practice," as claimed by Blue Shield, the court held that Blue Shield had violated the Sherman Act. Ibid. 5 That section provides, in pertinent part, that "[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal." 6 That section provides, in pertinent part: "Any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court . . . and shall recover three-fold the damages by him sustained, and the cost of suit, including a reasonable attorney's fee." 7 Petitioners have argued in this Court that under § 2 of the McCarran-Ferguson Act, 15 U.S.C. § 1012, their actions were exempt from the antitrust laws as part of the "business of insurance." In ruling on petitioners' motion to dismiss, the District Court concluded that respondent had adequately pleaded a boycott beyond the protection of the McCarran-Ferguson Act, 15 U.S.C. § 1013(b). Respondent points out that on a full factual record the issue was resolved against the petitioners in VACP, 624 F.2d, at 483-484. The Court of Appeals did not address this question in the present case, however, and we do not reach it here. 8 Addressing the "target area" limitation on antitrust standing recognized in several Courts of Appeals, see n. 14, infra, the court concluded that the policies underlying that limitation were not implicated by McCready's claim. 649 F.2d, at 231-232. The dissenting judge took a contrary view of the "target area" rule. He emphasized that McCready had not described her injury "as a design or goal of any antitrust violation," but "rather as a consequence thereof." Id., at 232. He viewed this as the determinative factor in the proper application of the "target area" test to the facts of this case: "In determining who has standing to sue, the courts must look at who the illegal act was aimed to injure. A bystander, who is not the intended victim of the antitrust violation but who is injured nonetheless, cannot sue under the antitrust laws. His injury is too remote." Id., at 233. In addition, the dissent argued that McCready was not within the sector of the economy "competitively endangered" by the alleged violation, agreeing with the District Court that "she operated in a market which was unrestrained so far as she was concerned." Id., at 234. Finally, the dissent reasoned: "The price of psychologists' services to her was not increased by any act of the defendants. The fact that her Blue Shield contract . . . would not reimburse her for those services had nothing to do with the price she paid for the services, which . . . were not artificially inflated by an antitrust violation. . . . ". . . There is not even a claim that her psychologists' bills are higher than they would have been had the conspiracy not existed." Id., at 235-236. 9 In a related context we commented that "[i]n the face of [the congressional antitrust] policy this Court should not add requirements to burden the private litigant beyond what is specifically set forth by Congress . . . ." Radovich v. National Football League, 352 U.S. 445, 454, 77 S.Ct. 390, 395, 1 L.Ed.2d 456 (1957). See also Radiant Burners, Inc. v. Peoples Gas Co., 364 U.S. 656, 659-660, 81 S.Ct. 365, 367-68, 5 L.Ed.2d 358 (1961) (per curiam ) (To state a claim under § 1 of the Sherman Act, "allegations adequate to show a violation and, in a private treble damage action, that plaintiff was damaged thereby are all the law requires"). 10 Permitting McCready to maintain this lawsuit will, of course, further certain basic objectives of the private enforcement scheme embodied in § 4. Only by requiring violators to disgorge the "fruits of their illegality" can the deterrent objectives of the antitrust laws be fully served. Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 494, 88 S.Ct. 2224, 2232, 20 L.Ed.2d 1231 (1968). See Pfizer Inc. v. India, 434 U.S. 308, 314, 98 S.Ct. 584, 588, 54 L.Ed.2d 563 (1978); Illinois Brick Co. v. Illinois, 431 U.S. 720, 746, 97 S.Ct. 2061, 2075, 52 L.Ed.2d 707 (1977). But in addition to allowing Blue Shield to retain a palpable profit as a result of its unlawful plan, denying standing to McCready and the class she represents would also result in the denial of compensation for injuries resulting from unlawful conduct. 11 If there is a subordinate theme to our opinions in Hawaii and Illinois Brick, it is that the feasibility and consequences of implementing particular damages theories may, in certain limited circumstances, be considered in determining who is entitled to prosecute an action brought under § 4. Where consistent with the broader remedial purposes of the antitrust laws, we have sought to avoid burdening § 4 actions with damages issues giving rise to the need for "massive evidence and complicated theories," where the consequence would be to discourage vigorous enforcement of the antitrust laws by private suits. Hanover Shoe, Inc. v. United Shoe Machinery Corp., supra, at 493, 88 S.Ct., at 2231. Thus we recognized that the task of disentangling overlapping damages claims is not lightly to be imposed upon potential antitrust litigants, or upon the judicial system. See Hawaii v. Standard Oil Co., 405 U.S. 251, 264, 92 S.Ct. 885, 892, 31 L.Ed.2d 184 (1972); Illinois Brick Co. v. Illinois, supra, 431 U.S., at 741-742, 97 S.Ct., at 2072-2073. In addition, while "[d]ifficulty of ascertainment [should not be] confused with right of recovery," Bigelow v. RKO Radio Pictures, Inc., 327 U.S. 251, 265, 66 S.Ct. 574, 580, 90 L.Ed. 652 (1946), § 4 plainly focuses on tangible economic injury. It may therefore be appropriate to consider whether a claim rests at bottom on some abstract conception or speculative measure of harm. See Hawaii v. Standard Oil Co., supra, 405 U.S., at 262-263, n. 14, 92 S.Ct., at 891-892, n. 14. But like the policy against duplicative recoveries, our cautious approach to speculative, abstract, or impractical damages theories has no application to McCready's suit. The nature of her injury is easily stated: As the result of an unlawful boycott, Blue Shield failed to pay the cost she incurred for the services of a psychologist. Her damages were fixed by the plan contract and, as the Court of Appeals observed, they could be "ascertained to the penny." 649 F.2d, at 231. 12 We addressed two issues of "remoteness" in Perkins v. Standard Oil Co., 395 U.S. 642, 89 S.Ct. 1871, 23 L.Ed.2d 599 (1969). That case involved an alleged violation of § 2 of the Clayton Act, as amended by the Robinson-Patman Act, 15 U.S.C. § 13. Focusing on the substantive terms of § 2, we found no warrant in its "language or purpose" to engraft an "artificial" limitation on the reach of the remedy to bar what the court below had termed a "fourth level" injury. 395 U.S., at 648, 89 S.Ct., at 1874. We also rejected the claim that one form of damages claimed by the defendant was not the proximate result of the alleged violation. Id., at 649, 89 S.Ct., at 1875. The Courts of Appeals have developed a more substantial jurisprudence on the subject of "remoteness," formulating various "tests" as aids in analysis. Among the tests employed by the lower courts are those that focus on the "directness" of the injury, e.g., Loeb v. Eastman Kodak Co., 183 F. 704, 709 (CA3 1910); Productive Inventions, Inc. v. Trico Products Corp., 224 F.2d 678 (CA2 1955); Volasco Products Co. v. Lloyd A. Fry Roofing Co., 308 F.2d 383 (CA6 1962); on its foreseeability, e.g., In re Western Liquid Asphalt Cases, 487 F.2d 191, 199 (CA9 1973); Twentieth Century Fox Film Corp. v. Goldwyn, 328 F.2d 190, 220 (CA9 1964); or on whether the injury is "arguably . . . within the zone of interests protected by the [antitrust laws]," e.g., Malamud v. Sinclair Oil Corp., 521 F.2d 1142, 1152 (CA6 1975). See also n. 14, infra ("target area" test). The Third Circuit has concluded that "§ 4 standing analysis is essentially a balancing test comprised of many constant and variable factors and that there is no talismanic test capable of resolving all § 4 standing problems." Bravman v. Basset Furniture Industries, Inc., 552 F.2d 90, 99 (1977). The Third Circuit has thus rejected the definitional approach, opting instead for an analysis of the "factual matrix" presented by each case. Ibid. We have no occasion here to evaluate the relative utility of any of these possibly conflicting approaches toward the problem of remote antitrust injury. 13 The traditional principle of proximate cause suggests the use of words such as "remote," "tenuous," "fortuitous," "incidental," or "consequential" to describe those injuries that will find no remedy at law. See, e.g., South Carolina Council of Milk Producers, Inc. v. Newton, 360 F.2d 414, 419 (CA4 1966). And the use of such terms only emphasizes that the principle of proximate cause is hardly a rigorous analytic tool. See, e.g., Palsgraf v. Long Island R. Co., 248 N.Y. 339, 162 N.E. 99 (1928); id., at 351-352, 162 N.E., at 103 (Andrews, J., dissenting) ("What is a cause in a legal sense, still more what is a proximate cause, depend in each case upon many considerations. . . . What we do mean by the word 'proximate' is, that because of convenience, of public policy, of a rough sense of justice, the law arbitrarily declines to trace a series of events beyond a certain point"). It bears affirming that in identifying the limits of an explicit statutory remedy, legislative intent is the controlling consideration. Cf. Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 377-378, 102 S.Ct. 1825, 1838-1839, 72 L.Ed.2d 182 (1982); Middlesex County Sewerage Authority v. National Sea Clammers Assn., 453 U.S. 1, 13, 101 S.Ct. 2615, 2622-23, 69 L.Ed.2d 435 (1981); Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 15-16, 100 S.Ct. 242, 245, 62 L.Ed.2d 146 (1979). 14 In so arguing, petitioners advert to the "target area" test of antitrust standing that prevails in the Courts of Appeals for the First, Second, and Fifth Circuits. See, e.g., Pan-Islamic Trade Corp. v. Exxon Corp., 632 F.2d 539, 546 (CA5 1980); Engine Specialties, Inc. v. Bombardier Ltd., 605 F.2d 1, 18-19 (CA1 1979); Calderone Enterprises Corp. v. United Artists Theatre Circuit, Inc., 454 F.2d 1292 (CA2 1971). Petitioners place special reliance on the following frequently cited formulation of the "target area" principle: "[I]n order to have 'standing' to sue for treble damages under § 4 of the Clayton Act, a person must be within the 'target area' of the alleged antitrust conspiracy, i.e., a person against whom the conspiracy was aimed, such as a competitor of the persons sued. Accordingly we have drawn a line excluding those who have suffered economic damage by virtue of their relationships with 'targets' or with participants in an alleged antitrust conspiracy, rather than being 'targets' themselves." Id., at 1295. 15 Nor does the "target area" test applied by the Courts of Appeals " 'imply that it must have been a purpose of the conspirators to injure the particular individual claiming damages.' " See Schwimmer v. Sony Corp. of America, 637 F.2d 41, 47-48 (CA2 1980), quoting Twentieth Century Fox Film Corp. v. Goldwyn, 328 F.2d, at 220. 16 Petitioners borrow selectively from Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701 (1977), in arguing that McCready's § 4 claim is "unrelated to any reduction in competition caused by the alleged boycott," because the injury she alleges "is the result of the terms of her insurance contract, and not the result of a reduction in competition." Brief for Petitioners 16. Extracting additional language from Brunswick, they argue that "McCready would have suffered the identical 'loss'—but no compensable 'injury'—as long as her employer, which acted independently in an unrestrained market, continued to purchase a group insurance contract that did not cover the services of clinical psychologists." Brief for Petitioners 16-17 (footnote omitted). 17 Nor do we think that her employer's decision to retain Blue Shield coverage despite its continued failure to reimburse for the services of a psychologist—or indeed, her employer's unexercised option to terminate that relationship—is an intervening cause of McCready's injury. Although her employer's decision to purchase the Blue Shield plan for her benefit was in some sense a factor that contributed independently to McCready's injury, her coverage under the Blue Shield plan may, at this stage of the litigation, properly be accepted as a given, and the proper focus in evaluating her entitlement to raise a § 4 damages claim is on Blue Shield's change in the terms of the plan to link reimbursement to a subscriber's choice of one group of psychotherapists over another. 18 649 F.2d, at 236 (Widener, J., dissenting). 19 Brunswick held that a claim of injury arising from the preservation or enhancement of competition is a claim "inimical to the purposes of [the antitrust] laws," 429 U.S., at 488, 97 S.Ct., at 697. Most obviously, McCready's claim is quite unlike the claim asserted by the plaintiff in Brunswick for she does not seek to label increased competition as a harm to her. Nevertheless, we agree with petitioners that the relationship between the claimed injury and that which is unlawful in the defendant's conduct, as analyzed in Brunswick, is one factor to be considered in determining the redressability of a particular form of injury under § 4. 20 Or at the least, Blue Shield sought to compel McCready to employ the services of a physician in addition to those of a psychologist. 21 Justice REHNQUIST, dissenting, is of course correct in asserting that the "injury suffered by the plaintiff must be of the type the antitrust laws were intended to forestall," post, at 486. But Justice REHNQUIST's dissent takes an unrealistically narrow view of those injuries with which the antitrust laws might be concerned, and offers not the slightest hint—beyond sheer ipse dixit —to help in determining what kinds of injury are not amenable to § 4 redress. For example, the dissent acknowledges that "a distributor who refused to go along with the retailers' conspiracy [to injure a disfavored retailer] and thereby lost the conspiring retailers' business would . . . have an action against those retailers," post, at 490. The dissent characterizes this circumstance as a "concerted refusal to deal," and is thus willing to acknowledge the existence of compensable injury. But the dissent's is not the only pattern of concerted refusals to deal. If a group of psychiatrists conspired to boycott a bank until the bank ceased making loans to psychologists, the bank would no doubt be able to recover the injuries suffered as a consequence of the psychiatrists' actions. And plainly, in evaluating the reasonableness under the antitrust laws of the psychiatrists' conduct, we would be concerned with its effects not only on the business of banking, but also on the business of the psychologists against whom that secondary boycott was directed. McCready and the banker and the distributor are in many respects similarly situated. McCready alleges that she has been the victim of a concerted refusal by psychiatrists to reimburse through the Blue Shield plan. Because McCready is a consumer, rather than some other type of market participant, the dissent finds itself unwilling to acknowledge that she might have suffered a form of injury of significance under the antitrust laws. But under the circumstances of this case, McCready's participation in the market for psychotherapeutic services provides precisely that significance. 1 In addition to the antitrust claim, McCready's complaint asserts a claim for breach of contract under the principles of pendent jurisdiction. App. 57-58. She also alleges that Blue Shield's policy contravened state law. Id., at 55-56. 2 The Court explains that those subscribers, such as McCready, who did not yield to Blue Shield's coercive pressures suffer from Blue Shield's sanctions by way of increased costs in obtaining the services of a psychologist. Those subscribers who did yield to Blue Shield's pressure suffer antitrust injury indirectly because of suppressed competition in the psychotherapy market. Ante, at 483-484. I do not understand the Court to conclude that Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977), would not bar recovery by a subscriber, as opposed to a psychologist, in the latter situation. 3 The Court suggests a third theory—that McCready has standing herself as a target of a concerted refusal to deal. See ante, at 484, n. 21; infra, at 490-491 4 In order to recover under this theory, it would seem that respondent must prove at trial that she actually refused to yield to the economic pressure created by Blue Shield's reimbursement policy. If she decided to obtain the services of a psychologist rather than a psychiatrist without knowing of Blue Shield's policy, it cannot be said that her "injury" was proximately related to petitioners' alleged anticompetitive conduct. If she discovered the policy only after she sought reimbursement, then it cannot be said that Blue Shield's policy had any effect on McCready's conduct as a consumer in the market for psychotherapeutic services. This, of course, is not to say that a person in all circumstances must have knowledge of a defendant's anticompetitive activities before one may challenge that activity. One may not be a victim of economic pressure, however, if one acted obliviously to that pressure. 5 By excluding psychologists from the market, psychiatrists may well be able to increase their charges for psychotherapeutic services, which in turn, may raise the insurance rates charged by Blue Shield. McCready, however, alleges no such injury to herself on this theory. 6 Because McCready obtained the services of a psychologist, it cannot be said that the psychologists were injured by the economic pressure Blue Shield placed on McCready and the class of subscribers she represents. See ante, at 475 7 As pointed out by the Court, a concerted refusal to deal may take many forms. Ante, at 484, n. 21. I would agree that the bank could sue in the Court's hypothetical because, as conceded by the Court, the bank's ability to compete with other banks would be adversely affected. By contrast, my disagreement with the Court is that it permits McCready to sue solely because of an injury to a level of the market in which she does not participate. Moreover, McCready does not allege that petitioners' conspiracy adversely affected competition between psychologists and psychiatrists in such a manner as to adversely affect the price or supply of psychotherapeutic services available to her as a consumer. Thus, McCready's case is clearly distinguishable from that of the bank's in the Court's hypothetical. 8 If McCready's injury were truly "inextricably intertwined" with any injury actually suffered by the psychologists, the risk of duplicative recovery and the practical problems inherent in distinguishing the loss suffered by her from the loss suffered by the psychologists may mean that either subscribers or psychologists, but not both, may recover. See Illinois Brick Co. v. Illinois, 431 U.S. 720, 97 S.Ct. 2061, 52 L.Ed.2d 707 (1977). 1 "Any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws may sue therefor in any district court of the United States in the district in which the defendant resides or is found or has an agent, without respect to the amount in controversy, and shall recover threefold the damages by him sustained, and the cost of suit, including a reasonable attorney's fee." 15 U.S.C. § 15. 2 In Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 97 S.Ct. 690, 50 L.Ed.2d 701, we held that antitrust injury was limited to " 'the type of loss that the claimed violations . . . would be likely to cause.' " Id., at 489, 97 S.Ct., at 697-698 (quoting Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 125, 89 S.Ct. 1562, 1577, 23 L.Ed.2d 129). I would expect that the alleged violation in this case would be most likely to cause knowledgeable members of the class of potential consumers of psychotherapeutic services to exercise either the first or the second option. It is fair to assume that the third situation—the one in which respondent finds herself—would be "unlikely" to result. 3 The subscriber may have to undergo more extensive treatment later if he forgoes treatment now and his illness worsens. Any consequential economic injury, however, would no more constitute antitrust injury than the economic injury suffered by a consumer who decides to forgo a purchase on the ground that the price of the goods or services was fixed at an artificially high level. 4 If treatment by a psychiatrist and treatment by a psychologist were fungible, then a subscriber who exercised this third option effectively would be paying twice for the psychotherapeutic service, once to the insurer in premiums and once to the psychologist in an unreimbursable payment. But the subscriber's exercise of this option presumably indicates that treatment by a psychologist is more valuable to him than treatment by a psychiatrist. If that be true, the subscriber is in the same situation as any policyholder who desires a service for which he has not purchased insurance. 5 If the subscriber would purchase a service that was covered by the Blue Shield policy, such as a surgical operation, then he would be reimbursed by Blue Shield for that payment. If respondent's antitrust claim is that petitioners have engaged in an unlawful boycott, it therefore is manifest that respondent is not the boycottee. For petitioners have not refused to deal with respondent—they offer her the same coverage as any other subscriber or potential subscriber. 6 Presumably, the charge (if any) of the referring physician would be reimbursable under the policy. In any event, the complaint does not claim damages based on any such unreimbursed charge.

78

457 U.S. 537 102 S.Ct. 2579 73 L.Ed.2d 202 UNITED STATES, Petitionerv.Raymond Eugene JOHNSON. No. 80-1608. Argued Feb. 24, 1982. Decided June 21, 1982. Syllabus Payton v. New York, 445 U.S. 573, 100 S.Ct. 1371, 63 L.Ed.2d 639, held that the Fourth Amendment prohibits the police from making a warrantless and nonconsensual entry into a suspect's home to make a routine felony arrest. Before Payton was decided, respondent was arrested on a federal charge by Secret Service agents who had entered his home without an arrest warrant. Subsequently, the Federal District Court denied respondent's pretrial motion to suppress incriminating statements he made after his arrest. This evidence was admitted at his trial and he was convicted. While his case was still pending on direct appeal, Payton was decided. On the strength of Payton, the Court of Appeals reversed the conviction, holding that Payton applied retroactively. Held : A decision of this Court construing the Fourth Amendment is to be applied retroactively to all convictions that were not yet final at the time the decision was rendered, except where a case would be clearly controlled by existing retroactivity precedents. Hence, Payton is to be applied retroactively to respondent's case. Pp. 542-563. (a) Respondent's case does not present a retrospectivity problem clearly controlled by existing precedent. Where a decision of this Court merely has applied settled principles to a new set of facts, it has been a foregone conclusion that the rule of the later case applies in earlier cases. Conversely, where the Court has declared a rule of criminal procedure to be "a clear break with the past," it almost invariably has found the new principle nonretroactive. Also, this Court has recognized full retroactivity as a necessary adjunct to a ruling that a trial court lacked authority to convict or punish the defendant in the first place. Respondent's case does not fit any of these categories, as Payton did not apply settled precedent to a new set of facts, did not announce an entirely new and unanticipated principle of law, and did not hold either that the trial court lacked authority to convict Payton or that the Fourth Amendment immunized his conduct from punishment. Pp. 548-554 (b) The retroactivity question presented here is fairly resolved by applying the Payton rule to all cases still pending on direct appeal at the time Payton was decided. To do so (1) provides a principle of decisionmaking consonant with this Court's original understanding in Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601, and Tehan v. United States ex rel. Shott, 382 U.S. 406, 86 S.Ct. 459, 15 L.Ed.2d 453, that all newly declared constitutional rules of criminal procedure would apply retrospectively at least to convictions not yet final when the rule was established; (2) comports with this Court's judicial responsibility "to do justice to each litigant on the merits of his own case," Desist v. United States, 394 U.S. 244, 259, 89 S.Ct. 1030, 1039, 22 L.Ed.2d 248 (Harlan, J., dissenting), and to "resolve all cases before us on direct review in light of our best understanding of governing constitutional principles," Mackey v. United States, 401 U.S. 667, 679, 91 S.Ct. 1160, 1173, 28 L.Ed.2d 404 (separate opinion of Harlan, J.); and (3) furthers the goal of treating similarly situated defendants similarly. Pp. 554-556. (c) There is no merit to the Government's arguments, based on United States v. Peltier, 422 U.S. 531, 95 S.Ct. 2313, 45 L.Ed.2d 374, against adoption of the above approach to the retroactivity question in this case. Pp. 557-562 9 Cir., 626 F.2d 753, affirmed. Elliott Schulder, Washington, D. C., for petitioner. John F. Walter, Los Angeles, Cal., for respondent. Justice BLACKMUN delivered the opinion of the Court. 1 In Payton v. New York, 445 U.S. 573, 100 S.Ct. 1371, 63 L.Ed.2d 639 (1980), this Court held that the Fourth Amendment1 prohibits the police from making a warrantless and nonconsensual entry into a suspect's home to make a routine felony arrest. The question before us in the present case is whether the rule announced in Payton applies to an arrest that took place before Payton was decided. 2 * Special Agents Hemenway and Pickering of the United States Secret Service suspected respondent Raymond Eugene Johnson and his codefendant, Oscar Joseph Dodd, of attempting to negotiate a misdelivered United States Treasury check.2 Proceeding without an arrest warrant, on May 5, 1977, the two agents went to respondent's Los Angeles home and waited outside. Shortly thereafter, respondent and his wife arrived and entered the house. 3 The agents drew their weapons, approached the doorway and knocked, identifying themselves by fictitious names. When respondent opened the door, he saw the two agents with their guns drawn and their badges raised. Respondent permitted the agents to enter the house. While one agent stood with respondent in the living room, the other searched the premises. The agents then advised respondent of his constitutional rights and interrogated him. When respondent revealed his involvement in the taking of the misdelivered check, the agents formally arrested him. Respondent later signed a written statement admitting his involvement with the check. 4 Before trial, respondent sought to suppress his oral and written statements as fruits of an unlawful arrest not supported by probable cause. The United States District Court for the Central District of California found respondent's arrest to be proper and admitted the evidence. App. 7. A jury then convicted respondent of aiding and abetting obstruction of correspondence, in violation of 18 U.S.C. §§ 2 and 1702.3 The imposition of respondent's sentence was suspended in favor of five years' probation. 5 By an unreported opinion filed December 19, 1978, the United States Court of Appeals for the Ninth Circuit affirmed the judgment of conviction. Acknowledging that "[i]t certainly would have been preferable had the agents obtained a warrant" for respondent's arrest before entering his residence, the court nonetheless ruled that "if probable cause exists for the arrest, [respondent's] constitutional rights were not violated by the warrantless arrest, even though there may have been time [for the agents] to have obtained a warrant for his arrest." App. to Pet. for Cert. 26a-27a. 6 On April 15, 1980, while respondent's petition for rehearing was still pending before the Ninth Circuit, this Court decided Payton v. New York, supra.4 On September 2, 1980, the Ninth Circuit granted respondent's petition for rehearing, withdrew its prior opinion, and on the strength of Payton, now reversed the judgment of conviction. 626 F.2d 753. "In light of the strong language by the Court in Payton emphasizing the special protection the Constitution affords to individuals within their homes," the Court of Appeals held that "the warrantless arrest of Johnson, while he stood within his home, after having opened the door in response to false identification by the agents, constituted a violation of his Fourth Amendment rights." Id., at 757. The Government petitioned for rehearing, arguing that the principles of Payton should not apply retroactively to an arrest that had occurred before Payton was decided. The Court of Appeals disagreed, denied the petition for rehearing, and amended its opinion to clarify that Payton did apply retroactively. App. to Pet. for Cert. 12a.5 7 The Government sought review in this Court. We granted certiorari to consider the retrospective effect, if any, of the Fourth Amendment rule announced in Payton. 454 U.S. 814, 102 S.Ct. 89, 70 L.Ed.2d 82 (1981).6 II 8 "[T]he federal constitution has no voice upon the subject" of retrospectivity. Great Northern R. Co. v. Sunburst Oil & Refining Co., 287 U.S. 358, 364, 53 S.Ct. 145, 148, 77 L.Ed. 360 (1932). Before 1965, when this Court decided Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601, "both the common law and our own decisions recognized a general rule of retrospective effect for the constitutional decisions of this Court . . . subject to [certain] limited exceptions." Robinson v. Neil, 409 U.S. 505, 507, 93 S.Ct. 876, 877, 35 L.Ed.2d 29 (1973), citing Norton v. Shelby County, 118 U.S. 425, 442, 6 S.Ct. 1121, 1125, 30 L.Ed. 178 (1886), and Chicot County Drainage Dist. v. Baxter State Bank, 308 U.S. 371, 60 S.Ct. 317, 84 L.Ed. 329 (1940).7 9 In Linkletter, however, the Court concluded "that the Constitution neither prohibits nor requires [that] retrospective effect" be given to any "new" constitutional rule. 381 U.S., at 629, 85 S.Ct., at 1737. Since Linkletter, the Court's announcement of a constitutional rule in the realm of criminal procedure has frequently been followed by a separate decision explaining whether, and to what extent, that rule applies to past, pending, and future cases. See generally Beytagh, Ten Years of Non-Retroactivity: A Critique and a Proposal, 61 Va.L.Rev. 1557 (1975). 10 Linkletter itself addressed the question whether the Fourth Amendment exclusionary rule of Mapp v. Ohio, 367 U.S. 643, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961), should apply to state convictions that had become final before Mapp was decided.8 At the outset, the Linkletter Court noted that cases still pending on direct review when Mapp was handed down had already received the benefit of Mapp's rule. See 381 U.S., at 622, n. 4, 85 S.Ct., at 1734, n. 4, citing Ker v. California, 374 U.S. 23, 83 S.Ct. 1623, 10 L.Ed.2d 726 (1963); Fahy v. Connecticut, 375 U.S. 85, 84 S.Ct. 229, 11 L.Ed.2d 171 (1963); and Stoner v. California, 376 U.S. 483 (1964). This limited retrospective application of Mapp was consistent with the common-law rule, recognized in both civil and criminal litigation, "that a change in law will be given effect while a case is on direct review." 381 U.S., at 627, 85 S.Ct., at 1736, citing United States v. Schooner Peggy, 1 Cranch 103, 2 L.Ed. 49 (1801). 11 To determine whether a particular ruling should also extend to cases that were already final, Linkletter directed courts to "weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation." 381 U.S., at 629, 85 S.Ct., at 1738. Employing that test, the Court concluded that the Mapp rule should not apply to convictions that had become final before Mapp was decided. 12 The following Term, in Tehan v. United States ex rel. Shott, 382 U.S. 406, 86 S.Ct. 459, 15 L.Ed.2d 453 (1966), the Court applied Linkletter's analysis to hold the Fifth Amendment rule of Griffin v. California, 380 U.S. 609, 85 S.Ct. 1229, 14 L.Ed.2d 106 (1965) (barring comment on a state defendant's failure to testify), nonretroactive to judgments of conviction made final before Griffin was decided. The Court again found no "question of the applicability of the Griffin rule to cases still pending on direct review at the time it was announced." 382 U.S., at 409, n. 3, 86 S.Ct., at 461, n. 3, citing O'Connor v. Ohio, 382 U.S. 286, 86 S.Ct. 445, 15 L.Ed.2d 337 (1965). Thus, after Linkletter and Shott, it appeared that all newly declared constitutional rules of criminal procedure would apply retrospectively at least to judgments of conviction not yet final when the rule was established. 13 In Johnson v. New Jersey, 384 U.S. 719, 86 S.Ct. 1772, 16 L.Ed.2d 882 (1966), and Stovall v. Denno, 388 U.S. 293, 87 S.Ct. 1967, 18 L.Ed.2d 1199 (1967), however, the Court departed from that basic principle. Those cases held that, in the interest of justice, the Court may balance three factors to determine whether a "new" constitutional rule should be retrospectively or prospectively applied: "(a) the purpose to be served by the new standards, (b) the extent of the reliance by law enforcement authorities on the old standards, and (c) the effect on the administration of justice of a retroactive application of the new standards." Id., at 297, 87 S.Ct., at 1970. See alsoJohnson v. New Jersey, 384 U.S., at 728, 86 S.Ct., at 1778. Because the outcome of that balancing process might call for different degrees of retroactivity in different cases, the Court concluded that "no distinction is justified between convictions now final . . . and convictions at various stages of trial and direct review." Stovall v. Denno, 388 U.S., at 300, 87 S.Ct., at 1971. See Johnson v. New Jersey, 384 U.S., at 732, 86 S.Ct., at 1780. 14 Because the balance of the three Stovall factors inevitably has shifted from case to case, it is hardly surprising that, for some, "the subsequent course of Linkletter became almost as difficult to follow as the tracks made by a beast of prey in search of its intended victim." Mackey v. United States, 401 U.S. 667, 676, 91 S.Ct. 1160, 1172, 28 L.Ed.2d 404 (1971) (separate opinion of Harlan, J.). At one extreme, the Court has regularly given complete retroactive effect to new constitutional rules whose major purpose "is to overcome an aspect of the criminal trial that substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts in past trials." Williams v. United States, 401 U.S. 646, 653, 91 S.Ct. 1148, 1152, 28 L.Ed.2d 388 (1971) (plurality opinion). See also id., at 653, n. 6, 91 S.Ct., at 1152, n. 6; Brown v. Louisiana, 447 U.S. 323, 328-330, 100 S.Ct. 2214, 2219-2220, 65 L.Ed.2d 159 (1980) (plurality opinion); Hankerson v. North Carolina, 432 U.S. 233, 243, 97 S.Ct. 2339, 2345, 53 L.Ed.2d 306 (1977); Gosa v. Mayden, 413 U.S. 665, 679, 93 S.Ct. 2926, 2935, 37 L.Ed.2d 873 (1973) (plurality opinion); Ivan V. v. City of New York, 407 U.S. 203, 205, 92 S.Ct. 1951, 1952, 32 L.Ed.2d 659 (1972). 15 At the other extreme, the Court has applied some standards only to future cases, denying the benefit of the new rule even to the parties before the Court. See, e.g., Morrissey v. Brewer, 408 U.S. 471, 490, 92 S.Ct. 2593, 2604, 33 L.Ed.2d 484 (1972) (establishing basic requirements applicable only to "future revocations of parole"). Cf. Johnson v. New Jersey, 384 U.S., at 733, 86 S.Ct., at 1780, citing England v. Louisiana State Board of Medical Examiners, 375 U.S. 411, 84 S.Ct. 461, 11 L.Ed.2d 440 (1964), and James v. United States, 366 U.S. 213, 81 S.Ct. 1052, 6 L.Ed.2d 246 (1961). As an intermediate position, the Court has applied a change in the law to all future litigants, but retroactively only to the parties at bar. See, e.g., Stovall v. Denno, 388 U.S., at 301, 87 S.Ct., at 1972; DeStefano v. Woods, 392 U.S. 631, 633, 88 S.Ct. 2093, 2095, 20 L.Ed.2d 1308 (1968); Adams v. Illinois, 405 U.S. 278, 284-285, 92 S.Ct. 916, 920, 31 L.Ed.2d 202 (1972) (plurality opinion); Michigan v. Payne, 412 U.S. 47, 93 S.Ct. 1966, 36 L.Ed.2d 736 (1973). 16 In a consistent stream of separate opinions since Linkletter, Members of this Court have argued against selective awards of retroactivity. Those opinions uniformly have asserted that, at a minimum, all defendants whose cases were still pending on direct appeal at the time of the law-changing decision should be entitled to invoke the new rule.9 In Desist v. United States, 394 U.S. 244, 256, 89 S.Ct. 1030, 1037, 22 L.Ed.2d 248 (1969) (dissenting opinion), and Mackey v. United States, 401 U.S., at 675, 91 S.Ct., at 1164 (separate opinion), Justice Harlan presented a comprehensive analysis in support of that principle. In his view, failure to apply a newly-declared constitutional rule at least to cases pending on direct review at the time of the decision violated three norms of constitutional adjudication. 17 First, Justice Harlan argued, the Court's "ambulatory retroactivity doctrine," id., at 681, 91 S.Ct., at 1174, conflicts with the norm of principled decisionmaking. "Some members of the Court, and I have come to regret that I was among them, initially grasped this doctrine as a way of limiting the reach of decisions that seemed to them fundamentally unsound. Others rationalized this resort to prospectivity as a 'technique' that provided an 'impetus . . . for the implementation of long overdue reforms, which otherwise could not be practicably effected.' " Id., at 676, 91 S.Ct., at 1172, citing Jenkins v. Delaware, 395 U.S. 213, 218, 89 S.Ct. 1677, 1680, 23 L.Ed.2d 253 (1969). "The upshot of this confluence of viewpoints," 401 U.S., at 676, 91 S.Ct., at 1172, was that the coalitions favoring nonretroactivity had realigned from case to case, inevitably generating a welter of "incompatible rules and inconsistent principles," Desist v. United States, 394 U.S., at 258, 89 S.Ct., at 1038. See also Michigan v. Payne, 412 U.S., at 61, 93 S.Ct., at 1973 (MARSHALL, J., dissenting) ("principled adjudication requires the Court to abandon the charade of carefully balancing countervailing considerations when deciding the question of retroactivity"). 18 Second, Justice Harlan found it difficult to accept the notion that the Court, as a judicial body, could apply a " 'new' constitutional rule entirely prospectively, while making an exception only for the particular litigant whose case was chosen as the vehicle for establishing that rule." Desist v. United States, 394 U.S., at 258, 89 S.Ct., at 1038 (dissenting opinion). A legislature makes its new rules "wholly or partially retroactive or only prospective as it deems wise." Mackey v. United States, 401 U.S., at 677, 91 S.Ct., at 1172 (Harlan, J., dissenting). This Court, however, 19 "announce[s] new constitutional rules . . . only as a correlative of our dual duty to decide those cases over which we have jurisdiction and to apply the Federal Constitution as one source of the matrix of governing legal rules. . . . Simply fishing one case from the stream of appellate review, using it as a vehicle for pronouncing new constitutional standards, and then permitting a stream of similar cases subsequently to flow by unaffected by that new rule constitute an indefensible departure from this model of judicial review." Id., at 678-679, 91 S.Ct., at 1172-1173. 20 Third, Justice Harlan asserted that the Court's selective application of new constitutional rules departed from the principle of treating similarly situated defendants similarly:10 21 "[W]hen another similarly situated defendant comes before us, we must grant the same relief or give a principled reason for acting differently. We depart from this basic judicial tradition when we simply pick and choose from among similarly situated defendants those who alone will receive the benefit of a 'new' rule of constitutional law." Desist v. United States, 394 U.S., at 258-259, 89 S.Ct., at 1038-1039 (dissenting opinion). 22 Justice Harlan suggested one simple rule to satisfy all three of his concerns. "I have concluded that Linkletter was right in insisting that all 'new' rules of constitutional law must, at a minimum, be applied to all those cases which are still subject to direct review by this Court at the time the 'new' decision is handed down." Id., at 258, 89 S.Ct., at 1039. "[A] proper perception of our duties as a court of law, charged with applying the Constitution to resolve every legal dispute within our jurisdiction on direct review, mandates that we apply the law as it is at the time, not as it once was." Mackey v. United States, 401 U.S., at 681, 91 S.Ct., at 1174 (separate opinion). 23 We now agree with Justice Harlan that " '[r]etroactivity' must be rethought," Desist v. United States, 394 U.S., at 258, 89 S.Ct., at 1038 (dissenting opinion). We therefore examine the circumstances of this case to determine whether it presents a retroactivity question clearly controlled by past precedents, and if not, whether application of the Harlan approach would resolve the retroactivity issue presented in a principled and equitable manner. III A. 24 At the outset, we must first ask whether respondent's case presents a retrospectivity problem clearly controlled by existing precedent. Re-examination of the post-Linkletter decisions convinces us that in three narrow categories of cases, the answer to the retroactivity question has been effectively determined, not by application of the Stovall factors, but rather, through application of a threshold test.11 25 First, when a decision of this Court merely has applied settled precedents to new and different factual situations, no real question has arisen as to whether the later decision should apply retrospectively. In such cases, it has been a foregone conclusion that the rule of the later case applies in earlier cases, because the later decision has not in fact altered that rule in any material way. See, e.g., Dunaway v. New York, 442 U.S. 200, 206, 99 S.Ct. 2248, 2253, 60 L.Ed.2d 824 (1979) (reviewing application of the rule in Brown v. Illinois, 422 U.S. 590, 95 S.Ct. 2254, 45 L.Ed.2d 416 (1975)); Spinelli v. United States, 393 U.S. 410, 412, 89 S.Ct. 584, 586, 21 L.Ed.2d 637 (1969) ("further explicat[ing]" the principles of Aguilar v. Texas, 378 U.S. 108, 84 S.Ct. 1509, 12 L.Ed.2d 723 (1964)); Desist v. United States, 394 U.S., at 263, 89 S.Ct., at 1041 (Harlan, J., dissenting). 26 Conversely, where the Court has expressly declared a rule of criminal procedure to be "a clear break with the past," Desist v. United States, 394 U.S., at 248, 89 S.Ct., at 1033, it almost invariably has gone on to find such a newly minted principle nonretroactive. See United States v. Peltier, 422 U.S. 531, 547, n. 5, 95 S.Ct. 2313, 2322, n. 5, 45 L.Ed.2d 374 (1975) (BRENNAN, J., dissenting) (collecting cases). In this second type of case, the traits of the particular constitutional rule have been less critical than the Court's express threshold determination that the " 'new' constitutional interpretatio[n] . . . so change[s] the law that prospectivity is arguably the proper course," Williams v. United States, 401 U.S., at 659, 91 S.Ct., at 1156 (plurality opinion). Once the Court has found that the new rule was unanticipated, the second and third Stovall factors—reliance by law enforcement authorities on the old standards and effect on the administration of justice of a retroactive application of the new rule—have virtually compelled a finding of nonretroactivity. See, e.g., Gosa v. Mayden, 413 U.S., at 672-673, 682-685, 93 S.Ct., at 2932-2933, 2937-2938 (plurality opinion); Michigan v. Payne, 412 U.S., at 55-57, 93 S.Ct., at 1970-1971.12 27 Third, the Court has recognized full retroactivity as a necessary adjunct to a ruling that a trial court lacked authority to convict or punish a criminal defendant in the first place. The Court has invalidated inconsistent prior judgments where its reading of a particular constitutional guarantee immunizes a defendant's conduct from punishment, see, e.g., United States v. United States Coin & Currency, 401 U.S. 715, 724, 91 S.Ct. 1041, 1046, 28 L.Ed.2d 434 (1971) (penalty against assertion of Fifth Amendment privilege against self-incrimination), or serves "to prevent [his] trial from taking place at all, rather than to prescribe procedural rules that govern the conduct of [that] trial," Robinson v. Neil, 409 U.S., at 509, 93 S.Ct., at 878 (double jeopardy). In such cases, the Court has relied less on the technique of retroactive application than on the notion that the prior inconsistent judgments or sentences were void ab initio. See, e.g., Moore v. Illinois, 408 U.S. 786, 800, 92 S.Ct. 2562, 2570, 33 L.Ed.2d 706 (1972) (retroactive application of Eighth Amendment ruling in Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972)); Ashe v. Swenson, 397 U.S. 436, 437, n. 1, 90 S.Ct. 1189, 1191, n. 1, 25 L.Ed.2d 469 (1970) (retroactive application of double jeopardy ruling in Benton v. Maryland, 395 U.S. 784, 89 S.Ct. 2056, 23 L.Ed.2d 707 (1969)). See also Gosa v. Mayden, 413 U.S., at 693, 93 S.Ct., at 2943 (MARSHALL, J., dissenting); Michigan v. Payne, 412 U.S., at 61, 93 S.Ct., at 1973 (MARSHALL, J., dissenting) (rulings are fully retroactive when the "Court has held that the trial court lacked jurisdiction in the traditional sense"). 28 Respondent's case neatly fits none of these three categories. First, Payton v. New York did not simply apply settled precedent to a new set of facts. InPayton, the Court acknowledged that the "important constitutional question presented" there had been "expressly left open in a number of our prior opinions." 445 U.S., at 574 and 575, n. 1, 100 S.Ct., at 1374, n. 1, citingUnited States v. Watson, 423 U.S. 411, 418, n. 6, 96 S.Ct. 820, 825, n. 6, 46 L.Ed.2d 598 (1976); Gerstein v. Pugh, 420 U.S. 103, 113, n. 13, 95 S.Ct. 854, 862, n. 13, 43 L.Ed.2d 54 (1975); Coolidge v. New Hampshire, 403 U.S. 443, 474-481, 91 S.Ct. 2022, 2042, 2045, 29 L.Ed.2d 564 (1971); and Jones v. United States, 357 U.S. 493, 499-500, 78 S.Ct. 1253, 1257, 2 L.Ed.2d 1514 (1958). 29 By the same token, however, Payton also did not announce an entirely new and unanticipated principle of law. In general, the Court has not subsequently read a decision to work a "sharp break in the web of the law," Milton v. Wainwright, 407 U.S. 371, 381, n. 2, 92 S.Ct. 2174, 2180, n. 2, 33 L.Ed.2d 1 (1972) (Stewart, J., dissenting), unless that ruling caused "such an abrupt and fundamental shift in doctrine as to constitute an entirely new rule which in effect replaced an older one," Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 498, 88 S.Ct. 2224, 2234, 20 L.Ed.2d 1231 (1968). Such a break has been recognized only when a decision explicitly overrules a past precedent of this Court, see, e.g., Desist v. United States, 394 U.S. 244, 89 S.Ct. 1048, 22 L.Ed.2d 248 (1969); Williams v. United States, 401 U.S. 646, 91 S.Ct. 1148, 28 L.Ed.2d 388 (1971), or disapproves a practice this Court arguably has sanctioned in prior cases, see, e.g., Gosa v. Mayden, 413 U.S., at 673, 93 S.Ct., at 2932 (plurality opinion); Adams v. Illinois, 405 U.S., at 283, 92 S.Ct., at 919; Johnson v. New Jersey, 384 U.S., at 731, 86 S.Ct., at 1779, or overturns a longstanding and widespread practice to which this Court has not spoken, but which a near-unanimous body of lower court authority has expressly approved. See, e.g., Gosa v. Mayden, 413 U.S., at 673, 93 S.Ct., at 2932 (plurality opinion) (applying nonretroactively a decision that "effected a decisional change in attitude that had prevailed for many decades"); Stovall v. Denno, 388 U.S., at 299-300, 87 S.Ct., at 1971. See also Chevron Oil Co. v. Huson, 404 U.S. 97, 107, 92 S.Ct. 349, 355, 30 L.Ed.2d 296 (1971); Cipriano v. City of Houma, 395 U.S. 701, 89 S.Ct. 1897, 23 L.Ed.2d 647 (1969); Milton v. Wainwright, 407 U.S., at 381-382, n. 2, 92 S.Ct., at 2179-2180, n. 2 (Stewart, J., dissenting) ("sharp break" occurs when "decision overrules clear past precedent . . . or disrupts a practice long accepted and widely relied upon"). 30 Payton did none of these. Payton expressly overruled no clear past precedent of this Court on which litigants may have relied. Nor did Payton disapprove an established practice that the Court had previously sanctioned. To the extent that the Court earlier had spoken to the conduct engaged in by the police officers in Payton, it had deemed it of doubtful constitutionality.13 The Court's own analysis in Payton makes it clear that its ruling rested on both long-recognized principles of Fourth Amendment law and the weight of historical authority as it had appeared to the Framers of the Fourth Amendment.14 Finally, Payton overturned no longstanding practice approved by a near-unanimous body of lower court authority.15 Payton therefore does not fall into that narrow class of decisions whose nonretroactivity is effectively preordained because they unmistakably signal "a clear break with the past," Desist v. United States, 394 U.S., at 248, 89 S.Ct., at 1033. 31 It is equally plain that Payton does not fall into the third category of cases that do not pose difficult retroactivity questions. Payton did not hold that the trial court lacked authority to convict or sentence Theodore Payton, nor did Payton's reading of the Fourth Amendment immunize Payton's conduct from punishment. The holding in Payton did not prevent the defendant's trial from taking place; rather, it reversed the New York Court of Appeals' judgment and remanded for a new trial to be conducted without unconstitutionally obtained evidence. B 32 Having determined that the retroactivity question here is not clearly controlled by our prior precedents, we next must ask whether that question would be fairly resolved by applying the rule in Payton to all cases still pending on direct appeal at the time when Payton was decided. Answering that question affirmatively would satisfy each of the three concerns stated in Justice Harlan's opinions in Desist and Mackey. 33 First, retroactive application of Payton to all previously nonfinal convictions would provide a principle of decisionmaking consonant with our original understanding of retroactivity in Linkletter and Shott. Moreover, such a principle would be one capable of general applicability, satisfying Justice Harlan's central concern: "Refusal to apply new constitutional rules to all cases arising on direct review . . . tends to cut this Court loose from the force of precedent, allowing us to restructure artificially those expectations legitimately created by extant law and thereby mitigate the practical force of stare decisis . . . a force which ought properly to bear on the judicial resolution of any legal problem." Mackey v. United States, 401 U.S., at 680-681, 91 S.Ct., at 1173-1174 (separate opinion). 34 Second, application of Payton to cases pending on direct review would comport with our judicial responsibilities "to do justice to each litigant on the merits of his own case," Desist v. United States, 394 U.S., at 259, 89 S.Ct., at 1039 (Harlan, J., dissenting), and to "resolve all cases before us on direct review in light of our best understanding of governing constitutional principles." Mackey v. United States, 401 U.S., at 679, 91 S.Ct., at 1173 (separate opinion of Harlan, J.). The Court of Appeals held that the circumstances of respondent's arrest violated Payton, and the Government does not dispute that contention. See n. 6, supra. It would be ironic indeed were we now to reverse a judgment applying Payton's rule, when in Payton itself, we reversed a directly contrary judgment of the New York Court of Appeals. As Justice Harlan noted in Desist : "If a 'new' constitutional doctrine is truly right, we should not reverse lower courts which have accepted it; nor should we affirm those which have rejected the very arguments we have embraced." 394 U.S., at 259, 89 S.Ct., at 1039. 35 Third, application of the Harlan approach to respondent's case would further the goal of treating similarly situated defendants similarly. The Government contends that respondent may not invoke Payton because he was arrested before Payton was decided. Yet it goes without saying that Theodore Payton also was arrested before Payton was decided, and he received the benefit of the rule in his case. Furthermore, at least one other defendant whose conviction was not final when Payton issued benefited from Payton's rule, although he, too, was arrested before Payton was decided.16 An approach that resolved all nonfinal convictions under the same rule of law would lessen the possibility that this Court might mete out different constitutional protection to defendants simultaneously subjected to identical police conduct.17 IV 36 Against adoption of this approach, the Government raises four arguments based on United States v. Peltier, 422 U.S. 531, 95 S.Ct. 2313, 45 L.Ed.2d 374 (1975). None is persuasive. 37 The Government first cites Peltier's holding: that the Fourth Amendment rule announced in Almeida-Sanchez v. United States, 413 U.S. 266, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973), should not apply retroactively to a case pending on appeal when Almeida-Sanchez was announced. By so holding, the Government suggests, Peltier declared a principle that controls the issue of retroactivity for all Fourth Amendment rulings.18 38 Upon examination, however, the retroactivity question posed here differs from that presented in Peltier. As the Government concedes, Payton overturned neither a statute nor any consistent judicial history approving nonconsensual, warrantless home entries. See Brief for United States 30, n. 18. Thus, its nonretroactivity is not preordained under the "clear break" principles stated above. In Peltier, in contrast, the Court noted that Almeida-Sanchez had invalidated a form of search previously sanctioned by "a validly enacted statute, supported by longstanding administrative regulations and continuous judicial approval." 422 U.S., at 541, 95 S.Ct., at 2319. See also Almeida-Sanchez v. United States, 413 U.S., at 278, 93 S.Ct., at 2542 (POWELL, J., concurring) ("While the question is one of first impression in this Court," the practice disapproved had "been consistently approved by the judiciary"); id., at 298-299, n. 10, 93 S.Ct., at 2552, n. 10 (WHITE, J., dissenting) (35 of 36 judges in 20 Court of Appeals cases had approved the invalidated practice). 39 Because Almeida-Sanchez had overturned a longstanding practice to which this Court had not spoken, but which a near-unanimous body of lower court authority had approved, it represented a "clear break" with the past. For that reason alone, under controlling retroactivity precedents, the nonretroactive application of Almeida-Sanchez would have been appropriate even if the case had involved no Fourth Amendment question. In that respect, Peltier resembles several earlier decisions that held "new" Fourth Amendment doctrine nonretroactive, not on the ground that all Fourth Amendment rulings apply only prospectively, but because the particular decisions being applied "so change[d] the law that prospectivity [was] arguably the proper course." Williams v. United States, 401 U.S., at 659, 91 S.Ct., at 1155 (plurality opinion) (refusing to apply retroactively Chimel v. California, 395 U.S. 752, 89 S.Ct. 2034, 23 L.Ed.2d 685 (1969), which overruled United States v. Rabinowitz, 339 U.S. 56, 70 S.Ct. 430, 94 L.Ed. 653 (1950), and Harris v. United States, 331 U.S. 145, 67 S.Ct. 1098, 91 L.Ed. 1399 (1947)). See also Desist v. United States, 394 U.S. 244, 89 S.Ct. 1048, 22 L.Ed.2d 248 (1969) (refusing to apply retroactively Katz v. United States, 389 U.S. 347, 88 S.Ct. 507, 19 L.Ed.2d 576 (1967), which overruled Goldman v. United States, 316 U.S. 129, 62 S.Ct. 993, 86 L.Ed. 1322 (1942), and Olmstead v. United States, 277 U.S. 438, 48 S.Ct. 564, 72 L.Ed. 944 (1928)). 40 The Government bases its second argument on Peltier's broad language: "If the purpose of the exclusionary rule is to deter unlawful police conduct then evidence obtained from a search should be suppressed only if it can be said that the law enforcement officer had knowledge, or may properly be charged with knowledge, that the search was unconstitutional under the Fourth Amendment" (emphasis added). 422 U.S., at 542, 95 S.Ct., at 2320. The Government reads this language to require that new Fourth Amendment rules must be denied retroactive effect in all cases except those in which law enforcement officers failed to act in good-faith compliance with then-prevailing constitutional norms. 41 The Government does not seriously suggest that the retroactivity of a given Fourth Amendment ruling should turn solely on the subjective state of a particular arresting officer's mind. Instead, it offers an "objective" test: that law enforcement officers "may properly be charged with knowledge" of all "settled" Fourth Amendment law. Under the Government's theory, because the state of Fourth Amendment law regarding warrantless home arrests was "unsettled" before Payton, that ruling should not apply retroactively even to cases pending on direct appeal when Payton was decided. See Brief for United States 14-19, 34-38. 42 Yet the Government's reading of Peltier would reduce its own "retroactivity test" to an absurdity. Under this view, the only Fourth Amendment rulings worthy of retroactive application are those in which the arresting officers violated pre-existing guidelines clearly established by prior cases. But as we have seen above, cases involving simple application of clear, pre-existing Fourth Amendment guidelines raise no real questions of retroactivity at all. Literally read, the Government's theory would automatically eliminate all Fourth Amendment rulings from consideration for retroactive application. 43 The Government's third claim is that Peltier's logic suggests that retroactive application of Fourth Amendment decisions like Payton—even to cases pending on direct review—would not serve the policies underlying the exclusionary rule. Cf. 422 U.S., at 536-542, 95 S.Ct., at 2317-2320. Yet viewed in the light ofPeltier's holding, this assertion also fails. Peltier suggested only that retroactive application of a Fourth Amendment ruling that worked a "sharp break" in the law, like Almeida-Sanchez, would have little deterrent effect, because law enforcement officers would rarely be deterred from engaging in a practice they never expected to be invalidated. See 422 U.S., at 541-542, 95 S.Ct., at 2319-2320. 44 This logic does not apply to a ruling like Payton, that resolved a previously unsettled point of Fourth Amendment law. Because this Court cannot rule on every unsettled Fourth Amendment question, years may pass before the Court finally invalidates a police practice of dubious constitutionality. See, e.g., Desist v. United States, 394 U.S., at 275, 89 S.Ct., at 1051 (Fortas, J., dissenting) (arguing that the "physical-trespass" wiretap rule of Olmstead v. United States, 277 U.S. 438, 48 S.Ct. 564, 72 L.Ed. 944 (1928), had been moribund for 17 years before it was formally overruled). Long before Payton, for example, this Court had questioned the constitutionality of warrantless home arrests. See n. 13, supra. Furthermore, the Court's opinions consistently had emphasized that, in light of the constitutional protection traditionally accorded to the privacy of the home, police officers should resolve any doubts regarding the validity of a home arrest in favor of obtaining a warrant. See, e.g., Johnson v. United States, 333 U.S. 10, 14, 68 S.Ct. 367, 369, 92 L.Ed. 436 (1948) ("Any assumption that evidence sufficient to support a magistrate's disinterested determination to issue a search warrant will justify the officers in making a search without a warrant would reduce the Amendment to a nullity and leave the people's homes secure only in the discretion of police officers"). 45 If, as the Government argues, all rulings resolving unsettled Fourth Amendment questions should be nonretroactive, then, in close cases, law enforcement officials would have little incentive to err on the side of constitutional behavior.19 Official awareness of the dubious constitutionality of a practice would be counterbalanced by official certainty that, so long as the Fourth Amendment law in the area remained unsettled, evidence obtained through the questionable practice would be excluded only in the one case definitively resolving the unsettled question. Failure to accord any retroactive effect to Fourth Amendment rulings would "encourage police or other courts to disregard the plain purport of our decisions and to adopt a let's-wait-until-it's-decided approach." Desist v. United States, 394 U.S., at 277, 89 S.Ct., at 1052 (Fortas, J., dissenting). 46 The Government finally argues that retroactive application of Payton, even to a case pending on direct appeal, would accomplish nothing but the discharge of a wrongdoer. Justice Harlan gave the answer to this assertion. "We do not release a criminal from jail because we like to do so, or because we think it wise to do so, but only because the government has offended constitutional principle in the conduct of his case. And when another similarly situated defendant comes before us, we must grant the same relief or give a principled reason for acting differently." Desist v. United States, 394 U.S., at 258, 89 S.Ct., at 1038 (dissenting opinion). Applying Payton to convictions that were not yet final when Payton issued would accomplish the first step toward "turning our backs on the ad hoc approach that has so far characterized our decisions in the retroactivity field and proceeding to administer the doctrine on principle." Jenkins v. Delaware, 395 U.S., at 224, 89 S.Ct., at 1683 (Harlan, J., dissenting). V 47 To the extent necessary to decide today's case, we embrace Justice Harlan's views in Desist and Mackey. We therefore hold that, subject to the exceptions stated below, a decision of this Court construing the Fourth Amendment is to be applied retroactively to all convictions that were not yet final at the time the decision was rendered. 48 By so holding, however, we leave undisturbed our precedents in other areas. First, our decision today does not affect those cases that would be clearly controlled by our existing retroactivity precedents. Second, because respondent's case arises on direct review, we need not address the retroactive reach of our Fourth Amendment decisions to those cases that still may raise Fourth Amendment issues on collateral attack.20 Cf. n. 10, supra. Third, we express no view on the retroactive application of decisions construing any constitutional provision other than the Fourth Amendment.21 Finally, all questions of civil retroactivity continue to be governed by the standard enunciated in Chevron Oil Co. v. Huson, 404 U.S., at 106-107, 92 S.Ct., at 355. See n. 12, supra. 49 Respondent's case was pending on direct appeal when Payton v. New York was decided. Because the Court of Appeals correctly held that the rule in Payton should apply to respondent's case, its judgment is affirmed.22 50 It is so ordered. 51 Justice BRENNAN, concurring. 52 I join the Court's opinion on my understanding that the decision leaves undisturbed our retroactivity precedents as applied to convictions final at the time of decision. See, e.g., Stovall v. Denno, 388 U.S. 293, 87 S.Ct. 1967, 18 L.Ed.2d 1199 (1967). 53 Justice WHITE, with whom THE CHIEF JUSTICE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 54 In my view, this case is controlled by United States v. Peltier, 422 U.S. 531, 95 S.Ct. 2313, 45 L.Ed.2d 374 (1975). Peltier established two propositions. First, retroactive application of a new constitutional doctrine is appropriate when that doctrine's major purpose is " 'to overcome an aspect of the criminal trial that substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts in past trials.' " Id., at 535, 95 S.Ct., at 2316, quoting Williams v. United States, 401 U.S. 646, 653, 91 S.Ct. 1148, 1152, 28 L.Ed.2d 388 (1971). Second, new extensions of the exclusionary rule do not serve this purpose and, therefore, will not generally be applied retroactively. There was surely nothing extraordinary about our ruling in Payton v. New York, 445 U.S. 573, 100 S.Ct. 1371, 63 L.Ed.2d 639 (1980), that would justify an exception to this general rule. 55 Peltier was the latest of a number of cases involving the question of whether rulings extending the reach of the exclusionary rule should be given retroactive effect. We noted there that "in every case in which the Court has addressed the retroactivity problem in the context of the exclusionary rule . . . the Court has concluded that any such new constitutional principle would be accorded only prospective application." 422 U.S., at 535, 95 S.Ct., at 2316. We suggested that there were two reasons for this consistent pattern of decisions and that these two reasons were directly related to the justifications for the exclusionary rule. 56 That rule has traditionally been understood to serve two purposes: first, it preserves "judicial integrity"; second, it acts as a deterrent to unconstitutional police conduct. Neither of these purposes, however, is furthered by retroactive application of new extensions of the rule. First, "if the law enforcement officers reasonably believed in good faith that evidence they had seized was admissible at trial, the 'imperative of judicial integrity' is not offended by the introduction into evidence of that material." Id., at 537, 95 S.Ct., at 2317. Second, a deterrence purpose can only be served when the evidence to be suppressed is derived from a search which the law enforcement officers knew or should have known was unconstitutional under the Fourth Amendment. Id., at 542, 95 S.Ct., at 2320. 57 In focusing on the purpose of the exclusionary rule in order to decide the question of retroactivity, the Court was following settled principles. In Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601 (1965), which the majority agrees is the first of the modern retroactivity cases, the Court set forth a three-pronged model for analysis of the retroactivity question presented there: 58 "[W]e must look to the purpose of the Mapp rule; the reliance placed upon the Wolf [Wolf v. People of State of Colorado, 338 U.S. 25, 69 S.Ct. 1359, 93 L.Ed. 1782 (1949) ] doctrine; and the effect on the administration of justice of a retrospective application of Mapp." Id., at 636, 85 S.Ct., at 1741. 59 This three-prong analysis was consistently applied in the cases which followed, Tehan v. United States ex rel. Shott, 382 U.S. 406, 419, 86 S.Ct. 459, 467, 15 L.Ed.2d 453 (1966); Johnson v. New Jersey, 384 U.S. 719, 727, 86 S.Ct. 1772, 1777, 16 L.Ed.2d 882 (1966); Stovall v. Denno, 388 U.S. 293, 297, 87 S.Ct. 1967, 1970, 18 L.Ed.2d 1199 (1967). Indeed, in Stovall, the Court specifically announced that these three considerations—purpose of the new rule, reliance on the old rule, and effect on the administration of justice—were generally to guide resolution of all retroactivity problems relating to constitutional rules of criminal procedure. In each of these cases, the purpose of the new rule was the first consideration. That this was not accidental was made absolutely clear in Desist v. United States, 394 U.S. 244, 249, 89 S.Ct. 1030, 1033, 22 L.Ed.2d 248 (1969): "Foremost among these factors is the purpose to be served by the new constitutional rule."* And as we went on to say there, "[t]his criterion strongly supports prospectivity for a decision amplifying the evidentiary exclusionary rule." Ibid. 60 Moreover, up until today's decision it was clear that these same principles governed the question of whether a new decision should retroactively apply to cases pending on appeal at the time of its announcement. Peltier itself was just this sort of a case: Peltier's case was on appeal at the time of the announcement of the decision in Almeida-Sanchez v. United States, 413 U.S. 266, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973). Indeed, we reversed the Court of Appeals' holding in that case that the "rule announced . . . in Almeida-Sanchez v. United States . . . should be applied to similar cases pending on appeal on the date the Supreme Court's decision was announced." United States v. Peltier, 500 F.2d 985, 986 (CA9 1974) (footnote omitted). I had thought that we long ago resolved the problem of the appearance of inequity that arises whenever we limit the retroactive reach of a new principle of law. As Justice BRENNAN stated for the Court in Stovall, supra, 388 U.S., at 301, 87 S.Ct., at 1972: 61 "Inequity arguably results from according the benefit of a new rule to the parties in the case in which it is announced but not to other litigants similarly situated in the trial or appellate process who have raised the same issue. But we regard the fact that the parties involved are chance beneficiaries as an insignificant cost for adherence to sound principles of decision-making." 62 All of these principles are well settled and require reversal of the judgment of the Court of Appeals. The majority, in an intricate and confusing opinion disagrees. Two reasons for its disagreement seem to be presented. 63 First, the majority discerns no consistent reading of our precedents that would control this case. Ante, at 554 ("Having determined that the retroactivity question here is not clearly controlled by our prior precedents . . ."). Given the clarity with which we have previously set out the applicable principles and the consistent application of those principles in cases involving extensions of the exclusionary rule, this is surely a strange conclusion. Eschewing the straight-forward reading of the cases set forth above, which looks primarily to the substantive purpose of the relevant rule of law, the majority replaces it with an exceedingly formal set of three categories. Ante, at 549-551. Because these categories turn out to be dicta only, they merit little comment. Suffice it to say that their inadequacy is obvious from even a moment's reflection: That category to which the majority agrees "the Court has regularly given complete retroactive effect" is nowhere included in this formal scheme—cases announcing new constitutional rules whose major purpose " 'is to overcome an aspect of the criminal trial that substantially impairs its truth-finding function and so raises serious questions about the accuracy of guilty verdicts in past trials.' " Ante, at 544, quoting Williams v. United States, 401 U.S., at 653, 91 S.Ct., at 1152 (plurality opinion). It is little wonder that the majority finds this case difficult, when it has failed to learn the most obvious lessons of the previous cases. 64 Second, the majority seems to think that the problems of principle that Justice Harlan struggled with in his dissent in Desist v. United States, supra, are unanswerable under any rule that fails to give the benefits of a new constitutional ruling to all criminal defendants whose cases are pending on appeal at the time of the announcement. These problems are not new and were, I believe, adequately answered by Justice BRENNAN in Stovall. The majority's approach, however, does not resolve these theoretical problems; it simply draws what is necessarily an arbitrary line in a somewhat different place than the Court had previously settled upon. Anything less than full retroactivity will necessarily appear unjust in some instances; it will provide different treatment to similarly situated individuals. The majority recognizes that the vagaries of the appellate process will cause this same problem to reappear under its proposed rule: "Even under our approach, it may be unavoidable that some similarly situated defendants will be treated differently." Ante, at 556-557, n. 17. We had previously held that the best way to deal with this problem of inherent arbitrariness was to abide by the substantive principles outlined in Stovall. The majority makes no better suggestion today and is fooling itself if it believes that its proposal is a reasoned response to this problem of arbitrariness, rather than an exercise in line-drawing. 65 The insubstantiality of the majority's analysis and proposal is well illustrated by its conclusion. Despite the appearance of having resolved the difficult problem of the apparent injustice of any rule of partial retroactivity, the Court announces at the end that its decision today applies only to decisions "construing the Fourth Amendment" and asserts that it is not disturbing any of our retroactivity precedents. Ante, at 562. That is, it returns from its abstract procedural approach to the substantive rule of law at issue. There are two problems with this, however. First, there is no connection between the analysis and the conclusion. Second, and more important, we already had a perfectly good rule for resolving retroactivity problems involving the Fourth Amendment. 66 Accordingly, I dissent. 1 The Fourth Amendment reads: "The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized." 2 On March 30, 1977, the United States Postal Service mistakenly delivered to Lena Kearney a Treasury check for $4,681.41, payable to Elihu Peterson. Kearney and her sister-in-law sought Dodd's assistance in cashing the check. Accompanied by respondent Johnson and another man, Dodd went to Kearney's residence to discuss methods of cashing the check. The three men eventually departed, taking the check with them. After Kearney and her sister-in-law related the foregoing events to Special Agent Hemenway, he obtained a warrant for Dodd's arrest. He, however, did not obtain a warrant to arrest respondent. See 626 F.2d 753, 754-755 (CA9 1980). 3 The jury acquitted respondent on a separate count of aiding and abetting the receipt of stolen Government property. See 18 U.S.C. §§ 2, 641. Respondent's codefendant Dodd was convicted on both counts. In an unreported decision, Dodd's conviction was affirmed summarily on appeal, and is not before us. See United States v. Dodd, 614 F.2d 777 (CA9 1980), rehearing denied, Mar. 5, 1980. 4 The Court noted probable jurisdiction in Payton on December 11, 1978. 439 U.S. 1044, 99 S.Ct. 718, 58 L.Ed.2d 703. On March 5, 1979, the Ninth Circuit deferred decision on respondent's petition for rehearing and rehearing en banc pending this Court's decision in Payton. App. 8. The Court heard argument in Payton on March 26, 1979, but restored the case to the calendar for reargument. See 441 U.S. 930, 99 S.Ct. 2049, 60 L.Ed.2d 658 (1979). On August 20, 1979, the Ninth Circuit reaffirmed respondent's conviction, in the process amending its initial opinion and denying respondent's petition for rehearing. App. to Pet. for Cert. 14a. Respondent timely filed a second petition for rehearing and suggestion for rehearing en banc, which was still pending in the Court of Appeals when Payton was decided. 5 In a decision issued three months before its initial ruling here, a different panel of the Ninth Circuit had anticipated Payton, holding that "absent exigent circumstances, police who have probable cause to arrest a felony suspect must obtain a warrant before entering a dwelling to carry out the arrest." United States v. Prescott, 581 F.2d 1343, 1350 (1978). Upon denial of the Government's petition for rehearing in respondent's case, the Court of Appeals made clear that its post-Payton reversal of respondent's conviction "rests chiefly upon basic principles common to our decision in Prescott and that of the Supreme Court in Payton." App. to Pet. for Cert. 13a. The court also noted that it had already held that its ruling in Prescott should apply retroactively. See United States v. Blake, 632 F.2d 731 (1980). 6 For the purposes of this case, the Government assumes the correctness of the Court of Appeals' ruling that, if applied to these facts, Payton would require exclusion of respondent's statements. Brief for United States 12-13, n. 6. We therefore need not examine the Court of Appeals' conclusion on that issue. 7 The pre-1965 requirement that all constitutional rules receive full retroactive application derived from the Blackstonian notion "that the duty of the court was not to 'pronounce a new law, but to maintain and expound the old one.' " Linkletter v. Walker, 381 U.S. 618, 622-623, 85 S.Ct. 1731, 1733-1734, 14 L.Ed.2d 601 (1965), citing 1 W. Blackstone, Commentaries 69 (15th ed. 1809). 8 "By final we mean where the judgment of conviction was rendered, the availability of appeal exhausted, and the time for petition for certiorari had elapsed [or a petition for certiorari finally denied, all] before our decision in Mapp v. Ohio." Linkletter v. Walker, 381 U.S., at 622, n. 5, 85 S.Ct., at 1734, n. 5. See also Tehan v. United States ex rel. Shott, 382 U.S. 406, 409, n. 3, 86 S.Ct. 459, 461, n. 3, 15 L.Ed.2d 453 (1966). 9 See, e.g., Brown v. Louisiana, 447 U.S. 323, 337, 100 S.Ct. 2214, 2224, 65 L.Ed.2d 159 (1980) (POWELL, J., with whom STEVENS, J., joined, concurring in judgment); Harlin v. Missouri, 439 U.S. 459, 460, 99 S.Ct. 709, 710, 58 L.Ed.2d 733 (1979) (POWELL, J., concurring in judgments); Hankerson v. North Carolina, 432 U.S. 233, 245, 97 S.Ct. 2339, 2346, 53 L.Ed.2d 306 (1977) (MARSHALL, J., concurring in judgment); id., at 246, 97 S.Ct., at 2346 (POWELL, J., concurring in judgment); United States v. Peltier, 422 U.S. 531, 543, 95 S.Ct. 2313, 2320, 45 L.Ed.2d 374 (1975) (Douglas, J., dissenting); Daniel v. Louisiana, 420 U.S. 31, 33, and n., 95 S.Ct. 704, 705, and n., 42 L.Ed.2d 790 (1975) (Douglas, J., dissenting); Michigan v. Tucker, 417 U.S. 433, 461, 94 S.Ct. 2357, 2372, 41 L.Ed.2d 182 (1974) (Douglas, J., dissenting); Michigan v. Payne, 412 U.S. 47, 58, 93 S.Ct. 1966, 1972, 36 L.Ed.2d 736 (1973) (Douglas, J., dissenting); id., at 59, 93 S.Ct., at 1973 (MARSHALL, J., dissenting); Adams v. Illinois, 405 U.S. 278, 286, 92 S.Ct. 916, 921, 31 L.Ed.2d 202 (1972) (Douglas, J., with whom MARSHALL, J., concurred, dissenting); Mackey v. United States, 401 U.S. 667, 675, 91 S.Ct. 1160, 1164, 28 L.Ed.2d 404 (1971) (separate opinion of Harlan, J.); id., at 713, 91 S.Ct., at 1170 (Douglas, J., with whom Black, J., concurred, dissenting); Williams v. United States, 401 U.S. 646, 665, 91 S.Ct. 1148, 1159, 28 L.Ed.2d 388 (1971) (MARSHALL, J., concurring in part and dissenting in part); Coleman v. Alabama, 399 U.S. 1, 19, 90 S.Ct. 1999, 2008, 26 L.Ed.2d 387 (1970) (Harlan, J., concurring in part and dissenting in part); Von Cleef v. New Jersey, 395 U.S. 814, 817, 89 S.Ct. 2051, 2053, 23 L.Ed.2d 728 (1969) (Harlan, J., concurring in result); Jenkins v. Delaware, 395 U.S. 213, 222, 89 S.Ct. 1677, 1682, 23 L.Ed.2d 253 (1969) (Harlan, J., dissenting); Desist v. United States, 394 U.S. 244, 255, 89 S.Ct. 1030, 1037, 22 L.Ed.2d 248 (1969) (Douglas, J., dissenting); id., at 256, 89 S.Ct., at 1037 (Harlan, J., dissenting); id., at 269, 89 S.Ct., at 1048 (Fortas, J., dissenting); Fuller v. Alaska, 393 U.S. 80, 82, 89 S.Ct. 61, 62, 21 L.Ed.2d 212 (1968) (Douglas, J., dissenting); DeStefano v. Woods, 392 U.S. 631, 635, 88 S.Ct. 2093, 2096, 20 L.Ed.2d 1308 (1968) (Douglas, J., with whom Black, J., joined, dissenting); Stovall v. Denno, 388 U.S. 293, 302, 87 S.Ct. 1967, 1972, 18 L.Ed.2d 1199 (1967) (Douglas, J., dissenting); id., at 303, 87 S.Ct., at 1973 (Black, J., dissenting); Johnson v. New Jersey, 384 U.S. 719, 736, 86 S.Ct. 1772, 1782, 16 L.Ed.2d 882 (1966) (Black, J., with whom Douglas, J., joined, dissenting); Whisman v. Georgia, 384 U.S. 895, 86 S.Ct. 1977, 16 L.Ed.2d 1001 (1966) (Douglas, J., dissenting); Tehan v. United States ex rel. Shott, 382 U.S., at 419, 86 S.Ct., at 467 (Black, J., with whom Douglas, J., joined, dissenting); Linkletter v. Walker, 381 U.S., at 640, 85 S.Ct., at 1743 (Black, J., with whom Douglas, J., joined, dissenting). 10 Evenhanded justice for similarly situated litigants was the principal theme sounded by the dissenting opinions of Justices Black and Douglas. See cases cited in n. 9, supra. The views of these Justices diverged from those of Justice Harlan, however, on the question whether equal treatment also requires retroactive application of newly announced constitutional rules to all cases arising on collateral attack. Compare Desist v. United States, 394 U.S., at 255, 89 S.Ct., at 1037 (Douglas, J., dissenting), with id., at 260-269, 89 S.Ct., at 1039-1044 (Harlan, J., dissenting). See also Adams v. Illinois, 405 U.S., at 287, and n. 4, 92 S.Ct., at 921, and n. 4 (Douglas, J., dissenting). Members of the Court continue to offer views on this troublesome question. Compare Hankerson v. North Carolina, 432 U.S., at 246, and n., 97 S.Ct., at 2346, and n. (MARSHALL, J., concurring in judgment), with id., at 248, 97 S.Ct., at 2347 (POWELL, J., concurring in judgment). 11 These cases therefore have not proved "readily susceptible of analysis under the Linkletter line of cases." Robinson v. Neil, 409 U.S. 505, 508, 93 S.Ct. 876, 878, 35 L.Ed.2d 29 (1973). The dissent's accusation that these categories exclude the "most obvious" line of cases—those announcing rules relating to the truth-finding function, post, at 567—misses our point. In those cases, the retroactivity decision has in fact turned on a traditional application of the Stovall factors, with the central issue in dispute often being the major purpose to be served by the new standard. Compare Brown v. Louisiana, 447 U.S. 323, 100 S.Ct. 2214, 65 L.Ed.2d 159 (1980) (plurality opinion), with id., at 337, 100 S.Ct., at 2225 (REHNQUIST, J., dissenting) (disagreeing over the "major purpose" of the unanimous six-person jury rule of Burch v. Louisiana, 441 U.S. 130, 99 S.Ct. 1623, 60 L.Ed.2d 96 (1979)). 12 In the civil context, in contrast, the "clear break" principle has usually been stated as the threshold test for determining whether or not a decision should be applied nonretroactively. See, e.g., Chevron Oil Co. v. Huson, 404 U.S. 97, 106, 92 S.Ct. 349, 355, 30 L.Ed.2d 296 (1971). Once it has been determined that a decision has "establish[ed] a new principle of law, either by overruling clear past precedent on which litigants may have relied . . . or by deciding an issue of first impression whose resolution was not clearly foreshadowed," the Court has gone on to examine the history, purpose, and effect of the new rule, as well as the inequity that would be imposed by its retroactive application. Id., at 106-107, 92 S.Ct., at 355-356. See also Hanover Shoe, Inc. v. United Shoe Machinery Corp., 392 U.S. 481, 499, 88 S.Ct. 2224, 2234, 20 L.Ed.2d 1231 (1968). 13 At least since Boyd v. United States, 116 U.S. 616, 630, 6 S.Ct. 524, 532, 29 L.Ed. 746 (1886), the Court had acknowledged that the Fourth Amendment accords special protection to the home. McDonald v. United States, 335 U.S. 451, 456, 69 S.Ct. 191, 193, 93 L.Ed. 153 (1948), stated that "the Constitution requires a magistrate to pass on the desires of the police before they violate the privacy of the home." See also Johnson v. United States, 333 U.S. 10, 13-15, 68 S.Ct. 367, 368-369, 92 L.Ed. 436 (1948). While ultimately declining to decide whether a warrant is necessary to effect a home arrest, Coolidge v. New Hampshire, 403 U.S. 443, 474-475, 91 S.Ct. 2022, 2042, 29 L.Ed.2d 564 (1971) (footnote omitted), had declared that "a search or seizure carried out on a suspect's premises without a warrant is per se unreasonable, unless the police can show that it falls within one of a carefully defined set of exceptions based on the presence of 'exigent circumstances.' " See also United States v. United States District Court, 407 U.S. 297, 313, 92 S.Ct. 2125, 2134, 32 L.Ed.2d 752 (1972) ("physical entry of the home is the chief evil against which the wording of the Fourth Amendment is directed"); United States v. Martinez-Fuerte, 428 U.S. 543, 561, 96 S.Ct. 3074, 3084, 49 L.Ed.2d 1116 (1976) ("the sanctity of private dwellings [is] ordinarily afforded the most stringent Fourth Amendment protection"). 14 The Payton Court relied on the " 'basic principle of Fourth Amendment law' that searches and seizures inside a home without a warrant are presumptively unreasonable." 445 U.S., at 586, 100 S.Ct., at 1380, citing Coolidge v. New Hampshire, 403 U.S., at 477, 91 S.Ct., at 2044. The Court further recognized that the express language of the Fourth Amendment "has drawn a firm line at the entrance to the house" in "terms that apply equally to seizures of property and to seizures of persons." 445 U.S., at 590, 100 S.Ct., at 1382. After examining the common-law understanding of an officer's authority to arrest a suspect in his own home, id., at 591-598, 100 S.Ct., at 1382-1386, the Court concluded that "the weight of authority as it appeared to the Framers [of the Fourth Amendment] was to the effect that a warrant was required [before a home arrest], or at the minimum that there were substantial risks in proceeding without one." Id., at 596, 100 S.Ct., at 1385. 15 While the practice invalidated in Payton had found support in some state courts, those decisions evinced "by no means the kind of virtual unanimity," id., at 600, 100 S.Ct., at 1387, required to make Payton a clear break with the past. In Payton, the Court noted that at the time of its decision, "[o]nly 24 of the 50 States currently sanction warrantless entries into the home to arrest, . . . and there is an obvious declining trend." Ibid. In California, where the present respondent's case arose, the State Supreme Court had held more than a year before respondent's arrest that, under the Fourth Amendment and its state constitutional counterpart, warrantless arrests within the home were per se unreasonable in the absence of exigent circumstances. See People v. Ramey, 16 Cal.3d 263, 275-276, 127 Cal.Rptr. 629, 636-637, 545 P.2d 1333, 1340-1341, cert. denied, 429 U.S. 929, 97 S.Ct. 335, 50 L.Ed.2d 299 (1976). Of the seven United States Courts of Appeals that had considered the question before Payton, five had expressed the view that warrantless home arrests were unconstitutional. 445 U.S., at 575, and n. 4, 100 S.Ct., at 1374, and n. 4. Three other Circuits had assumed, without expressly deciding, that such searches were unlawful. Ibid. After one of those decisions, in 1978, the Department of Justice instructed federal law enforcement agencies to follow the practice of procuring arrest warrants before entering a suspect's home to arrest him without exigent circumstances. Brief for United States 33, n. 20. In the Ninth Circuit, where respondent was arrested, it has been said that "law enforcement officials knew that th[e] circuit's law was unsettled but that there was some drift toward a warrant requirement." United States v. Blake, 632 F.2d, at 736. United States v. Phillips, 497 F.2d 1131, 1135 (CA9 1974), had suggested in dictum that warrants are required before officers may enter a private dwelling to effect an arrest. In United States v. Calhoun, 542 F.2d 1094, 1102 (CA9 1976), cert. denied sub nom. Stephenson v. United States, 429 U.S. 1064, 97 S.Ct. 792, 50 L.Ed.2d 781 (1977), it was observed that the Government had agreed that, absent exigent circumstances, a warrantless and nonconsensual entry into a suspect's home would be illegal. United States v. Prescott, 581 F.2d, at 1350, then squarely held such arrests unconstitutional. See n. 5, supra. 16 The New York Court of Appeals affirmed Payton's conviction along with that of Obie Riddick. See Payton v. New York, 445 U.S., at 578-579, 100 S.Ct., at 1376. This Court noted probable jurisdiction in Riddick's appeal, consolidated it with Payton's, then reversed both convictions. Id., at 603, 100 S.Ct., at 1388. In theory, the Court could have held Riddick's jurisdictional statement pending the disposition in Payton's case, then vacated and remanded the case for reconsideration in light of Payton. Such a course was taken in seven other nonfinal cases. See Gonzalez v. New York, 446 U.S. 902, 100 S.Ct. 1824, 64 L.Ed.2d 253 (1980); Brown v. Florida, 446 U.S. 902, 100 S.Ct. 1826, 64 L.Ed.2d 254 (1980); Busch v. Florida, 446 U.S. 902, 100 S.Ct. 1826, 64 L.Ed.2d 255 (1980); Vidal v. New York, 446 U.S. 903, 100 S.Ct. 1826, 64 L.Ed.2d 255 (1980); Gordon v. New York, 446 U.S. 903, 100 S.Ct. 1827, 64 L.Ed.2d 255 (1980); Gayle v. New York, 446 U.S. 905, 100 S.Ct. 1829, 64 L.Ed.2d 257 (1980); and Dunagan v. Illinois, 446 U.S. 905, 100 S.Ct. 1829, 64 L.Ed.2d 257 (1980). Alternatively, the Court could have given all these cases plenary review. Potential for unequal treatment is inherent in this process. As Justice Douglas "recalled," when the Court decided Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966): "some 80 cases were presented raising the same question. We took four of them and held the rest and then disposed of each of the four, applying the new procedural rule retroactively. But as respects the rest of the pending cases we denied any relief. . . . Yet it was sheer coincidence that those precise four were chosen. Any other single case in the group or any other four would have been sufficient for our purposes." Desist v. United States, 394 U.S., at 255, 89 S.Ct., at 1037 (dissenting opinion). The dissent argues that "we long ago resolved the problem of the appearance of inequity that arises whenever we limit the retroactive reach of a new principle of law." Post, at 566. But the dissent mischaracterizes both the problem and this Court's treatment of it. The problem is not merely the appearance of inequity, but the actual inequity that results when the Court chooses which of many similarly situated defendants should be the chance beneficiary of a retroactively applied rule. As the persistently voiced dissatisfaction with the Court's "ambulatory retroactivity doctrine" has revealed, see n. 9, supra, until now this Court has not "resolved" this problem so much as it has chosen to tolerate it. The time for toleration has come to an end. 17 We are aware, of course, that many considerations affect a defendant's progress through the judicial system, and that the speed of appellate review will differ from State to State, Circuit to Circuit, and case to case. Even under our approach, it may be unavoidable that some similarly situated defendants will be treated differently. Cf. Williams v. United States, 401 U.S., at 657, and n. 9, 91 S.Ct., at 1154, and n. 9 (plurality opinion). The Government suggests an approach, however, that virtually ensures that such anomalies will occur. The Government concedes that the Payton rule should apply to any pre-Payton case arising in a Circuit where the United States Court of Appeals already had held authoritatively that Payton -type searches were unlawful. Brief for United States 22-26. When respondent was arrested, two Courts of Appeals had invalidated warrantless home arrests conducted in the absence of exigent circumstances. See Dorman v. United States, 140 U.S.App.D.C. 313, 435 F.2d 385 (1970); United States v. Shye, 492 F.2d 886 (CA6 1974). Thus, under the Government's theory, the statements of a suspect arrested in the District of Columbia, on the same day as respondent was arrested in Los Angeles and under identical circumstances, should be excluded while respondent's statements should not. Moreover, under the Government's reasoning, this Court would be obliged to reverse a ruling of the Court of Appeals for the Ninth Circuit excluding those statements, but not an identical ruling from the District of Columbia Circuit in a parallel case. The dissent takes a different tack. Arguing that "inherent arbitrariness" arises whenever lines are drawn in this area, the dissent suggests that the "best way to deal with this problem" is to continue to make retroactivity decisions by picking and choosing from among similarly situated defendants. See post, at 568. By clinging to this view, the dissent, and not the Court, "is fooling itself." Ibid. This Court has no power to speed up or slow down the appellate process in the many tribunals throughout the country to ensure similar treatment of similarly situated defendants. The Court does, however, have the power to eliminate the obvious unfairness that results when it gives only the most conveniently situated defendant the retrospective benefit of a newly declared rule. 18 The dissent shares this mistaken impression. In support of its claim, the dissent cites Peltier's suggestion that every decision by this Court involving the exclusionary rule has been "accorded only prospective application." Post, at 564, citing 422 U.S., at 535, 95 S.Ct., at 2316. As Peltier recognized with discomfort, however, Linkletter itself—the first of the modern retroactivity cases—acknowledged the application of the Mapp exclusionary rule to cases that were pending on direct review at the time that Mapp was decided. See 422 U.S., at 535, n. 5, 95 S.Ct., at 2316, n. 5. 19 The record in this case, for example, does not explain why respondent's arresting officers failed to obtain a warrant for his arrest, when they did obtain a warrant to arrest his codefendant. See n. 2, supra. 20 After Stone v. Powell, 428 U.S. 465, 96 S.Ct. 3037, 49 L.Ed.2d 1067 (1976), the only cases raising Fourth Amendment challenges on collateral attack are those federal habeas corpus cases in which the State has failed to provide a state prisoner with an opportunity for full and fair litigation of his claim, analogous federal cases under 28 U.S.C. § 2255, and collateral challenges by state prisoners to their state convictions under postconviction relief statutes that continue to recognize Fourth Amendment claims. 21 The logic of our ruling, however, is not inconsistent with our precedents giving complete retroactive effect to constitutional rules whose purpose is to overcome an aspect of the criminal trial that substantially impairs its truth-finding function. See, e.g., Hankerson v. North Carolina, 432 U.S. 233, 97 S.Ct. 2339, 53 L.Ed.2d 306 (1977); Ivan V. v. City of New York, 407 U.S. 203, 92 S.Ct. 1951, 32 L.Ed.2d 659 (1972). Depending on the constitutional provision involved, additional factors may warrant giving a particular ruling retroactive effect beyond those cases pending on direct review. See Hankerson v. North Carolina, 432 U.S., at 248, n. 2, 97 S.Ct., at 2347, n. 2 (POWELL, J., concurring in judgment). Curiously, the dissent faults us not only for limiting our ruling to the only context properly presented by this case—the Fourth Amendment—but also for preserving, rather than overruling, clearly controlling retroactivity precedents. See post, at 568. The dissent then recasts those precedents in its own simplistic way, arguing that rules related to truth-finding automatically receive full retroactive effect, while implying that all other rules—including Fourth Amendment rules—should receive none. There are, however, two problems with this. First, the Court's decisions regularly giving complete retroactive effect to truth-finding rules have in no way required that newly declared Fourth Amendment rulings be denied all retroactive effect. For the reasons already stated, retroactive application of Fourth Amendment rules at least to cases pending on direct review furthers the policies underlying the exclusionary rule. Second, and more important, the Fourth Amendment "rule" urged by the dissent is far from a "perfectly good" one. Ibid. As we already have shown, that "rule" condones obviously inequitable treatment of similarly situated litigants and judicial injustice to individual litigants. 22 The question on which we granted certiorari encompassed one other issue: whether the Court of Appeals correctly concluded that its own decision in United States v. Prescott, 581 F.2d 1343 (1978), applies retroactively to respondent's arrest. See n. 5, supra. Because we hold that the principles of our decision in Payton apply retroactively to respondent's case, we need not disturb the Court of Appeals' ruling regarding the retroactive application of its own prior decision. * See also 394 U.S., at 251, 89 S.Ct., at 1035: "It is to be noted also that we have relied heavily on the factors of the extent of reliance and consequent burden on the administration of justice only when the purpose of the rule in question did not clearly favor either retroactivity or prospectivity."

01

457 U.S. 423 102 S.Ct. 2515 73 L.Ed.2d 116 MIDDLESEX COUNTY ETHICS COMMITTEE, etc., Petitionerv.GARDEN STATE BAR ASSOCIATION, et al. No. 81-460. Argued March 31, 1982. Decided June 21, 1982. Syllabus Under rules promulgated by the New Jersey Supreme Court pursuant to its authority under the State Constitution to license and discipline attorneys admitted to practice in the State, a claim of unethical conduct by an attorney is first considered by a local District Ethics Committee appointed by the Supreme Court. If a complaint is issued, the attorney whose conduct is challenged is served with the complaint and has 10 days to answer. Upon a determination that a prima facie case of unethical conduct exists, a formal hearing is held. The attorney charged may have counsel, discovery is available, and all witnesses are sworn. The Committee may ultimately dismiss the complaint, issue a private letter of reprimand, or forward a presentment to the statewide Disciplinary Review Board, which is also appointed by the Supreme Court. After a de novo review, the Board is required to make formal findings and recommendations to the Supreme Court, which reviews all decisions beyond a private reprimand and which permits briefing and oral argument for cases involving disbarment or suspension for more than one year. Respondent Hinds, a member of the New Jersey Bar, was served by petitioner, a local Ethics Committee, with a formal statement of charges of violating certain Supreme Court disciplinary rules. Instead of filing an answer to the charges, Hinds and the three respondent organizations of lawyers filed suit in Federal District Court, contending that the disciplinary rules violated their rights under the Federal Constitution. The court dismissed the complaint on the basis of the abstention principles of Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669. The Court of Appeals reversed on the ground that the disciplinary proceedings did not provide a meaningful opportunity to adjudicate constitutional claims, notwithstanding an affidavit stating that the New Jersey Supreme Court would directly consider Hinds' constitutional challenges and would consider whether such a procedure should be made explicit in the Supreme Court rules. Held : The federal courts should abstain from interfering with the ongoing disciplinary proceedings within the jurisdiction of the New Jersey Supreme Court. Pp. 431-437. (a) The policies underlying Younger are fully applicable to noncriminal judicial proceedings when important state interests are involved. Where such interests are involved, a federal court should abstain unless state law clearly bars the interposition of the constitutional claims. The pertinent inquiry is whether the state proceedings afford an adequate opportunity to raise the constitutional claims. Pp. 431-432. (b) The New Jersey Supreme Court considers its disciplinary proceedings, beginning with the filing of a complaint with the local Ethics Committee, as "judicial in nature." As such, the proceedings are of a character to warrant federal-court deference. Pp.432-434 (c) The State has an extremely important interest in maintaining and assuring the professional conduct of the attorneys it licenses. The State's interest in the present litigation is demonstrated by the fact that petitioner, an agency of the New Jersey Supreme Court, is the named defendant in the present suit and was the body which initiated the state proceedings against Hinds. The importance of the state interest in the pending state judicial proceedings and in the federal case calls Younger abstention into play. Pp. 434-435 (d) In light of the unique relationship between the New Jersey Supreme Court and the local Ethics Committee, and in view of the nature of the proceedings, it cannot be concluded that there was no "adequate opportunity" for Hinds to raise his constitutional claims. Any doubt as to this matter was laid to rest by the New Jersey Supreme Court's subsequent actions when, prior to the filing of the petition for certiorari in this Court, it sua sponte entertained the constitutional issues raised by Hinds. And there is no reason to disturb the District Court's unchallenged findings that there was no bad faith or harassment on petitioner's part and that the state disciplinary rules were not "flagrantly and patently" unconstitutional. Nor have any other extraordinary circumstances been presented to indicate that abstention would not be appropriate. Pp. 435-437. 3rd Cir., 643 F.2d 119 and 3rd Cir., 651 F.2d 154, reversed and remanded. Mary Ann Burgess, Asst. Atty. Gen., Trenton, N. J., for petitioner. Morton Stavis, Hoboken, N. J., for respondents. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to determine whether a federal court should abstain from considering a challenge to the constitutionality of disciplinary rules that are the subject of pending state disciplinary proceedings within the jurisdiction of the New Jersey Supreme Court. 454 U.S. 962, 102 S.Ct. 500, 72 L.Ed.2d 377 (1981). The Court of Appeals held that it need not abstain under Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971). We reverse. 2 * A. 3 The Constitution of New Jersey charges the State Supreme Court with the responsibility for licensing and disciplining attorneys admitted to practice in the State. Art. 6, § 2, ¶ 3.1 Under the rules established by the New Jersey Supreme Court, promulgated pursuant to its constitutional authority, a complaint moves through a three-tier procedure. First, local District Ethics Committees appointed by the State Supreme Court are authorized to receive complaints relating to claimed unethical conduct by an attorney. New Jersey Court Rule 1:20-2(d). At least two of the minimum of eight members of the District Ethics Committee must be nonattorneys. Complaints are assigned to an attorney member of the Committee to report and make a recommendation. Rule 1:20-2(h). The decision whether to proceed with the complaint is made by the person who chairs the Ethics Committee. If a complaint is issued by the Ethics Committee it must state the name of the complainant, describe the claimed improper conduct, cite the relevant rules, and state, if known, whether the same or a similar complaint has been considered by any other Ethics Committee. The attorney whose conduct is challenged is served with the complaint and has 10 days to answer.2 4 Unless good cause appears for referring the complaint to another Committee member, each complaint is referred to the member of the Committee who conducted the initial investigation for review and further investigation, if necessary. The Committee member submits a written report stating whether a prima facie indication of unethical or unprofessional conduct has been demonstrated. The report is then evaluated by the chairman of the Ethics Committee to determine whether a prima facie case exists. Absent a prima facie showing, the complaint is summarily dismissed. If a prima facie case is found, a formal hearing on the complaint is held before three or more members of the Ethics Committee, a majority of whom must be attorneys. The lawyer who is charged with unethical conduct may have counsel, discovery is available, and all witnesses are sworn. The panel is required to prepare a written report with its findings of fact and conclusions. The full Committee, following the decision of the panel, has three alternatives. The Committee may dismiss the complaint, prepare a private letter of reprimand, or prepare a presentment to be forwarded to the Disciplinary Review Board. Rule 1:20-2(o).3 5 The Disciplinary Review Board, a state-wide board which is also appointed by the Supreme Court, consists of nine members, at least five of whom must be attorneys and at least three of whom must be nonattorneys. The Board makes a de novo review. Rule 1:20-3(d)(3).4 The Board is required to make formal findings and recommendations to the New Jersey Supreme Court. 6 All decisions of the Disciplinary Review Board beyond a private reprimand are reviewed by the New Jersey Supreme Court. Briefing and oral argument are available in the Supreme Court for cases involving disbarment or suspension for more than one year. Rule 1:20-4. B 7 Respondent Lennox Hinds, a member of the New Jersey Bar, served as executive director of the National Conference of Black Lawyers at the time of his challenged conduct. Hinds represented Joanne Chesimard in a civil proceeding challenging her conditions of confinement in jail. In 1977 Chesimard went to trial in state court for the murder of a policeman. Respondent Hinds was not a counsel of record for Chesimard in the murder case. However, at the outset of the criminal trial Hinds took part in a press conference, making statements critical of the trial and of the trial judge's judicial temperament and racial insensitivity. In particular, Hinds referred to the criminal trial as "a travesty," a "legalized lynching," and "a kangaroo court." 8 One member of the Middlesex County Ethics Committee read news accounts of Hinds' comments and brought the matter to the attention of the Committee. In February 1977 the Committee directed one of its members to conduct an investigation. A letter was written to Hinds, who released the contents of the letter to the press. The Ethics Committee on its own motion then suspended the investigation until the conclusion of the Chesimard criminal trial. 9 After the trial was completed the Committee investigated the complaint and concluded that there was probable cause to believe that Hinds had violated DR 1-102(A)(5) of the Disciplinary Rules of the Code of Professional Responsibility.5 That section provides that "[a] lawyer shall not . . . [e]ngage in conduct that is prejudicial to the administration of justice." Respondent Hinds also was charged with violating DR 7-107(D), which prohibits extrajudicial statements by lawyers associated with the prosecution or defense of a criminal matter.6 The Committee then served a formal statement of charges on Hinds. 10 Instead of filing an answer to the charges in accordance with the New Jersey Bar disciplinary procedures, Hinds and the three respondent organizations filed suit in the United States District Court for the District of New Jersey contending that the disciplinary rules violated respondents' First Amendment rights. In addition, respondents charged that the disciplinary rules were facially vague and overbroad. The District Court granted petitioner's motion to dismiss based on Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971), concluding that "[t]he principles of comity and federalism dictate that the federal court abstain so that the state is afforded the opportunity to interpret its rules in the face of a constitutional challenge." App. to Pet. for Cert. 53a-54a. At respondents' request the District Court reopened the case to allow respondents an opportunity to establish bad faith, harassment, or other extraordinary circumstance which would constitute an exception to Younger abstention. Dombrowski v. Pfister, 380 U.S. 479, 85 S.Ct. 1116, 14 L.Ed.2d 22 (1965). After two days of hearings the District Court found no evidence to justify an exception to the Younger abstention doctrine and dismissed the federal-court complaint. 11 A divided panel of the United States Court of Appeals for the Third Circuit reversed on the ground that the state bar disciplinary proceedings did not provide a meaningful opportunity to adjudicate constitutional claims. 643 F.2d 119 (1981). The court reasoned that the disciplinary proceedings in this case are unlike the state judicial proceedings to which the federal courts usually defer. The Court of Appeals majority viewed the proceedings in this case as administrative, "nonadjudicative" proceedings analogous to the preindictment stage of a criminal proceeding.7 12 On petition for rehearing petitioner attached an affidavit from the Clerk of the New Jersey Supreme Court which stated that the New Jersey Supreme Court would directly consider Hinds' constitutional challenges and that the court would consider whether such a procedure should be made explicit in the Supreme Court rules. On reconsideration a divided panel of the Third Circuit declined to alter its original decision, stating that the relevant facts concerning abstention are those that existed at the time of the District Court's decision. 651 F.2d 154 (1981).8 13 Pending review in this Court, the New Jersey Supreme Court has heard oral arguments on the constitutional challenges presented by respondent Hinds and has adopted a rule allowing for an aggrieved party in a disciplinary hearing to seek interlocutory review of a constitutional challenge to the proceedings.9 II A. 14 Younger v. Harris, supra, and its progeny espouse a strong federal policy against federal-court interference with pending state judicial proceedings absent extraordinary circumstances. The policies underlying Younger abstention have been frequently reiterated by this Court. The notion of "comity" includes "a proper respect for state functions, a recognition of the fact that the entire country is made up of a Union of separate state governments, and a continuance of the belief that the National Government will fare best if the States and their institutions are left free to perform their separate functions in their separate ways." Id., at 44, 91 S.Ct., at 750.10 Minimal respect for the state processes, of course, precludes any presumption that the state courts will not safeguard federal constitutional rights. 15 The policies underlying Younger are fully applicable to noncriminal judicial proceedings when important state interests are involved. Moore v. Sims, 442 U.S. 415, 423, 99 S.Ct. 2371, 2377, 60 L.Ed.2d 994 (1979); Huffman v. Pursue, Ltd., 420 U.S. 592, 604-605, 95 S.Ct. 1200, 1208, 43 L.Ed.2d 482 (1975). The importance of the state interest may be demonstrated by the fact that the noncriminal proceedings bear a close relationship to proceedings criminal in nature, as in Huffman, supra. Proceedings necessary for the vindication of important state policies or for the functioning of the state judicial system also evidence the state's substantial interest in the litigation. Trainor v. Hernandez, 431 U.S. 434, 97 S.Ct. 1911, 52 L.Ed.2d 486 (1977); Juidice v. Vail, 430 U.S. 327, 97 S.Ct. 1211, 51 L.Ed.2d 376 (1977). Where vital state interests are involved, a federal court should abstain "unless state law clearly bars the interposition of the constitutional claims." Moore, 442 U.S., at 426, 99 S.Ct., at 2379. "[T]he . . . pertinent inquiry is whether the state proceedings afford an adequate opportunity to raise the constitutional claims . . . ." Id., at 430, 99 S.Ct., at 2380. See also Gibson v. Berryhill, 411 U.S. 564, 93 S.Ct. 1689, 36 L.Ed.2d 488 (1973). 16 The question in this case is threefold: first, do state bar disciplinary hearings within the constitutionally prescribed jurisdiction of the State Supreme Court constitute an ongoing state judicial proceeding; second, do the proceedings implicate important state interests; and third, is there an adequate opportunity in the state proceedings to raise constitutional challenges. B 17 The State of New Jersey, in common with most States,11 recognizes the important state obligation to regulate persons who are authorized to practice law. New Jersey expresses this in a state constitutional provision vesting in the New Jersey Supreme Court the authority to fix standards, regulate admission to the bar, and enforce professional discipline among members of the bar. N.J. Const., Art. 6, § 2, ¶ 3. The Supreme Court of New Jersey has recognized that the local District Ethics Committees act as the arm of the court in performing the function of receiving and investigating complaints and holding hearings. Rule 1:20-2; In re Logan, 70 N.J. 222, 358 A.2d 787 (1976). The New Jersey Supreme Court has made clear that filing a complaint with the local Ethics and Grievance Committee "is in effect a filing with the Supreme Court . . . ." Toft v. Ketchum, 18 N.J. 280, 284, 113 A.2d 671, 674, cert. denied, 350 U.S. 887 (1955). "From the very beginning a disciplinary proceeding is judicial in nature, initiated by filing a complaint with an ethics and grievance committee."12 18 N.J., at 284, 113 A.2d, at 674. It is clear beyond doubt that the New Jersey Supreme Court considers its bar disciplinary proceedings as "judicial in nature."13 As such, the proceedings are of a character to warrant federal-court deference. The remaining inquiries are whether important state interests are implicated so as to warrant federal-court abstention and whether the federal plaintiff has an adequate opportunity to present the federal challenge. C 18 The State of New Jersey has an extremely important interest in maintaining and assuring the professional conduct of the attorneys it licenses. States traditionally have exercised extensive control over the professional conduct of attorneys. See n. 11, supra. The ultimate objective of such control is "the protection of the public, the purification of the bar and the prevention of a re-occurrence." In re Baron, 25 N.J. 445, 449, 136 A.2d 873, 875 (1957). The judiciary as well as the public is dependent upon professionally ethical conduct of attorneys and thus has a significant interest in assuring and maintaining high standards of conduct of attorneys engaged in practice. See In re Stein, 1 N.J. 228, 237, 62 A.2d 801, 805 (1949), quoting In re Cahill, 66 N.J.L. 527, 50 A. 119 (1901). The State's interest in the professional conduct of attorneys involved in the administration of criminal justice is of special importance. Finally, the State's interest in the present litigation is demonstrated by the fact that the Middlesex County Ethics Committee, an agency of the Supreme Court of New Jersey, is the named defendant in the present suit and was the body which initiated the state proceedings against respondent Hinds. 19 The importance of the state interest in the pending state judicial proceedings and in the federal case calls Younger abstention into play. So long as the constitutional claims of respondents can be determined in the state proceedings and so long as there is no showing of bad faith, harassment, or some other extraordinary circumstance that would make abstention inappropriate, the federal courts should abstain. D 20 Respondent Hinds contends that there was no opportunity in the state disciplinary proceedings to raise his federal constitutional challenge to the disciplinary rules. Yet Hinds failed to respond to the complaint filed by the local Ethics Committee and failed even to attempt to raise any federal constitutional challenge in the state proceedings. Under New Jersey's procedure, its Ethics Committees constantly are called upon to interpret the state disciplinary rules. Respondent Hinds points to nothing existing at the time the complaint was brought by the local Committee to indicate that the members of the Ethics Committee, the majority of whom are lawyers, would have refused to consider a claim that the rules which they were enforcing violated federal constitutional guarantees. Abstention is based upon the theory that " '[t]he accused should first set up and rely upon his defense in the state courts, even though this involves a challenge of the validity of some statute, unless it plainly appears that this course would not afford adequate protection.' " Younger v. Harris, 401 U.S., at 45, 91 S.Ct., at 750, quoting Fenner v. Boykin, 271 U.S. 240, 244, 46 S.Ct. 492, 493, 70 L.Ed. 927 (1926). 21 In light of the unique relationship between the New Jersey Supreme Court and the local Ethics Committee, and in view of the nature of the proceedings, it is difficult to conclude that there was no "adequate opportunity" for respondent Hinds to raise his constitutional claims.14 Moore, 442 U.S., at 430, 99 S.Ct., at 2380. 22 Whatever doubt, if any, that may have existed about respondent Hinds' ability to have constitutional challenges heard in the bar disciplinary hearings was laid to rest by the subsequent actions of the New Jersey Supreme Court. Prior to the filing of the petition for certiorari in this Court the New Jersey Supreme Court sua sponte entertained the constitutional issues raised by respondent Hinds. Respondent Hinds therefore has had abundant opportunity to present his constitutional challenges in the state disciplinary proceedings.15 23 There is no reason for the federal courts to ignore this subsequent development. In Hicks v. Miranda, 422 U.S. 332, 95 S.Ct. 2281, 45 L.Ed.2d 223 (1975), we held that "where state criminal proceedings are begun against the federal plaintiffs after the federal complaint is filed but before any proceedings of substance on the merits have taken place in federal court, the principles of Younger v. Harris should apply in full force." Id., at 349, 95 S.Ct., at 2291. An analogous situation is presented here; the principles of comity and federalism which call for abstention remain in full force. Thus far in the federal-court litigation the sole issue has been whether abstention is appropriate. No proceedings have occurred on the merits and therefore no federal proceedings on the merits will be terminated by application of Younger principles. It would trivialize the principles of comity and federalism if federal courts failed to take into account that an adequate state forum for all relevant issues has clearly been demonstrated to be available prior to any proceedings on the merits in federal court. 422 U.S., at 350, 95 S.Ct., at 2292.16 24 Respondents have not challenged the findings of the District Court that there was no bad faith or harassment on the part of petitioner and that the state rules were not " 'flagrantly and patently' " unconstitutional. Younger, supra, 401 U.S., at 53, 91 S.Ct., at 754, quoting Watson v. Buck, 313 U.S. 387, 402, 61 S.Ct. 962, 967, 85 L.Ed. 1416 (1941). See App. to Pet. for Cert. 50a-52a. We see no reason to disturb these findings, and no other extraordinary circumstances have been presented to indicate that abstention would not be appropriate.17 III 25 Because respondent Hinds had an "opportunity to raise and have timely decided by a competent state tribunal the federal issues involved," Gibson v. Berryhill, 411 U.S., at 577, 93 S.Ct., at 1697, and because no bad faith, harassment, or other exceptional circumstances dictate to the contrary, federal courts should abstain from interfering with the ongoing proceedings. Accordingly, the judgment of the United States Court of Appeals for the Third Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. 26 Reversed and remanded. Justice BRENNAN, concurring in the judgment. 27 For the reasons stated by Justice MARSHALL, I join the judgment in this case. I agree that federal courts should show particular restraint before intruding into an ongoing disciplinary proceeding by a state court against a member of the State's bar, where there is an adequate opportunity to raise federal issues in that proceeding. The traditional and primary responsibility of state courts for establishing and enforcing standards for members of their bars and the quasi-criminal nature of bar disciplinary proceedings, In re Ruffalo, 390 U.S. 544, 551, 88 S.Ct. 1222, 1226, 20 L.Ed.2d 117 (1968), call for exceptional deference by the federal courts. See Gipson v. New Jersey Supreme Court, 558 F.2d 701, 703-704 (CA3 1977); Erdmann v. Stevens, 458 F.2d 1205, 1209-1210 (CA2 1972). I continue to adhere to my view, however, that Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971), is in general inapplicable to civil proceedings. See Huffman v. Pursue, Ltd., 420 U.S. 592, 613, 95 S.Ct. 1200, 1212, 43 L.Ed.2d 482 (1975) (BRENNAN, J., dissenting). 28 Justice MARSHALL, with whom Justice BRENNAN, Justice BLACKMUN, and Justice STEVENS join, concurring in the judgment. 29 I agree with much of the general language in the Court's opinion discussing the importance of the State's interest in regulating the professional conduct of its attorneys. However, I believe that the question whether Younger abstention would have been appropriate at the time that the District Court or the Court of Appeals considered this issue is not as simple as the Court's opinion might be read to imply. As the Court acknowledges, absent an ongoing judicial proceeding in which there is an adequate opportunity for a party to raise federal constitutional challenges, Younger is inapplicable. Ante, at 432. See also Gibson v. Berryhill, 411 U.S. 564, 577, 93 S.Ct. 1689, 1697, 36 L.Ed.2d 488 (1973). Here, it is unclear whether, at the time the lower courts addressed this issue, there was an adequate opportunity in the state disciplinary proceedings to raise a constitutional challenge to the disciplinary rules. Furthermore, it is unclear whether proceedings before the Ethics Committee are more accurately viewed as prosecutorial rather than judicial in nature. 30 I agree with the Court that we may consider events subsequent to the decisions of the courts below because the federal litigation has addressed only the question whether abstention is appropriate. Thus far, there have been no proceedings on the merits in federal court. Ante, at 436-437. After the Court of Appeals rendered its decision and denied petitioner's petition for rehearing, the New Jersey Supreme Court certified the complaint against respondent Hinds to itself. App. to Pet. for Cert. 62a. Now, there are ongoing judicial proceedings in the New Jersey Supreme Court in which Hinds has been given the opportunity to raise his constitutional challenges. As a result, Younger abstention, at least with respect to Hinds, is appropriate at this time. For this reason only, I join the judgment of the Court. 1 Article 6, § 2, ¶ 3, provides: "The Supreme Court shall make rules governing the administration of all courts in the State and, subject to the law, the practice and procedure in all such courts. The Supreme Court shall have jurisdiction over the admission to the practice of law and the discipline of persons admitted." 2 For a more detailed explanation of the disciplinary procedure of the District Ethics Committees, see Rule 1:20-2. As noted below, the procedure, as amended in 1981, now provides that a charged attorney may raise constitutional questions in the District Committees. Any constitutional challenges are to be set forth in the answer to the complaint. Rule 1:20-2(j) now provides: "All constitutional questions shall be withheld for consideration by the Supreme Court as part of its review of the final decision of the Disciplinary Review Board. Interlocutory relief may be sought only in accordance with R. 1:20-4(d)(i)." 3 Each District Ethics Committee appoints one member of the bar to serve as Secretary. The Secretary maintains records of the proceedings. The Secretary also transmits copies of all documents filed to the Division of Ethics and Professional Services. Rule 1:20-2(c). 4 Subsequent to the initiation of the disciplinary hearing involved in this case, Rule 1:20-3(e) was amended to provide: "Constitutional challenges to the proceedings not raised before the District Committee shall be preserved, without Board action, for Supreme Court consideration as part of its review of the matter on the merits. Interlocutory relief may be sought only in accordance with Rule 1:20-4(d)(i)." 5 The Disciplinary Rules of the Code of Professional Responsibility and Code of Judicial Conduct of the American Bar Association, with amendment and supplementation, have been adopted by the New Jersey Supreme Court as the applicable standard of conduct for members of the bar and the judges of New Jersey. New Jersey Court Rule 1:14. 6 DR 7-107 deals with "Trial Publicity" and states: "(D) During the selection of a jury or the trial of a criminal matter, a lawyer or law firm associated with the prosecution or defense of a criminal matter shall not make or participate in making an extra-judicial statement that he expects to be disseminated by means of public communication and that relates to the trial, parties, or issues in the trial or other matters that are reasonably likely to interfere with a fair trial . . . ." 7 The majority concluded that the hearings are designed to elicit facts, not legal arguments, as indicated by the presence of nonlawyers. The court also found that the ability to raise constitutional claims before the Ethics Committee does not constitute a meaningful opportunity to have constitutional questions adjudicated. No formal opinion is filed by the District Ethics Committee. The Third Circuit distinguished Gipson v. New Jersey Supreme Court, 558 F.2d 701 (CA3 1977), on the ground that in Gipson the attorney being disciplined was already subject to the state-court action at the time the federal proceeding had been initiated. Judge Adams, concurring, emphasized that state courts have the primary responsibility to discipline their bar and, in general, the federal judiciary is to exercise no supervisory powers. Judge Weis, dissenting, argued that respondents have full opportunity in the New Jersey proceedings to raise constitutional issues, concluding that the disciplinary proceedings are not a series of separate segments before independent bodies but are part of a whole. Judge Weis also concluded that there was nothing to prevent the Ethics Committee from considering constitutional claims. 8 The panel majority noted that no rule existed at the time of the District Court's decision to assure the Court of Appeals that the New Jersey Supreme Court would consider the constitutional claims. The court also concluded that the possibility of a formal procedure of the New Jersey court for consideration of constitutional claims does not moot this case because the underlying dispute as to the validity of the rules still remains. Judge Weis, again dissenting, concluded that no justiciable controversy remained as to the issue in the Court of Appeals and recommended that the case be remanded and dismissed as moot. 9 Rule 1:20-4(d) states: "(i) Interlocutory Review. An aggrieved party may file a motion for leave to appeal with the Supreme Court to seek interlocutory review of a constitutional challenge to proceedings pending before the District Ethics Committee or the Disciplinary Review Board. The motion papers shall conform to R. 2:8-1. Leave to appeal may be granted only when necessary to prevent irreparable injury. If leave to appeal is granted, the record below may, in the discretion of the Court, be supplemented by the filing of briefs and oral argument. "(ii) Final Review. In any case in which a constitutional challenge to the proceedings has been properly raised below and preserved pending review of the merits of the disciplinary matter by the Supreme Court, the aggrieved party may, within 10 days of the filing of the report and recommendation of the Disciplinary Review Board, seek the review of the Court by proceeding in accordance with the applicable provisions of R. 1:19-8." 10 Samuels v. Mackell, 401 U.S. 66, 91 S.Ct. 764, 27 L.Ed.2d 688 (1971), concluded that the same comity and federalism principles govern the issuance of federal-court declaratory judgments concerning the state statute that is the subject of the ongoing state criminal proceeding. 11 See M. Shoaf, State Disciplinary Enforcement Systems Structural Survey (ABA National Center for Professional Responsibility 1980). The New Jersey allocation of responsibility is consistent with § 2.1 of the ABA Standards for Lawyer Discipline and Disability Proceedings (Proposed Draft 1978), which states that the "[u]ltimate and exclusive responsibility within a state for the structure and administration of the lawyer discipline and disability system and the disposition of individual cases is within the inherent power of the highest court of the state." The rationale for vesting responsibility with the judiciary is that the practice of law "is so directly connected and bound up with the exercise of judicial power and the administration of justice that the right to define and regulate it naturally and logically belongs to the judicial department." Id., commentary to § 2.1. 12 The New Jersey Supreme Court has concluded that bar disciplinary proceedings are neither criminal nor civil in nature, but rather are sui generis. In re Logan, 70 N.J. 222, 358 A.2d 787 (1976). See also ABA Standards for Lawyer Discipline and Disability Proceedings § 1.2 (Proposed Draft 1978). As recognized in Juidice v. Vail, 430 U.S. 327, 97 S.Ct. 1211, 51 L.Ed.2d 376 (1977), however, whether the proceeding "is labeled civil, quasi-criminal, or criminal in nature," the salient fact is whether federal-court interference would unduly interfere with the legitimate activities of the state. Id., 430 U.S., at 335-336, 97 S.Ct., at 1217. The instant case arose before the 1978 rule change. In 1978 the New Jersey Supreme Court established a Disciplinary Review Board charged with review of findings of District Ethics Committees. Nothing in this rule change, however, altered the nature of such proceedings. The responsibility under Art. 6, § 2, ¶ 3, remains with the New Jersey Supreme Court. 13 The role of local ethics or bar association committees may be analogized to the function of a special master. Anonymous v. Association of Bar of City of New York, 515 F.2d 427 (CA2), cert. denied, 423 U.S. 863, 96 S.Ct. 122, 46 L.Ed.2d 92 (1975). The essentially judicial nature of disciplinary actions in New Jersey has been recognized previously by the federal courts. In Gipson v. New Jersey Supreme Court, 558 F.2d 701 (1977), the United States Court of Appeals for the Third Circuit agreed that "incursions by federal courts into ongoing [New Jersey] disciplinary proceedings would be peculiarly disruptive of notions of comity." Id., at 704. 14 This case is distinguishable from Steffel v. Thompson, 415 U.S. 452, 462, 94 S.Ct. 1209, 1217, 39 L.Ed.2d 505 (1974), in which there was no ongoing state proceeding to serve as a vehicle for vindicating the constitutional rights of the federal plaintiff. This case is also distinguishable from Gerstein v. Pugh, 420 U.S. 103, 108, n.9, 95 S.Ct. 854, 860, n.9, 43 L.Ed.2d 54 (1975), in which the issue of the legality of a pretrial detention could not be raised in defense of a criminal prosecution. See also Juidice v. Vail, 430 U.S., at 337, 97 S.Ct., at 1218. 15 In addition, after the filing of the writ of certiorari the New Jersey Supreme Court amended the state bar disciplinary rules to expressly permit a motion directly to the New Jersey Supreme Court for interlocutory adjudication of constitutional issues. Rule 1:20-4(d)(i). See n. 9, supra. Even if interlocutory review is not granted, constitutional issues are preserved for consideration by the New Jersey Supreme Court. Rule 1:20-2(j). The New Jersey Supreme Court reviews all disciplinary actions except the issuance of private letters of reprimand. Rule 1:20-4. Rule 1:20-2(j), however, requires that all constitutional issues be withheld for consideration by the Supreme Court as part of its review of the decision of the Disciplinary Review Board. This appears to provide for Supreme Court review of constitutional challenges even when a private reprimand is made. 16 Indeed, the decision of the New Jersey Supreme Court to consider respondent Hinds' constitutional challenges indicates that the state court desired to give Hinds a swift judicial resolution of his constitutional claims. 17 It is not clear whether the Court of Appeals decided whether abstention would be proper as to the respondent organizations who are not parties to the state disciplinary proceedings. We leave this issue to the Court of Appeals on remand.

89

457 U.S. 569 102 S.Ct. 2597 73 L.Ed.2d 227 Richard S. SCHWEIKER, Secretary of Health and Human Services, Appellantv.George HOGAN, et al. No. 81-213. Argued March 24, 1982. Decided June 21, 1982. Syllabus Section 1903(f) of the Social Security Act provides that federal reimbursement to States electing to provide Medicaid benefits to the "medically needy" is available only if the income of those persons, after deduction of incurred medical expenses, is less than 133 1/3% of the state Aid to Families With Dependent Children (AFDC) payment level. Section 1903(f) specifically excepts from this rule the "categorically needy"—those receiving Supplemental Security Income (SSI) because of lack of income to meet their basic needs. As applied in Massachusetts, § 1903(f) results in a distribution of Medicaid benefits to recipients of SSI that is more generous than the distribution of such benefits to persons who are self-supporting. Appellees, each of whom (or his spouse) receives Social Security benefits in an amount that renders him ineligible for either SSI benefits or state supplementary payments, filed suit in Federal District Court, alleging that § 1903(f), as applied in Massachusetts, violates the equal protection component of the Fifth Amendment. Appellees asserted that, since 133 1/3% of the Massachusetts AFDC payment level is for them lower than the SSI payment level, they are ineligible for Medicaid until their income, after deduction of incurred medical expenses, is less than that of SSI payment recipients, and that because of the Social Security benefits which they receive, appellees thus have less income available for nonmedical expenses than individuals who—possibly because they never worked and receive no Social Security benefits—are dependent upon public assistance for support. The District Court entered judgment for appellees. Held: 1. There is no merit to appellees' contention that the Social Security Act itself compels the conclusion that, if Medicaid services are provided to the "medically needy," those persons may not be forced to incur medical expenses that would reduce their remaining income below the applicable public assistance standard. The legislative history of the Medicaid provisions of the Act does not justify a departure from the literal and clear language of § 1903(f). Nor does § 1903(f)'s literal language conflict with any other provision of the Act. Moreover, adherence to that section's language is consistent with its interpretation by the Secretary of Health and Human Services. Thus, the discrimination challenged in this case is required by the Social Security Act. Pp. 584-588. 2. As applied in Massachusetts, § 1903(f) does not violate constitutional principles of equal treatment. While powerful equities support appellees' claim of unfair treatment insofar as they receive less medical assistance and have less income remaining for their nonmedical needs than do SSI recipients, a belief that an Act of Congress may be inequitable or unwise is an insufficient basis on which to conclude that it is unconstitutional. The optional character of the congressional scheme—whereby participating States must provide Medicaid benefits to the categorically needy but may elect not to provide any benefits at all to the medically needy—does not itself violate constitutional principles of equality. Since a State may deny all benefits to the medically needy—while providing benefits to the categorically needy and rendering some persons who are on public assistance better off than others who are not—it may narrow the gap between the two classes by providing partial benefits to the medically needy, even though certain members of that class may remain in a position less fortunate than those on public assistance. The fact that Massachusetts has provided Medicaid benefits to the medically needy does not force it to make immediate medical need the sole standard in its entire Medicaid program. Pp. 588-593 501 F.Supp. 1129, reversed and remanded. George W. Jones, Chicago, Ill., for the appellant, pro hac vice, by special leave of Court. William H. Simon, Jamaica Plain, Mass., for appellees. Justice STEVENS delivered the opinion of the Court. 1 At issue in this case are the meaning and validity of § 1903(f) of the Social Security Act, 81 Stat. 898, as amended, 42 U.S.C. § 1396b(f). As applied in Massachusetts, that provision results in a distribution of Medicaid benefits to recipients of Supplemental Security Income (SSI)—a class of aged, blind, or disabled persons who lack sufficient income to meet their basic needs—that is more generous than the distribution of such benefits to persons who are self-supporting. Appellees are members of the latter class. Because they must incur medical expenses—for which they are never reimbursed—before they become eligible for Medicaid, they have less income available for their nonmedical needs than the recipients of SSI. The District Court concluded that this discrimination was irrational and held that § 1903(f) was unconstitutional. Hogan v. Harris, 501 F.Supp. 1129 (Mass.1980). We disagree and reverse. 2 The statutory provisions governing the Medicaid program are complex. See 42 U.S.C. § 1396 et seq. (1976 ed. and Supp. IV). We first consider the history of the specific provisions at issue in this case, then relate the circumstances that gave rise to the present controversy, and finally address the two legal issues that are presented. 3 * Section 1903(f) of the Social Security Act (Act) was enacted in 1968. To understand the present controversy, however, it is necessary to consider amendments to the Act made in 1965, 1967, and 1972. A. 4 The Medicaid program was established in 1965 in Title XIX of the Act "for the purpose of providing federal financial assistance to States that choose to reimburse certain costs of medical treatment for needy persons." Harris v. McRae, 448 U.S. 297, 301, 100 S.Ct. 2671, 2680, 65 L.Ed.2d 784. Section 1902(a)(10) of the Act, 42 U.S.C. § 1396a(a)(10), sets forth the basic scope of the program, which has not changed significantly from its enactment in 1965. See 79 Stat. 345. Participating States are required to provide Medicaid coverage to certain individuals—now described as the "categorically needy"; at their option States also may provide coverage (and receive partial federal reimbursement) to other individuals—described as the "medically needy." See Schweiker v. Gray Panthers, 453 U.S. 34, 37, 101 S.Ct. 2633, 2636, 69 L.Ed.2d 460.1 These classes are defined by reference to other federal assistance programs. 5 In 1965, federal assistance programs existed for the aged, the blind, the disabled, and families with dependent children.2 At that time, each of these programs was administered by the States, which established both the "standard of need" and the "level of benefits." See Jefferson v. Hackney, 406 U.S. 535, 92 S.Ct. 1724, 32 L.Ed.2d 285; Rosado v. Wyman, 397 U.S. 397, 90 S.Ct. 1207, 25 L.Ed.2d 442.3 In establishing the Medicaid program, Congress required participating States to provide medical assistance to individuals who received cash payments under one of these assistance programs. 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(A). The House Report explained: "These people are the most needy in the country and it is appropriate for medical care costs to be met, first, for these people."4 They are the "categorically needy." 6 Congress also provided that a participating State could offer Medicaid benefits to individuals who fell within one of the categories for which federal assistance was available but whose income made them ineligible for aid under those programs. These individuals were deemed "less needy"5 and could receive assistance only if their income and resources were insufficient "to meet the costs of necessary medical or remedial care and services." 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(C). In 1965, no limit was placed on the extent to which federal reimbursement was available for optional coverage that States elected to provide to these persons who might become "medically needy."6 7 Since States established the income limits for the categorical assistance programs, they also established the income limits for the "categorically needy" under the Medicaid program. In addition, participating States established the eligibility standards for the optional coverage provided to the "medically needy." In § 1902(a)(17) of the Act, 42 U.S.C. § 1396a(a)(17), however, Congress set forth certain requirements governing state standards for determining eligibility. In particular, Congress required States to "provide for flexibility in the application of such standards with respect to income by taking into account, except to the extent prescribed by the Secretary, the costs (whether in the form of insurance premiums or otherwise) incurred for medical care or any other type of remedial care recognized under State law." 79 Stat. 346; see 42 U.S.C. § 1396a(a)(17).7 8 Most States promptly elected to participate in the Medicaid program.8 Many of these States also chose to provide Medicaid coverage to the "medically needy." Within a year, Congress recognized that it was fiscally improvident to rely exclusively on the States to set income limits for both aspects of the Medicaid program. See H.R.Rep. No. 2224, 89th Cong., 2d Sess., 1-3 (1966). It cautioned States "to avoid unrealistic levels of income and resources for title XIX eligibility purposes." Id., at 3. B 9 In 1967, Congress placed a limit on federal participation in the Medicaid program. Representative Mills introduced a bill, sponsored by the Johnson administration, that would have made significant changes in both the Medicaid program and the categorical assistance programs. H.R. 5710, 90th Cong., 1st Sess. (1967). Under § 220 of H.R. 5710, a State participating in the Medicaid program would have been entitled to receive federal financial assistance for providing Medicaid benefits only to those persons whose income, after deduction of incurred medical expenses, was less than 150% of the highest of the State's categorical assistance standards of need.9 Section 202 of the bill would have required States to revise annually the standards of need under each of the categorical assistance programs to reflect changes in the costs of living and, in some circumstances, to pay 100% of the standard of need established under the programs. In support of this provision, the Secretary of the Department of Health, Education, and Welfare explained that "33 States provide less support for needy children [under the AFDC program] than the standards the States themselves have set as necessary to meet basic human needs."10 10 After extensive consideration, the House Ways and Means Committee reported out a substantially revised bill. H.R. 12080, 90th Cong., 1st Sess. (1967). The Committee Report described its primary proposed limitation on federal participation: 11 "Your committee is proposing . . . that Federal sharing will not be available for families whose income exceeds 1331/3 percent of the highest amount ordinarily paid to a family of the same size (without any income and resources) in the form of money payments under the AFDC program. (AFDC income limits are, generally speaking, the lowest that are used in the categorical assistance programs)." 1967 House Report, at 119. 12 As noted, see n. 10, supra, the amount of benefits paid in many States was less than the qualifying standard of need.11 The Committee Report explained the reasons for the move to limit federal participation in the Medicaid program. After noting that a few States had provided benefits beyond that anticipated by Congress, it stated: 13 "Your committee expected that the State plans submitted under title XIX would afford better medical care and services to persons unable to pay for adequate care. 14 It neither expected nor intended that such care would supplant health insurance presently carried or presently provided under collective bargaining agreements for individuals and families in or close to an average income range. Your committee is also concerned that the operation of some State plans may greatly reduce the incentives for persons aged 65 or over to participate in the supplementary medical insurance program [Medicare] of title XVIII of the Social Security Act, which was also established by the Social Security Amendments of 1965. The provisions of the bill are directed toward eliminating, insofar as Federal sharing is concerned, these clearly unintended and, in your committee's judgment, undesirable actual and potential effects of the legislation." Id., at 118. 15 In States that paid less than 75% of the AFDC standard of need, the House provision would have provided Medicaid benefits only to persons whose income, after deduction of incurred medical expenses, was less than the AFDC standard of need.12 16 The Committee proposal was severely criticized on the House floor.13 It nevertheless was passed by the House and sent to the Senate.14 The Senate returned a substantially different bill and the matter was referred to conference.15 17 The Conference Committee adopted the House 133 1/3% AFDC payment standard. H.R.Conf.Rep. No. 1030, 90th Cong., 1st Sess., 63 (1967), U.S.Code Cong. & Admin.News 1967, p. 2834. It added, however, an express exception for the categorically needy. Ibid. Opposition to the Conference proposal was voiced in both the House and the Senate.16 The 133 1/3% AFDC payment standard nevertheless was approved by Congress and enacted into law as § 1903(f) of the Social Security Act. See 81 Stat. 898, as amended, 42 U.S.C. § 1396b(f).17 C 18 In 1972, Congress replaced three of the four state-administered categorical assistance programs with a new federal program entitled Supplemental Security Income for the Aged, Blind, and Disabled (SSI), 42 U.S.C. § 1381 et seq. (1976 ed. and Supp.IV).18 The SSI program establishes a federally guaranteed minimum income for the aged, blind, and disabled. See Schweiker v. Wilson, 450 U.S. 221, 223, 101 S.Ct. 1074, 1077, 67 L.Ed.2d 186. Under the program, however, the States may (and in some cases must) raise that minimum standard and supplement the benefits provided by the Federal Government. See 42 U.S.C. § 1382e (1976 ed. and Supp.IV). Moreover, if supplemental payments are made to persons who would be eligible for SSI benefits except for the amount of their income, the State also may provide Medicaid benefits to those persons. See 42 U.S.C. § 1396a(a)(10)(C)(ii).19 II 19 The Commonwealth of Massachusetts has chosen to participate in the Medicaid program and to provide benefits—to the extent that federal financial assistance is available—to the "medically needy." The State also has elected to make supplementary payments to individuals who are eligible for SSI benefits or who would be eligible except for their income. Finally, the State has chosen to provide Medicaid benefits to those persons who receive supplemental payments. In Massachusetts, 133 1/3% of the appropriate state AFDC payment amount is less in some cases than the combined federal SSI and state supplementary payment level.20 20 Appellees filed this suit in 1980 in federal court, contending that § 1903(f) of the Act—as applied in Massachusetts violates the equal protection component of the Fifth Amendment.21 Each of the appellees is either aged, blind, or disabled, but they are not categorically needy. For each appellee or his spouse was employed at one time and paid "Social Security" taxes. Each appellee (or his spouse) currently receives Social Security benefits (Federal Old-Age, Survivors, and Disability Insurance, 42 U.S.C. § 401 et seq. (1976 ed. and Supp.IV)) in an amount that renders him ineligible for either SSI benefits or state supplementary payments. Appellees challenged the fact that, since 133 1/3% of the Massachusetts AFDC payment level is for them lower than the SSI payment level, they are ineligible for Medicaid until their income, after deduction of incurred medical expenses, is less than that of SSI payment recipients. By reason of the Social Security benefits that they receive, appellees thus have less income available for nonmedical expenses than individuals who possibly because they never worked and receive no Social Security benefits—are dependent upon public assistance for support.22 21 The District Court granted appellees' motion for partial summary judgment.23 It ruled that the Massachusetts Medicaid program was unconstitutional insofar as it forced Social Security recipients to incur medical expenses that reduced their remaining income to an amount below SSI payment levels. The court later declared explicitly that § 1903(f) of the Act, 42 U.S.C. § 1396b(f), is unconstitutional as applied in Massachusetts. App. to Juris. Statement 25a. We noted probable jurisdiction. 454 U.S. 891, 102 S.Ct. 385, 70 L.Ed.2d 204. III 22 In this Court, for the first time, appellees contend that the Social Security Act itself compels the conclusion that, if Medicaid services are provided to the "medically needy," those persons may not be forced to incur medical expenses that would reduce their remaining income below the applicable public assistance standard. Although appellees did not advance this argument in the District Court, they are not precluded from asserting it as a basis on which to affirm that court's judgment.24 "Where a party raises both statutory and constitutional arguments in support of a judgment, ordinarily we first address the statutory argument in order to avoid unnecessary resolution of the constitutional issue." Blum v. Bacon, 457 U.S. 132, 137, 102 S.Ct. 2355, 2359, 72 L.Ed.2d 728. See Harris v. McRae, 448 U.S., at 306-307, 100 S.Ct., at 2682-2683. 23 Appellees contend that a "fundamental Congressional purpose in the creation of the medically needy feature of Title XIX was to achieve equity between public assistance recipients and others similarly situated." Brief for Appellees 12. In support of this contention, appellees cite the requirement first imposed in 1965 that States "include reasonable standards (which shall be comparable for all groups ) for determining eligibility for and the extent of medical assistance under the plan . . .," 79 Stat. 346 (emphasis added), as amended, 42 U.S.C. § 1396a(a)(17), and note the statements in the legislative history that a State could not require an individual to use, for medical expenses, income "which would bring the individual below the test of eligibility under the State plan." See n. 7, supra. 24 Moreover, appellees contend that this "comparability requirement" was not changed by the enactment of § 1903(f) in 1968. Appellees argue that the separate bills passed in both the House and the Senate would have affected both the categorically and the medically needy.25 Only when the Conference Committee accepted the House provision and added an exception for the categorically needy, appellees argue, did the 1968 modification potentially change the comparability requirement between the two groups. Appellees assert that such a change was not intended; rather, they argue that the exception for the categorically needy was added only to ensure that they would not be adversely affected by § 1903(f). Appellees assert that the medically needy were not similarly excepted from the 133 1/3% rule in those States in which that figure was less than the applicable standard of need because, in 1967, those States did not have medically needy programs. 25 Thus, appellees urge that we construe § 1903(f) to require the medically needy to incur medical expenses until their income is 133 1/3% of the AFDC payment amount or—to maintain comparability—100% of the combined SSI-state supplementary payment level if that figure is higher. Appellees argue that the legislative history of the 1965 and 1967 Amendments to the Social Security Act justifies a departure from the literal language of § 1903(f) and the Secretary's interpretation of that provision. 26 We cannot agree. Congress explicitly stated in § 1903(f) that federal reimbursement for benefits provided to the medically needy was available only if the income of those persons, after the deduction of incurred medical expenses, was less than 133 1/3% of the state AFDC payment level. In specifically excepting the categorically needy from this rule, Congress recognized that this amount could be lower than categorical assistance eligibility levels. There is no basis in either the statute or the legislative history for appellees' argument that Congress implicitly "assumed" that those States in which 133 1/3% of the AFDC payment level was less than the applicable standard of need simply would not provide assistance to the medically needy. Even if this were true in 1967, the Medicaid program then was less than two years old; Congress was aware that many States were in the process of adopting Medicaid programs.26 To assume that Congress was unaware that § 1903(f)—which applied only to the medically needy—could operate in those States—which Congress knew existed—in which 133 1/3% of the AFDC payment amount was less than the applicable standard of need is to demean the intelligence of the Congress. We are not prepared to interpret a statute on the basis of an unsupported assumption that Congress had little idea of what it was doing.27 27 The literal and clear language of § 1903(f) does not conflict with any other provision of the Act. In both § 1902(a)(10) and § 1902(a)(17), see 79 Stat. 345-346, Congress required comparability among the various "categories" for which federal assistance was available, but these provisions did not require that the medically needy be treated comparably to the categorically needy in all respects. See n. 6, supra.28 Indeed, such a broad comparability requirement would be inconsistent with the fact that Congress provided in 1965 that the medically needy could be excluded entirely from the Medicaid program. Moreover, § 1903(f) is not inconsistent with the congressional intent, see n. 7, supra, that medical expenses be considered in determining, where appropriate, an individual's eligibility for Medicaid. In § 1903(f) Congress determined that federal assistance would not be available for payments made to individuals whose income, after deduction of incurred medical expenses, was greater than 133 1/3% of applicable state AFDC payments. Congress determined that, so long as an individual retained that level of income to meet basic needs, he need not receive reimbursement for medical expenses. That income level might appear unreasonably low, but it is the level that Congress chose. We find no inconsistency between § 1903(f) and § 1902(a)(17). 28 In sum, we see no reason to ignore the literal language of § 1903(f). Moreover, this analysis is consistent with the Secretary's interpretation of that statutory provision. "We have often noted that the interpretation of an agency charged with the administration of a statute is entitled to substantial deference." Blum v. Bacon, supra, at 141, 102 S.Ct., at 2361. We hold that the discrimination challenged in this case is required by the Social Security Act. IV 29 Appellees also contend—and the District Court held—that § 1903(f), as applied in Massachusetts, irrationally discriminates between the categorically and the medically needy.29 The unfairness of the statute stems from the fact that appellees receive less medical assistance, and have less income remaining for their nonmedical needs, than do SSI recipients. The unfairness is accentuated by the fact that the disfavored class consists largely of persons who worked and paid taxes to provide for their retirement while the favored class includes persons who may never have done so. Powerful equities unquestionably support the appellees' claim of unfair treatment. 30 A belief that an Act of Congress may be inequitable or unwise is of course an insufficient basis on which to conclude that it is unconstitutional. Moreover, the validity of a broad legislative classification is not properly judged by focusing solely on the portion of the disfavored class that is affected most harshly by its terms. Califano v. Jobst, 434 U.S. 47, 98 S.Ct. 95, 54 L.Ed.2d 228. In this case, Congress has differentiated between the categorically needy—a class of aged, blind, disabled, or dependent persons who have very little income—and other persons with similar characteristics who are self-supporting. Members of the former class are automatically entitled to Medicaid; members of the latter class are not eligible unless a State elects to provide benefits to the medically needy and unless their income, after consideration of medical expenses, is below state standards of eligibility.30 31 According to the congressional scheme, then, the medically needy may be excluded entirely from the Medicaid program. Before considering the constitutional constraints that may exist if a State chooses to provide benefits to that class, it is appropriate to confront the more basic question whether the optional character of the program for the medically needy is itself constitutionally permissible. 32 In establishing public assistance programs, Congress often has determined that the Federal Government cannot finance a program that provides meaningful benefits in equal measure to everyone. Both federal and state funds available for such assistance are limited. In structuring the Medicaid program, Congress chose to direct those limited funds to persons who were most impoverished and who—because of their physical characteristics—were often least able to overcome the effects of poverty. The legislative history of the 1965 Amendments makes clear that this group was not chosen for administrative convenience. "These people are the most needy in the country and it is appropriate for medical care costs to be met, first, for these people."31 A decision to allocate medical assistance benefits only to the poor does not itself violate constitutional principles of equality; in terms of their ability to provide for essential medical services, the wealthy and the poor are not similarly situated and need not be treated the same. It is rational to distribute public assistance benefits on the basis of the income and resources available to potential recipients. 33 In choosing to require coverage only for the categorically needy, Congress permitted States to exclude from the program many persons who—by reason of large medical expenses—often were just as much in need of medical assistance as the categorically needy. Yet Congress found these persons "less needy." 1965 House Report, at 66. By reason of the greater income available to them, as a class these persons generally are better able to provide for their medical needs. In the legislative history of the 1967 Amendments, see supra, at 577-580, and n. 14, Congress noted that these persons often are able to prepare for future medical expenses through private insurance or through participation in the Medicare program. 34 In Fullington v. Shea, 404 U.S. 963, 92 S.Ct. 345, 30 L.Ed.2d 282, this Court affirmed a decision of a three-judge District Court holding that it was constitutional for the State of Colorado to provide benefits to the categorically needy but not to the medically needy. We decided Fullington summarily. It is clear that a decision to allocate scarce assistance benefits on the basis of an assumption that persons with greater incomes generally are better able to prepare for future medical needs is not inconsistent with constitutional principles of equal treatment. In other words, it is rational to define need on the basis of income, even though some persons with greater income—who have been unable or unwilling to save enough of their earnings to prepare for future medical needs—may actually be in greater need of assistance than those with less gross income. Although some "medically needy" persons have less income available for nonmedical expenses than those who receive categorical assistance, the broad legislative classification does not involve the type of arbitrariness that is constitutionally offensive.32 35 Appellees do not challenge the decision in Fullington. They do not contend that it is irrational to deny benefits entirely to the medically needy. Thus, they do not challenge the line drawn by Congress to separate the class that receives favored treatment from the class that does not. Appellees attack only the manner in which one of the separate classes is affected by the program. They argue that if medical benefits are made available to a class of persons who are not categorically needy, it is constitutionally impermissible to deny them benefits if their income, after the deduction of incurred medical expenses, is lower than that of an individual who receives public assistance. 36 In view of the unchallenged decision in Fullington, appellees' constitutional argument is self-defeating. The injury that they regard as inconsistent with constitutional principles of equal treatment could be avoided by denying them all Medicaid benefits, thus placing them in a worse position financially than they are in now. No interest in "equality" could be furthered by such a result. If a State may deny all benefits to the medically needy—while providing benefits to the categorically needy and rendering some persons who are on public assistance better off than others who are not—a State surely may narrow the gap between the two classes by providing partial benefits to the medically needy, even though certain members of that class may remain in a position less fortunate than those on public assistance. 37 The validity of the distinction between the categorically needy and the medically needy is not undermined by § 1903(f), because the impact of that provision falls entirely on persons who are not within the categorically needy class. See n. 30, supra. The function of the 1331/3% AFDC payment rule is to place a limit on the availability of reimbursement for potential members of the "medically needy" class. That rule prevents some persons (although not the appellees) from qualifying as medically needy; it also determines the extent to which the medically needy are reimbursed for their medical expenses. Yet appellees do not challenge the fact that, among persons who do not receive public assistance, some are treated differently from others. In other words, they do not complain of any discrimination within the class (all persons who are not categorically needy) in which the rule performs its entire function.33 Nor do they argue that Congress chose an eligibility level that is unrelated to ability to provide for medical needs. 38 The fact that Massachusetts, unlike the State of Colorado in Fullington, has provided Medicaid benefits to the medically needy and in doing so has defined eligibility for persons who are not categorically needy on the basis of incurred medical expenses—does not force it to make immediate medical need the sole standard in its entire Medicaid program. Massachusetts in essence has determined that those individuals whose gross income is greater than public assistance levels are ineligible for Medicaid, unless medical expenses in any computation period reduce available income to 133 1/3% of the state AFDC payment level. By adding the qualifying clause, which the State of Colorado did not, Massachusetts did not offend any constitutional interest in equality. Accordingly, without endorsing the wisdom of the particular standard that Congress selected—a matter that is not for us to consider—we conclude that it violates no constitutional command. The judgment of the District Court is reversed. The case is remanded for further proceedings consistent with this opinion. 39 It is so ordered. 1 But see n. 18, infra. 2 These programs were entitled: Old Age Assistance (OAA), 42 U.S.C. § 301 et seq. (1970 ed.); Aid to the Blind, § 1201 et seq.; Aid to the Permanently and Totally Disabled, § 1351 et seq.; and Aid to Families with Dependent Children (AFDC), § 601 et seq. See also 42 U.S.C. §§ 1381-1385 (1970 ed.). These programs are of course fundamentally different from Old Age, Survivors, and Disability Insurance (OASDI or Social Security), 42 U.S.C. § 401 et seq. 3 In many States, the "level of benefits" did not raise an individual's income to the "standard of need." The standard of need determined eligibility for some benefits; often the benefits provided, however, were merely a fraction of the difference between the individual's income and the defined standard of need. See Jefferson v. Hackney. The standards of need also typically varied from program to program. 4 H.R.Rep. No. 213, 89th Cong., 1st Sess., 66 (1965) (1965 House Report). 5 Ibid. See also S.Rep. No. 404, 89th Cong., 1st Sess., pt. 1, p. 77 (1965) (1965 Senate Report), U.S.Code Cong. & Admin.News 1965, p. 1943. 6 The 1965 Act contained certain requirements governing the comparative treatment of different beneficiaries under the Act. It provided that the medical assistance afforded to an individual who qualified under any categorical assistance program could not be different from that afforded to an individual who qualified under any other program. 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(B)(i). In other words, the amount, duration, and scope of medical assistance provided to an individual who qualified to receive assistance for the aged could not be different from the amount, duration, and scope of benefits provided to an individual who qualified to receive assistance for the blind. "This will assure comparable treatment for all of the needy under the federally aided categories of assistance and will eliminate some of the unevenness which has been apparent in the treatment of the medical needs of various groups of the needy." 1965 House Report, at 66. See also 1965 Senate Report, at 77, U.S.Code Cong. & Admin.News 1965, p. 2017. A similar "comparability" requirement among the aged, blind, disabled, and dependent applied to the optional distribution of benefits to the "medically needy." If a State elected to provide benefits to one group, it was obligated to provide benefits to the others, and "the determination of financial eligibility must be on a basis that is comparable as among the people who, except for their income and resources, would be recipients of money for maintenance under the other public assistance programs." 1965 House Report, at 67; see also 1965 Senate Report, at 77, U.S.Code Cong. & Admin.News 1965, p. 2017. 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(C)(i). In addition, the benefits provided to each categorical group of the medically needy were required to be equal in amount, duration, and scope. 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(C)(ii). In its provision for "comparability among the various categorical groups of needy people," 1965 House Report, at 67, the Act required comparability in the criteria used to determine eligibility for each group. 79 Stat. 346, as amended, 42 U.S.C. § 1396a(a)(17). See also 1965 House Report, at 67; 1965 Senate Report, at 77, U.S.Code Cong. & Admin.News 1965, p. 2018 ("Although States may set a limitation on income and resources which individuals may hold and be eligible for aid, they must do so by maintaining a comparability among the various categorical groups of needy people"). Finally, the Act provided that the assistance provided to the "medically needy" could not be greater in amount, duration, or scope than the assistance provided to the "categorically needy." 79 Stat. 345, as amended, 42 U.S.C. § 1396a(a)(10)(B)(ii). "This was included in order to make sure that the most needy in a State receive no less comprehensive care than those who are not as needy." 1965 House Report, at 67; see also 1965 Senate Report, at 77, U.S.Code Cong. & Admin.News 1965, p. 2017. 7 In its discussion of this portion of the statute, the 1965 House Report, at 68, explains: "The bill also contains a provision designed to correct one of the weaknesses identified in the medical assistance for the aged program. Under the current provisions of Federal law, some States have enacted programs which contain a cutoff point on income which determines the financial eligibility of the individual. Thus, an individual with an income just under the specified limit may qualify for all of the aid provided under the State plan. Individuals, however, whose income exceeds the limitation adopted by the State are found ineligible for the medical assistance provided under the State plan even though the excess of the individual's income may be small when compared with the cost of the medical care needed. In order that all States shall be flexible in the consideration of an individual's income, your committee bill requires that the States standards for determining eligibility for and the extent of medical assistance shall take into account, except to the extent prescribed by the Secretary, the cost—whether in the form of insurance premiums or otherwise—incurred for medical care or any other type of remedial care recognized under State law. Thus, before an individual is found ineligible for all or part of the cost of his medical needs, the State must be sure that the income of the individual has been measured in terms of both the State's allowance for basic maintenance needs and the cost of the medical care he requires." See also 1965 Senate Report, at 78-79, U.S.Code Cong. & Admin.News 1965, pp. 2018-2019. To this extent, the House Report mirrors the statutory language. In further describing this provision, however, the 1965 House Report, at 68, immediately continues: "The State may require the use of all the excess income of the individual toward his medical expenses, or some proportion of that amount. In no event, however, with respect to either this provision or that described below with reference to the use of deductibles for certain items of medical service, may a State require the use of income or resources which would bring the individual below the test of eligibility under the State plan. If the test of eligibility should be $2,000 a year, an individual with income in excess of that amount shall not be required to use his income to the extent he has remaining less than $2,000. This action would reduce the individual below the level determined by the State as necessary for his maintenance." See also 1965 Senate Report, at 79. This additional comment has no direct foundation in the statutory language of § 1902(a)(17). See 42 U.S.C. § 1396a(a)(17). 8 See H.R.Rep. No. 544, 90th Cong., 1st Sess., 117 (1967) (1967 House Report). 9 This provision, of course, would have had no effect on the "categorically needy," since their income was necessarily less than 150% of the highest categorical assistance standard of need. 10 President's Proposals for Revision in the Social Security System: Hearings on H.R. 5710 before the House Committee on Ways and Means, 90th Cong., 1st Sess., 118 (1967). In January 1965, there were 21 States that paid less than 75% of the standard of need established for a family of four under the State's AFDC program. Id., at 119. 11 The proposed bill also provided another limit on federal participation. It included a provision that set "a figure of 1331/3 percent of the average per capita income of a State as the upper limit on Federal sharing when applied to a family of four under the title XIX program." 1967 House Report, at 119. It is noteworthy that these proposals were not an insignificant part of what was—admittedly—a complex bill. In setting forth at the outset the "principal purposes of the bill," the House Report provides: "Fifth, to modify the program of medical assistance to establish certain limits on Federal participation in the program and to add flexibility in administration, the bill would— "(a) Impose a limitation on Federal matching at an income level related to payments for families receiving aid to families with dependent children or to the per capita income of the State, if lower." Id., at 5. 12 If the House bill applied to both the categorically needy and the medically needy, it could have resulted in the denial of Medicaid benefits to certain categorically needy individuals who although eligible for assistance under the State's standard of need—had an income that was higher than 133 1/3% of the amount the State actually paid to a qualifying individual with no income. The House bill did not, however, alter § 1902(a)(10) of the Act, 42 U.S.C. § 1396a(a)(10), which required participating States to provide Medicaid benefits to all of the categorically needy. 13 See 113 Cong.Rec. 23065 (1967) (remarks of Rep. King); id., at 23077 (remarks of Rep. Burke); id., at 23082 (remarks of Rep. Vanik); id., at 23084 (remarks of Rep. Bingham); id., at 23087 (remarks of Rep. Halpern); id., at 23093 (remarks of Rep. Ryan); id., at 23104 (remarks of Rep. Bingham); id., at 23125 (remarks of Rep. Boland); id., at 23128 (remarks of Rep. Kastenmeier). In particular, see id., at 23131 (remarks of Rep. Farbstein); id., at 23083 (remarks of Rep. Gilbert); id., at 23092 (remarks of Rep. Burton). 14 Representative Mills defended the bill against criticism that its treatment of those with income above the categorical assistance limit was unfair. He noted that it was "only because of what we walked into with this program that the committee has seen fit to put limits on it," id., at 23093, and added: "I do not think it is fair to tax people through the general funds of the Treasury to pay for the medical costs of those who undoubtedly have the means to buy insurance and to defray their own medical costs." Ibid. See also id., at 23061-23062 (remarks of Rep. Byrnes); id., at 23084-23085 (remarks of Rep. Hanley); id., at 23090 (remarks of Rep. Stratton); id., at 23090, 23091 (remarks of Rep. McCarthy); id., at 23105 (remarks of Rep. Taft); id., at 22783 (remarks of Rep. Quillen). 15 In hearings before the Senate Finance Committee, an HEW official recommended that the administration's proposal be adopted. He criticized the House bill and noted that, in States such as Indiana and Texas, 133% of the AFDC payment amount was less than the AFDC standard of need. Social Security Amendments of 1967: Hearings on H.R. 12080 before the Senate Committee on Finance, 90th Cong., 1st Sess., 280 (1967). He pointed out that such a standard could result in exclusion of some of the categorically needy, which he suggested probably had not been intended. Ibid. Senator Robert Kennedy also criticized the House proposal, noting that medically needy individuals would not be eligible for Medicaid in some States until their income, after deduction of incurred medical expenses, was less than the standards of need established for the categorically needy. Id., at 784. The Finance Committee subsequently proposed a bill that provided participating States with federal assistance for Medicaid expenditures made on behalf of any person whose income after the deduction of medical expenses was less than 150% of the OAA standard, which generally was the highest of the cash assistance standards. See S.Rep. No. 744, 90th Cong., 1st Sess., 177 (1967), U.S.Code Cong. & Admin.News 1967, p. 2834. The Senate bill also introduced a new formula for computing the amount of federal reimbursement under the Medicaid program that was designed to reduce federal matching funds for payments to the medically needy. Id., at 176-177. The proposals encountered resistance on the Senate floor. Senator Javits, speaking in support of an amendment offered by Senator Kuchel that would have substituted the proposals of the administration, criticized the Finance Committee bill on the ground that it discriminated against the medically needy. See 113 Cong.Rec. 33168, 33169 (1967). In response, Senator Long acknowledged that the bill discriminated against the medically needy, but explained that it "encourages the State to concentrate its medical assistance for those who are most in need, those who qualify for public welfare assistance." Id., at 33169, 33171. The Senate rejected the Kuchel amendment and adopted the Finance Committee bill. 16 See id., at 36380 (1967) (remarks of Rep. Burton); id., at 36381 (remarks of Rep. Gilbert); id., at 36385 (remarks of Rep. Reid); id., at 36387 (remarks of Rep. Ryan); id., at 36389 (remarks of Rep. Farbstein). In the Senate, Robert Kennedy complained that in Mississippi the 133 1/3% limitation amounted to an income level, after medical expenses had been incurred, of $80 per month for a family of four. Id., at 36784. Senator Mondale quoted the testimony in the Senate Hearings, see n. 15, supra, that in some States the 133 1/3% AFDC payment amount was less than the standard of need established under even the AFDC program. 113 Cong.Rec. 36819 (1967). 17 Title 42 U.S.C. § 1396b(f) provides: "(f) Limitation on Federal participation in medical assistance "(1)(A) Except as provided in paragraph (4), payment under the preceding provisions of this section shall not be made with respect to any amount expended as medical assistance in a calendar quarter, in any State, for any member of a family the annual income of which exceeds the applicable income limitation determined under this paragraph. "(B)(i) Except as provided in clause (ii) of this subparagraph, the applicable income limitation with respect to any family is the amount determined, in accordance with standards prescribed by the Secretary, to be equivalent to 1331/3 percent of the highest amount which would ordinarily be paid to a family of the same size without any income or resources, in the form of money payments, under the plan of the State approved under Part A of subchapter IV of this chapter. . . . . "(2) In computing a family's income for purposes of paragraph (1), there shall be excluded any costs (whether in the form of insurance premiums or otherwise) incurred by such family for medical care or for any other type of remedial care recognized under State law. . . . . "(4) The limitations on payment imposed by the preceding provisions of this subsection shall not apply with respect to any amount expended by a State as medical assistance for any individual— "(A) who is receiving aid or assistance under any plan of the State approved under subchapter I, X, XIV or XVI, or part A of subchapter IV, or with respect to whom supplemental security income benefits are being paid under subchapter XVI of this chapter, or "(B) who is not receiving such aid or assistance, and with respect to whom such benefits are not being paid, but (i) is eligible to receive such aid or assistance, or to have such benefits paid with respect to him, or (ii) would be eligible to receive such aid or assistance, or to have such benefits paid with respect to him if he were not in a medical institution, or "(C) with respect to whom there is being paid, or who is eligible, or would be eligible if he were not in a medical institution, to have paid with respect to him, a State supplementary payment and is eligible for medical assistance equal in amount, duration, and scope to the medical assistance made available to individuals described in section 1396a(a)(10)(A) of this title, but only if the income of such individual (as determined under section 1382a of this title, but without regard to subsection (b) thereof) does not exceed 300 percent of the supplemental security income benefit rate established by section 1382(b)(1) of this title, at the time of the provision of the medical assistance giving rise to such expenditure." 18 The SSI program is funded and administered by the Federal Government. As its name indicates, it replaced the categorical assistance programs for the aged, the blind, and the disabled. The AFDC program continues to be administered by the States and is only partially funded by the Federal Government. In some States the number of individuals eligible for SSI was significantly greater than the number of persons who had been eligible under the state-administered categorical assistance programs. See Schweiker v. Gray Panthers, 453 U.S. 34, 38, 101 S.Ct. 2633, 2637, 69 L.Ed.2d 460. Since recipients of categorical welfare assistance are also entitled to Medicaid benefits, the expansion of general welfare accomplished by the SSI program increased Medicaid obligations for some States. To guarantee that States would not, for that reason, withdraw from the Medicaid program, Congress offered what has become known as the "§ 209(b) option." Under it, States may elect to provide Medicaid assistance only to those individuals who would have been eligible under the state Medicaid plan in effect on January 1, 1972. See id., at 38-39, 10 S.Ct. at 2637-2638. Thus, in some States, Medicaid is not automatically available for all of the "categorically needy." Massachusetts is not a § 209(b) State. 19 There is a limit on federal participation in this aspect of the program. A State is entitled to federal financial assistance for providing Medicaid benefits to a state supplementary payment recipient only if his gross income is less than 300% of the applicable SSI income limitation. See 42 U.S.C. § 1396b(f)(4)(C); n. 17, supra. 20 There is no statutory requirement that state AFDC payment amounts be comparable to state supplemental benefits. 21 See Bolling v. Sharpe, 347 U.S. 497, 499, 74 S.Ct. 693, 694, 98 L.Ed. 884. Appellees also contended that certain state statutory provisions violated the Equal Protection Clause of the Fourteenth Amendment. 22 Appellees alleged that federal and state provisions require an individual to apply for and to accept all Social Security benefits for which he is eligible as a condition of application for SSI and Medicaid benefits. See 42 U.S.C. § 1382(e)(2). Appellees' grievances are best illustrated by the situation of appellee Hunter. The District Court found that Hunter had worked for 41 years and had paid Social Security taxes during that period. As a result, he received at the time of trial $534 per month in Social Security benefits, $20 of which apparently was disregarded in computing eligibility for SSI and state supplementary payments. As a result of his income, Hunter was ineligible for either SSI or state supplemental payments; the "standard of need" under those programs was $513 per month. If he had qualified, he of course would also have been eligible for Medicaid. Since the applicable AFDC payment amount in Massachusetts was $300, Hunter was ineligible for Medicaid until his income, after deduction of incurred medical expenses, was no higher than $400. Hunter regularly incurred over $200 each month in medical expenses; thus, by reason of his Social Security benefits, he had less income available for nonmedical needs ($400 per month) than he would have had on public assistance ($513 per month). In his case, a Social Security payment of $1 less each month ($534 less $20 less $1) would apparently have rendered him fully eligible for Medicaid. See Hogan v. Harris, 501 F.Supp. 1129, 1132 (Mass.1980). In other words, if his gross income were reduced by $1, he would receive over $100 in additional medical benefits and have that additional amount of income available for nonmedical needs. 23 The District Court certified a class "consisting of all (i) present and future Social Security recipients; (ii) who reside or will reside in Massachusetts; (iii) who are or will be disabled or 65 years old or older; (iv) who are or will be ineligible because of the amount of their incomes for Massachusetts Supplemental Security Income payments; and (v) who have or will have, as determined in accordance with the applicable Massachusetts Medicaid regulations, medical expenses not subject to payment by a third party which exceed the difference between their countable incomes under the Massachusetts Medicaid regulations and the applicable Massachusetts Supplemental Security Income standard." App. to Juris. Statement 23a-24a. 24 "It is well accepted . . . that without filing a cross-appeal or cross-petition, an appellee may rely upon any matter appearing in the record in support of the judgment below." Blum v. Bacon, 457 U.S., at 137, n.5, 102 S.Ct., at 2359, n.5. The statutory argument raised by the appellees, although not presented in the District Court, may be decided on the basis of the record developed in that court. 25 See n. 12, supra. Since the limitation in the Senate bill was set at 150% of the OAA assistance standard, by definition it would not likely have affected the categorically needy. In any event, appellees contend that both bills were consistent with a comparability requirement. 26 See 1967 House Report, at 117-118. 27 Moreover, appellees' "congressional ignorance" argument rests on another unsupportable premise. Appellees assume that the House bill—which they admit was vigorously debated—had a "comparable" effect on the categorically and the medically needy. That bill, however, did not propose an amendment to § 1902(a)(10) of the Act, 42 U.S.C. § 1396a(a)(10), which required that Medicaid coverage be provided to all the categorically needy. It is much more likely—in light of § 1902(a)(10)—that the House assumed that its proposed limits on federal participation in the Medicaid program would affect only the medically needy. See Hearings on H.R. 12080, supra n.15, at 280 (describing the possibility that the House bill would affect the categorically needy as a "drafting error"). This assumption was made explicit by the Conference Committee, which chose the House standard but added—with little discussion—a direct exception for the categorically needy. 28 Relying on 42 U.S.C. §§ 1396a(a)(10)(C)(i) and 1396a(a)(17), courts have concluded that certain treatment of the medically needy must be comparable to that afforded to the categorically needy. See Caldwell v. Blum, 621 F.2d 491 (CA2 1980), cert. denied, 452 U.S. 909, 101 S.Ct. 3039, 69 L.Ed.2d 412; Fabula v. Buck, 598 F.2d 869 (CA4 1979); Greklek v. Toia, 565 F.2d 1259 (CA2 1977), cert. denied, sub nom. Blum v. Toomey, 436 U.S. 962, 98 S.Ct. 3081, 57 L.Ed.2d 1128; Aitchison v. Berger, 404 F.Supp. 1137 (S.D.N.Y. 1975), aff'd, 538 F.2d 307 (CA2 1976), cert. denied, 429 U.S. 890, 97 S.Ct. 246, 50 L.Ed.2d 172. Whatever the scope of the requirement of comparability between the categorically and the medically needy, it is clear that the Act does not require the income of medically needy persons—after the deduction of incurred medical expenses—to be at least comparable to that of the categorically needy. 29 The discriminatory impact challenged in this case arises solely from the fact that Massachusetts has chosen to supplement SSI payments to an extent that exceeds 133 1/3% of state AFDC payment levels. It is not disputed that 133 1/3% of the Massachusetts AFDC payment level is higher than federal SSI benefit levels. See 45 Fed.Reg. 31782 (1980); 46 Fed.Reg. 27076 (1981). 30 Although the arguments in this case have focused on two classes, in fact there are three: (1) the categorically needy; and (2) all others, (a) some of whom have medical expenses that reduce their remaining income to a level that qualifies them as medically needy, and (b) some of whom are neither categorically needy nor medically needy. 31 1965 House Report, at 66. 32 See Schweiker v. Wilson, 450 U.S. 221, 238, 101 S.Ct. 1074, 1084, 67 L.Ed.2d 186 ("This Court has granted a 'strong presumption of constitutionality' to legislation conferring monetary benefits, Mathews v. De Castro, 429 U.S. 181, at 185, 97 S.Ct. 431, at 434, 50 L.Ed.2d 389, because it believes that Congress should have discretion in deciding how to expend necessarily limited resources"). The fact that the recipient of a governmental benefit—such as an indigent defendant who is represented by a public defender—may in some cases be better off after receiving the benefit than a wealthier person who did not qualify to receive it does not undermine the validity of the basis for determining eligibility. 33 The fact that the amount of benefits payable to persons within the medically needy class is determined on the basis of income remaining after medical expenses have been incurred does not impeach the rationality of defining the basic distinction between the categorically needy and all others on the basis of income before medical expenses are considered.

12

457 U.S. 440 102 S.Ct. 2525 73 L.Ed.2d 130 CONNECTICUT, et al., Petitionersv.Winnie TEAL et al. No. 80-2147. Argued March 29, 1982. Decided June 21, 1982. Syllabus Respondent black employees of a Connecticut state agency were promoted provisionally to supervisors. To attain permanent status as supervisors, they had to participate in a selection process that required, as a first step, a passing score on a written examination. Subsequently, an examination was given to 48 black and 259 white candidates. Fifty-four percent of the black candidates passed, this being approximately 68 percent of the passing rate for the white candidates. Respondent black employees failed the examination and were thus excluded from further consideration for permanent supervisory positions. They then brought an action in Federal District Court against petitioners (the State of Connecticut and certain state agencies and officials), alleging that petitioners had violated Title VII of the Civil Rights Act of 1964 by requiring, as an absolute condition for consideration for promotion, that applicants pass a written test that disproportionately excluded blacks and was not job related. In the meantime, before trial, petitioners made promotions from the eligibility list, the overall result being that 22.9 percent of the black candidates were promoted but only 13.5 percent of the white candidates. Petitioners urged that this "bottom-line" result, more favorable to blacks than to whites, was a complete defense to the suit. The District Court agreed and entered judgment for petitioners, holding that the "bottom line" percentages precluded the finding of a Title VII violation and that petitioners were not required to demonstrate that the promotional examination was job related. The Court of Appeals reversed, holding that the District Court erred in ruling that the examination results alone were insufficient to support a prima facie case of disparate impact in violation of Title VII. Held : Petitioners' nondiscriminatory "bottom line" does not preclude respondents from establishing a prima facie case nor does it provide petitioners with a defense to such a case. Pp. 445-456. (a) Despite petitioners' nondiscriminatory "bottom line," respondents' claim of disparate impact from the examination, a pass-fail barrier to employment opportunity, states a prima facie case of employment discrimination under § 703(a)(2) of Title VII, which makes it an unlawful employment practice for an employer to "limit, segregate, or classify his employees" in any way which would deprive "any individual of employment opportunities" because of race, color, religion, sex, or national origin. To measure disparate impact only at the "bottom line" ignores the fact that Title VII guarantees these individual black respondents the opportunity to compete equally with white workers on the basis of job-related criteria. Respondents' rights under § 703(a)(2) have been violated unless petitioners can demonstrate that the examination in question was not an artificial, arbitrary, or unnecessary barrier but measured skills related to effective performance as a supervisor. Pp. 445-451. (b) No special haven for discriminatory tests is offered by § 703(h) of Title VII, which provides that it shall not be an unlawful employment practice for an employer to act upon results of an ability test if such test is "not designed, intended, or used to discriminate" because of race, color, religion, sex, or national origin. A nonjob-related test that has a disparate impact and is used to "limit" or "classify" employees is "used to discriminate" within the meaning of Title VII, whether or not it was "designed or intended" to have this effect and despite an employer's efforts to compensate for its discriminatory effect. Pp. 451-452 (c) The principal focus of § 703(a)(2) is the protection of the individual employee, rather than the protection of the minority group as a whole. To suggest that the "bottom line" may be a defense to a claim of discrimination against an individual employee confuses unlawful discrimination with discriminatory intent. Resolution of the factual question of intent is not what is at issue in this case, but rather petitioners seek to justify discrimination against the black respondents on the basis of petitioners' favorable treatment of other members of these respondents' racial group. Congress never intended to give an employer license to discriminate against some employees on the basis of race or sex merely because he favorably treats other members of the employees' group. Pp. 452-456 2nd Cir., 645 F.2d 133, affirmed and remanded. Bernard F. McGovern, Jr., Hartford, Conn., for petitioners. Thomas W. Bucci, Bridgeport, Conn., for respondents. Justice BRENNAN delivered the opinion of the Court. 1 We consider here whether an employer sued for violation of Title VII of the Civil Rights Act of 19641 may assert a "bottom-line" theory of defense. Under that theory, as asserted in this case, an employer's acts of racial discrimination in promotions—effected by an examination having disparate impact would not render the employer liable for the racial discrimination suffered by employees barred from promotion if the "bottom-line" result of the promotional process was an appropriate racial balance. We hold that the "bottom line" does not preclude respondent employees from establishing a prima facie case, nor does it provide petitioner employer with a defense to such a case. 2 * Four of the respondents, Winnie Teal, Rose Walker, Edith Latney, and Grace Clark, are black employees of the Department of Income Maintenance of the State of Connecticut.2 Each was promoted provisionally to the position of Welfare Eligibility Supervisor and served in that capacity for almost two years. To attain permanent status as supervisors, however, respondents had to participate in a selection process that required, as the first step, a passing score on a written examination. This written test was administered on December 2, 1978, to 329 candidates. Of these candidates, 48 identified themselves as black and 259 identified themselves as white. The results of the examination were announced in March 1979. With the passing score set at 65,3 54.17 percent of the identified black candidates passed. This was approximately 68 percent of the passing rate for the identified white candidates.4 The four respondents were among the blacks who failed the examination, and they were thus excluded from further consideration for permanent supervisory positions. In April 1979, respondents instituted this action in the United States District Court for the District of Connecticut against petitioners, the State of Connecticut, two state agencies, and two state officials. Respondents alleged, inter alia, that petitioners violated Title VII by imposing, as an absolute condition for consideration for promotion, that applicants pass a written test that excluded blacks in disproportionate numbers and that was not job related. 3 More than a year after this action was instituted, and approximately one month before trial, petitioners made promotions from the eligibility list generated by the written examination. In choosing persons from that list, petitioners considered past work performance, recommendations of the candidates' supervisors and, to a lesser extent, seniority. Petitioners then applied what the Court of Appeals characterized as an affirmative-action program in order to ensure a significant number of minority supervisors.5 Forty-six persons were promoted to permanent supervisory positions, 11 of whom were black and 35 of whom were white. The overall result of the selection process was that, of the 48 identified black candidates who participated in the selection process, 22.9 percent were promoted and of the 259 identified white candidates, 13.5 percent were promoted.6 It is this "bottom-line" result, more favorable to blacks than to whites, that petitioners urge should be adjudged to be a complete defense to respondents' suit. 4 After trial, the District Court entered judgment for petitioners. App. to Pet. for Cert. 18a. The court treated respondents' claim as one of disparate impact under Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975), and Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 53 L.Ed.2d 786 (1977). However, the court found that, although the comparative passing rates for the examination indicated a prima facie case of adverse impact upon minorities, the result of the entire hiring process reflected no such adverse impact. Holding that these "bottom-line" percentages precluded the finding of a Title VII violation, the court held that the employer was not required to demonstrate that the promotional examination was job related. App. to Pet. for Cert. 22a-24a, 26a. The United States Court of Appeals for the Second Circuit reversed, holding that the District Court erred in ruling that the results of the written examination alone were insufficient to support a prima facie case of disparate impact in violation of Title VII. 645 F.2d 133 (1981). The Court of Appeals stated that where "an identifiable pass-fail barrier denies an employment opportunity to a disproportionately large number of minorities and prevents them from proceeding to the next step in the selection process," that barrier must be shown to be job related. Id., at 138. We granted certiorari, 454 U.S. 813, 102 S.Ct. 89, 70 L.Ed.2d 82 (1981), and now affirm. II A. 5 We must first decide whether an examination that bars a disparate number of black employees from consideration for promotion, and that has not been shown to be job related, presents a claim cognizable under Title VII. Section 703(a)(2) of Title VII provides in pertinent part: 6 "It shall be an unlawful employment practice for an employer— 7 * * * * * 8 "(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's race, color, religion, sex, or national origin." 78 Stat. 255, as amended, 42 U.S.C. § 2000e-2(a)(2). 9 Respondents base their claim on our construction of this provision in Griggs v. Duke Power Co., supra. Prior to the enactment of Title VII, the Duke Power Co. restricted its black employees to the labor department. Beginning in 1965, the company required all employees who desired a transfer out of the labor department to have either a high school diploma or to achieve a passing grade on two professionally prepared aptitude tests. New employees seeking positions in any department other than labor had to possess both a high school diploma and a passing grade on these two examinations. Although these requirements applied equally to white and black employees and applicants, they barred employment opportunities to a disproportionate number of blacks. While there was no showing that the employer had a racial purpose or invidious intent in adopting these requirements, this Court held that they were invalid because they had a disparate impact and were not shown to be related to job performance: 10 "[Title VII] proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation. The touchstone is business necessity. If an employment practice which operates to exclude Negroes cannot be shown to be related to job performance, the practice is prohibited." 401 U.S., at 431, 91 S.Ct., at 853. 11 Griggs and its progeny have established a three-part analysis of disparate-impact claims. To establish a prima facie case of discrimination, a plaintiff must show that the facially neutral employment practice had a significantly discriminatory impact. If that showing is made, the employer must then demonstrate that "any given requirement [has] a manifest relationship to the employment in question," in order to avoid a finding of discrimination. Griggs, supra, at 432, 91 S.Ct., at 854. Even in such a case, however, the plaintiff may prevail, if he shows that the employer was using the practice as a mere pretext for discrimination. See Albemarle Paper Co., supra, 422 U.S., at 425, 95 S.Ct., at 2375; Dothard, supra, 433 U.S., at 329, 97 S.Ct., at 2726-2727.7 12 Griggs recognized that in enacting Title VII, Congress required "the removal of artificial, arbitrary, and unnecessary barriers to employment" and professional development that had historically been encountered by women and blacks as well as other minorities. 401 U.S., at 431, 91 S.Ct., at 853. See also Dothard v. Rawlinson, supra.8 McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), explained that 13 "Griggs was rightly concerned that childhood deficiencies in the education and background of minority citizens, resulting from forces beyond their control, not be allowed to work a cumulative and invidious burden on such citizens for the remainder of their lives." Id., at 806. 14 Petitioners' examination, which barred promotion and had a discriminatory impact on black employees, clearly falls within the literal language of § 703(a)(2), as interpreted by Griggs. The statute speaks, not in terms of jobs and promotions, but in terms of limitations and classifications that would deprive any individual of employment opportunities.9 A disparate-impact claim reflects the language of § 703(a)(2) and Congress' basic objectives in enacting that statute: "to achieve equality of employment opportunities and remove barriers that have operated in the past to favor an identifiable group of white employees over other employees." 401 U.S., at 429-430, 91 S.Ct., at 852-853 (emphasis added). When an employer uses a nonjob-related barrier in order to deny a minority or woman applicant employment or promotion, and that barrier has a significant adverse effect on minorities or women, then the applicant has been deprived of an employment opportunity "because of . . . race, color, religion, sex, or national origin." In other words, § 703(a)(2) prohibits discriminatory "artificial, arbitrary, and unnecessary barriers to employment," 401 U.S., at 431, 91 S.Ct., at 853, that "limit . . . or classify . . . applicants for employment . . . in any way which would deprive or tend to deprive any individual of employment opportunities." (Emphasis added.) 15 Relying on § 703(a)(2), Griggs explicitly focused on employment "practices, procedures, or tests," 401 U.S., at 430, 91 S.Ct., at 853, that deny equal employment "opportunity," id., at 431, 91 S.Ct., at 853. We concluded that Title VII prohibits "procedures or testing mechanisms that operate as 'built-in headwinds' for minority groups." Id., at 432, 91 S.Ct., at 854. We found that Congress' primary purpose was the prophylactic one of achieving equality of employment "opportunities" and removing "barriers" to such equality. Id., at 429-430, 91 S.Ct., at 852-853. See Albemarle Paper Co. v. Moody, 422 U.S., at 417, 95 S.Ct., at 2371. The examination given to respondents in this case surely constituted such a practice and created such a barrier. 16 Our conclusion that § 703(a)(2) encompasses respondents' claim is reinforced by the terms of Congress' 1972 extension of the protections of Title VII to state and municipal employees. See n. 8, supra. Although Congress did not explicitly consider the viability of the defense offered by the state employer in this case, the 1972 amendments to Title VII do reflect Congress' intent to provide state and municipal employees with the protection that Title VII, as interpreted by Griggs, had provided to employees in the private sector: equality of opportunity and the elimination of discriminatory barriers to professional development. The Committee Reports and the floor debates stressed the need for equality of opportunity for minority applicants seeking to obtain governmental positions. E.g., S.Rep. No. 92-415, p. 10 (1971); 118 Cong.Rec. 1815 (1972) (remarks of Sen. Williams). Congress voiced its concern about the widespread use by state and local governmental agencies of "invalid selection techniques" that had a discriminatory impact. S.Rep. No. 92-415, supra, at 10; H.R.Rep. No. 92-238, p. 17 (1971); 117 Cong.Rec. 31961 (1971) (remarks of Rep. Perkins).10 17 The decisions of this Court following Griggs also support respondents' claim. In considering claims of disparate impact under § 703(a)(2) this Court has consistently focused on employment and promotion requirements that create a discriminatory bar to opportunities. This Court has never read § 703(a)(2) as requiring the focus to be placed instead on the overall number of minority or female applicants actually hired or promoted. Thus Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 53 L.Ed.2d 786 (1977), found that minimum statutory height and weight requirements for correctional counselors were the sort of arbitrary barrier to equal employment opportunity for women forbidden by Title VII. Although we noted in passing that women constituted 36.89 percent of the labor force and only 12.9 percent of correctional counselor positions, our focus was not on this "bottom line." We focused instead on the disparate effect that the minimum height and weight standards had on applicants: classifying far more women than men as ineligible for employment. Id., at 329-330, and n. 12, 97 S.Ct., at 2726-2727, and n. 12. Similarly, in Albemarle Paper Co. v. Moody, supra, the action was remanded to allow the employer to attempt to show that the tests that he had given to his employees for promotion were job related. We did not suggest that by promoting a sufficient number of the black employees who passed the examination, the employer could avoid this burden. See 422 U.S., at 436, 95 S.Ct., at 2380. See also New York Transit Authority v. Beazer, 440 U.S. 568, 584, 99 S.Ct. 1355, 1365, 59 L.Ed.2d 587 (1979) ("A prima facie violation of the Act may be established by statistical evidence showing that an employment practice has the effect of denying members of one race equal access to employment opportunities") (emphasis added). 18 In short, the District Court's dismissal of respondents' claim cannot be supported on the basis that respondents failed to establish a prima facie case of employment discrimination under the terms of § 703(a)(2). The suggestion that disparate impact should be measured only at the bottom line ignores the fact that Title VII guarantees these individual respondents theopportunity to compete equally with white workers on the basis of job-related criteria. Title VII strives to achieve equality of opportunity by rooting out "artificial, arbitrary, and unnecessary" employer-created barriers to professional development that have a discriminatory impact upon individuals. Therefore, respondents' rights under § 703(a)(2) have been violated, unless petitioners can demonstrate that the examination given was not an artificial, arbitrary, or unnecessary barrier, because it measured skills related to effective performance in the role of Welfare Eligibility Supervisor. B 19 The United States, in its brief as amicus curiae, apparently recognizes that respondents' claim in this case falls within the affirmative commands of Title VII. But it seeks to support the District Court's judgment in this case by relying on the defenses provided to the employer in § 703(h).11 Section 703(h) provides in pertinent part: 20 "Notwithstanding any other provision of this subchapter, it shall not be an unlawful employment practice for an employer . . . to give and to act upon the results of any professionally developed ability test provided that such test, its administration or action upon the results is not designed, intended or used to discriminate because of race, color, religion, sex or national origin." 78 Stat. 257, as amended, 42 U.S.C. § 2000e-2(h). 21 The Government argues that the test administered by the petitioners was not "used to discriminate" because it did not actually deprive disproportionate numbers of blacks of promotions. But the Government's reliance on § 703(h) as offering the employer some special haven for discriminatory tests is misplaced. We considered the relevance of this provision in Griggs. After examining the legislative history of § 703(h), we concluded that Congress, in adding § 703(h), intended only to make clear that tests that were job related would be permissible despite their disparate impact. 401 U.S., at 433-436, 91 S.Ct., at 854-856. As the Court recently confirmed, § 703(h), which was introduced as an amendment to Title VII on the Senate floor, "did not alter the meaning of Title VII, but 'merely clarifie[d] its present intent and effect.' " American Tobacco Co. v. Patterson, 456 U.S. 63, 73, n. 11, 102 S.Ct. 1534, 1539, n. 11, 71 L.Ed.2d 748 (1982), quoting 110 Cong.Rec. 12723 (1964) (remarks of Sen. Humphrey). A nonjob-related test that has a disparate racial impact, and is used to "limit" or "classify" employees, is "used to discriminate" within the meaning of Title VII, whether or not it was "designed or intended" to have this effect and despite an employer's efforts to compensate for its discriminatory effect. See Griggs, 401 U.S., at 433, 91 S.Ct., at 854. 22 In sum, respondents' claim of disparate impact from the examination, a pass-fail barrier to employment opportunity, states a prima facie case of employment discrimination under § 703(a)(2), despite their employer's nondiscriminatory "bottom line," and that "bottom line" is no defense to this prima facie case under § 703(h). III 23 Having determined that respondents' claim comes within the terms of Title VII, we must address the suggestion of petitioners and some amici curiae that we recognize an exception, either in the nature of an additional burden on plaintiffs seeking to establish a prima facie case or in the nature of an affirmative defense, for cases in which an employer has compensated for a discriminatory pass-fail barrier by hiring or promoting a sufficient number of black employees to reach a nondiscriminatory "bottom line." We reject this suggestion, which is in essence nothing more than a request that we redefine the protections guaranteed by Title VII.12 24 Section 703(a)(2) prohibits practices that would deprive or tend to deprive "any individual of employment opportunities." The principal focus of the statute is the protection of the individual employee, rather than the protection of the minority group as a whole. Indeed, the entire statute and its legislative history are replete with references to protection for the individual employee. See, e.g., §§ 703(a)(1), (b), (c), 704(a), 78 Stat. 255-257, as amended, 42 U.S.C. §§ 2000e-2(a)(1), (b), (c), 2000e-3(a); 110 Cong.Rec. 7213 (1964) (interpretive memorandum of Sens. Clark and Case) ("discrimination is prohibited as to any individual"); id., at 8921 (remarks of Sen. Williams) ("Every man must be judged according to his ability. In that respect, all men are to have an equal opportunity to be considered for a particular job"). 25 In suggesting that the "bottom line" may be a defense to a claim of discrimination against an individual employee, petitioners and amici appear to confuse unlawful discrimination with discriminatory intent. The Court has stated that a nondiscriminatory "bottom line" and an employer's good-faith efforts to achieve a nondiscriminatory work force, might in some cases assist an employer in rebutting the inference that particular action had been intentionally discriminatory: "Proof that [a] work force was racially balanced or that it contained a disproportionately high percentage of minority employees is not wholly irrelevant on the issue of intent when that issue is yet to be decided." Furnco Construction Corp. v. Waters, 438 U.S. 567, 580, 98 S.Ct. 2943, 2951, 57 L.Ed.2d 957 (1978). See also Teamsters v. United States, 431 U.S. 324, 340, n. 20, 97 S.Ct. 1843, 1856-1857, n. 20, 52 L.Ed.2d 396 (1977). But resolution of the factual question of intent is not what is at issue in this case. Rather, petitioners seek simply to justify discrimination against respondents on the basis of their favorable treatment of other members of respondents' racial group. Under Title VII, "[a] racially balanced work force cannot immunize an employer from liability for specific acts of discrimination." Furnco Construction Corp. v. Waters, 438 U.S., at 579, 98 S.Ct., at 2950-2951. 26 "It is clear beyond cavil that the obligation imposed by Title VII is to provide an equal opportunity for each applicant regardless of race, without regard to whether members of the applicant's race are already proportionately represented in the work force. See Griggs v. Duke Power Co., 401 U.S., at 430 [91 S.Ct., at 853]; McDonald v. Santa Fe Trail Transportation Co., 427 U.S. 273, 279 [96 S.Ct. 2574, 2578, 49 L.Ed.2d 493] (1976)." Ibid. (emphasis in original). 27 It is clear that Congress never intended to give an employer license to discriminate against some employees on the basis of race or sex merely because he favorably treats other members of the employees' group. We recognized in Los Angeles Dept. of Water & Power v. Manhart, 435 U.S. 702, 98 S.Ct. 1370, 55 L.Ed.2d 657 (1978), that fairness to the class of women employees as a whole could not justify unfairness to the individual female employee because the "statute's focus on the individual is unambiguous." Id., at 708, 98 S.Ct., at 1375. Similarly, in Phillips v. Martin Marietta Corp., 400 U.S. 542, 91 S.Ct. 496, 27 L.Ed.2d 613 (1971) (per curiam), we recognized that a rule barring employment of all married women with preschool children, if not a bona fide occupational qualification under § 703(e), violated Title VII, even though female applicants without preschool children were hired in sufficient numbers that they constituted 75 to 80 percent of the persons employed in the position plaintiff sought. 28 Petitioners point out that Furnco, Manhart, and Phillips involved facially discriminatory policies, while the claim in the instant case is one of discrimination from a facially neutral policy. The fact remains, however, that irrespective of the form taken by the discriminatory practice, an employer's treatment of other members of the plaintiffs' group can be "of little comfort to the victims of . . . discrimination." Teamsters v. United States, supra, 431 U.S., at 342, 97 S.Ct., at 1858. Title VII does not permit the victim of a facially discriminatory policy to be told that he has not been wronged because other persons of his or her race or sex were hired. That answer is no more satisfactory when it is given to victims of a policy that is facially neutral but practically discriminatory. Every individual employee is protected against both discriminatory treatment and "practices that are fair in form, but discriminatory in operation." Griggs v. Duke Power Co., 401 U.S., at 431, 91 S.Ct., at 853. Requirements and tests that have a discriminatory impact are merely some of the more subtle, but also the more pervasive, of the "practices and devices which have fostered racially stratified job environments to the disadvantage of minority citizens." McDonnell Douglas Corp. v. Green, 411 U.S., at 800, 93 S.Ct., at 1823. IV 29 In sum, petitioners' nondiscriminatory "bottom line" is no answer, under the terms of Title VII, to respondents' prima facie claim of employment discrimination. Accordingly, the judgment of the Court of Appeals for the Second Circuit is affirmed, and this case is remanded to the District Court for further proceedings consistent with this opinion. 30 It is so ordered. 31 Justice POWELL, with whom THE CHIEF JUSTICE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 32 In past decisions, this Court has been sensitive to the critical difference between cases proving discrimination under Title VII, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp.IV), by a showing of disparate treatment or discriminatory intent and those proving such discrimination by a showing of disparate impact. Because today's decision blurs that distinction and results in a holding inconsistent with the very nature of disparate-impact claims, I dissent. 33 * Section 703(a)(2) of Title VII, 42 U.S.C. § 2000e-2(a)(2), provides that it is an unlawful employment practice for an employer to 34 "limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's race, color, religion, sex, or national origin." 35 Although this language suggests that discrimination occurs only on an individual basis, in Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S.Ct. 849, 854, 28 L.Ed.2d 158 (1971), the Court held that discriminatory intent on the part of the employer against an individual need not be shown when "employment procedures or testing mechanisms . . . operate as 'built-in headwinds' for minority groups and are unrelated to measuring job capability." Thus, the Court held that the "disparate impact" of an employer's practices on a racial group can violate § 703(a)(2) of Title VII. In Griggs and each subsequent disparate-impact case, however, the Court has considered, not whether the claimant as an individual had been classified in a manner impermissible under § 703(a)(2), but whether an employer's procedures have had an adverse impact on the protected group to which the individual belongs. 36 Thus, while disparate-treatment cases focus on the way in which an individual has been treated, disparate-impact cases are concerned with the protected group. This key distinction was explained in Furnco Construction Corp. v. Waters, 438 U.S. 567, 581-582, 98 S.Ct. 2943, 2952, 57 L.Ed.2d 957 (1978) (MARSHALL, J., concurring in part): 37 "It is well established under Title VII that claims of employment discrimination because of race may arise in two different ways. Teamsters v. United States, 431 U.S. 324, 335-336, n. 15 [97 S.Ct. 1843, 1854-1855, n. 15, 52 L.Ed.2d 396 (1977). An individual may allege that he has been subjected to 'disparate treatment' because of his race, or that he has been the victim of a facially neutral practice having a 'disparate impact' on his racial group."1 38 In keeping with this distinction, our disparate-impact cases consistently have considered whether the result of an employer's total selection process had an adverse impact upon the protected group.2 If this case were decided by reference to the total process—as our cases suggest that it should be—the result would be clear. Here 22.9% of the blacks who entered the selection process were ultimately promoted, compared with only 13.5% of the whites. To say that this selection process had an unfavorable "disparate impact" on blacks is to ignore reality. 39 The Court, disregarding the distinction drawn by our cases, repeatedly asserts that Title VII was designed to protect individual, not group, rights. It emphasizes that some individual blacks were eliminated by the disparate impact of the preliminary test. But this argument confuses the aim of Title VII with the legal theories through which its aims were intended to be vindicated. It is true that the aim of Title VII is to protect individuals, not groups. But in advancing this commendable objective, Title VII jurisprudence has recognized two distinct methods of proof. In one set of cases—those involving direct proof of discriminatory intent—the plaintiff seeks to establish direct, intentional discrimination against him. In that type of case, the individual is at the forefront throughout the entire presentation of evidence. In disparate-impact cases, by contrast, the plaintiff seeks to carry his burden of proof by way of inference—by showing that an employer's selection process results in the rejection of a disproportionate number of members of a protected group to which he belongs. From such a showing a fair inference then may be drawn that the rejected applicant, as a member of that disproportionately excluded group, was himself a victim of that process' " 'built-in headwinds.' " Griggs, supra, at 432, 91 S.Ct., at 854. But this method of proof—which actually defines disparate-impact theory under Title VII—invites the plaintiff to prove discrimination by reference to the group rather than to the allegedly affected individual.3 There can be no violation of Title VII on the basis of disparate impact in the absence of disparate impact on a group.4 40 In this case respondent black employees seek to benefit from a conflation of "discriminatory treatment" and "disparate impact" theories. But they cannot have it both ways. Having undertaken to prove discrimination by reference to one set of group figures (used at a preliminary point in the selection process), these respondents then claim that nondiscrimination cannot be proved by viewing the impact of the entire process on the group as a whole. The fallacy of this reasoning—accepted by the Court—is transparent. It is to confuse the individualistic aim of Title VII with the methods of proof by which Title VII rights may be vindicated. The respondents, as individuals, are entitled to the full personal protection of Title VII. But, having undertaken to prove a violation of their rights by reference to group figures, respondents cannot deny petitioners the opportunity to rebut their evidence by introducing figures of the same kind. Having pleaded a disparate-impact case, the plaintiff cannot deny the defendant the opportunity to show that there was no disparate impact. As the Court of Appeals for the Third Circuit noted in EEOC v. Greyhound Lines, Inc., 635 F.2d 188, 192 (1980): 41 "[N]o violation of Title VII can be grounded on the disparate impact theory without proof that the questioned policy or practice has had a disproportionate impact on the employer's workforce. This conclusion should be as obvious as it is tautological: there can be no disparate impact unless there is [an ultimate] disparate impact." 42 Where, under a facially neutral employment process, there has been no adverse effect on the group—and certainly there has been none here—Title VII has not been infringed. II 43 The Court's position is no stronger in case authority than it is in logic. None of the cases relied upon by the Court controls the outcome of this case.5 Indeed, the disparateimpact cases do not even support the propositions for which they are cited. For example, the Court cites Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 53 L.Ed.2d 786 (1977) (holding impermissible minimum statutory height and weight requirements for correctional counselors), and observes that "[a]lthough we noted in passing that women constituted 36.89 percent of the labor force and only 12.9 percent of correctional counselor positions, our focus was not on this 'bottom line.' We focused instead on the disparate effect that the minimum height and weight standards had on applicants: classifying far more women than men as ineligible for employment." Ante, at 450. In Dothard, however, the Court was not considering a case in which there was any difference between the discriminatory effect of the employment standard and the number of minority members actually hired. The Dothard Court itself stated: 44 "[T]o establish a prima facie case of discrimination, a plaintiff need only show that the facially neutral standards in question select applicants for hire in a discriminatory pattern. Once it is shown that the employment standards are discriminatory in effect, the employer must meet 'the burden of showing that any given requirement [has] . . . a manifest relationship to the employment in question.' " 433 U.S., at 329, 97 S.Ct., at 2726-2727 (emphasis added). 45 The Dothard Court did not decide today's case. It addressed only a case in which the challenged standards had a discriminatory impact at the bottom line—the hiring decision. And the Dothard Court's "focus," referred to by the Court, is of no help in deciding the instant case.6 46 The Court concedes that the other major cases on which it relies, Furnco, Los Angeles Dept. of Water & Power v. Manhart, 435 U.S. 702, 98 S.Ct. 1370, 55 L.Ed.2d 657 (1978), and Phillips v. Martin Marietta Corp., 400 U.S. 542, 91 S.Ct. 496, 27 L.Ed.2d 613 (1971) (per curiam) "involved facially discriminatory policies, while the claim in the instant case is one of discrimination from a facially neutral policy." Ante, at 455. The Court nevertheless applies the principles derived from those cases to the case at bar. It does so by reiterating the view that Title VII protects individuals, not groups, and therefore that the manner in which an employer has treated other members of a group cannot defeat the claim of an individual who has suffered as a result of even a facially neutral policy. As appealing as this sounds, it confuses the distinction—uniformly recognized until today—between disparate impact and disparate treatment. See, supra, at 457-458. Our cases, cited above, have made clear that discriminatory-impact claims cannot be based on how an individual is treated in isolation from the treatment of other members of the group. Such claims necessarily are based on whether the group fares less well than other groups under a policy, practice, or test. Indeed, if only one minority member has taken a test, a disparate-impact claim cannot be made, regardless of whether the test is an initial step in the selection process or one of several factors considered by the employer in making an employment decision.7 III 47 Today's decision takes a long and unhappy step in the direction of confusion. Title VII does not require that employers adopt merit hiring or the procedures most likely to permit the greatest number of minority members to be considered for or to qualify for jobs and promotions. See Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 258-259, 101 S.Ct. 1089, 1096-1097, 67 L.Ed.2d 207 (1981); Furnco, 438 U.S., at 578, 98 S.Ct., at 2950. Employers need not develop tests that accurately reflect the skills of every individual candidate; there are few if any tests that do so. Yet the Court seems unaware of this practical reality, and perhaps oblivious to the likely consequences of its decision. By its holding today, the Court may force employers either to eliminate tests or rely on expensive, job-related, testing procedures, the validity of which may or may not be sustained if challenged. For state and local governmental employers with limited funds, the practical effect of today's decision may well be the adoption of simple quota hiring.8 This arbitrary method of employment is itself unfair to individual applicants, whether or not they are members of minority groups. And it is not likely to produce a competent work force. Moreover, the Court's decision actually may result in employers employing fewer minority members. As Judge Newman noted in Brown v. New Haven Civil Service Board, 474 F.Supp. 1256, 1263 (Conn.1979): 48 "[A]s private parties are permitted under Title VII itself to adopt voluntary affirmative action plans, . . . Title VII should not be construed to prohibit a municipality's using a hiring process that results in a percentage of minority policemen approximating their percentage of the local population, instead of relying on the expectation that a validated job-related testing procedure will produce an equivalent result, yet with the risk that it might lead to substantially less minority hiring." 49 Finding today's decision unfortunate in both its analytical approach and its likely consequences, I dissent. 1 Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp. IV). 2 The black respondents were joined as plaintiffs by four white employees on a pendent claim that the written test violated provisions of state law that require promotional exams to be job related. That claim is not before us. See 645 F.2d 133, 135, n. 3 (CA2 1981). 3 The mean score on the examination was 70.4 percent. However, because the black candidates had a mean score 6.7 percentage points lower than the white candidates, the passing score was set at 65, apparently in an attempt to lessen the disparate impact of the examination. See id., at 135, and n. 4. 4 The following table shows the passing rates of various candidate groups: Passing Candidate No. Receiving Rate Group Number Passing Score (%) Black 48 26 54.17 Hispanic 4 3 75.00 Indian 3 2 66.67 White 259 206 79.54 Unidentified 15 9 60.00 --- --- ----- Total 329 246 74.77 Petitioners do not contest the District Court's implicit finding that the examination itself resulted in disparate impact under the "eighty percent rule" of the Uniform Guidelines on Employee Selection Procedures adopted by the Equal Employment Opportunity Commission. See App. to Pet. for Cert. 18a, 23a, and n. 2. Those guidelines provide that a selection rate that "is less than [80 percent] of the rate for the group with the highest rate will generally be regarded . . . as evidence of adverse impact." 29 CFR § 1607.4D (1981). 5 Petitioners contest this characterization of their selection procedure. We have no need, however, to resolve this dispute in the context of the present controversy. 6 The actual promotion rate of blacks was thus close to 170 percent that of the actual promotion rate of whites. 7 Petitioners apparently argue both that the nondiscriminatory "bottom line" precluded respondents from establishing a prima facie case and, in the alternative, that it provided a defense. 8 The legislative history of the 1972 amendments to Title VII, 86 Stat. 103-113, is relevant to this case because those amendments extended the protection of the Act to respondents here by deleting exemptions for state and municipal employers. See 86 Stat. 103. That history demonstrates that Congress recognized and endorsed the disparate-impact analysis employed by the Court in Griggs. Both the House and Senate Reports cited Griggs with approval, the Senate Report noting: "Employment discrimination as viewed today is a . . . complex and pervasive phenomenon. Experts familiar with the subject now generally describe the problem in terms of 'systems' and 'effects' rather than simply intentional wrongs." S.Rep. No. 92-415, p. 5 (1971). See also H.R.Rep. No. 92-238, p. 8 (1971), U.S.Code Cong. & Admin.News 1972, p. 2137. In addition, the section-by-section analyses of the 1972 amendments submitted to both Houses explicitly stated that in any area not addressed by the amendments, present case law—which as Congress had already recognized included our then recent decision in Griggs —was intended to continue to govern. 118 Cong.Rec. 7166, 7564 (1972). 9 In contrast, the language of § 703(a)(1), 42 U.S.C. § 2000e-2(a)(1), if it were the only protection given to employees and applicants under Title VII, might support petitioners' exclusive focus on the overall result. That subsection makes it an unlawful employment practice "to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions or privileges of employment, because of such individual's race, color, religion, sex, or national origin." 10 The Committee Reports in both Houses, and Senator Williams, principal sponsor of the Senate bill that was ultimately enacted in large part, relied upon a report of the United States Commission on Civil Rights, which Senator Williams placed in the Congressional Record. See H.R.Rep. No. 92-238, p. 17 (1971); S.Rep. No. 92-415, p. 10 (1971); 118 Cong.Rec. 1815-1819 (1972). The Commission concluded that serious "[b]arriers to equal opportunity" existed for state and local government employees. Two of the three barriers cited were "recruitment and selection devices which are arbitrary, unrelated to job performance, and result in unequal treatment of minorities," and promotions made on the basis of "criteria unrelated to job performance and on discriminatory supervisory ratings." U. S. Commission on Civil Rights, For All the People . . . By All the People—A Report on Equal Opportunity in State and Local Government Employment 119 (1969), reprinted in 118 Cong.Rec. 1817 (1972). 11 The Government's brief is submitted by the Department of Justice, which shares responsibility for federal enforcement of Title VII with the Equal Employment Opportunity Commission (EEOC). The EEOC declined to join this brief. See Brief for United States as Amicus Curiae 1, and n. 12 Petitioners suggest that we should defer to the EEOC Guidelines in this regard. But there is nothing in the Guidelines to which we might defer that would aid petitioners in this case. The most support petitioners could conceivably muster from the Uniform Guidelines on Employee Selection Procedures, 29 CFR pt. 1607 (1981) (now issued jointly by the EEOC, the Office of Personnel Management, the Department of Labor, and the Department of Justice, see 29 CFR § 1607.1A (1981)), is neutrality on the question whether a discriminatory barrier that does not result in a discriminatory overall result constitutes a violation of Title VII. Section 1607.4C of the Guidelines, relied upon by petitioners, states that as a matter of "administrative and prosecutorial discretion, in usual circumstances," the agencies will not take enforcement action based upon the disparate impact of any component of a selection process if the total selection process results in no adverse impact. (Emphasis added.) The agencies made clear that the "guidelines do not address the underlying question of law," and that an individual "who is denied the job because of a particular component in a procedure which otherwise meets the 'bottom line' standard . . . retains the right to proceed through the appropriate agencies, and into Federal court." 43 Fed.Reg. 38291 (1978). See 29 CFR § 1607.16I (1981). In addition, in a publication entitled Adoption of Questions and Answers to Clarify and Provide a Common Interpretation of the Uniform Guidelines on Employee Selection Procedures, the agencies stated: "Since the [bottom-line] concept is not a rule of law, it does not affect the discharge by the EEOC of its statutory responsibilities to investigate charges of discrimination, render an administrative finding on its investigation, and engage in voluntary conciliation efforts. Similarly, with respect to the other issuing agencies, the bottom line concept applies not to the processing of individual charges, but to the initiation of enforcement action." 44 Fed.Reg. 12000 (1979). 1 See also Teamsters v. United States, 431 U.S. 324, 335-336, n. 15, 97 S.Ct. 1843, 1854-1855, n. 15, 52 L.Ed.2d 396 (1977) (similar explanation). 2 See Dothard v. Rawlinson, 433 U.S. 321, 329, 97 S.Ct. 2720, 2726-2727, 53 L.Ed.2d 786 (1977) (statutory height and weight requirements operated as a bar to employment of disproportionate number of women); Albemarle Paper Co. v. Moody, 422 U.S. 405, 409-411, 95 S.Ct. 2362, 2367-2368, 45 L.Ed.2d 280 (1975) (seniority system allegedly locked blacks into lower paying jobs; applicants to skilled lines of progression were required to pass two tests); Griggs v. Duke Power Co., 401 U.S. 424, 431, 91 S.Ct. 849, 853, 28 L.Ed.2d 158 (1971) (tests were an absolute bar to transfers or hiring; the Court observed that all Congress requires is "the removal of artificial, arbitrary, and unnecessary barriers to employment . . .") (emphasis added). 3 Initially, the plaintiff bears the burden of establishing a prima facie case that Title VII has been infringed. See Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 252-253, 101 S.Ct. 1089, 1093-1094, 67 L.Ed.2d 207 (1981). In a disparate-impact case, this burden is met by showing that an employer's selection process results in the rejection of a disproportionate number of members of a protected group. See Teamsters v. United States, supra, 336-338, 97 S.Ct., at 1855-1856. Regardless of whether the plaintiff's prima facie case must itself focus on the defendant's overall selection process or whether it is sufficient that the plaintiff establish that at least one pass-fail barrier has resulted in disparate impact, the employer's presentation of evidence showing that its overall selection procedure does not operate in a discriminatory fashion certainly dispels any inference of discrimination. In such instances, at the close of the evidence, the plaintiff has failed to show disparate impact by a preponderance of the evidence. 4 The Equal Employment Opportunity Commission and other federal enforcement agencies have adopted the "bottom-line" principle—i.e., the process viewed as a whole—in deciding when to bring an action against an employer. See Uniform Guidelines on Employee Selection Procedures, 5 CFR § 300.103(c) (1981). 5 The Court concentrates on cases of questionable relevance. Most of the lower courts that have squarely considered the question have concluded that there can be no violation of Title VII on a disparate-impact basis when there is no disparate impact at the bottom line. See, e.g., EEOC v. Greyhound Lines, Inc., 635 F.2d 188 (CA3 1980); EEOC v. Navajo Refining Co., 593 F.2d 988 (CA10 1979); Friend v. Leidinger, 588 F.2d 61, 66 (CA4 1978); Rule v. International Assn. of Ironworkers, 568 F.2d 558 (CA8 1977); Smith v. Troyan, 520 F.2d 492, 497-498 (CA6 1975), cert. denied, 426 U.S. 934, 96 S.Ct. 2646, 49 L.Ed.2d 385 (1976); Williams v. City & County of San Francisco, 483 F.Supp. 335 (ND Cal.1979); Brown v. New Haven Civil Service Board, 474 F.Supp. 1256 (Conn.1979); Lee v. City of Richmond, 456 F.Supp. 756 (ED Va.1978). 6 The Court cites language from two other disparate-impact cases. The Court notes that in Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975), the Court "remanded to allow the employer to attempt to show that the tests . . . given . . . for promotion were job related." Ante, at 450. But the fact that the Court did so without suggesting "that by promoting a sufficient number of black employees who passed the examination, the employer could avoid this burden," ibid., can hardly be precedent for the negative of that proposition when the issue was neither presented in the facts of the case nor addressed by the Court. Similarly, New York Transit Authority v. Beazer, 440 U.S. 568, 99 S.Ct. 1355, 59 L.Ed.2d 587 (1979), provides little support despite the language quoted by the Court. See ante, at 450, quoting 440 U.S., at 584, 99 S.Ct., at 1365 (" 'A prima facie violation of the Act may be established by statistical evidence showing that an employment practice has the effect of denying members of one race equal access to employment opportunities.' ") (emphasis added by the Court). In Beazer, the Court ruled that the statistical evidence actually presented was insufficient to establish a prima facie case of discrimination, and in doing so it indicated that it would have found statistical evidence of the number of applicants and employees in a methadone program quite probative. See id., at 585, 99 S.Ct., at 1365. Beazer therefore does not justify the Court's speculation that the number of blacks and Hispanics actually employed were irrelevant to whether a case of disparate impact had been established under Title VII. 7 Courts have recognized that the probative value of statistical evidence varies with sample size in disparate-impact cases. See, e.g., Teamsters v. United States, 431 U.S., at 340 n. 20, 97 S.Ct., at 1856 n. 20 ("Considerations such as small sample size may, of course, detract from the value of such evidence . . ."); Mayor of Philadelphia v. Educational Equality League, 415 U.S. 605, 621, 94 S.Ct. 1323, 1333, 39 L.Ed.2d 630 (1974) ("[T]he District Court's concern for the smallness of the sample presented by the 13-member Panel was . . . well founded"); Rogillio v. Diamond Shamrock Chemical Co., 446 F.Supp. 423, 427-428 (SD Tex.1978) (sample of 10 too small); Dendy v. Washington Hospital Center, 431 F.Supp. 873, 876 (DC 1977) (sample must be "large enough to mirror the reality of the employment situation"). A sample of only one would have far too little probative value to establish a prima facie case of disparate impact. 8 Another possibility is that employers may integrate consideration of test results into one overall hiring decision based on that "factor" and additional factors. Such a process would not, even under the Court's reasoning, result in a finding of discrimination on the basis of disparate impact unless the actual hiring decisions had a disparate impact on the minority group. But if employers integrate test results into a single-step decision, they will be free to select only the number of minority candidates proportional to their representation in the work force. If petitioners had used this approach, they would have been able to hire substantially fewer blacks without liability on the basis of disparate impact. The Court hardly could have intended to encourage this.

12

457 U.S. 496 102 S.Ct. 2557 73 L.Ed.2d 172 Georgia PATSY, Petitioner,v.BOARD OF REGENTS OF the STATE OF FLORIDA, etc. No. 80-1874. Argued March 2, 1982. Decided June 21, 1982. Syllabus Petitioner filed an action in Federal District Court under 42 U.S.C. § 1983 for declaratory or injunctive relief or damages, alleging that respondent employer had denied her employment opportunities solely on the basis of her race and sex. The District Court granted respondent's motion to dismiss because petitioner had not exhausted available state administrative remedies. The Court of Appeals vacated, holding that a § 1983 plaintiff could be required to exhaust administrative remedies if certain specified conditions were met, and remanded the case to the District Court to determine whether exhaustion would be appropriate in the instant case. Held: Exhaustion of state administrative remedies is not a prerequisite to an action under § 1983. Pp. 500-516. (a) This conclusion is supported by the legislative histories of both § 1983 and 42 U.S.C. § 1997e (1976 ed., Supp. IV), which carves out a narrow exception to the general no-exhaustion rule established in this Court's prior decisions by creating a specific, limited exhaustion requirement for adult prisoners bringing actions pursuant to § 1983. A judicially imposed exhaustion requirement in cases other than adult prisoners' cases would be inconsistent with Congress' decision to adopt § 1997e, would usurp policy judgments that Congress has reserved for itself, and would also be inconsistent with the detailed exhaustion scheme embodied in § 1997e. Pp. 502-512 (b) Even if, as respondent argues, an exhaustion requirement would lessen the burden that § 1983 actions impose on federal courts, would further the goal of comity and improve federal-state relations, and would enable the state agency to enlighten the federal court's ultimate decision, these are policy considerations that alone cannot justify judicially imposed exhaustion unless exhaustion is consistent with congressional intent. Moreover, difficult questions concerning the design and scope of an exhaustion requirement, which might be answered swiftly and surely by legislation, would create costly, remedy-delaying and court-burdening litigation if answered by the judiciary in the context of diverse constitutional claims relating to thousands of different state agencies. Pp. 512-515. 634 F.2d 900, 5 Cir., reversed and remanded. Charles S. Sims, New York City, for petitioner. Mitchell D. Franks, Tallahassee, Fla., for respondent. [Amicus Curiae Information from page 497 intentionally omitted] Justice MARSHALL delivered the opinion of the Court. 1 This case presents the question whether exhaustion of state administrative remedies is a prerequisite to an action under 42 U.S.C. § 1983 (1976 ed., Supp.IV). Petitioner Georgia Patsy filed this action, alleging that her employer, Florida International University (FIU), had denied her employment opportunities solely on the basis of her race and sex. By a divided vote, the United States Court of Appeals for the Fifth Circuit found that petitioner was required to exhaust "adequate and appropriate" administrative remedies, and remanded the case to the District Court to consider the adequacy of the administrative procedures. Patsy v. Florida International University, 634 F.2d 900 (1981) (en banc). We granted certiorari, 454 U.S. 813, 102 S.Ct. 88, 70 L.Ed.2d 81, and reverse the decision of the Court of Appeals. 2 * Petitioner alleges that even though she is well qualified and has received uniformly excellent performance evaluations from her supervisors, she has been rejected for more than 13 positions at FIU.1 She further claims that FIU has unlawfully filled positions through intentional discrimination on the basis of race and sex. She seeks declaratory and injunctive relief or, in the alternative, damages.2 3 The United States District Court for the Southern District of Florida granted respondent Board of Regents' motion to dismiss because petitioner had not exhausted available administrative remedies. On appeal, a panel of the Court of Appeals reversed, and remanded the case for further proceedings. Patsy v. Florida International University, 612 F.2d 946 (1980). The full court then granted respondent's petition for rehearing and vacated the panel decision. 4 The Court of Appeals reviewed numerous opinions of this Court holding that exhaustion of administrative remedies was not required, and concluded that these cases did not preclude the application of a "flexible" exhaustion rule. 634 F.2d, at 908. After canvassing the policy arguments in favor of an exhaustion requirement, the Court of Appeals decided that a § 1983 plaintiff could be required to exhaust administrative remedies if the following minimum conditions are met: (1) an orderly system of review or appeal is provided by statute or agency rule; (2) the agency can grant relief more or less commensurate with the claim; (3) relief is available within a reasonable period of time; (4) the procedures are fair, are not unduly burdensome, and are not used to harass or discourage those with legitimate claims; and (5) interim relief is available, in appropriate cases, to prevent irreparable injury and to preserve the plaintiff's rights during the administrative process. Where these minimum standards are met, a court must further consider the particular administrative scheme, the nature of the plaintiff's interest, and the values served by the exhaustion doctrine in order to determine whether exhaustion should be required. Id., at 912-913. The Court of Appeals remanded the case to the District Court to determine whether exhaustion would be appropriate in this case. II 5 The question whether exhaustion of administrative remedies should ever be required in a § 1983 action has prompted vigorous debate and disagreement. See, e.g., Turner, When Prisoners Sue: A Study of Prisoner Section 1983 Cases in the Federal Courts, 92 Harv.L.Rev. 610 (1979); Note, 8 Ind.L.Rev. 565 (1975); Comment, 41 U.Chi.L.Rev. 537 (1974). Our resolution of this issue, however, is made much easier because we are not writing on a clean slate. This Court has addressed this issue, as well as related issues, on several prior occasions. 6 Respondent suggests that our prior precedents do not control our decision today, arguing that these cases can be distinguished on their facts or that this Court did not "fully" consider the question whether exhaustion should be required. This contention need not detain us long. Beginning with McNeese v. Board of Education, 373 U.S. 668, 671-673, 83 S.Ct. 1433, 1435-1436, 10 L.Ed.2d 622 (1963), we have on numerous occasions rejected the argument that a § 1983 action should be dismissed where the plaintiff has not exhausted state administrative remedies. See Barry v. Barchi, 443 U.S. 55, 63, n. 10, 99 S.Ct. 2642, 2648, n. 10, 61 L.Ed.2d 365 (1979); Gibson v. Berryhill, 411 U.S. 564, 574, 93 S.Ct. 1689, 1695, 36 L.Ed.2d 488 (1973); Carter v. Stanton, 405 U.S. 669, 671, 92 S.Ct. 1232, 1234, 31 L.Ed.2d 569 (1972); Wilwording v. Swenson, 404 U.S. 249, 251, 92 S.Ct. 407, 409, 30 L.Ed.2d 418 (1971); Houghton v. Shafer, 392 U.S. 639, 640, 88 S.Ct. 2119, 2120, 20 L.Ed.2d 1319 (1968); King v. Smith, 392 U.S. 309, 312, n. 4, 88 S.Ct. 2128, 2130, n. 4, 20 L.Ed.2d 1118 (1968); Damico v. California, 389 U.S. 416, 88 S.Ct. 526, 19 L.Ed.2d 647 (1967). Cf. Steffel v. Thompson, 415 U.S. 452, 472-473, 94 S.Ct. 1209, 1222, 39 L.Ed.2d 505 (1974) ("When federal claims are premised on [§ 1983]—as they are here—we have not required exhaustion of state judicial or administrative remedies, recognizing the paramount role Congress has assigned to the federal courts to protect constitutional rights"). Respondent may be correct in arguing that several of these decisions could have been based on traditional exceptions to the exhaustion doctrine. Nevertheless, this Court has stated categorically that exhaustion is not a prerequisite to an action under § 1983, and we have not deviated from that position in the 19 years sinceMcNeese. Therefore, we do not address the question presented in this case as one of first impression. III 7 Respondent argues that we should reconsider these decisions and adopt the Court of Appeals' exhaustion rule, which was based on McKart v. United States, 395 U.S. 185, 89 S.Ct. 1657, 23 L.Ed.2d 194 (1969). This Court has never announced a definitive formula for determining whether prior decisions should be overruled or reconsidered. However, in Monell v. New York City Dept. of Social Services, 436 U.S. 658, 695-701, 98 S.Ct. 2018, 2038-2041, 56 L.Ed.2d 611 (1978), we articulated four factors that should be considered. Two of these factors—whether the decisions in question misconstrued the meaning of the statute as revealed in its legislative history and whether overruling these decisions would be inconsistent with more recent expressions of congressional intent—are particularly relevant to our decision today.3 Both concern legislative purpose, which is of paramount importance in the exhaustion context because Congress is vested with the power to prescribe the basic procedural scheme under which claims may be heard in federal courts. Of course, courts play an important role in determining the limits of an exhaustion requirement and may impose such a requirement even where Congress has not expressly so provided. However, the initial question whether exhaustion is required should be answered by reference to congressional intent; and a court should not defer the exercise of jurisdiction under a federal statute unless it is consistent with that intent.4 Therefore, in deciding whether we should reconsider our prior decisions and require exhaustion of state administrative remedies, we look to congressional intent as reflected in the legislative history of the predecessor to § 1983 and in recent congressional activity in this area. 8 * In determining whether our prior decisions misconstrued the meaning of § 1983, we begin with a review of the legislative history to § 1 of the Civil Rights Act of 1871, 17 Stat. 13, the precursor to § 1983.5 Although we recognize that the 1871 Congress did not expressly contemplate the exhaustion question, we believe that the tenor of the debates over § 1 supports our conclusion that exhaustion of administrative remedies in § 1983 actions should not be judicially imposed. 9 The Civil Rights Act of 1871, along with the Fourteenth Amendment it was enacted to enforce, were crucial ingredients in the basic alteration of our federal system accomplished during the Reconstruction Era. During that time, the Federal Government was clearly established as a guarantor of the basic federal rights of individuals against incursions by state power. As we recognized in Mitchum v. Foster, 407 U.S. 225, 242, 92 S.Ct. 2151, 32 L.Ed.2d 705 (1972) (quoting Ex parte Virginia, 100 U.S. 339, 346, 25 L.Ed. 676 (1880)), "[t]he very purpose of § 1983 was to interpose the federal courts between the States and the people, as guardians of the people's federal rights—to protect the people from unconstitutional action under color of state law, 'whether that action be executive, legislative, or judicial.' " 10 At least three recurring themes in the debates over § 1 cast serious doubt on the suggestion that requiring exhaustion of state administrative remedies would be consistent with the intent of the 1871 Congress. First, in passing § 1, Congress assigned to the federal courts a paramount role in protecting constitutional rights. Representative Dawes expressed this view as follows: 11 "The first remedy proposed by this bill is a resort to the courts of the United States. Is that a proper place in which to find redress for any such wrongs? If there be power to call into courts of the United States an offender against these rights, privileges, and immunities, and hold him to an account there, either civilly or criminally, for their infringement, I submit to the calm and candid judgment of every member of this House that there is no tribunal so fitted, where equal and exact justice would be more likely to be meted out in temper, in moderation, in severity, if need be, but always according to the law and the fact, as that great tribunal of the Constitution." Cong.Globe, 42d Cong., 1st Sess., 476 (1871) (hereinafter Globe). 12 See also id., at 332 (remarks of Rep. Hoar); id., at 375 (remarks of Rep. Lowe); id., at 448-449 (remarks of Rep. Butler); id., at 459 (remarks of Rep. Coburn).6 13 The 1871 Congress intended § 1 to "throw open the doors of the United States courts" to individuals who were threatened with, or who had suffered, the deprivation of constitutional rights, id., at 376 (remarks of Rep. Lowe), and to provide these individuals immediate access to the federal courts notwithstanding any provision of state law to the contrary. For example, Senator Edmunds, who introduced the bill in the Senate, stated in his closing remarks that the bill was similar in principle to an earlier act upheld by this Court in Prigg v. Pennsylvania, 16 Pet. 539, 10 L.Ed. 1060 (1842): 14 "[T]he Supreme Court decided . . . that it was the solemn duty of Congress under the Constitution to secure to the individual, in spite of the State, or with its aid, as the case might be, precisely the rights that the Constitution gave him, and that there should be no intermediate authority to arrest or oppose the direct performance of this duty by Congress." Globe 692 (emphasis added). 15 Similarly, Representative Elliott viewed the issue as whether "the Government of the United States [has] the right, under the Constitution, to protect a citizen in the exercise of his vested rights as an American citizen by . . . the assertion of immediate jurisdiction through its courts, without the appeal or agency of the State in which the citizen is domiciled." Id., at 389 (emphasis added). See, e.g., id., at 459 (remarks of Rep. Coburn); id., at 807 (remarks of Rep. Garfield); id., at 609 (remarks of Sen. Pool); Globe App. 141 (remarks of Rep. Shanks).7 16 A second theme in the debates further suggests that the 1871 Congress would not have wanted to impose an exhaustion requirement. A major factor motivating the expansion of federal jurisdiction through §§ 1 and 2 of the bill was the belief of the 1871 Congress that the state authorities had been unable or unwilling to protect the constitutional rights of individuals or to punish those who violated these rights. See, e.g., Globe 321 (remarks of Rep. Stoughton) ("The State authorities and local courts are unable or unwilling to check the evil or punish the criminals"); id., at 374 (remarks of Rep. Lowe) ("the local administrations have been found inadequate or unwilling to apply the proper corrective"); id., at 459 (remarks of Rep. Coburn); id., at 609 (remarks of Sen. Pool); id., at 687 (remarks of Sen. Shurz); id., at 691 (remarks of Sen. Edmunds); Globe App. 185 (remarks of Rep. Platt).8 Of primary importance to the exhaustion question was the mistrust that the 1871 Congress held for the factfinding processes of state institutions. See, e.g., Globe 320 (testimony of Hon. Thomas Settle, Justice of the North Carolina Supreme Court, before the House Judiciary Committee) ("The defect lies not so much with the courts as with the juries"); id., at 394 (remarks of Rep. Rainey); Globe App. 311 (remarks of Rep. Maynard). This Congress believed that federal courts would be less susceptible to local prejudice and to the existing defects in the factfinding processes of the state courts. See, e.g., Globe 322 (remarks of Rep. Stoughton); id., at 459 (remarks of Rep. Coburn).9 This perceived defect in the States' factfinding processes is particularly relevant to the question of exhaustion of administrative remedies: exhaustion rules are often applied in deference to the superior factfinding ability of the relevant administrative agency. See, e.g., McKart v. United States, 395 U.S., at 192-196, 89 S.Ct., at 1661-1662. 17 A third feature of the debates relevant to the exhaustion question is the fact that many legislators interpreted the bill to provide dual or concurrent forums in the state and federal system, enabling the plaintiff to choose the forum in which to seek relief. Cf. Monroe v. Pape, 365 U.S. 167, 183, 81 S.Ct. 473, 482, 5 L.Ed.2d 492 (1961) ("The federal remedy is supplementary to the state remedy, and the latter need not be first sought and refused before the federal one is invoked"). For example, Senator Thurman noted: 18 "I object to [§ 1], first, because of the centralizing tendency of transferring all mere private suits, as well as the punishment of offenses, from the State into the Federal courts. I do not say that this section gives to the Federal courts exclusive jurisdiction. I do not suppose that it is so understood. It leaves it, I presume, in the option of the person who imagines himself to be injured to sue in the State court or in the Federal court, an option that he who has been the least injured, but who has some malice to gratify, will be the most likely to avail himself of." Globe App. 216. 19 See also Globe 578, 694-695 (remarks of Sen. Edmunds); id., at 334 (remarks of Rep. Hoar); id., at 514 (remarks of Rep. Farnworth); Globe App. 85 (remarks of Rep. Bingham) ("Admitting that the States have concurrent power to enforce the Constitution of the United States within their respective limits, must we wait for their action?"). 20 This legislative history supports the conclusion that our prior decisions, holding that exhaustion of state administrative remedies is not a prerequisite to an action under § 1983, did not misperceive the statutory intent: it seems fair to infer that the 1871 Congress did not intend that an individual be compelled in every case to exhaust state administrative remedies before filing an action under § 1 of the Civil Rights Act. We recognize, however, that drawing such a conclusion from this history alone is somewhat precarious: the 1871 Congress was not presented with the question of exhaustion of administrative remedies, nor was it aware of the potential role of state administrative agencies. Therefore, we do not rely exclusively on this legislative history in deciding the question presented here. Congress addressed the question of exhaustion under § 1983 when it recently enacted 42 U.S.C. § 1997e (1976 ed., Supp. IV). The legislative history of § 1997e provides strong evidence of congressional intent on this issue. B 21 The Civil Rights of Institutionalized Persons Act, 42 U.S.C. § 1997 et seq. (1976 ed., Supp. IV), was enacted primarily to ensure that the United States Attorney General has "legal standing to enforce existing constitutional rights and Federal statutory rights of institutionalized persons." H.R.Conf.Rep.No.96-897, p. 9 (1980) (Conf.Rep.), U.S.Code Cong. & Admin.News 1980, p. 787. In § 1997e, Congress also created a specific, limited exhaustion requirement for adult prisoners bringing actions pursuant to § 1983. Section 1997e and its legislative history demonstrate that Congress understood that exhaustion is not generally required in § 1983 actions, and that it decided to carve out only a narrow exception to this rule. A judicially imposed exhaustion requirement would be inconsistent with Congress' decision to adopt § 1997e and would usurp policy judgments that Congress has reserved for itself. 22 In considering whether an exhaustion requirement should be incorporated into the bill, Congress clearly expressed its belief that a decision to require exhaustion for certain § 1983 actions would work a change in the law. Witnesses testifying before the Subcommittee that drafted the bill discussed the decisions of this Court holding that exhaustion was not required. See, e.g., Hearings on H.R. 2439 and H.R. 5791 before the Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the House Committee on the Judiciary, 95th Cong., 1st Sess., 20 (1977) (1977 Hearings); id., at 47, 69, 77, 323; Hearings on H.R. 10 before the Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the House Committee on the Judiciary, 96th Cong., 1st Sess., 48 (1979) (1979 Hearings). During these hearings, Representative Kastenmeier, Chairman of this Subcommittee, stated: 23 "Another thing that I think requires some discussion within the committee, and is a point of argument, . . . is whether there ought to be an exhaustion of remedies requirement. 24 ". . . In fact, I think it has been pointed out that if [we] were to require it, particularly in 1983, that would constitute regression from the current state of the law. It would set the law back, because presently it is clearly held, that is the Supreme Court has held, that in 1983 civil rights suits the litigant need not necessarily fully exhaust State remedies." 1977 Hearings 57-58. 25 See also id., at 272 (remarks of Rep. Drinan) (Representative Railsback "grounds his bill on doing something which the Supreme Court has consistently refused to do, namely require exhaustion of remedies"); 1979 Hearings 26 (remarks of Rep. Kastenmeier) (adopting § 1997e "was resisted as a possible encroachment on civil liberties; that is to say, in the free, unimpeded resort to 1983"). 26 The debates over adopting an exhaustion requirement also reflect this understanding. See, e.g., 124 Cong.Rec. 11988 (1978) (remarks of Rep. Volkmer and Rep. Kastenmeier); id., at 15445 (remarks of Rep. Ertel); id., at 23180 (remarks of Rep. Wiggins) ("it is settled law that an exhaustion of administrative remedies is not required as a precondition of maintaining a 1983 action"); 125 Cong.Rec. 12496 (1979) (remarks of Rep. Butler) ("Under existing law there is no requirement that a complainant first ask the State prison system to help him"). With the understanding that exhaustion generally is not required, Congress decided to adopt the limited exhaustion requirement of § 1997e in order to relieve the burden on the federal courts by diverting certain prisoner petitions back through state and local institutions, and also to encourage the States to develop appropriate grievance procedures. See, e.g., Conf.Rep. 9; 124 Cong.Rec. 11976 (1978) (remarks of Rep. Kastenmeier); id., at 11976, 11983 (remarks of Rep. Railsback); id., at 15442 (remarks of Rep. Kastenmeier); id., at 15445 (remarks of Rep. Ertel); id., at 23176 (remarks of Rep. Kastenmeier); id., at 23179-23180 (remarks of Rep. Butler); id., at 23180 (remarks of Rep. Ertel). Implicit in this decision is Congress' conclusion that the no-exhaustion rule should be left standing with respect to other § 1983 suits. 27 A judicially imposed exhaustion requirement would also be inconsistent with the extraordinarily detailed exhaustion scheme embodied in § 1997e. Section 1997e carves out a narrow exception to the general no-exhaustion rule to govern certain prisoner claims, and establishes a procedure to ensure that the administrative remedies are adequate and effective. The exhaustion requirement is expressly limited to § 1983 actions brought by an adult convicted of a crime. 42 U.S.C. § 1997e(a)(1) (1976 ed., Supp. IV).10 Section 1997e(b)(1) instructs the Attorney General to "promulgate minimum standards for the development and implementation of a plain, speedy, and effective system" of administrative remedies, and § 1997e(b)(2) specifies certain minimum standards that must be included.11 A court may require exhaustion of administrative remedies only if "the Attorney General has certified or the court has determined that such administrative remedies are in substantial compliance with the minimum acceptable standards promulgated under subsection (b)." § 1997e(a)(2). Before exhaustion may be required, the court must further conclude that it "would be appropriate and in the interests of justice." § 1997e(a)(1).12 Finally, in those § 1983 actions meeting all the statutory requirements for exhaustion, the district court may not dismiss the case, but may only "continue such case for a period of not to exceed ninety days in order to require exhaustion." Ibid. This detailed scheme is inconsistent with discretion to impose, on an ad hoc basis, a judicially developed exhaustion rule in other cases. 28 Congress hoped that § 1997e would improve prison conditions by stimulating the development of successful grievance mechanisms. See, e.g., Conf.Rep. 9; H.R.Rep.No.96-80, p. 4 (1979); 1979 Hearings 4 (remarks of Rep. Railsback); 124 Cong.Rec. 11976 (1978) (remarks of Rep. Railsback); 125 Cong.Rec. 12492 (1979) (remarks of Rep. Drinan); 126 Cong.Rec. 10780 (1980) (remarks of Rep. Kastenmeier). To further this purpose, Congress provided for the deferral of the exercise of federal jurisdiction over certain § 1983 claims only on the condition that the state prisons develop adequate procedures. This purpose would be frustrated by judicial discretion to impose exhaustion generally: the States would have no incentive to adopt grievance procedures capable of certification, because prisoner § 1983 cases could be diverted to state administrative remedies in any event. 29 In sum, the exhaustion provisions of the Act make sense, and are not superfluous, only if exhaustion could not be required before its enactment and if Congress intended to carve out a narrow exception to this no-exhaustion rule. The legislative history of § 1997e demonstrates that Congress has taken the approach of carving out specific exceptions to the general rule that federal courts cannot require exhaustion under § 1983. It is not our province to alter the balance struck by Congress in establishing the procedural framework for bringing actions under § 1983. C 30 Respondent and the Court of Appeals argue that exhaustion of administrative remedies should be required because it would further various policies. They argue that an exhaustion requirement would lessen the perceived burden that § 1983 actions impose on federal courts;13 would further the goal of comity and improve federal-state relations by postponing federal-court review until after the state administrative agency had passed on the issue;14 and would enable the agency, which presumably has expertise in the area at issue, to enlighten the federal court's ultimate decision. 31 As we noted earlier, policy considerations alone cannot justify judicially imposed exhaustion unless exhaustion is consistent with congressional intent. See supra, at 501-502, and n. 4. Furthermore, as the debates over incorporating the exhaustion requirement in § 1997e demonstrate, the relevant policy considerations do not invariably point in one direction, and there is vehement disagreement over the validity of the assumptions underlying many of them.15 The very difficulty of these policy considerations, and Congress' superior institutional competence to pursue this debate, suggest that legislative not judicial solutions are preferable. Cf. Diamond v. Chakrabarty, 447 U.S. 303, 317, 100 S.Ct. 2204, 2212, 65 L.Ed.2d 144 (1980); Steelworkers v. Bouligny, Inc., 382 U.S. 145, 150, 153, 86 S.Ct. 272, 274-276, 15 L.Ed.2d 217 (1965). 32 Beyond the policy issues that must be resolved in deciding whether to require exhaustion, there are equally difficult questions concerning the design and scope of an exhaustion requirement. These questions include how to define those categories of § 1983 claims in which exhaustion might be desirablesS how to unify and centralize the standards for judging the kinds of administrative procedures that should be exhausted;16 what tolling requirements and time limitations should be adopted;17 what is the res judicata and collateral estoppel effect of particular administrative determinations; what consequences should attach to the failure to comply with procedural requirements of administrative proceedings; and whether federal courts could grant necessary interim injunctive relief and hold the action pending exhaustion, or proceed to judgment without requiring exhaustion even though exhaustion might otherwise be required, where the relevant administrative agency is either powerless or not inclined to grant such interim relief. These and similar questions might be answered swiftly and surely by legislation, but would create costly, remedy-delaying, and court-burdening litigation if answered incrementally by the judiciary in the context of diverse constitutional claims relating to thousands of different state agencies.18 33 The very variety of claims, claimants, and state agencies involved in § 1983 cases argues for congressional consideration of the myriad of policy considerations, and may explain why Congress, in deciding whether to require exhaustion in certain § 1983 actions brought by adult prisoners, carved out such a narrow, detailed exception to the no-exhaustion rule. After full debate and consideration of the various policy arguments, Congress adopted § 1997e, taking the largest class of § 1983 actions and constructing an exhaustion requirement that differs substantially from the McKart-type standard urged by respondent and adopted by the Court of Appeals. See n. 18, supra. It is not for us to say whether Congress will or should create a similar scheme for other categories of § 1983 claims or whether Congress will or should adopt an altogether different exhaustion requirement for nonprisoner § 1983 claims.19 IV 34 Based on the legislative histories of both § 1983 and § 1997e, we conclude that exhaustion of state administrative remedies should not be required as a prerequisite to bringing an action pursuant to § 1983. We decline to overturn our prior decisions holding that such exhaustion is not required. The decision of the Court of Appeals is reversed, and the case is remanded for proceedings consistent with this opinion. 35 It is so ordered. 36 Justice O'CONNOR, with whom Justice REHNQUIST joins, concurring. 37 As discussed in Justice POWELL's dissenting opinion, as well as in the opinion of the court below, considerations of sound policy suggest that a § 1983 plaintiff should be required to exhaust adequate state administrative remedies before filing his complaint. At the very least, prior state administrative proceedings would resolve many claims, thereby decreasing the number of § 1983 actions filed in the federal courts, which are now straining under excessive caseloads. However, for the reasons set forth in the Court's opinion, this Court already has ruled that, in the absence of additional congressional legislation, exhaustion of administrative remedies is not required in § 1983 actions. Perhaps Congress' enactment of the Civil Rights of Institutionalized Persons Act, 42 U.S.C. § 1997 et seq. (1976 ed., Supp.IV), which creates a limited exhaustion requirement for prisoners bringing § 1983 suits, will prompt it to reconsider the possibility of requiring exhaustion in the remainder of § 1983 cases. Reluctantly, I concur. 38 Justice WHITE, concurring in part. 39 I fully agree with the Court that our frequent and unequivocal statements on exhaustion cannot be explained or distinguished away as the Fifth Circuit attempted to do. For nearly 20 years and on at least 10 occasions, this Court has clearly held that no exhaustion of administrative remedies is required in a § 1983 suit. Ante, at 500. Whether or not this initially was a wise choice, these decisions are stare decisis, and in a statutory case, a particularly strong showing is required that we have misread the relevant statute and its history. I have no difficulty in concluding that on the issue of exhaustion, unlike the question of municipal immunity faced in Monell v. New York City Dept. of Social Services, 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611 (1978), the Court has not previously misapprehended the meaning of the 1871 debates in rejecting an exhaustion rule in McNeese v. Board of Education, 373 U.S. 668, 671-673, 83 S.Ct. 1433, 1435-1436, 10 L.Ed.2d 622 (1963), and adhering to that position ever since. Our precedents and the legislative history are sufficient to support reversal, and I accordingly join the judgment in all but Part III-B of the opinion of the Court. 40 In Part III-B, the Court unnecessarily and unwisely ventures further to find support where none may be had. The wisdom of a general no-exhaustion rule in § 1983 suits was not at issue when Congress considered and passed the Civil Rights of Institutionalized Persons Act, 42 U.S.C. § 1997 et seq. (1976 ed., Supp. IV). As Justice POWELL persuasively points out in his dissenting opinion, and as reflected in the title of the Act, congressional attention was narrowly focused on procedures concerning the legal rights of prisoners and other institutionalized persons. Unsurprisingly, the legislation which emerged addressed only the specific problem under investigation; it indicates neither approval of a no-exhaustion rule nor an intent to preclude us from reconsidering the issue. 41 As the Court acknowledges, ante, at 513, the policy arguments cut in both directions. The Court concludes that "the very difficulty of these policy considerations, and Congress' superior institutional competence . . . suggest that legislative not judicial decisions are preferable." To be sure, exhaustion is a statutory issue and the dispositive word on the matter belongs to Congress. It does not follow, however, that, were the issue not foreclosed by earlier decisions, we would be institutionally incompetent to formulate an exhaustion rule. The lack of an exhaustion requirement in § 1983 actions is itself an exception to the general rule, judicially formulated, that exhaustion of administrative remedies is required in a civil action. Myers v. Bethlehem Shipbuilding Corp., 303 U.S. 41, 58 S.Ct. 459, 82 L.Ed. 638 (1938); McKart v. United States, 395 U.S. 185, 89 S.Ct. 1657, 23 L.Ed.2d 194 (1969). Unlike other statutory questions, exhaustion is "a rule of judicial administration," Myers v. Bethlehem Shipping Corp., supra, at 50, 58 S.Ct., at 463, and unless Congress directs otherwise, rightfully subject to crafting by judges. Our resolution of this case as governed by stare decisis, reinforced by the legislative history of § 1983, should not be taken as undercutting the general exhaustion principle of long standing. The result today is also fully consistent with our decisions that a defendant in a civil or administrative enforcement proceeding may not enjoin and sidetrack that proceeding by resorting to a § 1983 action in federal court, Huffman v. Pursue, Ltd., 420 U.S. 592, 95 S.Ct. 1200, 43 L.Ed.2d 482 (1975); Juidice v. Vail, 430 U.S. 327, 97 S.Ct. 1211, 51 L.Ed.2d 376 (1977); Trainor v. Hernandez, 431 U.S. 434, 97 S.Ct. 1911, 52 L.Ed.2d 486 (1977); Moore v. Sims, 442 U.S. 415, 99 S.Ct. 2371, 60 L.Ed.2d 994 (1979), and that a federal action should be stayed pending determination of state-law issues central to the constitutional dispute. Railroad Comm'n v. Pullman Co., 312 U.S. 496, 61 S.Ct. 643, 85 L.Ed. 971 (1941). On this understanding, I join all but Part III-B of the opinion of the Court.* 42 Justice POWELL, with whom THE CHIEF JUSTICE joins as to Part II, dissenting. 43 The Court holds that the limitations on federal judicial power embodied in the Eleventh Amendment and in the doctrine of sovereign immunity are not jurisdictional. I consider this holding to be a serious departure from established constitutional doctrine. 44 I dissent also from the Court's rejection of the rule of "flexible" exhaustion of state administrative remedies developed and stated persuasively by the Court of Appeals for the Fifth Circuit, sitting en banc. In disagreeing with the 17 judges of the Court of Appeals who adopted the flexible exhaustion principle, this Court places mistaken reliance on the Civil Rights of Institutionalized Persons Act, 42 U.S.C. § 1997 et seq. (1976 ed., Supp. IV). I disagree with both portions of the Court's holding and therefore dissent. I. The Eleventh Amendment.1 45 In this "reverse discrimination" action, petitioner, an employee of the Florida International University, brought suit under 42 U.S.C. § 1983 against the Board of Regents of the State of Florida.2 She did not name the individual Regents as defendants. She sued for $500,000 in damages, and for injunctive and other equitable relief. See ante, at 498-499, n. 2. The Board filed a motion to dismiss, arguing that petitioner's suit was premature in light of her failure to exhaust available administrative remedies. The District Court agreed and granted the motion to dismiss. 46 On petitioner's appeal, the Board added the bar of the Eleventh Amendment to its defense.3 It argued that as an instrumentality of the State, the Board could not be subjected to suit in federal court absent a waiver of immunity.4 And it asserted that there had been no waiver. Although the Board of Regents was created as a body corporate with power "to sue and be sued . . . to plead and be impleaded in all courts of law and equity," Fla.Stat. § 240.205(4)(1) (1981), it is well established that language such as this does not operate to waive the defense of the Eleventh Amendment.5 In reply, petitioner argued that whether or not the statute creating the Board amounted to a waiver—and petitioner believed that it did the Eleventh Amendment simply was irrelevant to the equitable claims she had lodged against the State. See Reply Brief for Petitioner 3-4. 47 Neither the Court of Appeals panel nor the Court of Appeals en banc addressed the Board's Eleventh Amendment defense. They directed their attention solely to the question of exhaustion of administrative remedies. The panel held that there was no exhaustion requirement in § 1983 suits and remanded to the District Court for consideration of the Board's Eleventh Amendment argument. Patsy v. Florida International University, 612 F.2d 946 (1980). The Court of Appeals, sitting en banc, reversed, holding that § 1983 plaintiffs must exhaust available and reasonable administrative remedies. Patsy v. Florida International University, 634 F.2d 900 (1981). Again the court did not consider the Board's Eleventh Amendment defense. 48 The Eleventh Amendment question was raised before this Court, at the first opportunity after the Court of Appeals' decision, in the Board's response to the petition for writ of certiorari. The Board argued, as it had on appeal, that it was an arm of the State and that it had not waived its immunity from suit in federal court.6 Again petitioner answered that at most the Eleventh Amendment defense would bar her claim for damages. And, even as to this claim, petitioner now argued that the Amendment would not bar damages if the Board could meet the claim out of its "own funds"—e.g., from gifts and bequests—rather than from the state treasury. These arguments were repeated at oral argument.7 B 49 The Court views the jurisdictional question presented by the Eleventh Amendment as if it were of little or no importance. Its entire discussion of the question is relegated to a conclusory note at the end of the opinion. See ante, at 515-516, n. 19. The Court concedes that the Amendment and the bar of sovereign immunity are "jurisdictional," but only in the sense that the State may raise the claim at any point in the proceedings. The statement is then made that the Amendment is not jurisdictional "in the sense that it must be raised and decided by this Court on its own motion." Ibid.8 The Court cites to no authority in support of this statement,9 and it would be surprising if any existed. The reason that the Eleventh Amendment question may be raised at any point in the proceedings is precisely because it places limits on the basic authority of federal courts to entertain suits against a State. The history and text of the Eleventh Amendment, the principle of sovereign immunity exemplified by it, and the well-established precedents of this Court make clear that today's decision misconceives our jurisdiction and the purpose of this Amendment. 50 A basic principle of our constitutional system is that the federal courts are courts of limited jurisdiction. Their authority extends only to those matters within the judicial power of the United States as defined by the Constitution. In language that could not be clearer, the Eleventh Amendment removes from the judicial power, as set forth in Art. III, suits "commenced or prosecuted against one of the United States." When an Amendment to the Constitution states in plain language that "the judicial power of the United States shall not be construed to extend" to suits against a State, from what source does the Court today derive its jurisdiction? The Court's "back-of-the-hand" treatment of this threshold issue offers no answer. Questions of jurisdiction and of the legitimate exercise of power are fundamental in our federal constitutional system.10 C 51 The Eleventh Amendment was adopted as a response to this Court's assumption of original jurisdiction in a suit brought against the State of Georgia. Chisholm v. Georgia, 2 Dall. 419, 1 L.Ed. 440 (1793). Relying upon express language in Art. III extending the judicial power to controversies between a State and citizens of another State, the Court found that it had jurisdiction. The decision is said to have created a shock throughout the country. See Hans v. Louisiana, 134 U.S. 1, 11, 10 S.Ct. 504, 505, 33 L.Ed. 842 (1890). The Amendment was adopted shortly thereafter, and the Court understood that it had been overruled: " 'the amendment being constitutionally adopted, there could not be exercised any jurisdiction, in any case, past or future, in which a State was sued by the citizens of another State, or by citizens or subjects of any foreign state.' " Ibid. 52 In light of the history and wording of the Amendment, the Court has viewed the Amendment as placing explicit limits on the judicial power as defined by Art. III. See Nevada v. Hall, 440 U.S. 410, 421, 99 S.Ct. 1182, 1188, 59 L.Ed.2d 416 (1979). But more than that, and beyond the express provisions of the Amendment, the Court has recognized that the Amendment stands for a principle of sovereign immunity by which the grant of authority in Art. III itself must be measured.11 Thus, in Hans v. Louisiana, supra, the Court held that the federal judicial power did not extend to a suit against a nonconsenting State by one of its own citizens. Although the Eleventh Amendment by its terms does not apply to such suits, the Court found that the language of the Amendment was but an illustration of a larger principle: Federal jurisdiction over suits against a State, absent consent, "was not contemplated by the Constitution when establishing the judicial power of the United States." Id., 134 U.S. at 15, 10 S.Ct., at 507.12 See Smith v. Reeves, 178 U.S. 436, 20 S.Ct. 919, 44 L.Ed. 1140 (1900). 53 Similarly, in Ex parte New York, 256 U.S. 490, 41 S.Ct. 588, 65 L.Ed. 1057 (1921), the Court found that despite the Eleventh Amendment's specific reference to suits in "law or equity," the principle of sovereign immunity exemplified by the Amendment would not permit the extension of federal admiralty jurisdiction over a nonconsenting State. The Court applied the same approach in Monaco v. Mississippi, 292 U.S. 313, 54 S.Ct. 745, 78 L.Ed. 1282 (1934), in which the Court refused to take jurisdiction over a suit against a State by a foreign state. On its face, Art. III provided jurisdiction over suits "between a State . . . and foreign States." Nor did the Eleventh Amendment specifically exempt the States from suit by a foreign state. Nevertheless, the Court concluded that the judicial power of the United States, granted by Art. III, did not extend so far: "We think that Madison correctly interpreted Clause one of § 2 of Article III of the Constitution as making provision for jurisdiction of a suit against a State by a foreign State in the event of the State's consent but not otherwise." Id., at 330, 54 S.Ct., at 751. 54 In this case a resident of the State of Florida has sued a Board exercising a major function of the State's sovereign authority. As prior decisions have held, whether this case is viewed only under the Eleventh Amendment—with its explicit limitation on federal jurisdiction—or under Art. III, the analysis must be the same. Absent consent, the "judicial power of the United States," as defined by Art. III and the Eleventh Amendment, simply does not extend to suits against one of the States by a citizen of that State:13 55 "That a State may not be sued without its consent is a fundamental rule of jurisprudence having so important a bearing upon the construction of the Constitution of the United States that it has become established by repeated decisions of this court that the entire judicial power granted by the Constitution does not embrace authority to entertain a suit brought by private parties against a State without consent given : not one brought by citizens of another State, or by citizens or subjects of a foreign State, because of the Eleventh Amendment; and not even one brought by its own citizens, because of the fundamental rule of which the Amendment is but an exemplification." Ex parte New York, supra, at 497, 41 S.Ct., at 589 (emphasis added). 56 The Court does not distinguish these unquestioned precedents. They are wholly and inexplicably ignored. Quite simply the Court today disregards controlling decisions and the explicit limitation on federal-court jurisdiction in Art. III and the Eleventh Amendment. The Court does recognize that the Eleventh Amendment is jurisdictional "in the sense" that the State may raise the bar of the Amendment for the first time on appeal. Yet the Court misses the point of this statement. The reason that the bar of the Amendment may be raised at any time—as the Court previously has explained—is precisely because it is jurisdictional: 57 "The objection to petitioner's suit as a violation of the Eleventh Amendment was first made and argued . . . in this Court. This was in time, however. The Eleventh Amendment declares a policy and sets forth an explicit limitation on federal judicial power of such compelling force that this Court will consider the issue arising under this Amendment . . . even though urged for the first time in this Court." Ford Motor Co. v. Department of Treasury of Indiana, 323 U.S. 459, 467, 65 S.Ct. 347, 352, 89 L.Ed. 389 (1945).14 58 Despite these precedents, and apparently because of an unexplained anxiety to reach the exhaustion issue decided by the Court of Appeals, this Court remands the issue of its own jurisdiction to the courts below. D 59 I believe that the Eleventh Amendment question must be addressed and that the answer could hardly be clearer. This is an action under § 1983.15 Petitioner seeks relief from the Board of Regents of the State of Florida, a major instrumentality or agency of the State. Petitioner's argument that the statute incorporating the Board should be understood to waive the Eleventh Amendment is foreclosed by numerous decisions of this Court and is unsupported by State law. See, e.g., Florida Dept. of Health v. Florida Nursing Home Assn., 450 U.S. 147, 101 S.Ct. 1032, 67 L.Ed.2d 132 (1981); n. 5, supra. Similarly, petitioner's suggestion that the Eleventh Amendment does not bar her equitable claims against the Board must be rejected. The Amendment applies to suits "in law or equity." All suits against an unconsenting State—whether for damages or injunctive relief—are barred. See Cory v. White, 457 U.S. 85, 102 S.Ct. 2325, 72 L.Ed.2d 694.16 Finally, the rule in Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 59 L.Ed. 714 (1908), permitting a federal court to order state officials to obey federal law in the future, is simply irrelevant to this case.17 Petitioner did not sue the members of the Board of Regents. She sued the Board itself, an arm of the State of Florida. 60 In my view, the Eleventh Amendment—and the principle of sovereign immunity exemplified by the Amendment and embodied in Art. III—clearly bar the suit in this case. The Court's refusal to address the question of its own jurisdiction violates well-established precedents of this Court as well as the basic premise that federal courts are courts of limited jurisdiction. Even had the parties neglected to address the Eleventh Amendment question, it would have been our responsibility to consider it on our own motion. In fact, the question has been fully briefed to the Court of Appeals and raised in this Court. See n. 8, supra. Cf. Sosna v. Iowa, 419 U.S. 393, 396, n. 2, 95 S.Ct. 553, 555, n. 2, 42 L.Ed.2d 532 (1975). I would dismiss this suit and vacate the decision of the Court of Appeals for lack of jurisdiction. 61 II. Exhaustion of Remedies. 62 In view of my belief that this case should be dismissed on jurisdictional grounds, I address the exhaustion question only briefly. Seventeen judges joined in the Court of Appeals' persuasive opinion adopting a rule of "flexible" exhaustion of administrative remedies in § 1983 suits. Other Courts of Appeals have adopted a similar rule. See, e.g., Eisen v. Eastman, 421 F.2d 560 (CA2 1969); Secret v. Brierton, 584 F.2d 823 (CA7 1978). The opinion for the en banc court carefully reviewed the exhaustion doctrine in general and as applied to § 1983 actions. It found that the prior decisions of this Court did not clearly decide the question.18 See Barry v. Barchi, 443 U.S. 55, 63, n. 10, 99 S.Ct. 2642, 2648, n. 10, 61 L.Ed.2d 365 (1979); Gibson v. Berryhill, 411 U.S. 564, 575, n. 14, 93 S.Ct. 1689, 1695, n. 14, 36 L.Ed.2d 488 (1973). And it concluded that the exhaustion of adequate and appropriate state administrative remedies would promote the achievement of the rights protected by § 1983. 63 I agree with the Court of Appeals' opinion. The requirement that a § 1983 plaintiff exhaust adequate state administrative remedies was the accepted rule of law until quite recently. See Eisen v. Eastman, supra, at 567. The rule rests on sound considerations. It does not defeat federal-court jurisdiction, it merely defers it.19 It permits the States to correct violations through their own procedures, and it encourages the establishment of such procedures. It is consistent with the principles of comity that apply whenever federal courts are asked to review state action or supersede state proceedings. See Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971). 64 Moreover, and highly relevant to the effective functioning of the overburdened federal court system, the rule conserves and supplements scarce judicial resources. In 1961, the year that Monroe v. Pape, 365 U.S. 167, 81 S.Ct. 473, 5 L.Ed.2d 492, was decided, only 270 civil rights actions were begun in the federal district courts. Annual Report of the Director of the Administrative Office of the U. S. Courts 238 (1961). In 1981, over 30,000 such suits were commenced.20 Annual Report of the Director of the Administrative Office of the U.S. Courts 63, 68 (1981). The result of this unprecedented increase in civil rights litigation is a heavy burden on the federal courts to the detriment of all federal-court litigants, including others who assert that their constitutional rights have been infringed. 65 The Court argues that past decisions of the Court categorically hold that there is no exhaustion requirement in § 1983 suits. But as the Court of Appeals demonstrates, and as the Court recognizes, many of these decisions can be explained as applications of traditional exceptions to the exhaustion requirement. See McNeese v. Board of Education, 373 U.S. 668, 82 S.Ct. 1433, 10 L.Ed.2d 622 (1963). Other decisions speak to the question in an offhand and conclusory fashion without full briefing and argument. See Wilwording v. Swenson, 404 U.S. 249, 251, 92 S.Ct. 407, 409, 30 L.Ed.2d 418 (1971) (unargued per curiam ); Damico v. California, 389 U.S. 416, 88 S.Ct. 526, 19 L.Ed.2d 647 (1967) (unargued per curiam ). Moreover, a categorical no-exhaustion rule would seem inconsistent with the decision in Younger v. Harris, supra, prescribing abstention when state criminal proceedings are pending. At least where administrative proceedings are pending, Younger would seem to suggest the appropriateness of exhaustion. Cf. Gibson v. Berryhill, 411 U.S., at 574-575, 93 S.Ct., at 1695-1696. Yet the Court today adopts a flat rule without exception. 66 The Court seeks to support its no-exhaustion rule with indications of congressional intent. Finding nothing directly on point in the history of the Civil Rights Act itself, the Court places primary reliance on the recent Civil Rights of Institutionalized Persons Act, 42 U.S.C. § 1997 et seq. (1976 ed., Supp. IV). This legislation was designed to authorize the Attorney General to initiate civil rights actions on behalf of institutionalized persons. § 1997a. The Act also placed certain limits on the existing authority of the Attorney General to intervene in suits begun by institutionalized persons. See § 1997c. In addition, in § 1997e, the Act sets forth an exhaustion requirement but only for § 1983 claims brought by prisoners. 67 On the basis of the exhaustion provision in § 1997e, and remarks primarily by Representative Kastenmeier, the Court contends that Congress has endorsed a general no-exhaustion rule. The irony in this reasoning should be obvious. A principal concern that prompted the Department of Justice to support, and the Congress to adopt, § 1997e was the vast increase in § 1983 suits brought by state prisoners in federal courts. There has been a year-by-year increase in these suits since the mid-1960's. The increase in fiscal 1981 over fiscal 1980 was some 26%, resulting in a total of 15,639 such suits filed in 1981 as compared with 12,397 in 1980. The 1981 total constituted over 8.6% of the total federal district court civil docket. Although most of these cases present frivolous claims, many are litigated through the courts of appeals to this Court. The burden on the system fairly can be described as enormous with few, if any, benefits that would not be available in meritorious cases if exhaustion of appropriate state administrative remedies were required prior to any federal-court litigation. It was primarily this problem that prompted enactment of § 1997e.21 68 Moreover, it is clear from the legislative history that Congress simply was not addressing the exhaustion problem in any general fashion. The concern focused on the problem of prisoner petitions. The new Act had a dual purpose in this respect. In addition to requiring prior exhaustion of adequate state remedies, Congress wished to authorize the Attorney General to act when necessary to protect the constitutional rights of prisoners, but at the same time minimize the need for federal action of any kind by requiring prior exhaustion. Both sponsors of the Act in the Senate made this clear. Senator Hatch explained § 1997e as follows: 69 "In actions relating to alleged violations of the constitutional rights of prisoners, such persons may be required to exhaust internal grievance procedures before the Attorney General can become involved pursuant to [the Act]." 126 Cong.Rec. 3716 (1980) (emphasis added).22 70 Senator Bayh, the author of the Act, described the exhaustion provision in similar terms: "[I]n the event of a prison inmate's rights being alleged to be violated . . . then before the Justice Department could intervene or initiate suits, the prison inmate or class of inmates would have to pursue all of their administrative remedies within the State law before the Justice Department could intervene under the provisions of [the Act]." Id., at 3970. 71 In short, in enacting the Civil Rights of Institutionalized Persons Act Congress was focusing on the powers of the Attorney General, and the particular question of prisoners' suits, not on the general question of exhaustion in § 1983 actions. Also revealing as to the limited purpose of § 1997e is Congress' consistent refusal to adopt legislation imposing a general no-exhaustion requirement. Thus, for example, in 1979, a bill was introduced into the Senate providing: 72 "No court of the United States shall stay or dismiss any civil action brought under this Act on the ground that the party bringing such action failed to exhaust the remedies available in the courts or the administrative agencies of any State." S. 1983, 96th Cong., 1st Sess., § 5 (1979). 73 The bill was never reported out of committee. 74 The requirement that plaintiffs exhaust available and adequate administrative remedies—subject to well-developed exceptions—is firmly established in virtually every area of the law. This is dictated in § 1983 actions by common sense, as well as by comity and federalism, where adequate state administrative remedies are available. 75 If the exhaustion question were properly before us, I would affirm the Court of Appeals. 1 Because this case is here on a motion to dismiss, we accept as true the factual allegations in petitioner's amended complaint. In her initial complaint, petitioner named FIU as the defendant. Relying on Byron v. University of Florida, 403 F.Supp. 49 (N.D.Fla.1975), the District Court granted FIU's motion to dismiss, holding that the Board of Regents and not the individual university had the capacity to sue and be sued under Florida law. The District Court granted petitioner leave to amend, and she amended her complaint to name the Board of Regents "on behalf of" FIU. 2 Petitioner requested the District Court to "[r]equire Defendants to remedy the discrimination practiced upon Plaintiff by promoting her to the next available position consistent with those previously applied for and for which she is qualified or in the alternative, to require the Defendants to pay to the Plaintiff the sum of $500,000 as actual and exemplary damages." Record 47. Petitioner also requested that the District Court "order further equitable and injunctive relief as it deems appropriate and necessary to correct the conditions of discrimination complained of herein." Id., at 48. 3 The other factors discussed in Monell —whether the decisions in question constituted a departure from prior decisions and whether overruling these decisions would frustrate legitimate reliance on their holdings—do not support overruling these decisions. McNeese was not a departure from prior decisions—this Court had not previously addressed the application of the exhaustion rule to § 1983 actions. Overruling these decisions might injure those § 1983 plaintiffs who had forgone or waived their state administrative remedies in reliance on these decisions. 4 Congressional intent is important in determining the application of the exhaustion doctrine to cases in which federal administrative remedies are available, as well as to those in which state remedies are available. Of course, exhaustion is required where Congress provides that certain administrative remedies shall be exclusive. See Myers v. Bethlehem Shipbuilding Corp., 303 U.S. 41, 58 S.Ct. 459, 82 L.Ed. 638 (1938). Even where the statutory requirement of exhaustion is not explicit, courts are guided by congressional intent in determining whether application of the doctrine would be consistent with the statutory scheme. In determining whether exhaustion of federal administrative remedies is required, courts generally focus on the role Congress has assigned to the relevant federal agency, and tailor the exhaustion rule to fit the particular administrative scheme created by Congress. See McKart v. United States, 395 U.S. 185, 193-195, 89 S.Ct. 1657, 1662-1663, 23 L.Ed.2d 194 (1969). With state administrative remedies, the focus is not so much on the role assigned to the state agency, but the role of the state agency becomes important once a court finds that deferring its exercise of jurisdiction is consistent with statutory intent. 5 Some of the debates relating to § 2, which created certain federal crimes in addition to those defined in § 2 of the 1866 Civil Rights Act, 14 Stat. 27, aimed primarily at the Ku Klux Klan, are also relevant to our discussion of § 1. 6 Opponents of the bill also recognized this purpose and complained that the bill would usurp the States' power, centralize the government, and perhaps ultimately destroy the States. See, e.g., Globe 337, 338 (remarks of Rep. Whitthorne); id., at 352 (remarks of Rep. Beck); id., at 361 (remarks of Rep. Swann); id., at 365 (remarks of Rep. Arthur); id., at 385 (remarks of Rep. Lewis); id., at 429, 431 (remarks of Rep. McHenry); id., at 454 (remarks of Rep. Cox); id., at 510, 511 (remarks of Rep. Eldridge); Cong.Globe, 42d Cong., 1st Sess., App. 46 (1871) (remarks of Rep. Kerr) (hereinafter Globe App.); id., at 216 (remarks of Sen. Thurman); id., at 243 (remarks of Sen. Bayard). 7 Opponents criticized this provision on this very ground. For example, Representative Storm lamented: "[Section one] does not even give the State courts a chance to try questions, or to show whether they will try the questions that might come before them under the first section of the fourteenth amendment, fairly or not. It takes the whole question away from them in the beginning." Id., at 86. See also Globe 416 (remarks of Rep. Biggs) ("for the violation of the rights, privileges, and immunities of the citizen a civil remedy is to be had by proceedings in the Federal courts, State authorization in the premises to the contrary notwithstanding"); id., at 337 (remarks of Rep. Whitthorne); id., at 373 (remarks of Rep. Archer); Globe App. 216 (remarks of Sen. Thurman). 8 This view was expressed in the Presidential message urging the passing of corrective legislation. See Globe 244 ("That the power to correct these evils is beyond the control of State authorities I do not doubt") (message of President Grant). The inability of state authorities to protect constitutional rights was also expressed in the findings of the House Judiciary Committee, which had been directed to investigate the situation. See id., at 320. The resolution introduced by Senator Sherman instructing the Senate Judiciary Committee to report a bill expressed a similar view. See Globe App. 210 (state "courts are rendered utterly powerless by organized perjury to punish crime"). 9 Opponents viewed the bill as a declaration of mistrust for state tribunals. See, e.g., Globe 361 (remarks of Rep. Swann); id., at 397 (remarks of Rep. Rice); id., at 454 (remarks of Rep. Cox); Globe App. 216 (remarks of Sen. Thurman). Representative McHenry found particularly offensive the removal of the fact-finding function from the local institutions. See Globe 429. 10 Representative Kastenmeier explains why juveniles were not included in § 1997e: "I think very candidly we should admit that the first reluctance to resort to this mechanism embodied in [§ 1997e] was resisted as a possible encroachment on civil liberties; that is to say, in the free, unimpeded resort to 1983; because it does deflect 1983 petitions back into—temporarily in any event—back into the State system. Therefore, to the extent that it is even so viewed, notwithstanding the limited form of [§ 1997e], that it should also extend to juveniles was rejected." 1979 Hearings 26. 11 Section 1997e(b)(2) states: "The minimum standards shall provide— "(A) for an advisory role for employees and inmates of any jail, prison, or other correctional institution (at the most decentralized level as is reasonably possible), in the formulation, implementation, and operation of the system; "(B) specific maximum time limits for written replies to grievances with reasons thereto at each decision level within the system; "(C) for priority processing of grievances which are of an emergency nature, including matters in which delay would subject the grievant to substantial risk of personal injury or other damages; "(D) for safeguards to avoid reprisals against any grievant or participant in the resolution of a grievance; and "(E) for independent review of the disposition of grievances, including alleged reprisals, by a person or other entity not under the direct supervision or direct control of the institution." 12 The Committee Reports state that Congress did not intend that every § 1983 action brought by an adult prisoner in institutions with appropriate grievance procedures be delayed pending exhaustion: "It is the intent of the Congress that the court not find such a requirement appropriate in those situations in which the action brought pursuant to [§ 1983] raises issues which cannot, in reasonable probability, be resolved by the grievance resolution system, including cases where imminent danger to life is alleged. Allegations unrelated to conditions of confinement, such as those which center on events outside of the institution, would not appropriately be continued for resolution by the grievance resolution system." Conf.Rep. 15. See also H.R.Rep.No.96-80, p. 25 (1979); S.Rep.No.96-416, p. 34 (1979). 13 Of course, this burden alone is not sufficient to justify a judicial decision to alter congressionally imposed jurisdiction. See Thermtron Products, Inc. v. Hermansdorfer, 423 U.S. 336, 344, 96 S.Ct. 584, 589, 46 L.Ed.2d 542 (1976); Steelworkers v. Bouligny, Inc., 382 U.S. 145, 150-151, 86 S.Ct. 272, 274-275, 15 L.Ed.2d 217 (1965). In any event, it is by no means clear that judicial discretion to impose an exhaustion requirement in § 1983 actions would lessen the caseload of the federal courts, at least in the short run. See infra, at 513-514, and n. 18. 14 The application of these federalism principles to actions brought pursuant to § 1983 has prompted criticism by several commentators. See, e.g., Koury, Section 1983 and Civil Comity: Two for the Federalism Seesaw, 25 Loyola L.Rev. 659 (1979); Note, 39 N.Y.U.L.Rev. 838 (1964). 15 For example, there is serious disagreement over whether judicial or administrative procedures offer § 1983 plaintiffs the swiftest, least costly, and most reliable remedy. See, e.g., 1977 Hearings 263-264; id., at 232-233; Note, 68 Colum.L.Rev. 1201, 1207 (1968). Similarly, there is debate over whether the specialization of federal courts in constitutional law is more important than the specialization of administrative agencies in their areas of expertise, and over whether the symbolic and institutional function of federal courts in defining, legitimizing, and enforcing constitutional claims outweighs the educational function that state and local agencies can serve. See, e.g., Whitman, Constitutional Torts, 79 Mich.L.Rev. 5, 23 (1980); Note, 68 Colum.L.Rev., supra, at 1208. Finally, it is uncertain whether the present "free market" system, under which litigants are free to pursue administrative remedies if they truly appear to be cheaper, more efficient, and more effective, is more likely to induce the creation of adequate remedies than a McKart-type standard under which plaintiffs have no initial choice. See, e.g., Note, 8 Ind.L.Rev. 565 (1975). Cf. 1977 Hearings 21, 34, 51; Hearings on S.1393 before the Subcommittee on the Constitution of the Senate Committee on the Judiciary, 95th Cong., 1st Sess., 442 (1977). 16 Section 1997e resolved this problem by directing the Attorney General to promulgate minimum standards and to establish a procedure by which prison administrative remedies could be reviewed and certified. §§ 1997e(b) and (c). If a procedure has not been certified, the court is directed to compare the procedure with the Attorney General's standards and to continue the case pending exhaustion only if the procedure is in substantial compliance with the standards of the Attorney General. § 1997e(a)(2). 17 Unless the doctrine that statutes of limitations are not tolled pending exhaustion were overruled, see Board of Regents v. Tomanio, 446 U.S. 478, 100 S.Ct. 1790, 64 L.Ed.2d 440 (1980), a judicially imposed exhaustion requirement might result in the effective repeal of § 1983. Congress avoided this problem in § 1997e by directing the court to merely continue the case for a period not to exceed 90 days. 18 The initial bill proposing to include an exhaustion requirement in § 1997e provided: "Relief shall not be granted by a district court in an action brought pursuant to [§ 1983] by an individual involuntarily confined in any State institution . . ., unless it appears that the individual has exhausted such plain, speedy, and efficient State administrative remedy as is available." H.R. 5791, 95th Cong., 1st Sess., 4 (1977). Congress declined to adopt this McKart-type standard after witnesses testified that this procedure would bog down the courts in massive procedural litigation thereby frustrating the purpose of relieving the caseloads of the federal courts, that state procedures are often not effective and take too much time, and that the court would have to judge a myriad of state procedures without much guidance. See, e.g., 1977 Hearings 34-35, 51, 164-165, 169-170, 263-264, 323; 1979 Hearings 48-49. 19 The question was posed from the bench at oral argument whether the Eleventh Amendment might bar this suit on the ground that the Board of Regents is an arm of the State for purposes of the Eleventh Amendment. Tr. of Oral Arg. 20. Cf. Alabama v. Pugh, 438 U.S. 781, 98 S.Ct. 3057, 57 L.Ed.2d 1114 (1978). Compare Hopkins v. Clemson Agricultural College, 221 U.S. 636, 31 S.Ct. 654, 55 L.Ed. 890 (1911), with Florida Dept. of Health v. Florida Nursing Home Assn., 450 U.S. 147, 101 S.Ct. 1032, 67 L.Ed.2d 132 (1981). The District Court dismissed this action on the pleadings, and no Eleventh Amendment issue had been raised. The Board of Regents first raised this issue in its brief to the original panel on appeal, but did not argue it in its brief on rehearing en banc. Neither the original panel nor the en banc court addressed this issue. Although the State mentioned a possible Eleventh Amendment defense in its response in opposition to the petition for certiorari, it did not brief the issue or press it at oral argument. Indeed, counsel for respondent urged that we affirm the Court of Appeals solely on its exhaustion holding. Tr. of Oral Arg. 24, 27. We have noted that "the Eleventh Amendment defense sufficiently partakes of the nature of a jurisdictional bar" that it may be raised by the State for the first time on appeal. Edelman v. Jordan, 415 U.S. 651, 678, 94 S.Ct. 1347, 1363, 39 L.Ed.2d 662 (1974). However, because of the importance of state law in analyzing Eleventh Amendment questions and because the State may, under certain circumstances, waive this defense, we have never held that it is jurisdictional in the sense that it must be raised and decided by this Court on its own motion. Cf. Mt. Healthy City Bd. of Ed. v. Doyle, 429 U.S. 274, 279, 97 S.Ct. 568, 572, 50 L.Ed.2d 471 (1977). Where, as here, the Board of Regents expressly requested that we address the exhaustion question and not pass on its potential Eleventh Amendment immunity, and, as a consequence, the parties have not briefed the issue, we deem it appropriate to address the issue that was raised and decided below and vigorously pressed in this Court. Nothing in this opinion precludes the Board of Regents from raising its Eleventh Amendment claim on remand. The District Court is in the best position to address in the first instance the competing questions of fact and state law necessary to resolve the Eleventh Amendment issue, and at this stage it has the discretion to permit amendments to the pleadings that might cure any potential Eleventh Amendment problems. * In my view, this case does not present a serious Eleventh Amendment issue. The Florida statute authorizing suits against the Board of Regents, Fla.Stat. § 240.205 (1981), is clear on its face. I see no reason to read a broad waiver to sue and be sued in "all courts of law and equity" as meaning all but federal courts. Nor am I aware of anything in Florida law that suggests a more limited meaning was intended than indicated by the unequivocal terms of the statute. Certainly, none of our cases have gone so far as to hold that federal courts must be expressly mentioned for an effective Eleventh Amendment waiver. The statutes at issue in cases recited by Justice POWELL, post, at 522-523, n. 5, presented more equivocal embodiments of state intent. For example, in Florida Dept. of Health v. Florida Nursing Home Assn., 450 U.S. 147, 101 S.Ct. 1032, 67 L.Ed.2d 132 (1981) (per curiam), the authorization to sue and be sued was limited to contract actions and, unlike the instant provision, did not extend to "all courts of law and equity." The same is true of the interstate compact involved in Petty v. Tennessee-Missouri Bridge Comm'n, 359 U.S. 275, 79 S.Ct. 785, 3 L.Ed.2d 804 (1959). The decision in Kennecott Copper Corp. v. Tax Comm'n, 327 U.S. 573, 66 S.Ct. 745, 90 L.Ed. 862 (1946), which involved a statute providing for suit in "any court of competent jurisdiction," turned on the incongruity of federal courts' interpreting state tax laws and the fact that "Utah employs explicit language to indicate, in other litigation, its consent to suits in federal courts." Id., at 579, 66 S.Ct., at 313. Thus, while I do not object to the Court's leaving the Eleventh Amendment issue for further consideration by the lower courts—at least where, as here, there is no logical priority in resolving Eleventh Amendment immunity before exhaustion—I find the issue sufficiently clear to be answered here and now. The statute means what it says. 1 The Eleventh Amendment provides: "The Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State, or by Citizens or Subjects of any Foreign State." 2 As the Court notes, see ante, at 498, n. 1, petitioner originally named the Florida International University as defendant. Because the Florida International University lacks the capacity to sue or be sued, the District Court found that it was not a proper defendant. Petitioner was permitted to amend her complaint, and she simply substituted the Board of Regents. In addition to racial discrimination, petitioner also claimed that she had been discriminated against on the basis of her sex. 3 The Court repeatedly has held that the defense of the Eleventh Amendment may be raised for the first time on appeal. See Edelman v. Jordan, 415 U.S. 651, 678, 94 S.Ct. 1347, 1363, 39 L.Ed.2d 662 (1974) ("Eleventh Amendment defense sufficiently partakes of the nature of a jurisdictional bar so that it need not be raised in the trial court"). The Board's brief on appeal was divided into three parts. Part III was devoted to the argument that "the Eleventh Amendment precludes subject matter jurisdiction over plaintiff's complaint." Brief for Defendant-Appellee in No. 79-2965 (CA5), p. 17. A lengthy statutory addendum was attached in support of the arguments advanced in this section of the brief. After the case was scheduled for rehearing en banc, the parties filed short—i.e. 4- and 10-page—supplemental briefs to be considered in addition to the main briefs already submitted to the Court of Appeals. The supplemental briefs did not add to the discussion of the Eleventh Amendment issue. But the question was placed before the Court of Appeals en banc, as it had been placed before the panel, through the thorough discussion in the main briefs. This Court's explanation for not addressing the Eleventh Amendment issue is that it was not considered below. See ante, at 515-516, n. 19. But contrary to the implication in the Court's explanation, the issue—as shown here—was urged by the Board and argued here. 4 The Board of Regents of the Division of Universities of the Department of Education is established by the Florida Education Code as a part of the State University System. Fla.Stat. § 240.2011 (1981). The Board consists of the Commissioner of Education and 12 citizens appointed by the Governor. § 240.207. The Board has general supervisory authority over the State University System. § 240.209. Among its duties are the appointment of university presidents, the review of budget requests of each university in the state system, the preparation of an aggregated budget for the State University System, the development of a master plan, and the establishment of a systemwide personnel classification and pay plan. Ibid. The Board is an agency of the State of Florida. § 216.011. See Relyea v. State, 385 So.2d 1378 (Fla.App.1980). It may claim the defense of sovereign immunity in suits under state law. See ibid. Numerous Courts of Appeals have held state universities or state Boards of Regents immune from suit in federal court by reason of the Eleventh Amendment. See, e.g., Rutledge v. Arizona Board of Regents, 660 F.2d 1345, 1349 (CA9 1981); Brennan v. University of Kansas, 451 F.2d 1287 (CA10 1971); Ronwin v. Shapiro, 657 F.2d 1071 (CA9 1981). 5 See, e.g., Florida Dept. of Health v. Florida Nursing Home Assn., 450 U.S. 147, 150, 101 S.Ct. 1032, 1034, 67 L.Ed.2d 132 (1981); Petty v. Tennessee-Missouri Bridge Comm'n, 359 U.S. 275, 276-277, 79 S.Ct. 785, 3 L.Ed.2d 804 (1959) ("The conclusion that there has been a waiver of immunity will not be lightly inferred. . . . And where a public instrumentality is created with the right 'to sue and be sued' that waiver of immunity in the particular setting may be restricted to suits or proceedings of a special character in the state, not the federal courts"); Kennecott Copper Corp. v. State Tax Comm'n, 327 U.S. 573, 66 S.Ct. 745, 90 L.Ed. 862 (1946) (language in state statute providing for suit in "any court of competent jurisdiction" will not be understood as a waiver of the Eleventh Amendment); Ford Motor Co. v. Department of Treasury of Indiana, 323 U.S. 459, 65 S.Ct. 347, 89 L.Ed. 389 (1945) (same); Great Northern Life Insurance Co. v. Read, 322 U.S. 47, 54, 64 S.Ct. 873, 876, 88 L.Ed. 1121 (1944) ("a clear declaration of the state's intention to submit its fiscal problems to other courts than those of its own creation must be found"); Jagnandan v. Giles, 538 F.2d 1166, 1177 (CA5 1976). Cf. Edelman v. Jordan, 415 U.S., at 673, 94 S.Ct., at 1360 ("In deciding whether a State has waived its constitutional protection under the Eleventh Amendment, we will find waiver only where stated 'by the most express language or by such overwhelming implications from the text as [will] leave no room for any other reasonable construction' "). It is difficult to reconcile the Court's consistent requirement of an express waiver with the approach advocated by Justice WHITE. See ante, at 519, n. At oral argument here counsel for respondent stated that the Florida Legislature had not waived the Eleventh Amendment and had waived the defense of sovereign immunity "only in selected tort cases." Tr. of Oral Arg. 26. See Bragg v. Board of Public Instruction, 160 Fla. 590, 36 So.2d 222 (1948) ("The mere fact that the Board of Public Instruction is created as a body corporate with power to sue and be sued does not affect its immunity from tort"); Relyea v. State, supra (Board of Regents retains defense of sovereign immunity); Fla.Stat. § 111.071(1)(b), (4) (1981) (provision for payment by the State of civil rights judgments against state officers—including judgments under 42 U.S.C. § 1983 (1976 ed., Supp.IV)—does not waive sovereign immunity "or any other defense or immunity" to such lawsuits). Cf. Long v. Richardson, 525 F.2d 74, 79 (CA6 1975) (state university's immunity from suit under state law disposes of Eleventh Amendment question). 6 See Brief in Opposition 23 ("Should this Court grant the writ, the Board respectfully submits that review should be limited to the jurisdictional issues discussed below and this Court should vacate the Fifth Circuit's decision with instructions to dismiss [petitioner's] suit for lack of jurisdiction"). The Court, ante, at 516, n. 19, attaches importance to the statement at oral argument by counsel for the Board that the Board wanted the exhaustion issue decided. This must be viewed, however, in light of the Board's unsuccessful attempt to have this Court first decide the Eleventh Amendment issue. Moreover, a party's request—short of a binding waiver—cannot relieve this Court of its duty to resolve a jurisdictional question. 7 Tr. of Oral Arg. 25-28, 40-41. At oral argument, the Board's counsel stated that the Eleventh Amendment question had not been addressed in its main briefs to this Court "because of the grant of certiorari." Id., at 27. 8 In view of the Board's repeated efforts to raise the Eleventh Amendment question, and its specific request that this Court vacate the decision of the Court of Appeals for lack of jurisdiction, see n. 6, supra, it is hardly correct to say that the Court must now raise the question of jurisdiction on its own motion. Cf. Sosna v. Iowa, 419 U.S. 393, 396, n. 2, 95 S.Ct. 553, 555, n. 2, 42 L.Ed.2d 532 (1975). In any event, "we are obliged to inquire sua sponte whenever a doubt arises as to the existence of federal jurisdiction." Mt. Healthy City Bd. of Ed. v. Doyle, 429 U.S. 274, 278, 97 S.Ct. 568, 571, 50 L.Ed.2d 471 (1977). 9 The Court cites, with a "compare" signal, to Mt. Healthy City Bd. of Ed. v. Doyle, supra, at 279, 97 S.Ct., at 572. The Mt. Healthy Court in no way suggested that the Eleventh Amendment and the principle of sovereign immunity embodied in Art. III were less than jurisdictional. Indeed, the Court found it necessary to resolve the Eleventh Amendment question in that case prior to reaching the merits. On the contrary, the Court consistently has viewed the Amendment as jurisdictional. In Sosna v. Iowa, supra, at 396, n. 2, 95 S.Ct., 555, n. 2, the Court raised the question of the Eleventh Amendment even though the State had asserted the bar of the Amendment only in its answer to the complaint and had thereafter abandoned this defense. Unlike the Board of Regents in this case, the State of Iowa had not advanced the defense in this Court. Even so, the Sosna Court raised and addressed the question. These precedents are ignored by the Court today. 10 "Because of their unusual nature, and because it would not simply be wrong but indeed would be an unconstitutional invasion of the powers reserved to the states if the federal courts were to entertain cases not within their jurisdiction, the rule is well settled that the party seeking to invoke the jurisdiction of a federal court must demonstrate that the case is within the competence of that court." C. Wright & A. Miller, Federal Practice and Procedure § 3522, p. 45 (1975). 11 "[T]he Eleventh Amendment was introduced to clarify the intent of the Framers concerning the reach of the federal judicial power. . . . The Eleventh Amendment served effectively to reverse the particular holding in Chisholm, and, more generally, to restore the original understanding. . . . Thus, despite the narrowness of the language of the Amendment, its spirit has consistently guided this Court in interpreting the reach of the federal judicial power generally. . . ." Employees v. Missouri Public Health Dept., 411 U.S. 279, 291-292, 93 S.Ct. 1614, 1621-1622, 36 L.Ed.2d 251 (1973) (MARSHALL, J., concurring in result). 12 The Hans Court quoted at some length from the constitutional debates concerning the scope of Art. III. In the eighty-first number of the Federalist, for example, Hamilton sought to dispel the suggestion that Art. III extended federal jurisdiction over suits brought against one of the States: " 'It is inherent in the nature of sovereignty not to be amenable to the suit of an individual without its consent. This is the general sense and the general practice of mankind; and the exemption, as one of the attributes of sovereignty, is now enjoyed by the government of every State in the Union.' " As quoted in 134 U.S., at 13, 10 S.Ct., at 506 (emphasis in original). 13 Unlike other limitations on federal jurisdiction, the limitation imposed by the Eleventh Amendment and the doctrine of sovereign immunity may be waived by consent unequivocally expressed. This was the understanding of the doctrine at the time the Constitution was adopted, see n. 11, supra, and the Court has interpreted the "judicial power of the United States" as used in the Eleventh Amendment and Art. III accordingly. But the fact that the State or the United States may consent to federal jurisdiction, does not render the Eleventh Amendment or the doctrine of sovereign immunity embodied in Art. III "quasi" jurisdictional. Quite simply, where there has not been consent, there is no jurisdiction. See United States v. Sherwood, 312 U.S. 584, 586, 61 S.Ct. 767, 769, 85 L.Ed. 1058 (1941) ("The United States, as sovereign, is immune from suit save as it consents to be sued, . . . and the terms of its consent to be sued in any court define that court's jurisdiction to entertain the suit"); United States v. United States Fidelity & Guaranty Co., 309 U.S. 506, 514, 60 S.Ct. 653, 657, 84 L.Ed. 894 (1940) ("Consent alone gives jurisdiction to adjudge against a sovereign. Absent that consent, the attempted exercise of judicial power is void"). 14 See Edelman v. Jordan, 415 U.S., at 678, 94 S.Ct., at 1363; Sosna v. Iowa, 419 U.S., at 396, n. 2, 95 S.Ct., at 555, n. 2; Mt. Healthy City Bd. of Ed. v. Doyle, 429 U.S., at 278, 97 S.Ct., at 571. The Court has consistently viewed the Eleventh Amendment question as jurisdictional. See Great Northern Life Insurance Co. v. Read, 322 U.S., at 51, 64 S.Ct., at 875 ("A state's freedom from litigation was established as a constitutional right through the Eleventh Amendment") (emphasis added); Monaco v. Mississippi, 292 U.S. 313, 320, 54 S.Ct. 745, 746, 78 L.Ed. 1282 (1934) (Question is "whether this Court has jurisdiction to entertain a suit brought by a foreign State against a State without her consent") (emphasis added). 15 The States consented to a diminution of their sovereignty by ratifying the Fourteenth Amendment. In its exercise of the powers granted to it by § 5 of the Fourteenth Amendment, Congress may lift the bar of sovereign immunity. See Fitzpatrick v. Bitzer, 427 U.S. 445, 96 S.Ct. 2666, 49 L.Ed.2d 614 (1976). Thus, if petitioner had brought this suit under Title VII of the Civil Rights Act of 1964, there would have been no jurisdictional problem. But petitioner did not do so, and the Court has held that Congress has not removed the bar of sovereign immunity in § 1983 actions. See Quern v. Jordan, 440 U.S. 332, 99 S.Ct. 1139, 59 L.Ed.2d 358 (1979). 16 "It would be a novel proposition indeed that the Eleventh Amendment does not bar a suit to enjoin the State itself simply because no money judgment is sought. . . . [T]he Eleventh Amendment by its terms clearly applies to a suit seeking an injunction, a remedy available only from equity." Cory v. White, 457 U.S., at 90-91, 102 S.Ct., at 2329. 17 Under the theory of Ex parte Young the Eleventh Amendment does not bar suits against state officers because when a state officer "comes into conflict with the superior authority of [the] Constitution, . . . he is . . . stripped of his official or representative character." 209 U.S., at 159-160, 28 S.Ct., at 453. The rationale of that decision has no application to suits against the State or its agencies. Although an individual official may be viewed as acting on his own and without state authority when acting against federal law, the State—or an agency of the State—cannot act other than in its official state capacity. Similarly, an action for damages against the State, or an arm of the State, seeks damages that must be paid from the State's own coffers—whether the damages come directly from the State's general fund or from some other state fund. See Kennecott Copper Corp. v. State Tax Comm'n, 327 U.S. 573, 66 S.Ct. 745, 90 L.Ed. 862 (1946) (segregated funds of the State Tax Commission are state moneys subject to the Eleventh Amendment). Moreover, the fact that the Board is a corporate entity under state law does not permit application of the rule in Ex parte Young to the Board itself—as if the Board were an official. This Court repeatedly has held the Eleventh Amendment to bar suit against such state corporate agencies. See Florida Dept. of Health v. Florida Nursing Home Assn., 450 U.S. 147, 101 S.Ct. 1032, 67 L.Ed.2d 132 (1981); Great Northern Insurance Co. v. Read, supra; Ford Motor Co. v. Department of Treasury of Indiana, 323 U.S. 459, 65 S.Ct. 347, 89 L.Ed. 389 (1945); Kennecott Copper Corp. v. State Tax Comm'n, supra. Hopkins v. Clemson Agricultural College, 221 U.S. 636, 31 S.Ct. 654, 55 L.Ed. 890 (1911), is not to the contrary. In that case suit was brought against a state college in state court to recover damages caused by the college's construction of a dyke. Although the Court discussed the Eleventh Amendment in some detail, there was simply no Eleventh Amendment question in that case. It was clear before Hopkins that the Eleventh Amendment did not apply to bar review in this Court of any federal question presented in a suit against a State in state court. See Chandler v. Dix, 194 U.S. 590, 592, 24 S.Ct. 766, 767, 48 L.Ed. 1129 (1904). Cf. University of California Regents v. Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 570 (1978). Moreover, the Hopkins Court did not consider the college's activities in that case to be governmental. 221 U.S., at 647, 31 S.Ct., at 658. In short, no Eleventh Amendment question was presented to the Court. The opinion in Hopkins has never been cited by this Court for the proposition that the Eleventh Amendment is no bar to suit against a state corporate agency in federal court. See Florida Dept. of Health v. Florida Nursing Home Assn., supra; Alabama v. Pugh, 438 U.S. 781, 98 S.Ct. 3057, 57 L.Ed.2d 1114 (1978); Parden v. Terminal R. Co., 377 U.S. 184, 84 S.Ct. 1207, 12 L.Ed.2d 233 (1964). 18 "[I]n all the cases in which the Supreme Court has articulated its no-exhaustion rule, the state administrative remedies were sufficiently inadequate that exhaustion would not have been appropriate in any event." Developments in the Law, Section 1983 and Federalism, 90 Harv.L.Rev. 1133, 1274 (1977). 19 Cf. Fair Assessment in Real Estate Assn. v. McNary, 454 U.S. 100, 136, 102 S.Ct. 177, 196, 70 L.Ed.2d 271 (1981) (BRENNAN, J., concurring in judgment) (exhaustion requirement in § 1983 cases can be justified by "a somewhat lesser showing . . . where . . . we are concerned not with the displacement of the § 1983 remedy, but with the deferral of federal court consideration pending exhaustion of the state administrative process"). 20 Of the approximately 30,000 civil rights suits filed in fiscal year 1981, 15,639 were filed by state prisoners under § 1983. The remainder involved a variety of civil rights suits. Annual Report of the Director of the Administrative Office of the U. S. Courts 63, 68 (1981). See Parratt v. Taylor, 451 U.S. 527, 554, n. 13, 101 S.Ct. 1908, 1922, n. 13, 68 L.Ed.2d 420 (1981) (POWELL, J., concurring in result). 21 The exhaustion requirement in § 1997e only becomes effective if the Attorney General or a federal district court determines that the available prison grievance procedures comply with standards set forth in subsection (b) of § 1997e. As of this date, the Department of Justice has not certified the inmate grievance procedures of even a single State. 22 Senator Hatch offered the same explanation on several other occasions in the course of the debate. See 126 Cong.Rec. 9227 (1980) ("Section 7 would establish specific procedures that would be applicable before the Attorney General could enter into an action in behalf of an imprisoned or incarcerated person. Such person would first have had to fully exhaust all internal grievance mechanisms that existed in the institution in which he was confined"); Id., at 10005 ("Section 7(D) further clarifies that the administrative grievance procedures established in section 7 are only for the purposes of requiring prisoners to exhaust internal grievance mechanisms before the Attorney General can litigate on his behalf").

89

457 U.S. 596 102 S.Ct. 2613 73 L.Ed.2d 248 GLOBE NEWSPAPER COMPANY, Appellantv.SUPERIOR COURT FOR the COUNTY OF NORFOLK. No. 81-611. Argued March 29, 1982. Decided June 23, 1982. Syllabus Appellee Massachusetts trial court, relying on a Massachusetts statute providing for exclusion of the general public from trials of specified sexual offenses involving a victim under the age of 18, ordered the exclusion of the press and public from the courtroom during the trial of a defendant charged with rape of three minor girls. Appellant newspaper publisher challenged the exclusion order, and ultimately, after the trial had resulted in the defendant's acquittal, the Massachusetts Supreme Judicial Court construed the Massachusetts statute as requiring, under all circumstances, the exclusion of the press and public during the testimony of a minor victim in a sex-offense trial. Held : 1. The fact that the exclusion order expired with completion of the trial at which the defendant was acquitted does not render the controversy moot within the meaning of Art. III. The controversy is "capable of repetition, yet evading review," since it can reasonably be assumed that appellant will someday be subjected to another order relying on the Massachusetts statute and since criminal trials are typically of short duration. Pp.602-603 2. The Massachusetts statute, as construed by the Massachusetts Supreme Judicial Court, violates the First Amendment as applied to the States through the Fourteenth Amendment. Pp. 603-607. (a) To the extent that the First Amendment embraces a right of access to criminal trials, it is to ensure that the constitutionally protected "discussion of governmental affairs" is an informed one. The right of access to criminal trials in particular is properly afforded protection by the First Amendment both because such trials have historically been open to the press and public and because such right of access plays a particularly significant role in the functioning of the judicial process and the government as a whole. Pp. 603-606. (b) The right of access to criminal trials is not absolute, but the circumstances under which the press and public can be barred are limited. The State must show that denial of such right is necessitated by a compelling governmental interest and is narrowly tailored to serve that interest. Pp.606-607 3. The Massachusetts statute cannot be justified on the basis of either the State's interest in protecting minor victims of sex crimes from further trauma and embarrassment or its interest in encouraging such victims to come forward and testify in a truthful and credible manner. Pp. 607-610. (a) Compelling as the first interest is, it does not justify a mandatory closure rule. Such interest could be just as well served by requiring the trial court to determine on a case-by-case basis whether the State's legitimate concern for the minor victim's well-being necessitates closure. Such an approach ensures that the constitutional right of the press and public to gain access to criminal trials will not be restricted except where necessary to protect the State's interest. Pp. 607-609. (b) The second asserted interest is not only speculative in empirical terms but is also open to serious question as a matter of logic and common sense. Although the statute was construed to bar the press and public from the courtroom during a minor sex victim's testimony, the press is not denied access to the transcript, court personnel, or any other source that could provide an account of such testimony, and thus the statute cannot prevent the press from publicizing the substance of that testimony, as well as the victim's identity. Pp.609-610. 383 Mass. 838, 423 N.E.2d 773, reversed. James F. McHugh, III, Boston, Mass., for appellant. Mitchell J. Sikora, Jr., Boston, Mass., for appellee. Justice BRENNAN delivered the opinion of the Court. 1 Section 16A of Chapter 278 of the Massachusetts General Laws,1 as construed by the Massachusetts Supreme Judicial Court, requires trial judges, at trials for specified sexual offenses involving a victim under the age of 18, to exclude the press and general public from the courtroom during the testimony of that victim. The question presented is whether the statute thus construed violates the First Amendment as applied to the States through the Fourteenth Amendment. 2 * The case began when appellant, Globe Newspaper Co. (Globe), unsuccessfully attempted to gain access to a rape trial conducted in the Superior Court for the County of Norfolk, Commonwealth of Massachusetts. The criminal defendant in that trial had been charged with the forcible rape and forced unnatural rape of three girls who were minors at the time of trial—two 16 years of age and one 17. In April 1979, during hearings on several preliminary motions, the trial judge ordered the courtroom closed.2 Before the trial began, Globe moved that the court revoke this closure order, hold hearings on any future such orders, and permit appellant to intervene "for the limited purpose of asserting its rights to access to the trial and hearings on related preliminary motions." App. 12a-14a. The trial court denied Globe's motions,3 relying on Mass.Gen.Laws Ann., ch. 278, § 16A (West 1981), and ordered the exclusion of the press and general public from the courtroom during the trial. The defendant immediately objected to that exclusion order, and the prosecution stated for purposes of the record that the order was issued on the court's "own motion and not at the request of the Commonwealth." App. 18a. 3 Within hours after the court had issued its exclusion order, Globe sought injunctive relief from a justice of the Supreme Judicial Court of Massachusetts.4 The next day the justice conducted a hearing, at which the Commonwealth, "on behalf of the victims," waived "whatever rights it [might] have [had] to exclude the press." Id., at 28a.5 Nevertheless, Globe's request for relief was denied. Before Globe appealed to the full court, the rape trial proceeded and the defendant was acquitted. 4 Nine months after the conclusion of the criminal trial, the Supreme Judicial Court issued its judgment, dismissing Globe's appeal. Although the court held that the case was rendered moot by completion of the trial, it nevertheless stated that it would proceed to the merits, because the issues raised by Globe were "significant and troublesome, and . . . 'capable of repetition yet evading review.' " Globe Newspaper Co. v. Superior Court, 379 Mass. 846, 848, 401 N.E.2d 360, 362 (1980), quoting Southern Pacific Terminal Co. v. ICC, 219 U.S. 498, 515, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911). As a statutory matter, the court agreed with Globe that § 16A did not require the exclusion of the press from the entire criminal trial. The provision was designed, the court determined, "to encourage young victims of sexual offenses to come forward; once they have come forward, the statute is designed to preserve their ability to testify by protecting them from undue psychological harm at trial." 379 Mass., at 860, 401 N.E.2d, at 369. Relying on these twin purposes, the court concluded that § 16A required the closure of sex-offense trials only during the testimony of minor victims; during other portions of such trials, closure was "a matter within the judge's sound discretion." Id., at 864, 401 N.E.2d, at 371. The court did not pass on Globe's contentions that it had a right to attend the entire criminal trial under the First and Sixth Amendments, noting that it would await this Court's decision—then pending—in Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980).6 5 Globe then appealed to this Court. Following our decision in Richmond Newspapers, we vacated the judgment of the Supreme Judicial Court, and remanded the case for further consideration in light of that decision. Globe Newspaper Co. v. Superior Court, 449 U.S. 894, 101 S.Ct. 259, 66 L.Ed.2d 124 (1980). 6 On remand, the Supreme Judicial Court, adhering to its earlier construction of § 16A, considered whether our decision in Richmond Newspapers required the invalidation of the mandatory closure rule of § 16A. 383 Mass. 838, 423 N.E.2d 773 (1981).7 In analyzing the First Amendment issue,8 the court recognized that there is "an unbroken tradition of openness" in criminal trials. Id., at 845, 423 2d, at 778. But the court discerned "at least one notable exception" to this tradition: "In cases involving sexual assaults, portions of trials have been closed to some segments of the public, even when the victim was an adult." Id., at 846, 423 N.E.2d, at 778. The court also emphasized that § 16A's mandatory closure rule furthered "genuine State interests," which the court had identified in its earlier decision as underlying the statutory provision. These interests, the court stated, "would be defeated if a case-by-case determination were used." Id., at 851, 423 N.E.2d, at 779. While acknowledging that the mandatory-closure requirement results in a "temporary diminution" of "the public's knowledge about these trials," the court did not think "that Richmond Newspapers require[d] the invalidation of the requirement, given the statute's narrow scope in an area of traditional sensitivity to the needs of victims." Id., at 851, 423 N.E.2d, at 781. The court accordingly dismissed Globe's appeal.9 7 Globe again sought review in this Court. We noted probable jurisdiction. 454 U.S. 1051, 102 S.Ct. 594, 70 L.Ed.2d 586 (1981). For the reasons that follow, we reverse, and hold that the mandatory-closure rule contained in § 16A violates the First Amendment.10 II 8 In this Court, Globe challenges that portion of the trial court's order, approved by the Supreme Judicial Court of Massachusetts, that holds that § 16A requires, under all circumstances, the exclusion of the press and general public during the testimony of a minor victim in a sex-offense trial. Because the entire order expired with the completion of the rape trial at which the defendant was acquitted, we must consider at the outset whether a live controversy remains. Under Art. III, § 2, of the Constitution, our jurisdiction extends only to actual cases or controversies. Nebraska Press Assn. v. Stuart, 427 U.S. 539, 546, 96 S.Ct. 2791, 2796, 49 L.Ed.2d 683 (1976). "The Court has recognized, however, that jurisdiction is not necessarily defeated simply because the order attacked has expired, if the underlying dispute between the parties is one 'capable of repetition, yet evading review.' " Ibid., quoting Southern Pacific Terminal Co. v. ICC, 219 U.S., at 515, 31 S.Ct., at 283. 9 The controversy between the parties in this case is indeed "capable of repetition, yet evading review." It can reasonably be assumed that Globe, as the publisher of a newspaper serving the Boston metropolitan area, will someday be subjected to another order relying on § 16A's mandatory closure rule. See Gannett Co. v. DePasquale, 443 U.S. 368, 377-378, 99 S.Ct. 2898, 2904, 61 L.Ed.2d 608 (1979); Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 563, 100 S.Ct., at 2820 (plurality opinion). And because criminal trials are typically of "short duration," ibid., such an order will likely "evade review, or at least considered plenary review in this Court." Nebraska Press Assn. v. Stuart, supra, 427 U.S., at 547, 96 S.Ct., at 2797. We therefore conclude that the controversy before us is not moot within the meaning of Art. III, and turn to the merits. III A. 10 The Court's recent decision in Richmond Newspapers firmly established for the first time that the press and general public have a constitutional right of access to criminal trials. Although there was no opinion of the Court in that case, seven Justices recognized that this right of access is embodied in the First Amendment, and applied to the States through the Fourteenth Amendment. 448 U.S., at 558-581, 100 S.Ct., at 2818-2830 (plurality opinion); id., at 584-598, 100 S.Ct., at 2832-2839 (BRENNAN, J., concurring in judgment); id., at 598-601, 100 S.Ct., at 2839-2841 (STEWART, J., concurring in judgment); id., at 601-604, 100 S.Ct., at 2841-2842 (BLACKMUN, J., concurring in judgment).11 11 Of course, this right of access to criminal trials is not explicitly mentioned in terms in the First Amendment.12 But we have long eschewed any "narrow, literal conception" of the Amendment's terms, NAACP v. Button, 371 U.S. 415, 430, 83 S.Ct. 328, 336, 9 L.Ed.2d 405 (1963), for the Framers were concerned with broad principles, and wrote against a background of shared values and practices. The First Amendment is thus broad enough to encompass those rights that, while not unambiguously enumerated in the very terms of the Amendment, are nonetheless necessary to the enjoyment of other First Amendment rights. Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 579-580, and n. 16, 100 S.Ct., at 2828-2829, and n. 16 (plurality opinion) (citing cases); id., at 587-588, and n. 4, 100 S.Ct., at 2833-2834, and n. 4 (BRENNAN, J., concurring in judgment). Underlying the First Amendment right of access to criminal trials is the common understanding that "a major purpose of that Amendment was to protect the free discussion of governmental affairs,"Mills v. Alabama, 384 U.S. 214, 218, 86 S.Ct. 1434, 1436, 16 L.Ed.2d 484 (1966). By offering such protection, the First Amendment serves to ensure that the individual citizen can effectively participate in and contribute to our republican system of self-government. See Thornhill v. Alabama, 310 U.S. 88, 95, 60 S.Ct. 736, 740, 84 L.Ed. 1093 (1940); Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 587-588, 100 S.Ct., at 2833-2834 (BRENNAN, J., concurring in judgment). See also id., at 575, 100 S.Ct., at 2826 (plurality opinion) (the "expressly guaranteed freedoms" of the First Amendment "share a common core purpose of assuring freedom of communication on matters relating to the functioning of government"). Thus to the extent that the First Amendment embraces a right of access to criminal trials, it is to ensure that this constitutionally protected "discussion of governmental affairs" is an informed one. 12 Two features of the criminal justice system, emphasized in the various opinions in Richmond Newspapers, together serve to explain why a right of access to criminal trials in particular is properly afforded protection by the First Amendment. First, the criminal trial historically has been open to the press and general public. "[A]t the time when our organic laws were adopted, criminal trials both here and in England had long been presumptively open." Richmond Newspapers, Inc. v. Virginia, supra, at 569, 100 S.Ct., at 2823 (plurality opinion). And since that time, the presumption of openness has remained secure. Indeed, at the time of this Court's decision in In re Oliver, 333 U.S. 257, 68 S.Ct. 499, 92 L.Ed. 682 (1948), the presumption was so solidly grounded that the Court was "unable to find a single instance of a criminal trial conducted in camera in any federal, state, or municipal court during the history of this country." Id., at 266, 68 S.Ct., at 504 (footnote omitted). This uniform rule of openness has been viewed as significant in constitutional terms not only "because the Constitution carries the gloss of history," but also because "a tradition of accessibility implies the favorable judgment of experience." Richmond Newspapers, Inc. v. Virginia, supra, 448 U.S., at 589, 100 S.Ct., at 2834 (BRENNAN, J., concurring in judgment).13 13 Second, the right of access to criminal trials plays a particularly significant role in the functioning of the judicial process and the government as a whole. Public scrutiny of a criminal trial enhances the quality and safeguards the integrity of the factfinding process, with benefits to both the defendant and to society as a whole.14 Moreover, public access to the criminal trial fosters an appearance of fairness, thereby heightening public respect for the judicial process.15 And in the broadest terms, public access to criminal trials permits the public to participate in and serve as a check upon the judicial process—an essential component in our structure of self-government.16 In sum, the institutional value of the open criminal trial is recognized in both logic and experience. B 14 Although the right of access to criminal trials is of constitutional stature, it is not absolute. See Richmond Newspapers, Inc. v. Virginia, supra, at 581, n. 18, 100 S.Ct., at 2830, n. 18 (plurality opinion); Nebraska Press Assn. v. Stuart, 427 U.S., at 570, 96 S.Ct., at 2808. But the circumstances under which the press and public can be barred from a criminal trial are limited; the State's justification in denying access must be a weighty one. Where, as in the present case, the State attempts to deny the right of access in order to inhibit the disclosure of sensitive information, it must be shown that the denial is necessitated by a compelling governmental interest, and is narrowly tailored to serve that interest. See, e.g., Brown v. Hartlage, 456 U.S. 45, 53-54, 102 S.Ct. 1523, 1529, 71 L.Ed.2d 732 (1982); Smith v. Daily Mail Publishing Co., 443 U.S. 97, 101-103, 99 S.Ct. 2667, 2669-2671, 61 L.Ed.2d 399 (1979); NAACP v. Button, 371 U.S., at 438, 83 S.Ct., at 340.17 We now consider the state interests advanced to support Massachusetts' mandatory rule barring press and public access to criminal sex-offense trials during the testimony of minor victims. IV 15 The state interests asserted to support § 16A, though articulated in various ways, are reducible to two: the protection of minor victims of sex crimes from further trauma and embarrassment; and the encouragement of such victims to come forward and testify in a truthful and credible manner.18 We consider these interests in turn. 16 We agree with appellee that the first interest—safeguarding the physical and psychological well-being of a minor19—is a compelling one. But as compelling as that interest is, it does not justify a mandatory closure rule, for it is clear that the circumstances of the particular case may affect the significance of the interest. A trial court can determine on a case-by-case basis whether closure is necessary to protect the welfare of a minor victim.20 Among the factors to be weighed are the minor victim's age, psychological maturity and understanding, the nature of the crime, the desires of the victim,21 and the interests of parents and relatives. Section 16A, in contrast, requires closure even if the victim does not seek the exclusion of the press and general public, and would not suffer injury by their presence.22 In the case before us, for example, the names of the minor victims were already in the public record,23 and the record indicates that the victimsmay have been willing to testify despite the presence of the press.24 If the trial court had been permitted to exercise its discretion, closure might well have been deemed unnecessary. In short, § 16A cannot be viewed as a narrowly tailored means of accommodating the State's asserted interest: That interest could be served just as well by requiring the trial court to determine on a case-by-case basis whether the State's legitimate concern for the well-being of the minor victim necessitates closure. Such an approach ensures that the constitutional right of the press and public to gain access to criminal trials will not be restricted except where necessary to protect the State's interest.25 17 Nor can § 16A be justified on the basis of the Commonwealth's second asserted interest—the encouragement of minor victims of sex crimes to come forward and provide accurate testimony. The Commonwealth has offered no empirical support for the claim that the rule of automatic closure contained in § 16A will lead to an increase in the number of minor sex victims coming forward and cooperating with state authorities.26 Not only is the claim speculative in empirical terms, but it is also open to serious question as a matter of logic and common sense. Although § 16A bars the press and general public from the courtroom during the testimony of minor sex victims, the press is not denied access to the transcript, court personnel, or any other possible source that could provide an account of the minor victim's testimony. Thus § 16A cannot prevent the press from publicizing the substance of a minor victim's testimony, as well as his or her identity. If the Commonwealth's interest in encouraging minor victims to come forward depends on keeping such matters secret, § 16A hardly advances that interest in an effective manner. And even if § 16A effectively advanced the State's interest, it is doubtful that the interest would be sufficient to overcome the constitutional attack, for that same interest could be relied on to support an array of mandatory closure rules designed to encourage victims to come forward: Surely it cannot be suggested that minor victims of sex crimes are the only crime victims who, because of publicity attendant to criminal trials, are reluctant to come forward and testify. The State's argument based on this interest therefore proves too much, and runs contrary to the very foundation of the right of access recognized in Richmond Newspapers: namely, "that a presumption of openness inheres in the very nature of a criminal trial under our system of justice." 448 U.S., at 573, 100 S.Ct., at 2825 (plurality opinion). V 18 For the foregoing reasons, we hold that § 16A, as construed by the Massachusetts Supreme Judicial Court, violates the First Amendment to the Constitution.27 Accordingly, the judgment of the Massachusetts Supreme Judicial Court is 19 Reversed. 20 Justice O'CONNOR, concurring in the judgment. 21 In Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980), the Court held that the First Amendment protects the right of press and public to attend criminal trials. I do not interpret that decision to shelter every right that is "necessary to the enjoyment of other First Amendment rights." Ante, at 604. Instead, Richmond Newspapers rests upon our long history of open criminal trials and the special value, for both public and accused, of that openness. As the plurality opinion in Richmond Newspapers stresses, "it would be difficult to single out any aspect of government of higher concern and importance to the people than the manner in which criminal trials are conducted." 448 U.S., at 575, 100 S.Ct., at 2826. Thus, I interpret neither Richmond Newspapers nor the Court's decision today to carry any implications outside the context of criminal trials. 22 This case, however, does involve a criminal trial. Moreover, it involves a statute mandating automatic exclusion of the public from certain testimony. As the Court explains, Massachusetts has demonstrated no interest weighty enough to justify application of its automatic bar to all cases, even those in which the victim, defendant, and prosecutor have no objection to an open trial. Accordingly, I concur in the judgment. 23 Chief Justice BURGER, with whom Justice REHNQUIST joins, dissenting. 24 Historically our society has gone to great lengths to protect minors charged with crime, particularly by prohibiting the release of the names of offenders, barring the press and public from juvenile proceedings, and sealing the records of those proceedings. Yet today the Court holds unconstitutional a state statute designed to protect not the accused, but the minor victims of sex crimes. In doing so, it advances a disturbing paradox. Although states are permitted, for example, to mandate the closure of all proceedings in order to protect a 17-year-old charged with rape, they are not permitted to require the closing of part of criminal proceedings in order to protect an innocent child who has been raped or otherwise sexually abused. 25 The Court has tried to make its holding a narrow one by not disturbing the authority of state legislatures to enact more narrowly drawn statutes giving trial judges the discretion to exclude the public and the press from the courtroom during the minor victim's testimony. Ante, at 611, n. 27. I also do not read the Court's opinion as foreclosing a state statute which mandates closure except in cases where the victim agrees to testify in open court.1 But the Court's decision is nevertheless a gross invasion of state authority and a state's duty to protect its citizens—in this case minor victims of crime. I cannot agree with the Court's expansive interpretation of our decision in Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980), or its cavalier rejection of the serious interests supporting Massachusetts' mandatory closure rule. Accordingly, I dissent. 26 * The Court seems to read our decision in Richmond Newspapers, supra, as spelling out a First Amendment right of access to all aspects of all criminal trials under all circumstances. Ante, at 605, n. 13. That is plainly incorrect. In Richmond Newspapers, we examined "the right of access to places traditionally open to the public" and concluded that criminal trials were generally open to the public throughout this country's history and even before that in England. The opinions of a majority of the Justices emphasized the historical tradition of open criminal trials. 448 U.S., at 564-573, 100 S.Ct., at 2821-2825; id., at 589-591, 100 S.Ct., at 2834-2835 (BRENNAN, J., concurring in judgment); id., at 599, 100 S.Ct., at 2840 (STEWART, J., concurring in judgment); id., at 601, 100 S.Ct., at 2841 (BLACKMUN, J., concurring in judgment). The proper mode of analysis to be followed in determining whether there is a right of access was emphasized by Justice BRENNAN: "As previously noted, resolution of First Amendment public access claims in individual cases must be strongly influenced by the weight of historical practice and by an assessment of the specific structural value of public access in the circumstances." Id., at 597-598, 100 S.Ct., at 2838-2839. 27 Today Justice BRENNAN ignores the weight of historical practice. There is clearly a long history of exclusion of the public from trials involving sexual assaults, particularly those against minors. See, e.g., Harris v. Stephens, 361 F.2d 888 (CA8 1966), cert. denied, 386 U.S. 964, 87 S.Ct. 1040, 18 L.Ed.2d 113 (1967); Reagan v. United States, 202 F. 488 (CA9 1913); United States v. Geise, 158 F.Supp. 821, 17 Alaska 461 (Alaska), aff'd, 262 F.2d 151 (CA9 1958), cert. denied, 361 U.S. 842, 80 S.Ct. 94, 4 L.Ed.2d 80 (1959); Hogan v. State, 191 Ark. 437, 86 S.W.2d 931 (1935); State v. Purvis, 157 Conn. 198, 251 A.2d 178 (1968), cert. denied, 395 U.S. 928, 89 S.Ct. 1788, 23 L.Ed.2d 246 (1969); Moore v. State, 151 Ga. 648, 108 S.E. 47 (1921), appeal dism'd, 260 U.S. 702, 43 S.Ct. 98, 67 L.Ed. 471 (1922).2 Several States have longstanding provisions allowing closure of cases involving sexual assaults against minors.3 28 It would misrepresent the historical record to state that there is an "unbroken, uncontradicted history" of open proceedings in cases involving the sexual abuse of minors. Richmond Newspapers, supra, 448 U.S., at 573, 100 S.Ct., at 2825. Absent such a history of openness, the positions of the Justices joining reversal in Richmond Newspapers give no support to the proposition that closure of the proceedings during the testimony of the minor victim violates the First Amendment.4 II 29 The Court does not assert that the First Amendment right it discerns from Richmond Newspapers is absolute; instead, it holds that when a "State attempts to deny the right of access in order to inhibit the disclosure of sensitive information, it must be shown that the denial is necessitated by a compelling governmental interest, and is narrowly tailored to serve that interest." Ante, at 606-607. The Court's wooden application of the rigid standard it asserts for this case is inappropriate. The Commonwealth has not denied the public or the media access to information as to what takes place at trial. As the Court acknowledges, Massachusetts does not deny the press and the public access to the trial transcript or to other sources of information about the victim's testimony. Even the victim's identity is part of the public record, although the name of a 16-year-old accused rapist generally would not be a matter of public record. Mass.Gen.Laws Ann., ch. 119, § 60A (West Supp.1982-1983). The Commonwealth does not deny access to information, and does nothing whatever to inhibit its disclosure. This case is quite unlike others in which we have held unconstitutional state laws which prevent the dissemination of information or the public discussion of ideas. See, e.g., Brown v. Hartlage, 456 U.S. 45, 102 S.Ct. 1523, 71 L.Ed.2d 732 (1982); Smith v. Daily Mail Publishing Co., 443 U.S. 97, 99 S.Ct. 2667, 61 L.Ed.2d 399 (1979); Landmark Communications, Inc. v. Virginia, 435 U.S. 829, 98 S.Ct. 1534, 56 L.Ed.2d 1 (1978); Nebraska Press Assn. v. Stuart, 427 U.S. 539, 96 S.Ct. 2791, 49 L.Ed.2d 683 (1976); Cox Broadcasting Corp. v. Cohen, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975); NAACP v. Button, 371 U.S. 415, 83 S.Ct. 328, 9 L.Ed.2d 405 (1963). 30 The purpose of the Commonwealth in enacting § 16A was to give assurance to parents and minors that they would have this moderate and limited protection from the trauma, embarrassment, and humiliation of having to reveal the intimate details of a sexual assault in front of a large group of unfamiliar spectators—and perhaps a television audience—and to lower the barriers to the reporting of such crimes which might come from the victim's dread of public testimony. Globe Newspaper Co. v. Superior Court, 379 Mass. 846, 865, 401 N.E.2d 360, 372 (1980); 383 Mass. 838, 847-848, 423 N.E.2d 773, 779 (1981). 31 Neither the purpose of the law nor its effect is primarily to deny the press or public access to information; the verbatim transcript is made available to the public and the media and may be used without limit. We therefore need only examine whether the restrictions imposed are reasonable and whether the interests of the Commonwealth override the very limited incidental effects of the law on First Amendment rights. See Richmond Newspapers, 448 U.S., at 580-581, 100 S.Ct., at 2829-2830 (plurality opinion); id., at 600, 100 S.Ct., at 2840 (Stewart, J., concurring in judgment); Pell v. Procunier, 417 U.S. 817, 94 S.Ct. 2800, 41 L.Ed.2d 495 (1974); Saxbe v. Washington Post Co., 417 U.S. 843, 94 S.Ct. 2811, 41 L.Ed.2d 514 (1974); Cox v. New Hampshire, 312 U.S. 569, 61 S.Ct. 762, 85 L.Ed. 1049 (1941). Our obligation in this case is to balance the competing interests: the interests of the media for instant access, against the interest of the State in protecting child rape victims from the trauma of public testimony. In more than half the states, public testimony will include television coverage. III 32 For me, it seems beyond doubt, considering the minimal impact of the law on First Amendment rights and the overriding weight of the Commonwealth's interest in protecting child rape victims, that the Massachusetts law is not unconstitutional. The Court acknowledges that the press and the public have prompt and full access to all of the victim's testimony. Their additional interest in actually being present during the testimony is minimal. While denying it the power to protect children, the Court admits that the Commonwealth's interest in protecting the victimized child is a compelling interest. Ante, at 607. This meets the test of Richmond Newspapers, supra. 33 The law need not be precisely tailored so long as the state's interest overrides the law's impact on First Amendment rights and the restrictions imposed further that interest. Certainly this law, which excludes the press and public only during the actual testimony of the child victim of a sex crime, rationally serves the Commonwealth's overriding interest in protecting the child from the severe—possibly permanent psychological damage. It is not disputed that such injury is a reality.5 34 The law also seems a rational response to the undisputed problem of the underreporting of rapes and other sexual offenses. The Court rejects the Commonwealth's argument that § 16A is justified by its interest in encouraging minors to report sex crimes, finding the claim "speculative in empirical terms [and] open to serious question as a matter of logic and common sense." Ante, at 609-610. There is no basis whatever for this cavalier disregard of the reality of human experience. It makes no sense to criticize the Commonwealth for its failure to offer empirical data in support of its rule; only by allowing state experimentation may such empirical evidence be produced. "It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country." New State Ice Co. v. Liebmann, 285 U.S. 262, 311, 52 S.Ct. 371, 386, 76 L.Ed. 747 (1932) (Brandeis, J., dissenting). See also Chandler v. Florida, 449 U.S. 560, 579-580, 101 S.Ct. 802, 812, 66 L.Ed.2d 740 (1981); Reeves, Inc. v. Stake, 447 U.S. 429, 441, 100 S.Ct. 2271, 2279, 65 L.Ed.2d 244 (1980); Whalen v. Roe, 429 U.S. 589, 597, and n. 20, 97 S.Ct. 869, 875, and n. 20, 51 L.Ed.2d 64 (1977). 35 The Court also concludes that the Commonwealth's assertion that the law might reduce underreporting of sexual offenses fails "as a matter of logic and common sense." This conclusion is based on a misperception of the Commonwealth's argument and an overly narrow view of the protection the statute seeks to afford young victims. The Court apparently believes that the statute does not prevent any significant trauma, embarrassment, or humiliation on the part of the victim simply because the press is not prevented from discovering and publicizing both the identity of the victim and the substance of the victim's testimony. Ante, at 2622. Section 16A is intended not to preserve confidentiality, but to prevent the risk of severe psychological damage caused by having to relate the details of the crime in front of a crowd which inevitably will include voyeuristic strangers.6 In most states, that crowd may be expanded to include a live television audience, with reruns on the evening news. That ordeal could be difficult for an adult; to a child, the experience can be devastating and leave permanent scars.7 36 The Commonwealth's interests are clearly furthered by the mandatory nature of the closure statute. Certainly if the law were discretionary, most judges would exercise that discretion soundly and would avoid unnecessary harm to the child, but victims and their families are entitled to assurance of such protection. The legislature did not act irrationally in deciding not to leave the closure determination to the idiosyncracies of individual judges subject to the pressures available to the media. The victim might very well experience considerable distress prior to the court appearance, wondering, in the absence of such statutory protection, whether public testimony will be required. The mere possibility of public testimony may cause parents and children to decide not to report these heinous crimes. If, as psychologists report, the courtroom experience in such cases is almost as traumatic as the crime itself,8 a state certainly should be able to take whatever reasonable steps it believes are necessary to reduce that trauma. Furthermore, we cannot expect victims and their parents to be aware of all of the nuances of state law; a person who sees newspaper, or perhaps even television, reports of a minor victim's testimony may very well be deterred from reporting a crime on the belief that public testimony will be required. It is within the power of the state to provide for mandatory closure to alleviate such understandable fears and encourage the reporting of such crimes. IV 37 There is, of course, "a presumption of openness [that] inheres in the very nature of a criminal trial under our system of justice." But we have consistently emphasized that this presumption is not absolute or irrebuttable. A majority of the Justices in Richmond Newspapers acknowledged that closure might be permitted under certain circumstances. Justice Stewart's separate opinion pointedly recognized that exclusion of the public might be justified to protect "the sensibilities of a youthful prosecution witness . . . in a criminal trial for rape." 448 U.S., at 600, n. 5, 100 S.Ct., at 2840, n. 5.9 The Massachusetts statute has a relatively minor incidental impact on First Amendment rights and gives effect to the overriding state interest in protecting child rape victims. Paradoxically, the Court today denies the victims the kind of protection routinely given to juveniles who commit crimes. Many will find it difficult to reconcile the concern so often expressed for the rights of the accused with the callous indifference exhibited today for children who, having suffered the trauma of rape or other sexual abuse, are denied the modest protection the Massachusetts Legislature provided. 38 Justice STEVENS, dissenting. 39 The duration of a criminal trial generally is shorter than the time it takes for this Court's jurisdiction to be invoked and our judgment on the merits to be announced. As a result, our power to review pretrial or midtrial orders implicating the freedom of the press has rested on the exception to the mootness doctrine for orders "capable of repetition, yet evading review." See Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 563, 100 S.Ct. 2814, 2820, 65 L.Ed.2d 973; Gannett Co. v. DePasquale, 443 U.S. 368, 377-378, 99 S.Ct. 2898, 2904, 61 L.Ed.2d 608; Nebraska Press Assn. v. Stuart, 427 U.S. 539, 546-547, 96 S.Ct. 2791, 2796-2797, 49 L.Ed.2d 683. 40 Today the Court expands that exception in order to pass on the constitutionality of a statute that, as presently construed, has never been applied in a live controversy. In this case, unlike the three cases cited above, the governing state law was materially changed after the trial court's order had expired by its own terms. There consequently is no possibility " 'that the same complaining party will be subject to the same action again.' " Gannett Co. v. DePasquale, supra, 443 U.S., at 377, 99 S.Ct., at 2904 (quoting Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 348, 46 L.Ed.2d 350). 41 The fact that the Massachusetts Supreme Judicial Court narrowly construed—and then upheld in the abstract—the state statute that the trial court had read to mandate the closure of the entire trial bears on our review function in other respects. We have only recently recognized the First Amendment right of access to newsworthy matter. See ante, at 603; Richmond Newspapers, Inc. v. Virginia, supra, at 582, 100 S.Ct., at 2830 (STEVENS, J., concurring). In developing constitutional jurisprudence, there is a special importance in deciding cases on concrete facts. Cf. Minnick v. California Dept. of Corrections, 452 U.S. 105, 120-127, 101 S.Ct. 2211, 2219-2223, 68 L.Ed.2d 706; United States v. Raines, 362 U.S. 17, 21, 80 S.Ct. 519, 522, 4 L.Ed.2d 524. Only in specific controversies can the Court decide how this right of access to criminal trials can be accommodated with other societal interests, such as the protection of victims or defendants. The advisory opinion the Court announces today sheds virtually no light on how such rights should be accommodated. 42 The question whether the Court should entertain a facial attack on a statute that bears on the right of access cannot be answered simply by noting that the right has its source in the First Amendment. See, e.g., Bates v. State Bar of Arizona, 433 U.S. 350, 380-381, 97 S.Ct. 2691, 2707-2708, 53 L.Ed.2d 810; Young v. American Mini Theatres, Inc., 427 U.S. 50, 61, 96 S.Ct. 2440, 2447, 49 L.Ed.2d 310. For the right of access is plainly not coextensive with the right of expression that was vindicated in Nebraska Press Assn., supra.1 Because statutes that bear on this right of access do not deter protected activity in the way that other laws sometimes interfere with the right of expression, we should follow the norm of reviewing these statutes as applied rather than on their face. 43 It is not clear when, if ever, the Court will need to confront the question whether a mandatory partial-closure statute is unconstitutional. If the order hypothesized by the Supreme Judicial Court, instead of the trial court's order, had actually been entered in this case, and if the press had been given prompt access to a transcript of the testimony of the minor victims, appellant might not even have appealed. At the very least the press, the prosecutor, and defense counsel would have argued the constitutionality of the partial-closure order in the context of the facts relevant to such an order, and a different controversy would have been framed for appellate review. In future cases the trial courts may voluntarily follow the direction of Justice Wilkins and make specific findings demonstrating a compelling state interest supporting the mandated partial-closure order. See 383 Mass. 838, 852-853, 423 N.E.2d 773, 782 (concurring opinion). Or the record in future cases may plainly disclose a justification for a partial closure that the Court would consider acceptable. Thus, aside from the illumination provided by live controversies, a decision to review only orders actually entered pursuant to the Massachusetts statute would advance the policy of avoiding the premature and unnecessary adjudication of constitutional questions;2 it is at least conceivable that no such order may ever have to be justified by the conclusion of the legislature that the mandatory closure of the trial during the testimony of a minor victim of a sex crime is necessary to serve important state interests. 44 The Court does not hold that on this record a closure order limited to the testimony of the minor victims would have been unconstitutional. Rather, the Court holds only that if ever such an order is entered, it must be supported by adequate findings. Normally, if the constitutional deficiency is the absence of findings to support a trial order, the Court would either remand for factfinding, or examine the record itself, before deciding whether the order measured up to constitutional standards. The infeasibility of this course of action—since no such order was entered in this case and since the order that was entered has expired—further demonstrates that the Court's comment on the First Amendment issues implicated by the Massachusetts statute is advisory, hypothetical, and, at best, premature.3 45 I would dismiss the appeal. 1 Massachusetts Gen.Laws Ann., ch. 278, § 16A (West 1981), which provides in pertinent part: "At the trial of a complaint or indictment for rape, incest, carnal abuse or other crime involving sex, where a minor under eighteen years of age is the person upon, with or against whom the crime is alleged to have been committed, . . . the presiding justice shall exclude the general public from the court room, admitting only such persons as may have a direct interest in the case." 2 "The court caused a sign marked 'closed' to be placed on the courtroom door, and court personnel turned away people seeking entry." Globe Newspaper Co. v. Superior Court, 379 Mass. 846, 848, 401 N.E.2d 360, 362-363 (1980) (footnote omitted). 3 The court refused to permit Globe to file its motion to intervene and explicitly stated that it would not act on Globe's other motions. App. 17a-18a. 4 Globe's request was contained in a petition for extraordinary relief filed pursuant to Mass.Gen.Laws Ann., ch. 211, § 3 (West 1958 and Supp.1982-1983). 5 The Commonwealth's representative stated: "[O]ur position before the trial judge [was], and it is before this Court, that in some circumstances a trial judge, where the defendant is asserting his right to a constitutional, public trial, . . . may consider that as outweighing the otherwise legitimate statutory interests, particularly where the Commonwealth [acts] on behalf of the victims, and this is literally on behalf of the victims in the sense that they were consulted fully by the prosecutor in this case. The Commonwealth waives whatever rights it may have to exclude the press." App. 28a. Some time after the trial began, the prosecuting attorney informed the judge at a lobby conference that she had "spoke[n] with each of the victims regarding . . . excluding the press." Id., at 48a. The prosecuting attorney indicated that the victims had expressed some "privacy concerns" that were based on "their own privacy interests, as well as the fact that there are grandparents involved with a couple of these victims." Ibid. But according to the prosecuting attorney, the victims "wouldn't object to the press being included" if "it were at all possible to obtain a guarantee" that the press would not attempt to interview them or publish their names, photographs, or any personal information. Ibid. In fact, their names were already part of the public record. See 383 Mass. 838, 849, 423 N.E.2d 773, 780 (1981). It is not clear from the record, however, whether or not the victims were aware of this fact at the time of their discussions with the prosecuting attorney. 6 Justice Quirico dissented, being of the view that the mandatory closure rule of § 16A was not limited to the testimony of minor victims, but was applicable to the entire trial. 7 The court again noted that the First Amendment issue arising from the closure of the then-completed trial was " 'capable of repetition yet evading review.' " Id., at 841, n. 4, 423 N.E.2d, at 775, n. 4, quoting Southern Pacific Terminal Co. v. ICC, 219 U.S. 498, 515, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911). But in contrast to the view it had taken in its prior opinion, supra, at 2617, the court held that the case was not moot because of this possibility of repetition without opportunity for review. 8 The court found it unnecessary to consider Globe's argument that the mandatory closure rule violated the Sixth Amendment rights of the criminal defendant who had been acquitted in the rape trial. Those Sixth Amendment rights, the court stated, were "personal rights" that, "at least in the context of this case, [could] only be asserted by the original criminal defendant." 383 Mass., at 842, 423 N.E.2d, at 776 (footnote omitted). 9 Justice Wilkins filed a concurring opinion in which he expressed concern whether a statute constitutionally could require closure "without specific findings by the judge that the closing is justified by overriding or countervailing interests of the Commonwealth." Id., at 852, 423 N.E.2d, at 782. 10 We therefore have no occasion to consider Globe's additional argument that the provision violates the Sixth Amendment. 11 Justice POWELL took no part in the consideration or decision of Richmond Newspapers. But he had indicated previously in a concurring opinion in Gannett Co. v. DePasquale, 443 U.S. 368, 99 S.Ct. 2898, 61 L.Ed.2d 608 (1979), that he viewed the First Amendment as conferring on the press a right of access to criminal trials. Id., at 397-398, 99 S.Ct., at 2914. 12 "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances." U.S.Const., Amdt. 1. 13 Appellee argues that criminal trials have not always been open to the press and general public during the testimony of minor sex victims. Brief for Appellee 13-22. Even if appellee is correct in this regard, but see Gannett Co. v. DePasquale, supra, at 423, 99 S.Ct., at 2928 (BLACKMUN, J., concurring in part and dissenting in part), the argument is unavailing. In Richmond Newspapers, the Court discerned a First Amendment right of access to criminal trials based in part on the recognition that as a general matter criminal trials have long been presumptively open. Whether the First Amendment right of access to criminal trials can be restricted in the context of any particular criminal trial, such as a murder trial (the setting for the dispute in Richmond Newspapers ) or a rape trial, depends not on the historical openness of that type of criminal trial but rather on the state interests assertedly supporting the restriction. See Part III-B, infra. 14 See Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 569, 100 S.Ct., at 2823 (plurality opinion); id., at 596-597, 100 S.Ct., at 2838-2839 (BRENNAN, J., concurring in judgment); Gannett Co. v. DePasquale, 443 U.S., at 383, 99 S.Ct., at 2907; id., at 428-429, 99 S.Ct., at 2930-2931 (BLACKMUN, J., concurring in part and dissenting in part). 15 See Levine v. United States, 362 U.S. 610, 616, 80 S.Ct. 1038, 1042, 4 L.Ed.2d 989 (1960); In re Oliver, 333 U.S. 257, 268-271, 68 S.Ct. 499, 505-506, 92 L.Ed. 682 (1948); Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 570-571, 100 S.Ct., at 2824 (plurality opinion); id., at 595, 100 S.Ct., at 2837 (BRENNAN, J., concurring in judgment); Gannett Co. v. DePasquale, supra, 443 U.S., at 428-429, 99 S.Ct., at 2930-2931 (BLACKMUN, J., concurring in part and dissenting in part). 16 See Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 570-571, 100 S.Ct., at 2824 (plurality opinion); id., at 596, 100 S.Ct., at 2838 (BRENNAN, J., concurring in judgment); Gannett Co. v. DePasquale, 443 U.S., at 394, 99 S.Ct., at 2912 (BURGER, C. J., concurring); id., at 428, 99 S.Ct., at 2930 (BLACKMUN, J., concurring in part and dissenting in part). 17 Of course, limitations on the right of access that resemble "time, place, and manner" restrictions on protected speech, see Young v. American Mini Theatres, Inc., 427 U.S. 50, 63, n. 18, 96 S.Ct. 2440, 2448, n.18, 49 L.Ed.2d 310 (1976), would not be subjected to such strict scrutiny. See Richmond Newspapers, Inc. v. Virginia, 448 U.S., at 581-582, n. 18, 100 S.Ct., at 2830, n.18 (plurality opinion); id., at 598, n. 23, 100 S.Ct., at 2839 n. 23 (BRENNAN, J., concurring in judgment); id., at 600, 100 S.Ct., at 2840 (Stewart, J., concurring in judgment). 18 In its opinion following our remand, the Supreme Judicial Court of Massachusetts described the interests in the following terms: "(a) to encourage minor victims to come forward to institute complaints and give testimony . . .; (b) to protect minor victims of certain sex crimes from public degradation, humiliation, demoralization, and psychological damage . . .; (c) to enhance the likelihood of credible testimony from such minors, free of confusion, fright, or embellishment; (d) to promote the sound and orderly administration of justice . . .; (e) to preserve evidence and obtain just convictions." 383 Mass., at 848, 423 N.E.2d, at 779. 19 It is important to note that in the context of § 16A, the measure of the State's interest lies not in the extent to which minor victims are injured by testifying, but rather in the incremental injury suffered by testifying in the presence of the press and the general public. 20 Indeed, the plurality opinion in Richmond Newspapers suggested that individualized determinations are always required before the right of access may be denied: "Absent an overriding interest articulated in findings, the trial of a criminal case must be open to the public." 448 U.S., at 581, 100 S.Ct., at 2830 (footnote omitted) (emphasis added). 21 "[I]f the minor victim wanted the public to know precisely what a heinous crime the defendant had committed, the imputed legislative justifications for requiring the closing of the trial during the victim's testimony would in part, at least, be inapplicable." 383 Mass., at 853, 423 N.E.2d, at 782 (Wilkins, J., concurring). 22 It appears that while other States have statutory or constitutional provisions that would allow a trial judge to close a criminal sex-offense trial during the testimony of a minor victim, no other State has a mandatory provision excluding both the press and general public during such testimony. See, e.g., Ala.Code § 12-21-202 (1975); Ariz.Rule Crim.Proc. 9.3; Ga.Code § 81-1006 (1978); La.Rev.Stat.Ann. § 15:469.1 (West 1981); Miss.Const., Art. 3, § 26; N.H.Rev.Stat.Ann. § 632-A:8 (Supp.1981); N.Y.Jud.Law § 4 (McKinney 1968); N.C.Gen.Stat. § 15-166 (Supp.1981); N.D.Cent.Code § 27-01-02 (1974); Utah Code Ann. § 78-7-4 (1953); Vt.Stat.Ann., Tit. 12, § 1901 (1973); Wis.Stat. § 970.03(4) (1979-1980). See also Fla.Stat. § 918.16 (1979) (providing for mandatory exclusion of general public but not press during testimony of minor victims). Of course, we intimate no view regarding the constitutionality of these state statutes. 23 The Court has held that the government may not impose sanctions for the publication of the names of rape victims lawfully obtained from the public record. Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975). See also Smith v. Daily Mail Publishing Co., 443 U.S. 97, 99 S.Ct. 2667, 61 L.Ed.2d 399 (1979). 24 See n. 5, supra. 25 Of course, for a case-by-case approach to be meaningful, representatives of the press and general public "must be given an opportunity to be heard on the question of their exclusion." Gannett Co. v. DePasquale, 443 U.S., at 401, 99 S.Ct., at 2916 (POWELL, J., concurring). This does not mean, however, that for purposes of this inquiry the court cannot protect the minor victim by denying these representatives the opportunity to confront or cross-examine the victim, or by denying them access to sensitive details concerning the victim and the victim's future testimony. Such discretion is consistent with the traditional authority of trial judges to conduct in camera conferences. See Richmond Newspapers, Inc. v. Virginia, supra, at 598, n. 23, 100 S.Ct., at 2839, n. 23 (BRENNAN, J., concurring in judgment). Without such trial court discretion, a State's interest in safeguarding the welfare of the minor victim determined in an individual case to merit some form of closure, would be defeated before it could ever be brought to bear. 26 To the extent that it is suggested that, quite apart from encouraging minor victims to testify, § 16A improves the quality and credibility of testimony, the suggestion also is speculative. And while closure may have such an effect in particular cases, the Court has recognized that, as a general matter, "[o]penness in court proceedings may improve the quality of testimony." Gannett Co. v. DePasquale, supra, at 383, 99 S.Ct., at 2907 (emphasis added). In the absence of any showing that closure would improve the quality of testimony of all minor sex victims, the State's interest certainly cannot justify a mandatory closure rule. 27 We emphasize that our holding is a narrow one: that a rule of mandatory closure respecting the testimony of minor sex victims is constitutionally infirm. In individual cases, and under appropriate circumstances, the First Amendment does not necessarily stand as a bar to the exclusion from the courtroom of the press and general public during the testimony of minor sex-offense victims. But a mandatory rule, requiring no particularized determinations in individual cases, is unconstitutional. 1 It certainly cannot be said that the victims in this case consented to testifying in open court. During a lobby conference prior to trial, the prosecutor informed the trial judge that she had interviewed the victims, that they were concerned about publicity, and would agree to press attendance only if certain guarantees could be given: "Each of [the three victims] indicated that they had the same concerns and basically they are privacy concerns. "The difficulty of obtaining any kind of guarantee that the press would not print their names or where they go to school or any personal data or take pictures of them or attempt to interview them, those concerns come from their own privacy interests, as well as the fact that there are grandparents involved with a couple of these victims who do not know what happened and if they were to find out by reading the paper, everyone was concerned about what would happen then. And they stated that if it were at all possible to obtain a guarantee that this information would not be used, then they wouldn't object to the press being included. I explained that that is [a] very difficult guarantee to obtain because the Court cannot issue a conditional order, or anything like that, but I just wanted to put on the record what their concerns were and what they are afraid of." App. 48a. It is clear that the victims would "waive" the exclusion of the press only if the trial court gave them guarantees of strict privacy, guarantees that were probably beyond the authority of the court and which themselves would raise grave constitutional problems. See Oklahoma Publishing Co. v. District Court of Oklahoma County, 430 U.S. 308, 97 S.Ct. 1045, 51 L.Ed.2d 355 (1977); Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975). 2 Cf. Stamicarbon, N. V. v. American Cyanamid Co., 506 F.2d 532, 539-540 (CA2 1974), and cases cited therein. 3 See, e.g., Ala. Const., Art. VI, § 169 (1901) (repealed 1973); Fla.Stat. § 918.16 (1979); Ga.Code § 81-1006 (1978); Miss.Const., Art. 3, § 26; N.H.Rev.Stat.Ann. § 632-A:8 (Supp.1981); N.Y.Jud.Law § 4 (McKinney 1968); N.C.Gen.Stat. § 15-166 (Supp.1981); Utah Code Ann. § 78-7-4 (1953). 4 It is hard to find a limiting principle in the Court's analysis. The same reasoning might require a hearing before a trial judge could hold a bench conference or any in camera proceedings. 5 For a discussion of the traumatic effect of court proceedings on minor rape victims, see E. Hilberman, The Rape Victim 53-54 (1976); S. Katz & M. Mazur, Understanding the Rape Victim: A Synthesis of Research Findings 198-200 (1979), and studies cited therein. 6 As one commentator put it: "Especially in cases involving minors, the courts stress the serious embarrassment and shame of the victim who is forced to testify to sexual acts or whose intimate life is revealed in detail before a crowd of the idly curious." Berger, Man's Trial, Woman's Tribulation: Rape Cases in the Courtroom, 77 Colum.L.Rev. 1, 88 (1977). The victim's interest in avoiding the humiliation of testifying in open court is thus quite separate from any interest in preventing the public from learning of the crime. It is ironic that the Court emphasizes the failure of the Commonwealth to seal the trial transcript and bar disclosure of the victim's identity. The Court implies that a state law more severely encroaching upon the interests of the press and public would be upheld. 7 See Hilberman, supra ; L. Holmstrom & A. Burgess, The Victim of Rape: Institutional Reactions 222, 227 (1978); Berger, supra, at 88, 92-93; Libai, The Protection of the Child Victim of a Sexual Offense in the Criminal Justice System, 15 Wayne L.Rev. 977, 1021 (1969). Holmstrom and Burgess report that nearly half of all adult rape victims were disturbed by the public setting of their trials. Certainly the impact on children must be greater. 8 See Bohmer & Blumberg, Twice Traumatized: The Rape Victim and the Court, 58 Judicature 390 (1975); Katz & Mazur, supra ; Holmstrom & Burgess, supra ; Hilberman, supra ; Berger, supra. 9 See also 448 U.S., at 580-581, 100 S.Ct., at 2829-2830; id., at 582, 100 S.Ct., at 2830 (WHITE, J., concurring); id., at 584, 100 S.Ct., at 2831 (STEVENS, J., concurring); id., at 598, 100 S.Ct., at 2839 (BRENNAN, J., concurring in judgment). 1 For example, even though a reporter may have no right of access to a judge's side-bar conference, it surely does not follow that the judge could enjoin publication of what a reporter might have learned about such a conference. 2 "But the most fundamental principle of constitutional adjudication is not to face constitutional questions but to avoid them, if at all possible." United States v. Lovett, 328 U.S. 303, 320, 66 S.Ct. 1073, 1081, 90 L.Ed. 1252 (Frankfurter, J., concurring). 3 The "capable of repetition, yet evading review" exception to the mootness doctrine generally is compatible with our settled policy of avoiding the premature adjudication of constitutional questions, see Franks v. Bowman Transportation Co., 424 U.S. 747, 756, n. 8, 96 S.Ct. 1251, 1260, n. 8, 47 L.Ed.2d 444, for an order that is capable of repetition yet evading review generally is no less ripe for review the first time it is presented than it would be on subsequent occasions. But when the "order" that is presented for review the first time is formulated in the abstract, as was the ruling of the Supreme Judicial Court in this case, the policy requires the Court to defer review of such an order until it is entered in a live controversy.

23

457 U.S. 668 102 S.Ct. 2654 73 L.Ed.2d 300 FOREMOST INSURANCE COMPANY et al., Petitioners,v.Pansy F. RICHARDSON et al. No. 80-2134. Argued Jan. 12, 1982. Decided June 23, 1982. Rehearing Denied Oct. 4, 1982. See 459 U.S. 899, 103 S.Ct. 198. Syllabus An action to recover for the death of an occupant of a pleasure boat resulting from a collision with another pleasure boat on a river in Louisiana was instituted in Federal District Court on the asserted basis of admiralty jurisdiction under 28 U.S.C. § 1333(1). The court dismissed the complaint, holding that there must be some relationship with traditional maritime activity for an injury sustained on navigable water to fall within federal admiralty jurisdiction, and that commercial maritime activity (not present here) is necessary to satisfy this relationship. The Court of Appeals reversed. Held: In light of the need for uniform rules governing navigation, the potential impact on maritime commerce when two vessels collide on navigable waters, and the uncertainty and confusion that would necessarily accompany a jurisdictional test tied to the commercial use of a given boat, a complaint alleging a collision between two vessels—including pleasure boats—on navigable waters properly states a claim within the admiralty jurisdiction of the federal courts. The holding in Executive Jet Aviation, Inc. v. City of Cleveland, 409 U.S. 249, 93 S.Ct. 493, 34 L.Ed.2d 454, that claims arising from airplane accidents, although occurring in a maritime locality, are cognizable in admiralty only when the wrong bears a significant relationship to traditional maritime activity also applies to determinations of federal admiralty jurisdiction outside the context of aviation torts. However, there is no requirement that the maritime activity be an exclusively commercial one. The federal interest in protecting maritime commerce can be fully vindicated only if all operators of vessels on navigable waters—not just individuals actually engaged in commercial maritime activity—are subject to uniform rules of conduct. This interpretation is consistent with congressional activity as to legislation governing "vessels" without regard to whether they engage in commercial activity. Pp. 672-677. 641 F.2d 314, affirmed. Arthur H. Andrews, Baton Rouge, La., for petitioners. Dorsey C. Martin, III, Baton Rouge, for respondents. Justice MARSHALL delivered the opinion of the Court. 1 The issue presented in this case is whether the collision of two pleasure boats on navigable waters falls within the admiralty jurisdiction of the federal courts. See 28 U.S.C. § 1333. We granted certiorari to resolve the confusion in the lower courts respecting the impact of Executive Jet Aviation, Inc. v. City of Cleveland, 409 U.S. 249, 93 S.Ct. 493, 34 L.Ed.2d 454 (1972), on traditional rules for determining federal admiralty jurisdiction. 454 U.S. 813, 102 S.Ct. 88, 70 L.Ed.2d 81 (1981). The United States Court of Appeals for the Fifth Circuit held that an accident between two vessels in navigable waters bears a sufficient relationship to traditional maritime activity to fall within federal admiralty jurisdiction. We affirm. 2 * Two pleasure boats collided on the Amite River in Louisiana, resulting in the death of Clyde Richardson. The wife and children of the decedent brought this action in the United States District Court for the Middle District of Louisiana, alleging, inter alia, that petitioner Shirley Eliser had negligently operated the boat that collided with the vessel occupied by the decedent.1 Respondents also named petitioner Foremost Insurance Co., Eliser's insurer, as a defendant. Jurisdiction was claimed under 28 U.S.C. § 1333(1), which gives federal district courts exclusive jurisdiction over "[a]ny civil case of admiralty or maritime jurisdiction." Petitioners moved to dismiss, arguing that the complaint did not state a cause of action within the admiralty or maritime jurisdiction of the District Court. 3 In ruling on petitioners' motion, the District Court found the following facts to be undisputed:2 4 "(1) One boat was used for pleasure boating, such as boat riding and water skiing, and at the time of the accident the boat was actually pulling a skier on a zip sled; 5 "(2) The other boat was used exclusively for pleasure fishing and was described as a bass boat; 6 "(3) Neither boat had ever been used in any 'commercial maritime activity' before the accident; 7 "(4) At the time of the accident neither boat was involved in any 'commercial maritime activity' of any sort; 8 "(5) Neither of the two drivers of the boat were being paid to operate the boat nor was this activity in any way a part of their regular type of employment; 9 "(6) None of the passengers on either boat were engaged in any kind of 'traditional maritime activity' either before or at the time of the accident; 10 "(7) Neither of the boats involved were under hire in any traditional maritime form; 11 "(8) There is no evidence to indicate that any 'commercial activity', even in the broadest admiralty sense, had ever been previously engaged in by either of the boats in question, and in fact the two boats would have to be classified as 'purely pleasure craft', not in any way 'involved in commerce', and, 12 "(9) There was no other instrumentality involved in this accident that had even a minor relationship to 'admiralty' or 'commerce', i.e. a buoy, barge, oil drilling apparatus, etc." 470 F.Supp. 699, 700 (1979). 13 After reviewing decisions of this Court and the Fifth Circuit, as well as relevant commentary, the District Court found that there must be some relationship with traditional maritime activity for an injury sustained on navigable water to fall within federal admiralty jurisdiction. The District Court held that commercial maritime activity is necessary to satisfy this relationship, and granted petitioners' motion to dismiss the complaint for lack of subject-matter jurisdiction because the collision of these two pleasure boats did not involve any commercial activity. 14 The Court of Appeals reversed. 641 F.2d 314 (1981). The Court of Appeals agreed that Executive Jet, supra, and relevant Fifth Circuit decisions establish that "admiralty jurisdiction requires more than the occurrence of the tort on navigable waters that additionally there must be a significant relationship between the wrong and traditional maritime activity." 641 F.2d, at 315. It disagreed with the District Court, however, on the application of this principle to the undisputed facts of this case. Relying on the fact that the "Rules of the Road" govern all boats on navigable waters, and on the uncertainty that would accompany a finding of no admiralty jurisdiction in this case, the Court of Appeals held that "two boats, regardless of their intended use, purpose, size, and activity, are engaged in traditional maritime activity when a collision between them occurs on navigable waters." Id., at 316.3 II 15 Prior to our opinion in Executive Jet, there was little question that a complaint such as the one filed here stated a cause of action within federal admiralty jurisdiction. Indeed, the Executive Jet Court begins its opinion by observing that, under the traditional rule of admiralty jurisdiction, "[i]f the wrong occurred on navigable waters, the action is within admiralty jurisdiction." 409 U.S., at 253, 93 S.Ct., at 497 (citing Thomas v. Lane, 23 F.Cas. 957, 960 (No. 13,902) (CC Me.1813) (Story, J., on Circuit). See also The Plymouth, 3 Wall. 20, 36, 18 L.Ed. 125 (1866) ("Every species of tort, however occurring, and whether on board a vessel or not, if upon the high seas or navigable waters, is of admiralty cognizance"). Under this rule, an action arising out of a collision between two pleasure boats on navigable waters clearly falls within the admiralty jurisdiction of the district courts. When presented with this precise situation in the past, this Court has found it unnecessary even to discuss whether the district court's admiralty jurisdiction had been properly invoked, instead assuming the propriety of such jurisdiction merely because the accident occurred on navigable waters. Levinson v. Deupree, 345 U.S. 648, 651, 73 S.Ct. 914, 916, 97 L.Ed. 1319 (1953). See also Just v. Chambers, 312 U.S. 383, 61 S.Ct. 687, 85 L.Ed. 903 (1941) (injury to guest from carbon monoxide poisoning in the cabin of a pleasure boat). Cf. Coryell v. Phipps, 317 U.S. 406, 63 S.Ct. 291, 87 L.Ed. 363 (1943). In light of these decisions, we address here only the narrow question whether Executive Jet disapproved these earlier decisions sub silentio. In Executive Jet, this Court held that a suit for property damage to a jet aircraft that struck a flock of sea gulls upon takeoff and sank in the navigable waters of Lake Erie did not state a claim within the admiralty jurisdiction of the district courts. In reaching this conclusion, the Court observed that the mechanical application of the locality rule as the sole test for determining whether there is admiralty jurisdiction had been widely criticized by commentators, and that the federal courts and Congress had been compelled to make exceptions to this approach in the interests of justice in order to include certain torts with no maritime locality. The Court determined that claims arising from airplane accidents are cognizable in admiralty only when the wrong bears a significant relationship to traditional maritime activity. 409 U.S., at 268, 93 S.Ct., at 504. Given the realities of modern-day air travel, the Executive Jet Court held that, "in the absence of legislation to the contrary, there is no federal admiralty jurisdiction over aviation tort claims arising from flights by land-based aircraft between points within the continental United States." Id., at 274, 93 S.Ct., at 507. 16 The express holding of Executive Jet is carefully limited to the particular facts of that case. However, the thorough discussion of the theoretical and practical problems inherent in broadly applying the traditional locality rule has prompted several courts and commentators to construe Executive Jet as applying to determinations of federal admiralty jurisdiction outside the context of aviation torts. See, e.g., Kelly v. Smith, 485 F.2d 520 (CA5 1973); Calamari, The Wake of Executive Jet—A Major Wave or a Minor Ripple, 4 Maritime Law. 52 (1979). We believe that this is a fair construction. Although Executive Jet addressed only the unique problems associated with extending admiralty jurisdiction to aviation torts, much of the Court's rationale in rejecting a strict locality rule also applies to the maritime context. Indeed, the Executive Jet Court relied extensively on admiralty and maritime decisions of this Court and on congressional action extending admiralty jurisdiction to torts with a significant relationship to traditional maritime activity, but with no maritime locality.4 17 We recognize, as did the Court of Appeals, that the Executive Jet requirement that the wrong have a significant connection with traditional maritime activity is not limited to the aviation context. We also agree that there is no requirement that "the maritime activity be an exclusively commercial one." 641 F.2d, at 316. Because the "wrong" here involves the negligent operation of a vessel on navigable waters, we believe that it has a sufficient nexus to traditional maritime activity to sustain admiralty jurisdiction in the District Court. 18 We are not persuaded by petitioners' argument that a substantial relationship with commercial maritime activity is necessary because commercial shipping is at the heart of the traditional maritime activity sought to be protected by giving the federal courts exclusive jurisdiction over all admiralty suits. This argument is premised on the faulty assumption that, absent this relationship with commercial activity, the need for uniform rules to govern conduct and liability disappears, and "federalism" concerns dictate that these torts be litigated in the state courts. 19 Although the primary focus of admiralty jurisdiction is unquestionably the protection of maritime commerce, petitioners take too narrow a view of the federal interest sought to be protected. The federal interest in protecting maritime commerce cannot be adequately served if admiralty jurisdiction is restricted to those individuals actually engaged in commercial maritime activity. This interest can be fully vindicated only if all operators of vessels on navigable waters are subject to uniform rules of conduct. The failure to recognize the breadth of this federal interest ignores the potential effect of noncommercial maritime activity on maritime commerce. For example, if these two boats collided at the mouth of the St. Lawrence Seaway, there would be a substantial effect on maritime commerce, without regard to whether either boat was actively, or had been previously, engaged in commercial activity. Furthermore, admiralty law has traditionally been concerned with the conduct alleged to have caused this collision by virtue of its "navigational rules—rules that govern the manner and direction those vessels may rightly move upon the waters." Executive Jet, 409 U.S., at 270, 93 S.Ct., at 505. The potential disruptive impact of a collision between boats on navigable waters, when coupled with the traditional concern that admiralty law holds for navigation,5 compels the conclusion that this collision between two pleasure boats on navigable waters has a significant relationship with maritime commerce. 20 Yet, under the strict commercial rule proffered by petitioners, the status of the boats as "pleasure" boats, as opposed to "commercial" boats, would control the existence of admiralty jurisdiction. Application of this rule, however, leads to inconsistent findings or denials of admiralty jurisdiction similar to those found fatal to the locality rule in Executive Jet. Under the commercial rule, fortuitous circumstances such as whether the boat was, or had ever been, rented, or whether it had ever been used for commercial fishing, control the existence of federal-court jurisdiction. The owner of a vessel used for both business and pleasure might be subject to radically different rules of liability depending upon whether his activity at the time of a collision is found by the court ultimately assuming jurisdiction over the controversy to have been sufficiently "commercial." We decline to inject the uncertainty inherent in such line-drawing into maritime transportation. Moreover, the smooth flow of maritime commerce is promoted when all vessel operators are subject to the same duties and liabilities. Adopting the strict commercial rule would frustrate the goal of promoting the smooth flow of maritime commerce, because the duties and obligations of noncommercial navigators traversing navigable waters flowing through more than one State would differ "depending upon their precise location within the territorial jurisdiction of one state or another." 641 F.2d, at 316. 21 Finally, our interpretation is consistent with congressional activity in this area. First, Congress defines the term "vessel," for the purpose of determining the scope of various shipping and maritime transportation laws, to include all types of waterborne vessels, without regard to whether they engage in commercial activity. See, e.g., 1 U.S.C. § 3 (" 'vessel' includes every description of watercraft or other artificial contrivance used, or capable of being used, as a means of transportation on water"). Second, the federal "Rules of the Road," designed for preventing collisions on navigable waters, see, e.g., 94 Stat. 3415, 33 U.S.C. § 2001 et seq. (1976 ed., Supp.IV), apply to all vessels without regard to their commercial or noncommercial nature.6 Third, when it extended admiralty jurisdiction to injuries on land caused by ships on navigable waters, Congress directed that "[t]he admiralty and maritime jurisdiction of the United States shall extend to and include all cases of damage or injury . . . caused by a vessel on navigable water. . . ." Extension of Admiralty Jurisdiction Act, 62 Stat. 496, 46 U.S.C. § 740.7 22 In light of the need for uniform rules governing navigation, the potential impact on maritime commerce when two vessels collide on navigable waters, and the uncertainty and confusion that would necessarily accompany a jurisdictional test tied to the commercial use of a given boat, we hold that a complaint alleging a collision between two vessels on navigable waters properly states a claim within the admiralty jurisdiction of the federal courts. Therefore, the judgment of the Court of Appeals is 23 Affirmed. 24 Justice POWELL, with whom THE CHIEF JUSTICE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 25 No trend of decisions by this Court has been stronger—for two decades or more—than that toward expanding federal jurisdiction at the expense of state interests and state-court jurisdiction. Of course, Congress also has moved steadily and expansively to exercise its Commerce Clause and preemptive power to displace state and local authority. Often decisions of this Court and congressional enactments have been necessary in the national interest. The effect, nevertheless, has been the erosion of federalism—a basic principle of the Constitution and our federal Union. 26 Today's Court decision, an example of this trend, is not necessary to further any federal interest. On its face, it is inexplicable. The issue is whether the federal law of admiralty, rather than traditional state tort law, should apply to an accident on the Amite River in Louisiana between two small boats. "One was an eighteen foot pleasure boat powered by a 185 h.p. Johnson outboard motor that was being used for water skiing purposes at the time of the accident. The other was a sixteen-foot 'bass boat' powered by an outboard motor that was used exclusively for pleasure fishing." 470 F.Supp. 699, 700 (MD La.1979). It also is undisputed that both boats were used "exclusively for pleasure"; that neither had ever been used in any "commercial maritime activity"; that none of the persons aboard the boats had ever been engaged in any such activity; and that neither of the boats was used for hire. Ibid. The Court of Appeals conceded that "the place where the accident occurred is seldom, if ever, used for commercial activity." 641 F.2d 314, 316 (CA5 1981). 27 The absence of "commercial activity" on this waterway was held by the Court of Appeals to be immaterial. While recognizing that there was substantial authority to the contrary, the court held that federal admiralty law applied to this accident. This Court now affirms in a decision holding that "all operators of vessels on navigable waters are subject to uniform [federal] rules of conduct," conferring federal admiralty jurisdiction over all accidents. Ante, at 675 (emphasis deleted). In my view there is no substantial federal interest that justifies a rule extending admiralty jurisdiction to the edge of absurdity. I dissent. 28 * Executive Jet Aviation, Inc. v. City of Cleveland, 409 U.S. 249, 93 S.Ct. 493, 34 L.Ed.2d 454 (1972), established that admiralty jurisdiction does not extend to every accident on navigable waters. The Court today misconstrues Executive Jet. We emphasized in that case that it is "consistent with the history and purpose of admiralty to require . . . that the wrong bear a significant relationship to traditional maritime activity." Id., at 268, 93 S.Ct., at 504 (emphasis added). We acknowledged that "in a literal sense there may be some similarities between the problems posed for a plane downed on water and those faced by a sinking ship." Id., at 269, 93 S.Ct., at 505. But, recalling that "[t]he law of admiralty has evolved over many centuries," ibid., we noted that admiralty was "concerned with [matters such as] maritime liens, the general average,1 captures and prizes, limitation of liability, cargo damage, and claims for salvage." Id., at 270, 93 S.Ct., at 505. "It is clear, therefore, that neither the fact that a plane goes down on navigable waters nor the fact that the negligence 'occurs' while a plane is flying over such waters is enough to create such a relationship to traditional maritime activity as to justify the invocation of admiralty jurisdiction." Id., at 270-271, 93 S.Ct., at 505 (emphasis added). 29 Executive Jet's recognition that "[t]he law of admiralty has evolved over many centuries," id., at 269, 93 S.Ct., at 505, provides the appropriate understanding of that case's "traditional maritime activity" test. Admiralty is a specialized area of law that, since its ancient inception, has been concerned with the problems of seafaring commercial activity.2 As Professor Stolz has demonstrated, "[t]here can be no doubt that historically the civil jurisdiction of admiralty was exclusively concerned with matters arising from maritime commerce." Stolz, Pleasure Boating and Admiralty: Erie at Sea, 51 Calif.L.Rev. 661, 667 (1963). "The only valid criterion of the admiralty jurisdiction is the relation of the matter—whether it be tortious or contractual in nature—to maritime commerce." 7A J. Moore & A. Pelaez, Federal Practice, Admiralty ¶ .325[5], p. 3606 (2d ed. 1982) (emphasis in original).3 30 This case involves only pleasure craft. Neither of these boats had ever been used in any commercial activity. There is, therefore, no connection with any historic federal admiralty interest. In centuries past—long before modern means of transportation by land and air existed—rivers and oceans were the basic means of commerce, and the vessels that used the waterways were limited primarily to commercial and naval purposes.4 "Pleasure boating is basically a new phenomenon, the product of a technology that can produce small boats at modest cost and of an economy that puts such craft within the means of almost everyone."5 Stolz, supra, at 661. Thus, the "traditional" connection emphasized in Executive Jet is absent where pleasure boats are concerned. Moreover, even the Court today is hard put to identify an arguably substantial federal admiralty interest of any kind. I now comment briefly on the Court's reasoning. II 31 The Court's justification for extending federal admiralty jurisdiction to the use of millions6 of small pleasure boats on the countless rivers, streams, and inlets of our country is the need for "uniform rules of conduct." Ante, at 675. I agree, of course, that standard codes should govern traffic on waterways, just as it is crucial that certain uniform rules of traffic prevail on neighborhood streets as well as interstate highways. But this is no reason for admiralty jurisdiction to be extended to all boating activity. Congress has provided some rules governing water traffic, just as it has done for some land traffic. See 23 U.S.C. § 154 (55 m.p.h. speed limit). Yet no one suggests that federal jurisdiction is needed to prevent chaos in automobile traffic, or that only federal courts are qualified to try accident cases. 32 State courts are duty bound to apply federal as well as local "uniform rules of conduct." See Testa v. Katt, 330 U.S. 386, 67 S.Ct. 810, 91 L.Ed. 967 (1947). The Court does not suggest that state courts lack competency to apply federal as well as state law to this type of water traffic. And this Court stands ready, if necessary, to review state decisions to ensure that important issues of federal law are resolved correctly. As Judge Thornberry said in dissent in this case, "the desire for certainty cannot alone justify the assumption of federal control over matters of purely local concern . . . ." 641 F.2d, at 317. Consequently the Court's premise that there is a need for uniform traffic rules fails to support its conclusion that federal jurisdiction must be extended to cover the type of activity that typically involves small pleasure craft. 33 In an effort to rescue its logic, the Court refers to the "potential disruptive impact of a collision between boats on navigable waters . . . ." Ante, at 675. Yet this reasoning is countered by Executive Jet —a decision that the Court acknowledges to be a key authority for this case. For if "potential disruptive impact" on traffic in navigable waters provides a sufficient connection with "traditional maritime activity," then the crash of an airplane "in the navigable waters of Lake Erie," 409 U.S., at 250, 93 S.Ct., at 495, necessarily would support admiralty jurisdiction. The holding of Executive Jet is precisely to the contrary. The Court's reasoning in essence resurrects the locality rule that Executive Jet rejected, for any accident "located" on navigable waters has a "potential disruptive impact" on traffic there.7 34 Oral argument in this case revealed the degree to which the Court's decision displaces state authority. The Court posed a hypothetical in which children, for their own amusement, used rowboats to net crawfish from a stream. Two of the boats collide and sink near the water's edge, forcing the children to wade ashore. Counsel for respondents replied that this accident would fall within the admiralty jurisdiction of the federal courts, provided that the waterway was navigable. Tr. of Oral Arg. 24. Today the Court agrees. 35 For me, however, this example illustrates the substantial—and purposeless —expansion of federal authority and federal-court jurisdiction accomplished by the Court's holding. In this respect I agree with Chief Judge Haynsworth: 36 "The admiralty jurisdiction in England and in this country was born of a felt need to protect the domestic shipping industry in its competition with foreign shipping, and to provide a uniform body of law for the governance of domestic and foreign shipping, engaged in the movement of commercial vessels from state to state and to and from foreign states. The operation of small pleasure craft on inland waters which happen to be navigable has no more apparent relationship to that kind of concern than the operation of the same kind of craft on artificial inland lakes which are not navigable waters." Crosson v. Vance, 484 F.2d 840 (CA4 1973). 37 In the rowboat example, as in the case at bar, the Federal Government has little or no genuine interest in the resolution of a garden variety tort case. "Only the burdening of the federal courts and the frustration of the purposes of state tort law would be thereby served." Adams v. Montana Power Co., 528 F.2d 437, 440-441 (CA9 1975).8 38 The Court's opinion largely ignores the fact that expansions of federal admiralty jurisdiction are accompanied by application of substantive—and pre-empting —federal admiralty law. Southern Pacific Co. v. Jensen, 244 U.S. 205, 214-218, 37 S.Ct. 524, 528-529, 61 L.Ed. 1086 (1917); see Kossick v. United Fruit Co., 365 U.S. 731, 738-742, 81 S.Ct. 886, 891-894, 6 L.Ed.2d 56 (1961).9 "The chief objection to application of admiralty law to pleasure boating is that it implicitly prohibits the exercise of state legislative power in an area in which local legislatures have generally been thought competent and in which Congress cannot be expected either to be interested or to be responsive to local needs." Stolz, 51 Calif.L.Rev., at 664. For me, this federalism concern is the dominating issue in the case. I agree that "the law of pleasure boating will develop faster and more rationally if the creative capacities of the state courts and legislatures are freed of an imaginery [sic] federal concern with anything that floats on navigable waters." Id., at 719. 39 Federal courts should not displace state responsibility and choke the federal judicial docket on the basis of federal concerns that in truth are only "imaginary." In accord with the teaching of Executive Jet, I would not extend federal admiralty jurisdiction beyond its traditional roots and reason for existence. I dissent from the Court's decision to sever a historic doctrine from its historic justification. 1 The wife and children of the decedent also named respondent June Allen as a defendant. They alleged that Allen was operating the vessel at the time of the collision, and that the decedent's death was caused by either the negligence of Allen or that of petitioner Eliser. Allen counterclaimed, alleging that the decedent had been operating the boat, and that her injuries were caused by his negligence. The factual dispute concerning whether the decedent or Allen was operating the boat is irrelevant to the jurisdictional issue. However, because of the divergent interests and claims of respondent Allen and the respondent family of the decedent below, we refer only to the decedent's family when we use the term "respondents" throughout this opinion. 2 The District Court assumed that the Amite River is navigable at the site of the collision. Although the issue is not free from doubt, it appears from the opinion and the disposition of the Court of Appeals that the court found that the river is navigable at this site although seldom, if ever, used for commercial traffic. This opinion is premised on our understanding that the river at this point is navigable, see Brief for Petitioners 20, but we leave open the question whether petitioners have preserved the opportunity to argue this issue upon further development of facts in the District Court. 3 Judge Thornberry, concurring in part and dissenting in part, argued that federal admiralty jurisdiction could not be sustained if the river at the site of the accident, although navigable, did not also function as an integral or major "artery of commerce." 641 F.2d, at 317 (1981). 4 In addition to noting these examples where strict application of the locality rule would have deprived the courts of admiralty jurisdiction despite a clear connection to maritime activity, the Court noted the difficulties of extending jurisdiction to torts with a maritime locality, but absolutely no connection to maritime activity. See 409 U.S., at 255-256, 93 S.Ct., at 498 (disapproving decisions sustaining admiralty jurisdiction over claims by swimmers injured by other swimmers or submerged objects in shallow waters near shore); id., at 256-257, 93 S.Ct., at 498-499 (approving decisions requiring some connection with traditional maritime activity). 5 Not every accident in navigable waters that might disrupt maritime commerce will support federal admiralty jurisdiction. In Executive Jet, for example, we concluded that the sinking of the plane in navigable waters did not give rise to a claim in admiralty even though an aircraft sinking in the water could create a hazard for the navigation of commercial vessels in the vicinity. However, when this kind of potential hazard to maritime commerce arises out of activity that bears a substantial relationship to traditional maritime activity, as does the navigation of the boats in this case, admiralty jurisdiction is appropriate. 6 Petitioners argue that admiralty jurisdiction in the federal courts is unnecessary to ensure the uniform application of the Rules of the Road to boat navigation because state courts are bound by the construction federal courts give to statutes relating to navigation. Assuming that petitioners are correct, this fact does not negate the importance that Congress has attached to the federal interest in having all vessels operating on navigable waters governed by uniform rules and obligations, which is furthered by consistent application of federal maritime legislation under federal admiralty jurisdiction. 7 We refer to this language only to demonstrate that Congress did not require a commercial-activity nexus when it extended admiralty jurisdiction. We express no opinion on whether this Act could be construed to provide an independent basis for jurisdiction. 1 The doctrine of general average refers to rules for dividing the loss suffered when cargo must be thrown overboard in order to lighten a ship. See generally G. Gilmore & C. Black, The Law of Admiralty 244-271 (2d ed. 1975). 2 "Maritime courts, differing somewhat in name and somewhat in jurisdiction, have been established in all civilized nations at various periods in their history. The dates of their establishment may be said, because of the circumstances which brought them into being, to afford a very fair test of the advancement in civilization of their respective nations. "In every case their establishment has been due to the same cause, the necessities of commerce." T. Etting, The Admiralty Jurisdiction in America 7-8 (1879) (emphasis added). 3 See also Black, Admiralty Jurisdiction: Critique and Suggestions, 50 Colum.L.Rev. 259, 280 (1950) ("The main thing is that if the court of admiralty is to exist at all, it should exist because the business of river, lake, and ocean shipping calls for supervision by a tribunal enjoying a particular expertness in regard to the more complicated concerns of that business") (emphasis added); Swaim, Yes, Virginia, There is an Admiralty: The Rodrigue Case, 16 Loyola L.Rev. 43, 44 (1970) ("Maritime commerce—and nothing more—is the raison d'etre for the courts and rules of admiralty"); Bridwell & Whitten, Admiralty Jurisdiction: The Outlook for the Doctrine of Executive Jet, 1974 Duke L.J. 757, 793; Comment, 12 Cal. Western L.Rev. 535, 558, n. 133 (1976) ("The historical justification for admiralty law and courts is commercial. Its law was designed to meet commercial needs and practice"); Note, 34 Wash. & Lee L.Rev. 121, 139-140 (1977) ("Those pleasure craft torts occurring on commercially navigable waters must be considered in light of the historical design of admiralty jurisdiction to determine whether the exercise of jurisdiction furthers the commercial interests which admiralty courts were created to serve"). 4 At the beginning of the 19th century, "the commerce of the country was almost entirely limited to the foreign and coasting trade. The only roads which existed led from the woods to the principal towns on navigable waters. There was but one connected route from North to South at the commencement of the Revolution, and this was true also when the Constitution was framed. Even in 1796 the only roads with which the States were much concerned were those which led to navigable waters; the care of 'cross roads,' as the roads leading from State to State were called by one who had been a member of the Constitutional Convention, the States were unwilling to assume. 'Fifty miles back from the waters of the Atlantic the country was an unbroken jungle.' In the vigorous phrase used by Henry Clay, 'the country had scarcely any interior.' Turnpike roads did not come into general use until the nineteenth century." E. Prentice, The Federal Power over Carriers and Corporations 59-60 (1907) (footnotes omitted). 5 For this reason, the jurisdictional issue in this case is relatively new and, until today, has not been addressed by this Court. The Court's contrary suggestion, ante, at 672, relies on irrelevant dicta from decisions of the last century that do not involve pleasure craft. E.g., The Plymouth, 3 Wall. 20, 36, 18 L.Ed. 125 (1866) (holding admiralty jurisdiction does not include adjudication of a loss of packing-houses on a wharf that arose from fire on an adjacent merchant ship at anchor). The Court also cites cases apparently involving pleasure boats in which the jurisdictional question was not at issue. See Levinson v. Deupree, 345 U.S. 648, 651, 73 S.Ct. 914, 916, 97 L.Ed. 1319 (1953); Coryell v. Phipps, 317 U.S. 406, 63 S.Ct. 291, 87 L.Ed. 363 (1943); Just v. Chambers, 312 U.S. 383, 61 S.Ct. 687, 85 L.Ed. 903 (1941). "[W]hen questions of jurisdiction have been passed on in prior decisions sub silentio, this Court has never considered itself bound when a subsequent case finally brings the jurisdictional issue before us." Hagans v. Lavine, 415 U.S. 528, 535, n. 5, 94 S.Ct. 1372, 1378, n. 5, 39 L.Ed.2d 577 (1974). The jurisdictional issue has both a constitutional and a statutory element, since both Art. III and 28 U.S.C. § 1333 must support the exercise of jurisdiction in this case. The Court necessarily must find that both provisions are satisfied. Because construction of the statute is sufficient to support the result I would reach, I intimate no views on the constitutional extent of Art. III admiralty jurisdiction. 6 There were 14.3 million pleasure boats in the United States in 1980. See U.S. Dept. of Transportation, U.S. Coast Guard, Boating Statistics 1980, p. 8 (1981). 7 If a "potential disruptive effect" on interstate traffic in fact implicated a federal interest strong enough to support federal jurisdiction, then federal courts also should hear cases in which accidents disrupt similar land traffic. Cf. "71 Feared Dead as Plane Hits Bridge, Smashes Cars, Plunges Into Potomac," Washington Post, Jan. 14, 1982, p. A1, col. 1. According to the Court, the interest in expanding admiralty jurisdiction is supported by the difficulty of defining "pleasure boating." Ante, at 675-676. In view of the myriad of definitional tasks performed regularly by state and federal courts, determining in a particular case whether the boating at issue is essentially for pleasure rather than commerce rarely would present a difficult problem for any court. The Court also states that its action "is consistent with congressional activity in this area," ante, at 676, citing a number of federal statutes. This point is of course wholly irrelevant to the constitutional extent of admiralty jurisdiction. Moreover, the only statute cited having any relation to jurisdictional matters is the Extension of Admiralty Jurisdiction Act, 62 Stat. 496, 46 U.S.C. § 740. This Act provides: "The admiralty and maritime jurisdiction of the United States shall extend to and include all cases of damage or injury, to person or property, caused by a vessel on navigable water, notwithstanding that such damage or injury be done or consummated on land. "In any such case suit may be brought in rem or in personam according to the principles of law and rules of practice obtaining in cases where the injury or damage has been done and consummated on navigable water" (emphasis added). As its text makes plain, "[t]his Act was passed specifically to overrule cases, such as The Plymouth, supra, holding that admiralty does not provide a remedy for damage done to land structures by ships on navigable waters." Executive Jet, 409 U.S., at 260, 93 S.Ct., at 500. This purpose—and not any intent to expand or affect admiralty jurisdiction respecting pleasure boats—consistently appears in the Act's legislative history. See, e.g., S.Rep.No.1593, 80th Cong., 2d Sess., 1-6 (1948); H.R.Rep.No.1523, 80th Cong., 2d Sess., 1-6 (1948), U.S.Code Cong. & Admin.News 1948, p. 1898. See also Farnum, Admiralty Jurisdiction and Amphibious Torts, 43 Yale L.J. 34, 44-45 (1933); Note, 63 Harv.L.Rev. 861, 868 (1950); Note, The Extension of Admiralty Jurisdiction to Include Amphibious Torts, 37 Geo.L.J. 252 (1949); Note, Effects of Recent Legislation Upon the Admiralty Law, 17 Geo.Wash.L.Rev. 353 (1949). And this Court has never sustained the constitutionality of this Act. With respect, the Court's statutory arguments must be regarded as makeweights. 8 In construing the extent of 28 U.S.C. § 1333 admiralty jurisdiction, see n. 5, supra, I would prefer to leave to Congress an extension of federal authority of this magnitude. See n. 6, supra. Congress has the power to hold hearings and to weigh factors beyond the proper competency of a court. 9 "It should be emphasized . . . that, in the law of admiralty, the term 'jurisdiction' denotes both the power of a court to hear and dispose of a certain controversy, and also the power to prescribe rules of decision to be applied by those courts considering the controversy. This is so because a court of admiralty sits solely to administer and apply the maritime law." Swaim, supra n. 3, at 43 (footnotes and emphasis omitted).

78

457 U.S. 624 102 S.Ct. 2629 73 L.Ed.2d 269 James EDGAR, Appellantv.MITE CORPORATION and Mite Holdings, Inc. No. 80-1188. Argued Nov. 30, 1981. Decided June 23, 1982. Syllabus The Illinois Business Take-Over Act requires a tender offeror to notify the Secretary of State and the target company of its intent to make a tender offer and the terms of the offer 20 days before the offer becomes effective. During that time the offeror may not communicate its offer to the shareholders, but the target company is free to disseminate information to its shareholders concerning the impending offer. The Act also requires any takeover offer to be registered with the Secretary of State. A target company is defined as a corporation of which Illinois shareholders own 10% of the class of securities subject to the takeover offer or for which any two of the following conditions are met: the corporation has its principal office in Illinois, is organized under Illinois laws, or has at least 10% of its stated capital and paid-in surplus represented within the State. An offer becomes registered 20 days after a registration is filed with the Secretary of State unless he calls a hearing to adjudicate the fairness of the offer. Appellee MITE Corp., a corporation organized under Delaware laws with its principal office in Connecticut, initiated a tender offer for all outstanding shares of Chicago Rivet & Machine Co., an Illinois corporation, by filing with the Securities and Exchange Commission the schedule required by the Williams Act. MITE, however, did not comply with the Illinois Act, and brought an action in Federal District Court seeking a declaratory judgment that the Illinois Act was pre-empted by the Williams Act and violated the Commerce Clause, and also seeking injunctive relief. The District Court issued a preliminary injunction prohibiting enforcement of the Illinois Act against MITE's tender offer. MITE then published its offer. Subsequently, the District Court issued the requested declaratory judgment and a permanent injunction. Shortly, thereafter, MITE and Chicago Rivet entered into an agreement whereby both MITE's tender offer and an offer made by Chicago Rivet before the District Court entered its judgment were withdrawn and MITE was given a specified time to make another offer. Ultimately, MITE decided not to make another offer. The Court of Appeals affirmed the District Court. Held : The judgment is affirmed. 633 F.2d 486, affirmed. Justice WHITE delivered the opinion of the Court with respect to Parts I, II, and V-B, concluding that: 1 1. The case is not moot. Because the Secretary of State has indicated his intention to enforce the Illinois Act against MITE, a reversal of the District Court's judgment would expose MITE to civil and criminal liability for making an offer in violation of the Act. P. 630. 2 2. The Illinois Act is unconstitutional under the Commerce Clause, because it imposes burdens on interstate commerce that are excessive in light of the local interests the Act purports to further. Pike v. Bruce Church, Inc., 397 U.S. 137, 90 S.Ct. 844, 25 L.Ed.2d 174. Illinois' asserted interests in protecting resident security holders and regulating the internal affairs of companies incorporated under Illinois law are insufficient to outweigh such burdens. Pp. 643-646. 3 Russell C. Grimes, Jr., Chicago, Ill., for appellant. 4 Eugene D. Berman, New York City, for State of New York as amicus curiae, by special leave of Court. 5 Richard W. Hulbert, New York City, for appellees. 6 Stephen N. Shapiro, New York City, for S.E.C. as amicus curiae, by special leave of Court. 7 Justice WHITE delivered an opinion, Parts I, II, and V-B of which are the opinion of the Court.* 8 The issue in this case is whether the Illinois Business Take-Over Act, Ill.Rev.Stat., ch. 1211/2, ¶ 137.51 et seq. (1979), is unconstitutional under the Supremacy and Commerce Clauses of the Federal Constitution. 9 * Appellee MITE Corp. and its wholly owned subsidiary, MITE Holdings, Inc., are corporations organized under the laws of Delaware with their principal executive offices in Connecticut. Appellant James Edgar is the Secretary of State of Illinois and is charged with the administration and enforcement of the Illinois Act. Under the Illinois Act any takeover offer1 for the shares of a target company must be registered with the Secretary of State. Ill.Rev.Stat., ch. 1211/2, ¶ 137.54.A (1979). A target company is defined as a corporation or other issuer of securities of which shareholders located in Illinois own 10% of the class of equity securities subject to the offer, or for which any two of the following three conditions are met: the corporation has its principal executive office in Illinois, is organized under the laws of Illinois, or has at least 10% of its stated capital and paid-in surplus represented within the State. ¶ 137.52-10. An offer becomes registered 20 days after a registration statement is filed with the Secretary unless the Secretary calls a hearing. ¶ 137.54.E. The Secretary may call a hearing at any time during the 20-day waiting period to adjudicate the substantive fairness of the offer if he believes it is necessary to protect the shareholders of the target company, and a hearing must be held if requested by a majority of a target company's outside directors or by Illinois shareholders who own 10% of the class of securities subject to the offer. ¶ 137.57.A. If the Secretary does hold a hearing, he is directed by the statute to deny registration to a tender offer if he finds that it "fails to provide full and fair disclosure to the offerees of all material information concerning the take-over offer, or that the take-over offer is inequitable or would work or tend to work a fraud or deceit upon the offerees. . . ." ¶ 137.57.E. 10 On January 19, 1979, MITE initiated a cash tender offer for all outstanding shares of Chicago Rivet & Machine Co., a publicly held Illinois corporation, by filing a Schedule 14D-1 with the Securities and Exchange Commission in order to comply with the Williams Act.2 The Schedule 14D-1 indicated that MITE was willing to pay $28 per share for any and all outstanding shares of Chicago Rivet, a premium of approximately $4 over the then-prevailing market price. MITE did not comply with the Illinois Act, however, and commenced this litigation on the same day by filing an action in the United States District Court for the Northern District of Illinois. The complaint asked for a declaratory judgment that the Illinois Act was pre-empted by the Williams Act and violated the Commerce Clause. In addition, MITE sought a temporary restraining order and preliminary and permanent injunctions prohibiting the Illinois Secretary of State from enforcing the Illinois Act. 11 Chicago Rivet responded three days later by bringing suit in Pennsylvania, where it conducted most of its business, seeking to enjoin MITE from proceeding with its proposed tender offer on the ground that the offer violated the Pennsylvania Takeover Disclosure Law, Pa.Stat.Ann., Tit. 70, § 71 et seq. (Purdon Supp.1982-1983). After Chicago Rivet's efforts to obtain relief in Pennsylvania proved unsuccessful,3 both Chicago Rivet and the Illinois Secretary of State took steps to invoke the Illinois Act. On February 1, 1979, the Secretary of State notified MITE that he intended to issue an order requiring it to cease and desist further efforts to make a tender offer for Chicago Rivet. On February 2, 1979, Chicago Rivet notified MITE by letter that it would file suit in Illinois state court to enjoin the proposed tender offer. MITE renewed its request for injunctive relief in the District Court and on February 2 the District Court issued a preliminary injunction prohibiting the Secretary of State from enforcing the Illinois Act against MITE's tender offer for Chicago Rivet. 12 MITE then published its tender offer in the February 5 edition of the Wall Street Journal. The offer was made to all shareholders of Chicago Rivet residing throughout the United States. The outstanding stock was worth over $23 million at the offering price. On the same day Chicago Rivet made an offer for approximately 40% of its own shares at $30 per share.4 The District Court entered final judgment on February 9, declaring that the Illinois Act was preempted by the Williams Act and that it violated the Commerce Clause. Accordingly, the District Court permanently enjoined enforcement of the Illinois statute against MITE. Shortly after final judgment was entered, MITE and Chicago Rivet entered into an agreement whereby both tender offers were withdrawn and MITE was given 30 days to examine the books and records of Chicago Rivet. Under the agreement MITE was either to make a tender offer of $31 per share before March 12, 1979, which Chicago Rivet agreed not to oppose, or decide not to acquire Chicago Rivet's shares or assets. App. to Brief for Appellees 1a-4a. On March 2, 1979, MITE announced its decision not to make a tender offer. 13 The United States Court of Appeals for the Seventh Circuit affirmed sub nom. MITE Corp. v. Dixon, 633 F.2d 486 (1980). It agreed with the District Court that several provisions of the Illinois Act are pre-empted by the Williams Act and that the Illinois Act unduly burdens interstate commerce in violation of the Commerce Clause. We noted probable jurisdiction, 451 U.S. 968, 101 S.Ct. 2043, 68 L.Ed.2d 347 (1981), and now affirm. II 14 The Court of Appeals specifically found that this case was not moot, 633 F.2d, at 490, reasoning that because the Secretary has indicated he intends to enforce the Act against MITE, a reversal of the judgment of the District Court would expose MITE to civil and criminal liability5 for making the February 5, 1979, offer in violation of the Illinois Act. We agree. It is urged that the preliminary injunction issued by the District Court is a complete defense to civil or criminal penalties. While, as Justice STEVENS' concurrence indicates, that is not a frivolous question by any means; it is an issue to be decided when and if the Secretary of State initiates an action. That action would be foreclosed if we agree with the Court of Appeals that the Illinois Act is unconstitutional. Accordingly, the case is not moot. III 15 We first address the holding that the Illinois Takeover Act is unconstitutional under the Supremacy Clause. We note at the outset that in passing the Williams Act, which is an amendment to the Securities Exchange Act of 1934, Congress did not also amend § 28(a) of the 1934 Act, 15 U.S.C. § 78bb(a).6 In pertinent part, § 28(a) provides as follows: 16 "Nothing in this title shall affect the jurisdiction of the securities commission (or any agency or officer performing like functions) of any State over any security or any person insofar as it does not conflict with the provisions of this title or the rules and regulations thereunder." 48 Stat. 903. 17 Thus Congress did not explicitly prohibit States from regulating takeovers; it left the determination whether the Illinois statute conflicts with the Williams Act to the courts. Of course, a state statute is void to the extent that it actually conflicts with a valid federal statute; and 18 "[a] conflict will be found 'where compliance with both federal and state regulations is a physical impossibility . . .,' Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-143 [83 S.Ct. 1210, 1217, 10 L.Ed.2d 248] (1963), or where the state 'law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.' Hines v. Davidowitz, 312 U.S. 52, 67 [61 S.Ct. 399, 404, 85 L.Ed. 581] (1941); Jones v. Rath Packing Co., [430 U.S. 519,] 526, 540-541 [97 S.Ct. 1305, 1310, 1317, 51 L.Ed.2d 604 (1977)]. Accord, De Canas v. Bica, 424 U.S. 351, 363 [96 S.Ct. 933, 940, 47 L.Ed.2d 43] (1976)." Ray v. Atlantic Richfield Co., 435 U.S. 151, 158 [98 S.Ct. 988, 994, 55 L.Ed.2d 179] (1978). 19 Our inquiry is further narrowed in this case since there is no contention that it would be impossible to comply with both the provisions of the Williams Act and the more burdensome requirements of the Illinois law. The issue thus is, as it was in the Court of Appeals, whether the Illinois Act frustrates the objectives of the Williams Act in some substantial way. 20 The Williams Act, passed in 1968, was the congressional response to the increased use of cash tender offers in corporate acquisitions, a device that had "removed a substantial number of corporate control contests from the reach of existing disclosure requirements of the federal securities laws." Piper v. Chris-Craft Industries, Inc., 430 U.S. 1, 22 [97 S.Ct. 926, 939, 51 L.Ed.2d 124] (1977). The Williams Act filled this regulatory gap. The Act imposes several requirements. First, it requires that upon the commencement of the tender offer, the offeror file with the SEC, publish or send to the shareholders of the target company, and furnish to the target company detailed information about the offer. 15 U.S.C. § 78n(d)(1); 17 CFR § 240.24d-3 (1981). The offeror must disclose information about its background and identity; the source of the funds to be used in making the purchase; the purpose of the purchase, including any plans to liquidate the company or make major changes in its corporate structure; and the extent of the offeror's holdings in the target company. 15 U.S.C. § 78m(d)(1) (1976 ed., Supp.IV); 17 CFR § 240.13d-1 (1981). See also n. 2, supra. Second, stockholders who tender their shares may withdraw them during the first 7 days of a tender offer and if the offeror has not yet purchased their shares, at any time after 60 days from the commencement of the offer. 15 U.S.C. § 78n(d)(5).7 Third, all shares tendered must be purchased for the same price; if an offering price is increased, those who have already tendered receive the benefit of the increase. 15 U.S.C. § 78n(d)(7).8 21 There is no question that in imposing these requirements, Congress intended to protect investors. Piper v. Chris-Craft Industries, Inc., supra, at 35, 97 S.Ct. at 946; Rondeau v. Mosinee Paper Corp., 422 U.S. 49, 58, 95 S.Ct. 2069, 2075, 45 L.Ed.2d 12 (1975); S.Rep.No.550, 90th Cong., 1st Sess., 3-4 (1967) (Senate Report). But it is also crystal clear that a major aspect of the effort to protect the investor was to avoid favoring either management or the takeover bidder. As we noted in Piper, the disclosure provisions originally embodied in S. 2731 "were avowedly pro-management in the target company's efforts to defeat takeover bids." 430 U.S., at 30, 97 S.Ct., at 943. But Congress became convinced "that takeover bids should not be discouraged because they serve a useful purpose in providing a check on entrenched but inefficient management." Senate Report, at 3.9 It also became apparent that entrenched management was often successful in defeating takeover attempts. As the legislation evolved, therefore, Congress disclaimed any "intention to provide a weapon for management to discourage takeover bids," Rondeau v. Mosinee Paper Corp., supra, 422 U.S., at 58, 95 S.Ct., at 2075, and expressly embraced a policy of neutrality. As Senator Williams explained: "We have taken extreme care to avoid tipping the scales either in favor of management or in favor of the person making the takeover bids." 113 Cong.Rec. 24664 (1967). This policy of "evenhandedness," Piper v. Chris-Craft Industries, Inc., supra, 430 U.S., at 31, 97 S.Ct., at 944, represented a conviction that neither side in the contest should be extended additional advantages vis-a-vis the investor, who if furnished with adequate information would be in a position to make his own informed choice. We, therefore, agree with the Court of Appeals that Congress sought to protect the investor not only by furnishing him with the necessary information but also by withholding from management or the bidder any undue advantage that could frustrate the exercise of an informed choice. 633 F.2d, at 496. 22 To implement this policy of investor protection while maintaining the balance between management and the bidder, Congress required the latter to file with the Commission and furnish the company and the investor with all information adequate to the occasion. With that filing, the offer could go forward, stock could be tendered and purchased, but a stockholder was free within a specified time to withdraw his tendered shares. He was also protected if the offer was increased. Looking at this history as a whole, it appears to us, as it did to the Court of Appeals, that Congress intended to strike a balance between the investor, management, and the takeover bidder. The bidder was to furnish the investor and the target company with adequate information but there was no "inten[tion] to do . . . more than give incumbent management an opportunity to express and explain its position." Rondeau v. Mosinee Paper Corp., supra, 422 U.S., at 58, 95 S.Ct., at 2075. Once that opportunity was extended, Congress anticipated that the investor, if he so chose, and the takeover bidder should be free to move forward within the time frame provided by Congress. IV 23 The Court of Appeals identified three provisions of the Illinois Act that upset the careful balance struck by Congress and which therefore stand as obstacles to the accomplishment and execution of the full purposes and objectives of Congress. We agree with the Court of Appeals in all essential respects. 24 * The Illinois Act requires a tender offeror to notify the Secretary of State and the target company of its intent to make a tender offer and the material terms of the offer 20 business days before the offer becomes effective. Ill.Rev.Stat., ch. 1211/2, &Par; 137.54.E, 137.54.B (1979). During that time, the offeror may not communicate its offer to the shareholders. ¶ 137.54.A. Meanwhile, the target company is free to disseminate information to its shareholders concerning the impending offer. The contrast with the Williams Act is apparent. Under that Act, there is no pre-commencement notification requirement; the critical date is the date a tender offer is "first published or sent or given to security holders." 15 U.S.C. § 78n(d)(1). See also 17 CFR § 240.14d-2 (1981). 25 We agree with the Court of Appeals that by providing the target company with additional time within which to take steps to combat the offer, the precommencement notification provisions furnish incumbent management with a powerful tool to combat tender offers, perhaps to the detriment of the stockholders who will not have an offer before them during this period.10 These consequences are precisely what Congress determined should be avoided, and for this reason, the precommencement notification provision frustrates the objectives of the Williams Act. 26 It is important to note in this respect that in the course of events leading to the adoption of the Williams Act, Congress several times refused to impose a precommencement disclosure requirement. In October 1965, Senator Williams introduced S. 2731, a bill which would have required a bidder to notify the target company and file a public statement with the Securities and Exchange Commission at least 20 days before commencement of a cash tender offer for more than 5% of a class of the target company's securities. 111 Cong.Rec. 28259 (1965). The Commission commented on the bill and stated that "the requirement of a 20-day advance notice to the issuer and the Commission is unnecessary for the protection of security holders. . . ." 112 Cong.Rec. 19005 (1966). Senator Williams introduced a new bill in 1967, S. 510, which provided for a confidential filing by the tender offeror with the Commission five days prior to the commencement of the offer. S. 510 was enacted as the Williams Act after elimination of the advance disclosure requirement. As the Senate Report explained: 27 "At the hearings it was urged that this prior review was not necessary and in some cases might delay the offer when time was of the essence. In view of the authority and responsibility of the Securities and Exchange Commission to take appropriate action in the event that inadequate or misleading information is disseminated to the public to solicit acceptance of a tender offer, the bill as approved by the committee requires only that the statement be on file with the Securities and Exchange Commission at the time the tender offer is first made to the public." Senate Report, at 4. 28 Congress rejected another precommencement notification proposal during deliberations on the 1970 amendments to the Williams Act.11 B 29 For similar reasons, we agree with the Court of Appeals that the hearing provisions of the Illinois Act frustrate the congressional purpose by introducing extended delay into the tender offer process. The Illinois Act allows the Secretary of State to call a hearing with respect to any tender offer subject to the Act, and the offer may not proceed until the hearing is completed. Ill.Rev.Stat., ch. 1211/2, &Par; 137.57.A and B (1979). The Secretary may call a hearing at any time prior to the commencement of the offer, and there is no deadline for the completion of the hearing. &Par; 137.57.C and D. Although the Secretary is to render a decision within 15 days after the conclusion of the hearing, that period may be extended without limitation. Not only does the Secretary of State have the power to delay a tender offer indefinitely, but incumbent management may also use the hearing provisions of the Illinois Act to delay a tender offer. The Secretary is required to call a hearing if requested to do so by, among other persons, those who are located in Illinois "as determined by post office address as shown on the records of the target company and who hold of record or beneficially, or both, at least 10% of the outstanding shares of any class of equity securities which is the subject of the takeover offer." ¶ 137.57.A. Since incumbent management in many cases will control, either directly or indirectly, 10% of the target company's shares, this provision allows management to delay the commencement of an offer by insisting on a hearing. As the Court of Appeals observed, these provisions potentially afford management a "powerful weapon to stymie indefinitely a takeover." 633 F.2d, at 494.12 In enacting the Williams Act, Congress itself "recognized that delay can seriously impede a tender offer" and sought to avoid it. Great Western United Corp. v. Kidwell, 577 F.2d 1256, 1277 (CA5 1978); Senate Report, at 4.13 30 Congress reemphasized the consequences of delay when it enacted the Hart-Scott-Rodino Antitrust Improvements Act of 1976, Pub.L. 94-435, 90 Stat. 1397, 15 U.S.C. § 12 et seq. 31 "[I]t is clear that this short waiting period [the 10-day period for proration provided for by § 14(d)(6) of the Securities Exchange Act, which applies only after a tender offer is commenced] was founded on congressional concern that a longer delay might unduly favor the target firm's incumbent management, and permit them to frustrate many pro-competitive cash tenders. This ten-day waiting period thus underscores the basic purpose of the Williams Act—to maintain a neutral policy towards cash tender offers, by avoiding lengthy delays that might discourage their chances for success." H.R.Rep.No.94-1373, p. 12 (1976),14 U.S.Code Cong. & Admin.News, pp. 2572, 2644. 32 As we have said, Congress anticipated that investors and the takeover offeror would be free to go forward without unreasonable delay. The potential for delay provided by the hearing provisions upset the balance struck by Congress by favoring management at the expense of stockholders. We therefore agree with the Court of Appeals that these hearing provisions conflict with the Williams Act. C 33 The Court of Appeals also concluded that the Illinois Act is pre-empted by the Williams Act insofar as it allows the Secretary of State of Illinois to pass on the substantive fairness of a tender offer. Under ¶ 137.57.E of the Illinois law, the Secretary is required to deny registration of a takeover offer if he finds that the offer "fails to provide full and fair disclosure to the offerees . . . or that the take-over offer is inequitable . . ." (emphasis added).15 The Court of Appeals understood the Williams Act and its legislative history to indicate that Congress intended for investors to be free to make their own decisions. We agree. Both the House and Senate Reports observed that the Act was "designed to make the relevant facts known so that shareholders have a fair opportunity to make their decision." H.R.Rep.No.1711, 90th Cong., 2d Sess., 4 (1968), U.S.Code Cong. & Admin.News, pp. 2811, 2813; Senate Report, at 3. Thus, as the Court of Appeals said, "[t]he state thus offers investor protection at the expense of investor autonomy—an approach quite in conflict with that adopted by Congress." 633 F.2d, at 494. V 34 The Commerce Clause provides that "Congress shall have Power . . . [t]o regulate Commerce . . . among the several States." U.S.Const., Art. I, § 8, cl. 3. "[A]t least since Cooley v. Board of Wardens, 53 U.S. (12 How.) 299, 13 L.Ed. 996 (1852), it has been clear that 'the Commerce Clause. . . . even without implementing legislation by Congress is a limitation upon the power of the States.' " Great Atlantic & Pacific Tea Co. v. Cottrell, 424 U.S. 366, 370-371, 96 S.Ct. 923, 927, 47 L.Ed.2d 55 (1976), quoting Freeman v. Hewitt, 329 U.S. 249, 252, 67 S.Ct. 274, 276, 91 L.Ed. 265 (1946). See also Lewis v. BT Investment Managers, Inc., 447 U.S. 27, 35, 100 S.Ct. 2009, 2014, 64 L.Ed.2d 702 (1980). Not every exercise of state power with some impact on interstate commerce is invalid. A state statute must be upheld if it "regulates evenhandedly to effectuate a legitimate local public interest, and its effects on interstate commerce are only incidental . . . unless the burden imposed on such commerce is clearly excessive in relation to the putative local benefits." Pike v. Bruce Church, Inc., 397 U.S. 137, 142, 90 S.Ct. 844, 847, 25 L.Ed.2d 174 (1970), citing Huron Cement Co. v. Detroit, 362 U.S. 440, 443, 80 S.Ct. 813, 815, 4 L.Ed.2d 852 (1960). The Commerce Clause, however, permits only incidental regulation of interstate commerce by the States; direct regulation is prohibited. Shafer v. Farmers Grain Co., 268 U.S. 189, 199, 45 S.Ct. 481, 485, 69 L.Ed. 909 (1925). See also Pike v. Bruce Church, Inc., supra, 397 U.S., at 142, 90 S.Ct., at 847. The Illinois Act violates these principles for two reasons. First, it directly regulates and prevents, unless its terms are satisfied, interstate tender offers which in turn would generate interstate transactions. Second, the burden the Act imposes on interstate commerce is excessive in light of the local interests the Act purports to further. 35 * States have traditionally regulated intrastate securities transactions,16 and this Court has upheld the authority of States to enact "blue-sky" laws against Commerce Clause challenges on several occasions. Hall v. Geiger-Jones Co., 242 U.S. 539, 37 S.Ct. 217, 61 L.Ed. 480 (1917); Caldwell v. Sioux Falls Stock Yards Co., 242 U.S. 559, 37 S.Ct. 224, 61 L.Ed. 493 (1917); Merrick v. N. W. Halsey & Co., 242 U.S. 568, 37 S.Ct. 227, 61 L.Ed. 498 (1917). The Court's rationale for upholding blue-sky laws was that they only regulated transactions occurring within the regulating States. "The provisions of the law . . . apply to dispositions of securities within the State and while information of those issued in other States and foreign countries is required to be filed . . ., they are only affected by the requirement of a license of one who deals with them within the State. . . . Such regulations affect interstate commerce in [securities] only incidentally." Hall v. Geiger-Jones Co., supra, 242 U.S., at 557-558, 37 S.Ct., at 223 (citations omitted). Congress has also recognized the validity of such laws governing intrastate securities transactions in § 28(a) of the Securities Exchange Act, 15 U.S.C. § 78bb(a), a provision "designed to save state blue-sky laws from pre-emption." Leroy v. Great Western United Corp., 443 U.S. 173, 182, n.13, 99 S.Ct. 2710, 2716, n.13, 61 L.Ed.2d 464 (1979). 36 The Illinois Act differs substantially from state blue-sky laws in that it directly regulates transactions which take place across state lines, even if wholly outside the State of Illinois. A tender offer for securities of a publicly held corporation is ordinarily communicated by the use of the mails or other means of interstate commerce to shareholders across the country and abroad. Securities are tendered and transactions closed by similar means. Thus, in this case, MITE Corp., the tender offeror, is a Delaware corporation with principal offices in Connecticut. Chicago Rivet is a publicly held Illinois corporation with shareholders scattered around the country, 27% of whom live in Illinois. MITE's offer to Chicago Rivet's shareholders, including those in Illinois, necessarily employed interstate facilities in communicating its offer, which, if accepted, would result in transactions occurring across state lines. These transactions would themselves be interstate commerce. Yet the Illinois law, unless complied with, sought to prevent MITE from making its offer and concluding interstate transactions not only with Chicago Rivet's stockholders living in Illinois, but also with those living in other States and having no connection with Illinois. Indeed, the Illinois law on its face would apply even if not a single one of Chicago Rivet's shareholders were a resident of Illinois, since the Act applies to every tender offer for a corporation meeting two of the following conditions: the corporation has its principal executive office in Illinois, is organized under Illinois laws, or has at least 10% of its stated capital and paid-in surplus represented in Illinois. Ill.Rev.Stat., ch. 1211/2, ¶ 137.52-10(2) (1979). Thus the Act could be applied to regulate a tender offer which would not affect a single Illinois shareholder. 37 It is therefore apparent that the Illinois statute is a direct restraint on interstate commerce and that it has a sweeping extraterritorial effect. Furthermore, if Illinois may impose such regulations, so may other States; and interstate commerce in securities transactions generated by tender offers would be thoroughly stifled. In Shafer v. Farmers Grain Co., supra, at 199, 45 S.Ct., at 485, the Court held that "a state statute which by its necessary operation directly interferes with or burdens [interstate] commerce is a prohibited regulation and invalid, regardless of the purpose with which it was enacted." See also Hughes v. Alexandria Scrap Corp., 426 U.S. 794, 806, 96 S.Ct. 2488, 2496, 49 L.Ed.2d 220 (1976). The Commerce Clause also precludes the application of a state statute to commerce that takes place wholly outside of the State's borders, whether or not the commerce has effects within the State. In Southern Pacific Co. v. Arizona, 325 U.S. 761, 775, 65 S.Ct. 1515, 1523, 89 L.Ed. 1915 (1945), the Court struck down on Commerce Clause grounds a state law where the "practical effect of such regulation is to control [conduct] beyond the boundaries of the state. . . ." The limits on a State's power to enact substantive legislation are similar to the limits on the jurisdiction of state courts. In either case, "any attempt 'directly' to assert extraterritorial jurisdiction over persons or property would offend sister States and exceed the inherent limits of the State's power." Shaffer v. Heitner, 433 U.S. 186, 197, 97 S.Ct. 2569, 2576, 53 L.Ed.2d 683 (1977). 38 Because the Illinois Act purports to regulate directly and to interdict interstate commerce, including commerce wholly outside the State, it must be held invalid as were the laws at issue in Shaffer v. Farmers Grain Co. and Southern Pacific. B 39 The Illinois Act is also unconstitutional under the test of Pike v. Bruce Church, Inc., 397 U.S., at 142, 90 S.Ct., at 847, for even when a state statute regulates interstate commerce indirectly, the burden imposed on that commerce must not be excessive in relation to the local interests served by the statute. The most obvious burden the Illinois Act imposes on interstate commerce arises from the statute's previously described nationwide reach which purports to give Illinois the power to determine whether a tender offer may proceed anywhere. 40 The effects of allowing the Illinois Secretary of State to block a nationwide tender offer are substantial. Shareholders are deprived of the opportunity to sell their shares at a premium. The reallocation of economic resources to their highest valued use, a process which can improve efficiency and competition, is hindered. The incentive the tender offer mechanism provides incumbent management to perform well so that stock prices remain high is reduced. See Easterbrook & Fischel, The Proper Role of a Target's Management in Responding to a Tender Offer, 94 Harv.L.Rev. 1161, 1173-1174 (1981); Fischel, Efficient Capital Market Theory, the Market for Corporate Control, and the Regulation of Cash Tender Offers, 57 Texas L.Rev. 1, 5, 27-28, 45 (1978); H.R.Rep.No.94-1373, p. 12 (1976). 41 Appellant claims the Illinois Act furthers two legitimate local interests. He argues that Illinois seeks to protect resident security holders and that the Act merely regulates the internal affairs of companies incorporated under Illinois law. We agree with the Court of Appeals that these asserted interests are insufficient to outweigh the burdens Illinois imposes on interstate commerce. 42 While protecting local investors is plainly a legitimate state objective, the State has no legitimate interest in protecting nonresident shareholders. Insofar as the Illinois law burdens out-of-state transactions, there is nothing to be weighed in the balance to sustain the law. We note, furthermore, that the Act completely exempts from coverage a corporation's acquisition of its own shares. Ill.Rev.Stat., ch. 1211/2, ¶ 137.52-9(4) (1979). Thus Chicago Rivet was able to make a competing tender offer for its own stock without complying with the Illinois Act, leaving Chicago Rivet's shareholders to depend only on the protections afforded them by federal securities law, protections which Illinois views as inadequate to protect investors in other contexts. This distinction is at variance with Illinois' asserted legislative purpose, and tends to undermine appellant's justification for the burdens the statute imposes on interstate commerce. 43 We are also unconvinced that the Illinois Act substantially enhances the shareholders' position. The Illinois Act seeks to protect shareholders of a company subject to a tender offer by requiring disclosures regarding the offer, assuring that shareholders have adequate time to decide whether to tender their shares, and according shareholders withdrawal, proration, and equal consideration rights. However, the Williams Act provides these same substantive protections, compare Ill.Rev.Stat., ch. 1211/2, &Par; 137.59.C, D, and E (1979) (withdrawal, proration, and equal consideration rights), with 15 U.S.C. §§ 78n(d)(5), (6), and (7) and 17 CFR § 240.14d-7 (1981) (same). As the Court of Appeals noted, the disclosures required by the Illinois Act which go beyond those mandated by the Williams Act and the regulations pursuant to it may not substantially enhance the shareholders' ability to make informed decisions. 633 F.2d, at 500. It also was of the view that the possible benefits of the potential delays required by the Act may be outweighed by the increased risk that the tender offer will fail due to defensive tactics employed by incumbent management. We are unprepared to disagree with the Court of Appeals in these respects, and conclude that the protections the Illinois Act affords resident security holders are, for the most part, speculative. 44 Appellant also contends that Illinois has an interest in regulating the internal affairs of a corporation incorporated under its laws. The internal affairs doctrine is a conflict of laws principle which recognizes that only one State should have the authority to regulate a corporation's internal affairs—matters peculiar to the relationships among or between the corporation and its current officers, directors, and shareholders—because otherwise a corporation could be faced with conflicting demands. See Restatement (Second) of Conflict of Laws § 302, Comment b, pp. 307-308 (1971). That doctrine is of little use to the State in this context. Tender offers contemplate transfers of stock by stockholders to a third party and do not themselves implicate the internal affairs of the target company. Great Western United Corp. v. Kidwell, 577 F.2d, at 1280, n.53; Restatement, supra, § 302, Comment e, p. 310. Furthermore, the proposed justification is somewhat incredible since the Illinois Act applies to tender offers for any corporation for which 10% of the outstanding shares are held by Illinois residents, Ill.Rev.Stat., ch. 1211/2, ¶ 137.52-10 (1979). The Act thus applies to corporations that are not incorporated in Illinois and have their principal place of business in other States. Illinois has no interest in regulating the internal affairs of foreign corporations. 45 We conclude with the Court of Appeals that the Illinois Act imposes a substantial burden on interstate commerce which outweighs its putative local benefits. It is accordingly invalid under the Commerce Clause. The judgment of the Court of Appeals is 46 Affirmed. 47 Justice POWELL, concurring in part. 48 I agree with Justice MARSHALL that this case is moot. In view, however, of the decision of a majority of the Court to reach the merits, I join Parts I and V-B of the Court's opinion. 49 I join Part V-B because its Commerce Clause reasoning leaves some room for state regulation of tender offers. This period in our history is marked by conglomerate corporate formations essentially unrestricted by the antitrust laws. Often the offeror possesses resources, in terms of professional personnel experienced in takeovers as well as of capital, that vastly exceed those of the takeover target. This disparity in resources may seriously disadvantage a relatively small or regional target corporation. Inevitably there are certain adverse consequences in terms of general public interest when corporate headquarters are moved away from a city and State.* 50 The Williams Act provisions, implementing a policy of neutrality, seem to assume corporate entities of substantially equal resources. I agree with Justice STEVENS that the Williams Act's neutrality policy does not necessarily imply a congressional intent to prohibit state legislation designed to assure—at least in some circumstances—greater protection to interests that include but often are broader than those of incumbent management. 51 Justice O'CONNOR, concurring in part. 52 I agree with the Court that the case is not moot, and that portions of the Illinois Business Take-Over Act, Ill.Rev.Stat., ch. 1211/2, ¶ 137.51 et seq. (1979), are invalid under the Commerce Clause. Because it is not necessary to reach the pre-emption issue, I join only Parts I, II, and V of the Court's opinion, and would affirm the judgment of the Court of Appeals on that basis. 53 Justice STEVENS, concurring in part and concurring in the judgment. 54 The question whether this case is moot depends on the effect of the preliminary injunction entered on February 2, 1979, restraining the Illinois Secretary of State from enforcing the Illinois Business Take-Over Act while the injunction remained in effect. If, as Justice MARSHALL contends in his dissenting opinion, the injunction granted the MITE Corp. a complete immunity from state sanctions for any acts performed while the injunction was outstanding, I would agree that the case is moot. On the other hand, if the injunction did no more than it purported to do, setting aside the injunction would remove its protection and MITE would be subject to sanctions in the state courts. Those courts might regard the fact that an injunction was outstanding at the time MITE violated the Illinois statute as a defense to any enforcement proceeding, but unless the federal injunction was tantamount to a grant of immunity, there is no federal rule of law that would require the state courts to absolve MITE from liability. I believe, therefore, that to resolve the mootness issue—which, of course, is jurisdictional—we must answer the question that Justice MARSHALL's dissent raises. 55 Justice MARSHALL advances various reasons for adopting a rule that will give federal judges the power to grant complete immunity to persons who desire to test the constitutionality of a state statute. His proposed rule would treat any federal judge's preliminary injunction restraining enforcement of a state statute on federal grounds as a grant of immunity with respect to any conduct undertaken while the injunction was outstanding. Under the rule he proposes, "if the statute is later determined to be valid, the State will never be able to prosecute the individual that obtained the preliminary injunction for action taken while the injunction was in effect." Post, at 657, n. 1. For me, the question is not whether such a rule would be wise; the question is whether federal judges possess the power to grant such immunity. In my opinion they do not. 56 * The essential facts of this case are few and bear repeating. On February 2, 1979, MITE Corp. and MITE Holdings, Inc., obtained a preliminary injunction restraining the Illinois Secretary of State from invoking the provisions of the Illinois Business Take-Over Act to block MITE's intended takeover of Chicago Rivet & Machine Co. Three days later, without complying with the provisions of the Illinois statute, MITE published its offer in the Wall Street Journal. On February 9, 1979, the District Court entered a judgment declaring the Illinois statute unconstitutional; the court permanently enjoined the Secretary from enforcing the Illinois statute against MITE. 57 The State contends that the attempted takeover was subject to the provisions of the Illinois statute and that MITE violated the Act by failing to register with the Illinois Secretary of State. The State further argues that the Take-Over Act is consistent with federal law. For purposes of deciding the mootness issue, we must assume that these contentions are correct; a holding that this case is moot would mean that MITE is completely protected from any adverse action whether or not the statute is unconstitutional. Such a conclusion would be possible only if the District Court's preliminary injunction granted MITE absolute and permanent immunity from any prosecution—civil or criminal—brought to enforce the Illinois statute. 58 Neither the terms of the preliminary injunction nor prior equity practice provides any support for an interpretation of the District Court's order as a grant of total immunity from future prosecution. More fundamentally, federal judges have no power to grant such blanket dispensation from the requirements of valid legislative enactments. A. 59 An injunction restrains conduct. Its effect is normally limited to the parties named in the instrument. Since a preliminary injunction may be granted on a mere probability of success on the merits, generally the moving party must demonstrate confidence in his legal position by posting bond in an amount sufficient to protect his adversary from loss in the event that future proceedings prove that the injunction issued wrongfully.1 The bond, in effect, is the moving party's warranty that the law will uphold the issuance of the injunction. 60 These features of injunctive relief are inconsistent with a blanket grant of immunity, as this case demonstrates. The preliminary injunction did not purport to provide permanent immunity for violations of the statute that occurred during its effective period. It merely provided that the Secretary of State was enjoined from "issuing any cease and desist order or notice of hearing or from otherwise invoking, applying, or enforcing the Illinois Business Take-Over Act" against MITE. Record 16. It did not enjoin other parties who are authorized by the Act to enforce its provisions. Ill.Rev. ch. 1211/2, &Par; 137.62, 137.64 (1979). Moreover, the preliminary injunction was entered without any declaration that the Illinois statute was unconstitutional. There simply is no basis on which to conclude that the preliminary injunction issued by the District Court should be construed as having granted MITE permanent immunity from future proceedings brought under the Illinois statute. 61 In Steffel v. Thompson, 415 U.S. 452, 94 S.Ct. 1209, 39 L.Ed.2d 505, the Court unanimously held that an individual who wished to engage in "constitutionally protected activity" but was threatened with prosecution under a state criminal statute could obtain a declaratory judgment in federal court declaring the statute invalid. The Court did not suggest that, armed with such a judgment from a federal district court, the individual could violate the statute with impunity; indeed, it stated just the opposite: 62 " '[A] federal declaration of unconstitutionality reflects the opinion of the federal court that the statute cannot be fully enforced. If a declaration of total unconstitutionality is affirmed by this Court, it follows that this Court stands ready to reverse any conviction under the statute.' " Id., at 469-470, 94 S.Ct., at 1220-1221 (quoting Perez v. Ledesma, 401 U.S. 82, 124, 91 S.Ct. 674, 696, 27 L.Ed.2d 701 (separate opinion of BRENNAN, J.)).2 63 Justice WHITE attached possibly the greatest significance to a federal declaratory judgment, writing separately in Steffel that "I would anticipate that a final declaratory judgment entered by a federal court holding particular conduct of the federal plaintiff to be immune on federal constitutional grounds from prosecution under state law should be accorded res judicata effect in any later prosecution of that very conduct." 415 U.S., at 477, 94 S.Ct., at 1224 (concurring opinion). A declaratory judgment reversed on appeal, however, certainly would not have such res judicata effect. 64 An individual who is imminently threatened with prosecution for conduct that he believes is constitutionally protected should not be forced to act at his peril. One purpose of the federal declaratory judgment statute is to permit such an individual to test the legality of a state statute before engaging in conduct that is prohibited by its terms. See S.Rep.No.1005, 73d Cong., 2d Sess., 2-3 (1934). Recognition of this fact, however, does not determine the point at which an individual may act with absolute assurance that he may not be punished for his contemplated activity. The fact that a federal judge has entered a declaration that the law is invalid does not provide that assurance; every litigant is painfully aware of the possibility that a favorable judgment of a trial court may be reversed on appeal. To repeat the words of this Court in Steffel, the most that can be said is: " 'If a declaration of total unconstitutionality is affirmed by this Court, it follows that this Court stands ready to reverse any conviction under the statute.' " 415 U.S., at 470, 94 S.Ct., at 1221 (quoting Perez v. Ledesma, supra, 401 U.S., at 124, 91 S.Ct., at 696 (separate opinion of BRENNAN, J.)).3 65 Since a final judgment declaring a state statute unconstitutional would not grant immunity for actions taken in reliance on the court's decision, certainly a preliminary injunction—which on its face does nothing more than temporarily restrain conduct—should not accomplish that result. Neither the preliminary injunction nor the subsequent judgment declaring the statute unconstitutional can fairly be construed as a grant of absolute immunity from enforcement of the Illinois statute.4 B 66 My conclusions concerning the proper nature of injunctive and declaratory relief are not based upon arcane interpretations of common law. Federal courts are courts of limited jurisdiction.5 Before a federal court exercises any governmental power, it has a duty to determine its own jurisdiction to act. There simply is no constitutional or statutory authority that permits a federal judge to grant dispensation from a valid state law.6 67 As I have written before, the federal judiciary can continue to perform its vital function in our governmental structure only if it recognizes the limitations on its own legitimate authority. United States v. New York Telephone Co., 434 U.S. 159, 178, 98 S.Ct. 364, 375, 54 L.Ed.2d 376 (STEVENS, J., dissenting in part). A belief that a particular result appears reasonable or wise is an insufficient predicate for the exercise of federal judicial power. The District Court in this case entered both an injunction restraining certain conduct by the Illinois Secretary of State and a judgment declaring a state statute unconstitutional. It did not because it could not—grant immunity from the requirements of a valid state law.7 As a result, this Court has jurisdiction to consider whether the judgment and relief entered by the District Court were proper.8 II 68 On the merits, I agree with the Court that the Illinois Take-Over Act is invalid because it burdens interstate commerce. I therefore join Part V of its opinion. I am not persuaded, however, that Congress' decision to follow a policy of neutrality in its own legislation is tantamount to a federal prohibition against state legislation designed to provide special protection for incumbent management. Accordingly, although I agree with the Court's assessment of the impact of the Illinois statute, I do not join its pre-emption holding. 69 Justice MARSHALL, with whom Justice BRENNAN joins, dissenting. 70 The jurisdiction of this Court depends upon the existence of a live controversy. We may resolve a particular dispute only if the parties have a real interest in the outcome of that dispute. Otherwise, the case is moot, and must be dismissed. Roe v. Wade, 410 U.S. 113, 125, 93 S.Ct. 705, 712, 35 L.Ed.2d 147 (1973); SEC v. Medical Committee for Human Rights, 404 U.S. 403, 407, 92 S.Ct. 577, 580, 30 L.Ed.2d 560 (1972). In my view, this case should have been dismissed. The parties to this appeal have no adversary interest in the outcome of this case. Their positions would be the same whether the Court approved the Illinois Business Take-Over Act or struck it down. Because the Court finds that the Illinois Act is unconstitutional, there will be no further litigation. However, even if the Court had held that the Illinois Act is constitutional, and had lifted the permanent injunction that now restrains enforcement of the Act against MITE, there would be no basis for continued litigation. The Secretary stated that if the decision below were reversed, he would initiate enforcement proceedings against MITE in state court, seeking civil and criminal penalties for its failure to comply with the Illinois Act. But a preliminary injunction was in effect at the time the alleged violations occurred. As I explain below, I believe that this injunction would have barred the Secretary from seeking either civil or criminal penalties for violations of the Act that occurred during that period. MITE would have a complete defense to such an action. 71 * The Secretary argues that the case is not moot because the preliminary injunction would not be a complete defense to a state enforcement action. He contends that the preliminary injunction merely barred him from commencing an enforcement action during the period the injunction was in effect. Thus, if this Court had decided that the statute is constitutional and had lifted the permanent injunction, the State would have been able to commence an action seeking penalties for any violations that occurred during the period the preliminary injunction was in effect. In other words, argues the Secretary, the preliminary injunction only provided temporary security. It enabled MITE to go forward with the tender offer—subject to the risk that at some later stage, the constitutionality of the statute would be upheld, and the State would commence enforcement proceedings. 72 Federal courts undoubtedly have the power to issue a preliminary injunction that restrains enforcement of a state statute, subject to the condition that if the statute is later found to be valid, the State is free to seek penalties for violations that occurred during the period the injunction was in effect. In my view, however, federal courts also have the power to issue a preliminary injunction that offers permanent protection from penalties for violations of the statute that occurred during the period the injunction was in effect.1 Determining whether a particular injunction provides temporary or permanent protection becomes a question of interpretation. 73 I believe that in the ordinary case, unless the order contains specific language to the contrary, it should be presumed that an injunction secures permanent protection from penalties for violations that occurred during the period it was in effect; the burden should be on the State to show that the injunction provided only temporary security.2 A presumption in favor of permanent protection is likely to reflect the intentions of the court that granted the motion. In acting upon a request for an injunction, it will recognize that short-term protection is often only marginally better than no protection at all. Parties seek to restrain the enforcement of a state statute, not just because they want short-term protection, but because they desire permanent immunity for actions they take in reliance on the injunction. If they are contemplating action that might violate a state statute, they will take little solace from temporary immunity—when they know that if they decide to act, enforcement proceedings might be initiated at some later stage.3 74 Here, the preliminary injunction does not expressly state that it provides permanent immunity from penalties for violations of the Illinois Act that may occur during its effective period. The injunction provides only that the Secretary of State is enjoined from "issuing any cease and desist order or notice of hearing or from otherwise invoking, applying, or enforcing the Illinois Business Take-Over Act" against MITE. Record 16. However, I see no reason why the presumption in favor of permanent protection should not be applied here. In this context, as the District Court must have recognized, permanent protection was needed. MITE sought an injunction, not just because it desired protection from enforcement actions during the period it was actually making the tender offer, but also because it desired protection from such actions in the future. The Act provides for substantial civil and criminal penalties. MITE would have been reluctant to go forward with its offer, which entailed considerable expense, if there were some risk that it would be penalized later. Indeed, in the Schedule 14D-1 filed with the SEC, MITE expressly stated that it would not commence the tender offer unless it obtained injunctive relief. It also reserved the right to withdraw its offer if injunctive relief were initially granted, but later withdrawn. See Record, Plaintiff's Exhibit 14.4 75 Interpreting the injunction to provide permanent protection also ensures that MITE could never be penalized for acting in reliance on the injunction.5 MITE went forward with the tender offer, reasonably believing that the District Court's order provided complete immunity. Under the circumstances, it would be improper to permit the State to penalize action taken while the injunction was in effect. In the past, this Court has recognized that reasonable reliance on judicial pronouncements may constitute a valid defense to criminal prosecution. See, e.g., Marks v. United States, 430 U.S. 188, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977).6 76 In addition to arguing that the preliminary injunction should be interpreted to provide only temporary protection from a state enforcement action, the Secretary argues that resolution of the mootness issue in this case should be controlled by Leroy v. Great Western United Corp., 443 U.S. 173, 99 S.Ct. 2710, 61 L.Ed.2d 464 (1979). In that case, Great Western announced its intention to make a tender offer to purchase stock in another corporation. Idaho officials responsible for administering an Idaho statute governing corporate takeovers, see Idaho Code § 30-1501 et seq. (1980), objected to the offer and delayed its effective date. Great Western brought an action in Federal District Court, seeking a declaration that the Idaho takeover law was unconstitutional, and an injunction restraining Idaho officials from enforcing the statute. The District Court granted injunctive relief that enabled Great Western to complete the acquisition. This Court, in reviewing the case, held that the controversy was not moot. "[T]he question whether Great Western has violated Idaho's statute will remain open unless and until the District Court's judgment is finally affirmed." Id., at 178, 99 S.Ct., at 2714.7 77 Leroy v. Great Western United Corp. is easily distinguishable from this case. Unlike MITE, Great Western took actions that might have violated the state takeover statute before it obtained injunctive relief. If this Court had decided that the Idaho statute was valid, Idaho officials might have been able to seek penalties for those preinjunction violations.8 Leroy v. Great Western United Corp. can also be distinguished on the ground that the takeover offer in that case was successful. If the Idaho statute had been found to be valid, then Idaho officials would have been able to seek a rescission of the takeover.9 Here, since the acquisition was never completed, Illinois officials could not seek rescission.10 78 Finally, this case does not fall within the exception to the mootness doctrine for cases that "are capable of repetition, yet evading review." Unless a class action is involved, that exception applies only when the challenged action is too short to be fully litigated before its cessation, and when there is a reasonable expectation or a demonstrated probability that the same complaining party will be subject to the same action in the future. Illinois State Board of Elections v. Socialist Workers Party, 440 U.S. 173, 187, 99 S.Ct. 983, 991, 59 L.Ed.2d 230 (1979); Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 348, 46 L.Ed.2d 350 (1975). The second requirement has not been satisfied here. MITE has agreed not to renew its efforts to acquire Chicago Rivet. Thus, unless MITE breaches its agreement,11 the State will never again have occasion to prevent MITE from making a takeover offer for Chicago Rivet. In addition, there has been no showing that MITE plans to acquire another corporation with substantial connection to Illinois. Thus, there is no demonstrated probability that the State will have occasion to prevent MITE from making a takeover offer for some other corporation. II 79 The majority disposes of the mootness issue in a short paragraph. It concedes that the only possible basis for continued litigation in this case would be a state action for penalties. It further concedes that the preliminary injunction issued by the District Court may be a complete defense to an action for civil or criminal penalties. It argues, however, that the effect to be given the preliminary injunction should not be reached in this case. Rather, that question should be decided in a state enforcement action, if it is raised as a defense. Thus, contends the majority, the case is not moot. 80 I am completely unpersuaded by the majority's facile analysis. In deciding whether a case is moot, the Court must determine whether there is a live controversy. There is a live controversy in this case only if the State could seek penalties from MITE. Here, the State could not seek penalties from MITE. It may be true that the State could file a complaint if this Court were to lift the permanent injunction. However, this fact is not enough to keep the case alive where, as a matter of federal law, the complaint must be dismissed. If the action that the State plans to commence in state court lacks any merit—if MITE has an automatic defense to that action—then there simply is no controversy. 81 This case is made more difficult because the Court has never before decided what effect should be given to preliminary injunctions. But the fact that we must decide a novel question does not make the case any less moot. Certainly, if the Court had already held that a preliminary injunction provides permanent immunity, the case would be moot even though the State could go into state court and seek penalties. Such a suit, which would be clearly frivolous, could not keep the dispute alive. 82 The Court's refusal to confront the question whether a preliminary injunction would provide a complete defense is particularly ironic, given its recent decision in Lane v. Williams, 455 U.S. 624, 102 S.Ct. 1322, 71 L.Ed.2d 508 (1982). Respondents in that case had pleaded guilty in unrelated Illinois state-court prosecutions for burglary, an offense punishable by imprisonment and a mandatory 3-year parole term. Neither respondent was informed during his plea acceptance hearing that the negotiated sentence included the mandatory parole term. Each respondent completed his prison sentence but was reincarcerated for parole violation. While in custody, they filed petitions for federal habeas corpus, alleging that their guilty pleas were invalid because they were not informed of the mandatory parole requirement. The District Court decided to enter an order declaring the parole term void, and the United States Court of Appeals for the Seventh Circuit affirmed. By the time the cases reached this Court, both respondents had completed their sentences, and their parole terms had expired. This Court held that the claims for relief were moot. In reaching this conclusion, the Court determined that as a matter of Illinois law, no collateral consequences would flow from the parole revocations. Thus, there would be no point in declaring the parole terms void. In other words, the Court reached out to decide a question of state law in order to hold that the case was moot. Here, by contrast, the Court refuses to confront an important question of federal law—deciding instead that the question should be left to a state court—so that it can avoid holding that the case is moot. III 83 The parties to this appeal have no adversary interest in the resolution of the merits of this controversy. The majority acts without jurisdiction when it addresses the question whether the Illinois Business Take-Over Act is constitutional. Because I believe the case is moot, I would have vacated the judgment of the Court of Appeals, with instructions that it remand the case to the District Court with instructions to dismiss. 84 Justice REHNQUIST, dissenting. 85 I agree with Justice MARSHALL that this case does not present a justiciable controversy, but for a different reason. 86 MITE obtained an injunction in order to effect a cash tender offer for the stock of Chicago Rivet. The injunction restrained the Illinois Secretary of State from interfering with the Chicago Rivet tender offer by enforcing the Illinois Business Take-Over Act against MITE. Three days after the District Court issued a permanent injunction, MITE and Chicago Rivet reached an agreement and MITE withdrew its extant offer. Approximately one month later, MITE announced its decision not to make any tender offer. MITE is not presently engaging in activity that is regulated by the Illinois statute, and there is no indication that MITE intends to engage in any such activity in the future. Therefore, the facts that gave rise to this controversy over the constitutionality of Illinois' anti-takeover statutes no longer exist, and it is unlikely that they will be repeated in the future. As the tender offer has met its demise for reasons having nothing to do with the validity of the Illinois statute, the injunction is no longer necessary to accomplish the purposes for which it was obtained. MITE no longer needs an injunction in order to effect a tender offer for the shares of Chicago Rivet or any other corporation subject to the Illinois Act. Nor does MITE need the injunction in order to preclude the Secretary from rescinding a completed tender offer. 87 Despite these developments which have occurred after the District Court issued the injunction, the Court concludes that the present controversy between the Illinois Secretary of State and MITE over the constitutionality of the Illinois Business Take-Over Act is not moot. According to the Court, the Illinois Secretary of State's intention to bring an enforcement action against MITE keeps the present controversy alive. The possibility of a future enforcement action, however, is insufficient for me to conclude that the controversy that is before the Court is not moot.1 88 This Court has no power over a suit not pending before it. " 'Our power only extends over and is limited by the conditions of the case now before us.' " Oil Workers v. Missouri, 361 U.S. 363, 370, 80 S.Ct. 391, 396, 4 L.Ed.2d 373 (1960), quoting American Book Co. v. Kansas ex rel. Nichols, 193 U.S. 49, 52, 24 S.Ct. 394, 396, 48 L.Ed. 613 (1904). A case pending in this Court may not be kept alive simply because similar or identical issues are currently ripe for decision in a controversy between the same parties in another court. See Oil Workers v. Missouri, 361 U.S., at 370-371, 80 S.Ct., at 396; American Book Co. v. Kansas ex rel. Nichols, supra, 193 U.S., at 51, 24 S.Ct., at 395. A fortiori, this case may not be kept alive simply because there may exist a presently unripened controversy between these same parties over the constitutionality of the same Act. This is so even if our resolution of the merits of the instant case will resolve certain defenses that MITE could raise in an enforcement action were one to be brought by the Secretary. It follows that this case is not alive simply because a decision on the merits in this case will determine whether or not the Secretary's threatened enforcement action may ever ripen into a live controversy. 89 If an enforcement action were brought by the Secretary, "there is no way to know what the outcome of such a proceeding in the [Illinois] courts might be." Oil Workers v. Missouri, supra, 361 U.S., at 371, 80 S.Ct., at 396. The Illinois courts may well conclude that the injunction constitutes a defense either on state law grounds or upon the grounds suggested by Justice MARSHALL in his dissent. The Illinois courts may also agree with MITE that the Business Take-Over Act is pre-empted by the Williams Act or that Illinois' regulation of interstate tender offers runs afoul of the Commerce Clause. The possibility that this Court might disagree with the Illinois courts' ultimate resolution of the issues arising in a presently unripe, but threatened, enforcement action hardly justifies the Court's resolution of important constitutional issues in the abstract posture in which they are currently presented.2 90 The Secretary and MITE dispute the propriety of the injunction issued by the District Court in this case only with respect to a controversy that may ripen in another court. Because the controversy that is before the Court is no longer alive, I would vacate the judgment of the Court of Appeals and order that court to remand this case to the District Court with instructions to dismiss the complaint. See Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 348, 46 L.Ed.2d 350 (1975); United States v. Munsingwear, Inc., 340 U.S. 36, 39, 71 S.Ct. 104, 106, 95 L.Ed. 36 (1950). * THE CHIEF JUSTICE joins the opinion in its entirety; Justice BLACKMUN joins Parts I, II, III, and IV; Justice POWELL joins Parts I, and V-B; and Justice STEVENS and Justice O'CONNOR join Parts I, II, and V. 1 The Illinois Act defines "take-over offer" as "the offer to acquire or the acquisition of any equity security of a target company, pursuant to a tender offer. . . ." Ill.Rev.Stat., ch. 1211/2, ¶ 137.52-9 (1979). "A tender offer has been conventionally understood to be a publicly made invitation addressed to all shareholders of a corporation to tender their shares for sale at a specified price." Note, The Developing Meaning of "Tender Offer" Under the Securities Exchange Act of 1934, 86 Harv.L.Rev. 1250, 1251 (1973) (footnotes omitted). The terms "tender offer" and "takeover offer" are often used interchangeably. 2 The Williams Act, 82 Stat. 454, codified at 15 U.S.C. §§ 78m(d)-(e) and 78n(d)-(f), added new §§ 13(d), 13(e), and 14(d)-(f) to the Securities Exchange Act of 1934. Section 14(d)(1) of the Securities Exchange Act requires an offeror seeking to acquire more than 5% of any class of equity security by means of a tender offer to first file a Schedule 14D-1 with the Securities and Exchange Commission. The Schedule requires disclosure of the source of funds used to purchase the target shares, past transactions with the target company, and other material financial information about the offeror. In addition, the offeror must disclose any antitrust or other legal problems which might result from the success of the offer. 17 CFR § 240.14d-100 (1981). Section 14(d)(1) requires the offeror to publish or send a statement of the relevant facts contained in the Schedule 14D-1 to the shareholders of the target company. In addition, § 13(d), added by the Williams Act, requires a purchaser of any equity security registered pursuant to § 12 of the Securities Exchange Act, 15 U.S.C. § 78l, to file a Schedule 13D with the Commission within 10 days after its purchases have exceeded 5% of the outstanding shares of the security. Schedule 13D requires essentially the same disclosures as required by Schedule 14D-1. Compare 17 CFR § 240.13d-101 (1981) with 17 CFR § 240.14d-100 (1981). 3 In addition to filing suit in state court, Chicago Rivet filed a complaint with the Pennsylvania Securities Commission requesting the Commission to enforce the Pennsylvania Act against MITE. On January 31, 1979, the Pennsylvania Securities Commission decided that it would not invoke the Pennsylvania Takeover Disclosure Law. The next day, the United States District Court for the Western District of Pennsylvania, to which MITE had removed the state-court action, denied Chicago Rivet's motion for a temporary restraining order. 4 Chicago Rivet's offer for its own shares was exempt from the requirements of the Illinois Act pursuant to Ill.Rev.Stat., ch. 1211/2, ¶ 137.52-9(4) (1979). 5 The Secretary of State may bring an action for civil penalties for violations of the Illinois Act, Ill.Rev.Stat., ch. 1211/2, ¶ 137.65 (1979), and a person who willfully violates the Act is subject to criminal prosecution. ¶ 137.63. 6 There is no evidence in the legislative history that Congress was aware of state takeover laws when it enacted the Williams Act. When the Williams Act was enacted in 1968, only Virginia had a takeover statute. The Virginia statute, Va.Code § 13.1-528 (1978), became effective March 5, 1968; the Williams Act was enacted several months later on July 19, 1968. Takeover statutes are now in effect in 37 States. Sargent, On the Validity of State Takeover Regulation: State Responses to MITE and Kidwell, 42 Ohio St.L.J. 689, 690, n. 7 (1981). 7 The 7-day withdrawal period contained in the Williams Act has been extended to 15 business days by the Commission. 17 CFR § 240.14d-7(a)(1) (1981). 8 The Williams Act also provides that when the number of shares tendered exceeds the number of shares sought in the offer, those shares tendered during the first 10 days of the offer must be purchased on a pro rata basis. 15 U.S.C. § 78n(d)(6). The Act also contains a general antifraud provision, 15 U.S.C. § 78n(e), which has been interpreted to require disclosure of material information known to the offeror even if disclosure were not otherwise required. See, e.g., Sonesta International Hotels Corp. v. Wellington Associates, 483 F.2d 247, 250 (CA2 1973). 9 Congress also did not want to deny shareholders "the opportunities which result from the competitive bidding for a block of stock of a given company," namely, the opportunity to sell shares for a premium over their market price. 113 Cong.Rec. 24666 (1967) (remarks of Sen. Javits). 10 See n. 11 and accompanying text, infra. 11 H.R. 4285, 91st Cong., 2d Sess. (1970). The bill was not reported out of the Subcommittee. Instead, the Senate amendments to the Williams Act, which did not contain precommencement notification provisions, were adopted. Pub.L. 91-567, 84 Stat. 1497. The Securities and Exchange Commission has promulgated detailed rules governing the conduct of tender offers. Rule 14d-2(b), 17 CFR § 240.14d-2(b) (1981), requires that a tender offeror make its offer effective within five days of publicly announcing the material terms of the offer by disseminating specified information to shareholders and filing the requisite documents with the Commission. Otherwise the offeror must announce that it is withdrawing its offer. The events in this litigation took place prior to the effective date of Rule 14d-2(b), and because Rule 14d-2(b) operates prospectively only, see 44 Fed.Reg. 70326 (1979), it is not at issue in this case. 12 Delay has been characterized as "the most potent weapon in a tender-offer fight." Langevoort, State Tender-Offer Legislation: Interests, Effects, and Political Competency, 62 Cornell L.Rev. 213, 238 (1977). See also Wachtell, Special Tender Offer Litigation Tactics, 32 Bus.Law. 1433, 1437-1442 (1977); Wilner & Landy, The Tender Trap: State Takeover Statutes and Their Constitutionality, 45 Ford.L.Rev. 1, 9-10 (1976). 13 According to the Securities and Exchange Commission, delay enables a target company to: "(1) repurchase its own securities; "(2) announce dividend increases or stock splits; "(3) issue additional shares of stock; "(4) acquire other companies to produce an antitrust violation should the tender offer succeed; "(5) arrange a defensive merger; "(6) enter into restrictive loan agreements; and "(7) institute litigation challenging the tender offer." Brief for Securities and Exchange Commission as Amicus Curiae 10, n. 8. 14 Representative Rodino set out the consequences of delay in greater detail when he described the relationship between the Hart-Scott-Rodino Act and the Williams Act: "In the case of cash tender offers, more so than in other mergers, the equities include time and the danger of undue delay. This bill in no way intends to repeal or reverse the congressional purpose underlying the 1968 Williams Act, or the 1970 amendments to that act. . . . Lengthier delays will give the target firm plenty of time to defeat the offer, by abolishing cumulative voting, arranging a speedy defensive merger, quickly incorporating in a State with an antitakeover statute, or negotiating costly lifetime employment contracts for incumbent management. And the longer the waiting period, the more the target's stock may be bid up in the market, making the offer more costly—and less successful. Should this happen, it will mean that shareholders of the target firm will be effectively deprived of the choice that cash tenders give to them: Either accept the offer and thereby gain the tendered premium, or reject the offer. Generally, the courts have construed the Williams Act so as to maintain these two options for the target company's shareholders, and the House conferees contemplate that the courts will continue to do so." 122 Cong.Rec. 30877 (1976). 15 Appellant argues that the Illinois Act does not permit him to adjudicate the substantive fairness of a tender offer. Brief for Appellant 21-22. On this state-law issue, however, we follow the view of the Court of Appeals that ¶ 137.57.E allows the Secretary of State "to pass upon the substantive fairness of a tender offer. . . ." 633 F.2d 486, 493 (1980). 16 For example, the Illinois blue-sky law, Ill.Rev.Stat. ch. 1211/2, ¶ 137.1 et seq. (1979 and Supp.1980), provides that securities subject to the law must be registered "prior to sale in this State. . . ." ¶ 137.5. * The corporate headquarters of the great national and multinational corporations tend to be located in the large cities of a few States. When corporate headquarters are transferred out of a city and State into one of these metropolitan centers, the State and locality from which the transfer is made inevitably suffer significantly. Management personnel—many of whom have provided community leadership—may move to the new corporate headquarters. Contributions to cultural, charitable, and educational life—both in terms of leadership and financial support also tend to diminish when there is a move of corporate headquarters. 1 As provided by Federal Rule of Civil Procedure 65(c): "No restraining order or preliminary injunction shall issue except upon the giving of security by the applicant, in such sum as the court deems proper, for the payment of such costs and damages as may be incurred or suffered by any party who is found to have been wrongfully enjoined or restrained. No such security shall be required of the United States or of an officer or agency thereof." In Illinois damages apparently may be recovered for injuries caused by a preliminary injunction issued wrongfully by a state court even in the absence of an indemnity bond or abuse of process. See Ill.Rev.Stat., ch. 69, ¶ 12 (1979); Note, 73 Harv.L.Rev. 333, 347 (1959). 2 See also 415 U.S., at 480, 94 S.Ct., at 1226 (REHNQUIST, J., concurring) ("There is nothing in the [Declaratory Judgment] Act's history to suggest that Congress intended to provide persons wishing to violate state laws with a federal shield behind which they could carry on their contemplated conduct"); id., at 482, 94 S.Ct., at 1227 ("A declaratory judgment is simply a statement of rights, not a binding order supplemented by continuing sanctions"). 3 The fact that an unreviewed judgment does not provide absolute protection does not render the declaratory judgment of a district court or a court of appeals meaningless. As stated in Steffel: " 'Even where a declaration of unconstitutionality is not reviewed by this Court, the declaration may still be able to cut down the deterrent effect of an unconstitutional state statute. The persuasive force of the court's opinion and judgment may lead state prosecutors, courts, and legislators to reconsider their respective responsibilities toward the statute. Enforcement policies or judicial construction may be changed, or the legislature may repeal the statute and start anew.' " 415 U.S., at 470, 94 S.Ct., at 1221 (quoting Perez v. Ledesma, 401 U.S., at 125, 91 S.Ct., at 697 (separate opinion of BRENNAN, J.)). 4 In Liner v. Jafco, Inc., 375 U.S. 301, 84 S.Ct. 391, 11 L.Ed.2d 347, the respondent obtained an injunction from a state court that restrained picketing at a construction site. Petitioners moved to dissolve the injunction on the ground that the state court was without jurisdiction to adjudicate the controversy because the subject matter of the picketing was exclusively within the cognizance of the National Labor Relations Board. Petitioners' motion was denied by the state court and that decision was affirmed on appeal. This Court granted a petition for certiorari. While the case was pending in the state appellate court, construction at the site was completed. This Court nevertheless held that the issue of whether the injunction had issued properly was not moot because the respondent remained liable on an indemnity bond if the injunction had issued wrongfully. The Court stated: "The petitioners plainly have 'a substantial stake in the judgment . . .,' Fiswick v. United States, 329 U.S. 211, 222, 67 S.Ct. 224, 230, 91 L.Ed. 196, which exists apart from and is unaffected by the completion of construction. Their interest derives from the undertaking of respondent Jafco, Inc., in the injunction bond to indemnify them in damages if the injunction was 'wrongfully' sued out. Whether the injunction was wrongfully sued out turns solely upon the answer to the federal question which the petitioners have pressed from the beginning. If the answer of the Tennessee Court of Appeals to that question may not be challenged here, the petitioners have no recourse against Jafco on the bond." Id., 375 U.S., at 305-306, 84 S.Ct., at 394. In this case it does not appear that MITE is liable on an injunction bond. The posting of an indemnity bond, however, merely creates a right of action—that may or may not otherwise exist—for damages caused during the period that a wrongfully issued injunction was in effect. In this case, such rights of action exist under an independent state law that we must presume to be valid. As in Liner, these rights of action may be pursued "if the injunction was 'wrongfully' sued out"; and "[w]hether the injunction was wrongfully sued out turns solely upon the answer to the federal question which the petitioners have pressed from the beginning." 5 As stated by Chief Justice Marshall in Ex parte Bollman, 8 U.S. (4 Cranch) 75, 93, 2 L.Ed. 554: "As preliminary to any investigation of the merits of this motion, this court deems it proper to declare that it disdains all jurisdiction not given by the constitution, or by the laws of the United States. "Courts which originate in the common law possess a jurisdiction which must be regulated by the common law, until some statute shall change their established principles; but courts which are created by written law, and whose jurisdiction is defined by written law, cannot transcend that jurisdiction. It is unnecessary to state the reasoning on which this opinion is founded, because it has been repeatedly given by this court; and with the decisions heretofore rendered on this point, no member of the bench has, even for an instant, been dissatisfied." 6 I do not suggest that, if the state law is valid, a federal court lacks jurisdiction to enter an injunction restraining state officials from enforcing the statute. Such an injunction may be appropriate—and would be binding on the parties to permit the federal court to preserve its jurisdiction pending final decision on the constitutionality of the statute. United States v. Mine Workers, 330 U.S. 258, 289-290, 67 S.Ct. 677, 693-94, 91 L.Ed. 884. "Although only temporary, the injunction does prohibit state and local enforcement activities against the federal plaintiff pending final resolution of his case in the federal court." Doran v. Salem Inn, Inc., 422 U.S. 922, 931, 95 S.Ct. 2561, 2567, 45 L.Ed.2d 648. Such an injunction does not continue to be binding on the parties, however, if it is vacated on appeal; "an order issued by a court with jurisdiction over the subject matter and person must be obeyed by the parties until it is reversed by orderly and proper proceedings." United States v. Mine Workers, supra, 330 U.S., at 293, 67 S.Ct., at 695 (emphasis added). 7 A conflict between a federal rule of law and a state statute may nullify the state law. Although such invalidity may not be recognized or accepted until it is identified in litigation, in my opinion the conflict with a paramount rule of federal law nullifies a state law whether or not litigation is ever commenced. In other words, it is federal rules of law—and not the actions of federal judges—that may render a state law invalid. 8 Justice REHNQUIST concludes that this case is moot because the injunction restrains an enforcement proceeding that has not yet begun. If his view were accepted, an injunction against a threatened criminal proceeding, see Dombrowski v. Pfister, 380 U.S. 479, 85 S.Ct. 1116, 14 L.Ed.2d 22, would never be appropriate, for the controversy between the parties would not yet be "ripe." MITE sought an injunction not only to prevent the Illinois Secretary of State from interfering with its attempted takeover of Chicago Rivet, but also to bar the Secretary from proceeding against MITE for actions taken in violation of the statute. What is critical to the mootness question in this case is not that MITE abandoned the takeover before it was completed, but that MITE engaged in conduct that violated the terms of the Illinois statute. The extent of MITE's violation of state law cannot be determinative of its interest in avoiding an enforcement proceeding based on what MITE believed was constitutionally protected activity. Oil Workers v. Missouri, 361 U.S. 363, 80 S.Ct. 391, 4 L.Ed.2d 373, relied on by Justice REHNQUIST, does not compel a contrary result. In that case, the party subject to the injunction terminated the activity that had been enjoined. As a result, this Court refused to consider whether the injunction had issued properly, even though a resolution of that question would also have resolved other matters—based on similar questions of law pending in another proceeding between the same parties. In this case, the party subject to the injunction—the Illinois Secretary of State—has not abandoned his desire to do what the injunction currently restrains him from doing. 1 Unless the federal courts can grant preliminary injunctions that provide permanent protection, challenges to questionable state statutes may be deterred. A state statute may be either repugnant to the Constitution, or preempted by some federal law. Parties who wish to engage in conduct proscribed by state statutes may be reluctant to challenge their validity unless they can obtain permanent immunity from penalties. But there is a strong federal interest in encouraging such challenges: the Constitution itself provides that the Constitution and federal statutes shall be "the supreme Law of the Land." Grants of permanent immunity help ensure that federal law will remain paramount. Holding that federal courts have power to grant permanent protection would not substantially limit state power. In fact, the impact on state power will be relatively insignificant. A federal court may grant a preliminary injunction prohibiting the enforcement of a state statute only when there is substantial doubt about the validity of the statute, and when the party seeking relief is able to show that he will suffer irreparable injury if an injunction is not granted. It is true that under the rule I propose, if the statute is later determined to be valid, the State will never be able to prosecute the individual that obtained the preliminary injunction for action taken while the injunction was in effect. However, the State will be free to prosecute him for actions occurring either before or after the injunction, and will also be able to prosecute other persons who violated the statute. In other words, the State will be barred only from prosecuting the particular individual who requested the injunction for conduct undertaken during the pendency of the injunction. Moreover, it will be barred from prosecuting that individual, only because there was serious doubt about the constitutionality of the statute, and because he was able to show that he would suffer irreparable injury if an injunction was not granted. 2 It might be argued that because a party seeking a preliminary injunction must ordinarily post bond, there should be a presumption in favor of recovery of damages caused by a wrongfully issued preliminary injunction. However, the fact that an injunction bond is ordinarily required does not necessarily imply that the party against whom the injunction was issued is automatically entitled to damages. That party must still prove that damages are appropriate; the injunction bond merely provides security, when the party is able to make such a showing. It is true that when an injunction bond has been posted, and when the party challenging the injunction has a right to recover damages on the bond, the question whether an injunction was properly issued is not moot. See Liner v. Jafco, Inc., 375 U.S. 301, 84 S.Ct. 391, 11 L.Ed.2d 347 (1964). The District Court record does not reveal that a bond was posted in this case. Even if a bond had been posted, however, this case would probably be moot; I believe that the State would not have a cause of action for damages. If this Court had determined that the injunction was wrongfully entered, the State might argue that it was damaged because it was unable to recover penalties for violations of the Take-Over Act that occurred during the period the preliminary injunction was in effect. Such an argument should not prevail. Lost penalties do not constitute the sort of damages recoverable on a bond. In any event, as I suggest in this dissent, I believe that the preliminary injunction should be interpreted as protecting MITE from penalties. Thus, it should also protect MITE from liability for "damages" sustained by the State because it could not bring an action for penalties. If a bond had been posted, the State might be able to recover costs or nominal damages on the bond. However, where there is no other basis for challenging the validity of an injunction, the possibility of such recovery is not sufficient to keep a case alive. If it were, then almost no case challenging an injunction could become moot. See Washington Market Co. v. District of Columbia, 137 U.S. 62, 11 S.Ct. 4, 34 L.Ed. 572 (1890) (court costs); Hernandez v. European Auto Collision, Inc., 487 F.2d 378, 387 (CA2 1973) (nominal damages); Kerrigan v. Boucher, 450 F.2d 487 (CA2 1971) (nominal damages). 3 Cf. Steffel v. Thompson, 415 U.S. 452, 462, 94 S.Ct. 1209, 1217, 39 L.Ed.2d 505 (1974) (federal-court intervention is appropriate where the applicant for relief is situated "between the Scylla of intentionally flouting state law and the Charybdis of forgoing what he believes to be constitutionally protected activity in order to avoid becoming enmeshed in a criminal proceeding"). See also Hygrade Provision Co. v. Sherman, 266 U.S. 497, 500, 45 S.Ct. 141, 142, 69 L.Ed. 402 (1925); Terrace v. Thompson, 263 U.S. 197, 216, 44 S.Ct. 15, 18, 68 L.Ed. 255 (1923); Salem Inn, Inc. v. Frank, 501 F.2d 18, 21 (CA2 1974), aff'd in relevant part sub nom. Doran v. Salem Inn, Inc., 422 U.S. 922, 95 S.Ct. 2561, 45 L.Ed.2d 648 (1975). 4 I also find it significant that the District Court's final order granting a permanent injunction declares that the Illinois Act is "null and void and of no force and effect." App. to Juris. Statement 41a. A reasonable construction of the order granting a preliminary injunction is that it was also intended to render the act "null and void" while the injunction was in effect. 5 It is relevant to note that although MITE sought injunctive relief prior to engaging in any action that could subject it to civil or criminal penalties, the State never sought a stay of the District Court's injunction either in that court or in the Court of Appeals, and never expressed an intent to do so. 6 In Marks, a conviction for transporting obscene materials was overturned, where the materials were not obscene at the time of transportation, but were rendered obscene at the time of trial by an intervening decision of this Court. See also Cox v. Louisiana, 379 U.S. 559, 569-571, 85 S.Ct. 476, 483-484, 13 L.Ed.2d 487 (1965) (conviction for illegal picketing reversed where defendant had relied on permission from police officer); Raley v. Ohio, 360 U.S. 423, 437-439, 79 S.Ct. 1257, 1265-1267, 3 L.Ed.2d 1344 (1959) (conviction for refusal to testify before state commission reversed because witness had relied on opinion of commission chairman that he was privileged to remain silent); United States v. Mancuso, 139 F.2d 90 (CA3 1943) (defendant could not be held liable for ignoring induction notices issued while ex parte order staying induction was in effect). 7 The Court did not reach the question whether the Idaho statute was unconstitutional. It concluded that the action should have been dismissed on grounds of improper venue. 8 See Idaho Code §§ 30-1502 to 30-1504, 30-1510 (1980). 9 See Idaho Code § 30-1509 (1980) (allowing State to institute action for rescission). The Illinois Act also empowers the State to seek a court order rescinding sales that are unlawful under the Act. Ill.Rev.Stat., ch. 1211/2, ¶ 137.62 (1979). 10 It is true that a rescission action would have been predicated on acts that were taken under cover of the preliminary injunction. However, I believe that injunctions should ordinarily be interpreted only as providing permanent protection from penalties. The State should be barred from penalizing the offeror for acts that took place during the period the injunction was in effect. However, if a court determines that the state statute is valid, the State should be free to provide a remedy for the continuing effects of acts that violated the statute. In particular, a State should be permitted to dismantle a successful acquisition that violated a valid statute. 11 The possibility that MITE will breach its agreement does not bring this case within the "capable of repetition, yet evading review" exception. The likelihood that such a breach will occur is relatively small. The exception applies only when there is a reasonable expectation that the same action will occur in the future. 1 This case is unlike those in which this Court has found justiciable an action to enjoin a threatened criminal prosecution. The plaintiff in the present posture of this case no longer intends to engage in, or is presently engaging in, what is asserted to be federally protected activity. 2 Bus Employees v. Missouri, 374 U.S. 74, 83 S.Ct. 1657, 10 L.Ed.2d 763 (1963), and Super Tire Engineering Co. v. McCorkle, 416 U.S. 115, 94 S.Ct. 1694, 40 L.Ed.2d 1 (1974), are clearly distinguishable. In each case, subsequent developments did not moot the controversy because the challenged statute affected the challenging party's current or planned activities. There is no suggestion in the instant case that the Illinois Business Take-Over Act has such an effect on MITE. Nor do I believe that this case remains alive merely because it is the enjoined party who seeks appellate review. Otherwise, an enjoined party could always litigate the legal bases for the injunction even though the party who sought the injunction no longer needs the injunction for the purposes for which it was obtained. Cf. University of Texas v. Camenisch, 451 U.S. 390, 101 S.Ct. 1830, 68 L.Ed.2d 175 (1981).

78

457 U.S. 687 102 S.Ct. 2664 73 L.Ed.2d 314 Omar TAYLOR, Petitionerv.ALABAMA. No. 81-5152. Argued March 23, 1982. Decided June 23, 1982. Syllabus Petitioner was arrested on a grocery-store robbery charge without a warrant or probable cause, based on an uncorroborated informant's tip, and was taken to the police station, where he was given Miranda warnings, fingerprinted, questioned, and placed in a lineup. After being told that his fingerprints matched those on grocery items handled by one of the participants in the robbery and after a short visit with his girlfriend, petitioner signed a written confession. Over petitioner's objection, the confession was admitted into evidence at his trial in an Alabama state court, and he was convicted. The Alabama Court of Criminal Appeals reversed, holding that the confession should not have been admitted, but was in turn reversed by the Alabama Supreme Court. Held : Petitioner's confession should have been suppressed as the fruit of an illegal arrest. Brown v. Illinois, 422 U.S. 590, 95 S.Ct. 2254, 45 L.Ed.2d 416; Dunaway v. New York, 442 U.S. 200, 99 S.Ct. 2248, 60 L.Ed.2d 824. Pp. 689-694. (a) A confession obtained through custodial interrogation after an illegal arrest should be excluded unless intervening events break the causal connection between the arrest and the confession so that the confession is sufficiently an act of free will to purge the primary taint. Pp. 689-690. (b) Here, there was no meaningful intervening event. The illegality of the initial arrest was not cured by the facts that six hours elapsed between the arrest and confession; that the confession may have been "voluntary" for Fifth Amendment purposes because Miranda warnings were given; that petitioner was permitted a short visit with his girlfriend; or that the police did not physically abuse petitioner. Nor was the fact that an arrest warrant, based on a comparison of fingerprints, was filed after petitioner had been arrested and while he was being interrogated a significant intervening event, such warrant being irrelevant to whether the confession was the fruit of an illegal arrest. The initial fingerprints, which were themselves the fruit of the illegal arrest and were used to extract the confession, cannot be deemed sufficient "attenuation" to break the connection between the illegal arrest and the confession merely because they formed the basis for the arrest warrant. Pp. 690-693 399 So.2d 881 (Ala.1981), reversed and remanded. Robert M. Beno, Montgomery, Ala., for petitioner. Thomas R. Allison, Montgomery, Ala., for respondent. Justice MARSHALL delivered the opinion of the Court. 1 This case presents the narrow question whether petitioner's confession should have been suppressed as the fruit of an illegal arrest. The Supreme Court of Alabama held that the evidence was properly admitted. Because the decision below is inconsistent with our decisions in Dunaway v. New York, 442 U.S. 200, 99 S.Ct. 2248, 60 L.Ed.2d 824 (1979), and Brown v. Illinois, 422 U.S. 590, 95 S.Ct. 2254, 45 L.Ed.2d 416 (1975), we reverse. 2 * In 1978, a grocery store in Montgomery, Ala., was robbed. There had been a number of robberies in this area, and the police had initiated an intensive manhunt in an effort to apprehend the robbers. An individual who was at that time incarcerated on unrelated charges told a police officer that "he had heard that [petitioner] Omar Taylor was involved in the robbery." App. 4. This individual had never before given similar information to this officer, did not tell the officer where he had heard this information, and did not provide any details of the crime. This tip was insufficient to give the police probable cause to obtain a warrant or to arrest petitioner. 3 Nonetheless, on the basis of this information, two officers arrested petitioner without a warrant. They told petitioner that he was being arrested in connection with the grocery-store robbery, searched him, and took him to the station for questioning. Petitioner was given the warnings required by Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966). At the station, he was fingerprinted, readvised of his Miranda rights, questioned, and placed in a lineup. The victims of the robbery were unable to identify him in the lineup. The police told petitioner that his fingerprints matched those on some grocery items that had been handled by one of the participants in the robbery. After a short visit with his girlfriend and a male companion, petitioner signed a waiver-of-rights form and executed a written confession. The form and the signed confession were admitted into evidence. 4 Petitioner objected to the admission of this evidence at his trial. He argued that his warrantless arrest was not supported by probable cause, that he had been involuntarily transported to the police station, and that the confession must be suppressed as the fruit of this illegal arrest. The trial court overruled this objection, and petitioner was convicted. On appeal, the Alabama Court of Criminal Appeals reversed, 399 So.2d 875 (1980), holding that the facts of this case are virtually indistinguishable from those presented to this Court in Dunaway v. New York, supra, and that the confession should not have been admitted into evidence. The Alabama Supreme Court reversed the Court of Criminal Appeals, 399 So.2d 881 (1981), and we granted certiorari, 454 U.S. 963, 102 S.Ct. 502, 70 L.Ed.2d 378 (1981). II 5 In Brown v. Illinois, supra, and Dunaway v. New York, supra, the police arrested suspects without probable cause. The suspects were transported to police headquarters, advised of their Miranda rights, and interrogated. They confessed within two hours of their arrest. This Court held that the confessions were not admissible at trial, reasoning that a confession obtained through custodial interrogation after an illegal arrest should be excluded unless intervening events break the causal connection between the illegal arrest and the confession so that the confession is " 'sufficiently an act of free will to purge the primary taint.' " Brown v. Illinois, supra, at 602, 95 S.Ct., at 2261 (quoting Wong Sun v. United States, 371 U.S. 471, 486, 83 S.Ct. 407, 416, 9 L.Ed.2d 441 (1963)). See also Dunaway v. New York, supra, at 217, 99 S.Ct., at 2259. This Court identified several factors that should be considered in determining whether a confession has been purged of the taint of the illegal arrest: "[t]he temporal proximity of the arrest and the confession, the presence of intervening circumstances, . . . and, particularly, the purpose and flagrancy of the official misconduct." Brown v. Illinois, supra, at 603-604, 95 S.Ct., at 2261 (citations and footnote omitted); Dunaway v. New York, 442 U.S., at 218, 99 S.Ct., at 2259. The State bears the burden of proving that a confession is admissible. Ibid. 6 In Brown and Dunaway, this Court firmly established that the fact that the confession may be "voluntary" for purposes of the Fifth Amendment, in the sense that Miranda warnings were given and understood, is not by itself sufficient to purge the taint of the illegal arrest. In this situation, a finding of "voluntariness" for purposes of the Fifth Amendment is merely a threshold requirement for Fourth Amendment analysis. See Dunaway v. New York, supra, at 217, 99 S.Ct., at 2259. The reason for this approach is clear: "[t]he exclusionary rule, . . . when utilized to effectuate the Fourth Amendment, serves interests and policies that are distinct from those it serves under the Fifth" Amendment. Brown v. Illinois, 422 U.S., at 601, 95 S.Ct., at 2260. If Miranda warnings were viewed as a talisman that cured all Fourth Amendment violations, then the constitutional guarantee against unlawful searches and seizures would be reduced to a mere " 'form of words.' " Id., at 603, 95 S.Ct., at 2261 (quoting Mapp v. Ohio, 367 U.S. 643, 648, 81 S.Ct. 1684, 1687, 6 L.Ed.2d 1081 (1961)). 7 This case is a virtual replica of both Brown and Dunaway. Petitioner was arrested without probable cause in the hope that something would turn up, and he confessed shortly thereafter without any meaningful intervening event. The State's arguments to the contrary are unpersuasive. The State begins by focusing on the temporal proximity of the arrest and the confession. It observes that the length of time between the illegal arrest and the confession was six hours in this case, while in Brown and Dunaway the incriminating statements were obtained within two hours. However, a difference of a few hours is not significant where, as here, petitioner was in police custody, unrepresented by counsel, and he was questioned on several occasions, fingerprinted, and subjected to a lineup. The State has not even demonstrated the amount of this time that was spent in interrogation, arguing only that petitioner "had every opportunity to consider his situation, to organize his thoughts, to contemplate his constitutional rights, and to exercise his free will." Brief for Respondent 11. 8 The State points to several intervening events that it argues are sufficient to break the connection between the illegal arrest and petitioner's confession. It observes that petitioner was given Miranda warnings three times. As our foregoing discussion of Brown and Dunaway demonstrates, however, the State's reliance on the giving of Miranda warnings is misplaced. The State also observes that petitioner visited with his girlfriend and a male companion before he confessed. This claim fares no better. According to the officer and petitioner, these two visitors were outside the interrogation room where petitioner was being questioned. After petitioner signed a waiver-of-rights form, he was allowed to meet with these visitors. The State fails to explain how this 5- to 10-minute visit, after which petitioner immediately recanted his former statements that he knew nothing about the robbery and signed the confession, could possibly have contributed to his ability to consider carefully and objectively his options and to exercise his free will. This suggestion is particularly dubious in light of petitioner's uncontroverted testimony that his girlfriend was emotionally upset at the time of this visit.1 If any inference could be drawn, it would be that this visit had just the opposite effect. 9 The State points to an arrest warrant filed after petitioner had been arrested and while he was being interrogated as another significant "intervening event." While petitioner was in custody, the police determined that the fingerprints on some grocery items matched those that they had taken from petitioner immediately after his arrest. Based on this comparison, an arrest warrant was filed. The filing of this warrant, however, is irrelevant to whether the confession was the fruit of the illegal arrest. This case is not like Johnson v. Louisiana, 406 U.S. 356, 92 S.Ct. 1620, 32 L.Ed.2d 152 (1972), where the defendant was brought before a committing Magistrate who advised him of his rights and set bail. Here, the arrest warrant was filed ex parte, based on the comparison of the fingerprints found at the scene of the crime and petitioner's fingerprints, which had been taken immediately after his arrest. The initial fingerprints, which were themselves the fruit of petitioner's illegal arrest, see Davis v. Mississippi, 394 U.S. 721, 89 S.Ct. 1394, 22 L.Ed.2d 676 (1969), and which were used to extract the confession from petitioner, cannot be deemed sufficient "attenuation" to break the connection between the illegal arrest and the confession merely because they also formed the basis for an arrest warrant that was filed while petitioner was being interrogated.2 10 Finally, the State argues that the police conduct here was not flagrant or purposeful, and that we should not follow our decisions in Brown and Dunaway for that reason. However, we fail to see any relevant distinction between the conduct here and that in Dunaway. In this case, as in Dunaway, the police effectuated an investigatory arrest without probable cause, based on an uncorroborated informant's tip, and involuntarily transported petitioner to the station for interrogation in the hope that something would turn up. The fact that the police did not physically abuse petitioner, or that the confession they obtained may have been "voluntary" for purposes of the Fifth Amendment, does not cure the illegality of the initial arrest. Alternatively, the State contends that the police conduct here argues for adopting a "good faith" exception to the exclusionary rule. To date, we have not recognized such an exception, and we decline to do so here. III 11 In sum, petitioner's confession was the fruit of his illegal arrest. Under our decisions in Brown v. Illinois and Dunaway v. New York, the confession clearly should not have been admitted at his trial. Accordingly, we reverse the decision of the Alabama Supreme Court and remand this case for further proceedings not inconsistent with this opinion. 12 It is so ordered. 13 Justice O'CONNOR, with whom THE CHIEF JUSTICE, Justice POWELL, and Justice REHNQUIST join, dissenting. 14 The Court holds today that Omar Taylor's detailed confession was the fruit of an illegal arrest, and consequently, should be suppressed. Because I conclude that neither the facts nor the law supports the Court's analysis, I respectfully dissent. 15 * In the course of their investigation of the Moseley robbery, Montgomery police questioned Charles Martin, who was being held on unrelated rape and robbery charges. Martin stated that "he had heard that Omar Taylor was involved in the robbery of Moseley's Grocery," Tr. 6, but the police made no attempt to establish either Martin's credibility as an informant or the reliability of the information he provided.1 16 Based only on this tip, which did not provide probable cause, Sergeants Alford and Rutland arrested Taylor a little before 3 p. m. on January 4, 1979. At that time, they told him why he was being arrested and advised him of his Miranda rights, but asked him no questions regarding the robbery. Tr. 20, 24. When they arrived at the police station, the officers turned Taylor over to detectives. 17 After Taylor had been fingerprinted and signed a form acknowledging his Miranda rights, Detective Wilson questioned him for about 15 minutes, Tr. 48, and placed him in a lineup before one of the victims, Mrs. Moseley. Id., at 37-38. At the lineup, which lasted about an hour, id., at 48, Mrs. Moseley was unable to identify the petitioner. Following the lineup, Detective Wilson told Taylor that his fingerprints matched the fingerprints removed from grocery items handled by one of the robbers. Nevertheless, the petitioner denied knowledge of the robbery. 18 Toward 9 p. m. that evening, Detective Hicks readvised Taylor of his Miranda rights, Tr. 25, and Taylor once again read and signed a form setting forth his Miranda rights. Tr. 28, 125. At no time did Taylor ask for a lawyer or indicate that he did not want to talk to police. Id., at 28-29, 35, 40. During his 5- to 10-minute interview with Taylor, Detective Hicks confronted him with the fingerprint evidence. Id., at 36. Hicks urged the petitioner to cooperate with the police, but carefully refrained from making him any promises, stating that at most he could inform the judge of the petitioner's cooperation. Id., at 31, 34. Taylor continued to deny involvement in the robbery. Id., at 35-36. 19 Following this conversation, both the petitioner's girlfriend and his neighbor came to the police station and requested to speak with him. When Taylor indicated that he wanted to speak with his friends, Detective Hicks left them alone in his office for several minutes.2 After that meeting, the petitioner confessed to the crime, and signed a detailed written confession.3 20 Before trial, the petitioner moved to suppress his confession, arguing that it was the product of an illegal arrest, and that it had been obtained in violation of his Fifth and Sixth Amendment rights. The trial judge assumed that the arrest was illegal,4 but found that the confession was voluntary, consistent with the Fifth and Sixth Amendments, and that "there were enough intervening factors between the arrest and confession" to overcome the taint of the illegal arrest. Id., at 116. Accordingly, he admitted the confession. II 21 Although the Court misapprehends the facts of the present case, it has stated correctly the controlling substantive law. In the Court's words, "a confession obtained through custodial interrogation after an illegal arrest should be excluded unless intervening events break the causal connection between the illegal arrest and the confession so that the confession is 'sufficiently an act of free will to purge the primary taint.' " Ante, at 690 (quoting Brown v. Illinois, 422 U.S. 590, 602, 95 S.Ct. 2254, 2261, 45 L.Ed.2d 416 (1975)). 22 In Brown, this Court emphasized that "Miranda warnings are an important factor . . . in determining whether the confession [was] obtained by exploitation of an illegal arrest." Id., at 603, 95 S.Ct., at 226.5 The Court did not discount the significance of other factors, however, noting that "Miranda warnings, alone and per se, cannot always make the act sufficiently a product of free will to break, for Fourth Amendment purposes, the causal connection between the illegality and the confession." Ibid. Brown holds, therefore, that not only Miranda warnings, but also "[t]he temporal proximity of the arrest and the confession, the presence of intervening circumstances, and, particularly, the purpose and flagrancy of the official misconduct are all relevant." Id., at 603-604, 95 S.Ct., at 2261 (footnotes and citations omitted). 23 In light of those factors, the Brown Court reviewed the record and found that "Brown's first statement was separated from his illegal arrest by less than two hours, and [that] there was no intervening event of significance whatsoever." Id., at 604, 95 S.Ct., at 2262. Moreover, the police conduct in arresting Brown was particularly egregious. The "impropriety of the arrest was obvious," and the "manner in which Brown's arrest was effected gives the appearance of having been calculated to cause surprise, fright, and confusion." Id., at 605, 95 S.Ct., at 2262. The Court held that as a consequence the confession should have been suppressed. 24 Four Terms later, in Dunaway v. New York, 442 U.S. 200, 204, 99 S.Ct. 2248, 2252, 60 L.Ed.2d 824 (1979), this Court reaffirmed the Brown rule that in order to use at trial statements obtained following an arrest on less than probable cause 25 "the prosecution must show not only that the statements meet the Fifth Amendment voluntariness standard, but also that the causal connection between the statements and the illegal arrest is broken sufficiently to purge the primary taint of the illegal arrest." 26 Finding the facts in Dunaway to be "virtually a replica of the situation in Brown," id., at 218, 99 S.Ct., at 2259, the Court held that the petitioner's confession should have been suppressed. Critical to the Court's holding was its observation that the petitioner "confessed without any intervening event of significance." Ibid. See id., at 219, 99 S.Ct., at 2260 ("No intervening events broke the connection between petitioner's illegal detention and his confession"). III 27 Our task is to apply the law as articulated in Brown and Dunaway to the facts of this case. 28 The first significant consideration is that following his unlawful arrest, Taylor was warned on three separate occasions that he 29 "had a right to remain silent, [and] anything he said could be used against him in a court of law[;] he had the right to have an attorney present, [and] if he could not afford one, the State would appoint one for him[;] he could answer questions but he could stop answering at any time." Tr. 23. 30 Under Brown and Dunaway, these warnings must be counted as "an important factor . . . in determining whether the confession [was] obtained by exploitation of an illegal arrest," Brown v. Illinois, supra, at 603, 95 S.Ct., at 2261, though they are, standing alone, insufficient to prove that the primary taint of an illegal arrest had been purged. 31 Second, in contrast to the facts in Brown, the facts in the present case show that the petitioner was not subjected to intimidating police misconduct. In Brown, police had broken into the petitioner's house and searched it. When the petitioner later came home, two officers pointed their guns at him and arrested him, leading the Court to conclude that "[t]he manner in which [the petitioner's] arrest was effected gives the appearance of having been calculated to cause surprise, fright, and confusion." 422 U.S., at 605, 95 S.Ct., at 2262. By contrast, nothing in the record before us indicates that the petitioner's arrest was violent, or designed to "cause surprise, fright, and confusion." Instead, Montgomery officers approached Taylor, asked him his name, and told him that he was under arrest for the Moseley robbery. They then searched him, advised him of his rights, and took him to the police station. 32 Third, while in both Brown and Dunaway there was "no intervening event of significance whatsoever," 422 U.S., at 604, 95 S.Ct., at 2262, in the present case Taylor's girlfriend and neighbor came to the police station and asked to speak with him. Before meeting with his two friends, the petitioner steadfastly had denied involvement in the Moseley robbery. Immediately following the meeting, the petitioner gave a complete and detailed confession of his participation in the armed robbery. This meeting between the petitioner and his two friends, as described by the police in their testimony at the suppression hearing, plainly constituted an intervening circumstance. 33 Finally, the record reveals that the petitioner spent most of the time between his arrest and confession by himself.6 In Dunaway and Brown, by contrast, the defendants were interrogated continuously before they made incriminating statements. 34 In sum, when these four factors are considered together,7 it is obvious that there is no sufficient basis on which to overturn the trial court's finding that "there were enough intervening factors" to overcome the taint of the illegal arrest. In fact, I believe it is clear that the State carried its burden of proof. The petitioner was warned of his rights to remain silent and to have a lawyer present, and there is no dispute that he understood those rights or that he waived them voluntarily and without coercion. After receiving three sets of such warnings, he met with his girlfriend and neighbor, at his request. Following that meeting, at which no police officers were present, the petitioner decided to confess to his participation in the robbery. The petitioner's confession was not proximately caused by his illegal arrest, but was the product of a decision based both on knowledge of his constitutional rights and on the discussion with his friends. Accordingly, I respectfully dissent. 1 According to petitioner, his girlfriend became upset upon hearing the officer advise petitioner to cooperate. App. 16. Contrary to the allegations in the dissent, at no point did the officer contradict petitioner's version of his girlfriend's emotional state or petitioner's statement that his girlfriend was present at the time the officer advised him to cooperate. In fact, the testimony from both petitioner and the officer with respect to this visit are consistent. The officer testified only that he advised petitioner to cooperate between the time petitioner signed a rights form at the commencement of this interrogation period and the time that petitioner signed the statement of confession. Tr. 31, 136-137. He also testified that during this same interval, he allowed the short visit between petitioner and his girlfriend. Ibid. The District Court made no findings of fact with respect to these incidents. In any event, even assuming the accuracy of the dissent's version of the facts, compare post, at 695, and n. 2, with Tr. 31, 136-137, the dissent offers no explanation for its conclusion that this 5- to 10-minute visit should be viewed as an intervening event that purges the taint of the illegal arrest. 2 Petitioner also raises an ambiguous objection to the admission of fingerprint evidence at his trial. The trial court granted petitioner's motion to suppress the initial fingerprints as the fruit of his illegal arrest under Davis v. Mississippi, 394 U.S. 721, 89 S.Ct. 1394, 22 L.Ed.2d 676 (1969), and granted the State's motion to take petitioner's fingerprints at trial. The nature of petitioner's objection to the admission of any fingerprint evidence at trial is unclear, and it is also uncertain whether an objection to the procedure used for taking the second set of fingerprints has been properly preserved for our review. In any event, we need not reach this issue because we reverse the decision on the ground that the confession should not have been admitted. To the extent that petitioner still may challenge the fingerprinting procedure employed below, the state courts should be given the opportunity to address this challenge in the first instance. 1 The police, however, suspected Martin of complicity in the Moseley robbery, Tr. 15. It later developed that Martin had instigated, planned, and participated in the robbery. 2 The Court's rather different account of this meeting apparently stems from a decision to accept the testimony most favorable to the holding it wants to reach. That decision, however, runs counter to the longstanding practice of federal appellate courts to uphold the denial of the motion to suppress if, in the absence of any express findings by the district court, there is any reasonable view of the evidence to support it. See United States v. Payton, 615 F.2d 922, 923 (CA1), cert. denied, 446 U.S. 969, 100 S.Ct. 2950, 64 L.Ed.2d 830 (1980); United States v. Vicknair, 610 F.2d 372, 376, n. 4 (CA5), cert. denied, 449 U.S. 823, 101 S.Ct. 83, 66 L.Ed.2d 25 (1980). In the present case, the officer testified that Taylor's "girlfriend came to us and said she wanted to talk to Omar, and we told Omar she was outside and he wanted to talk to her. And at that time, we let him talk to her." Tr. 35. Detective Hicks specifically denied that he had urged Taylor to talk to his girlfriend. Id., at 35, 133-134. The detective acknowledged that he had told the petitioner that he could inform the judge of the petitioner's cooperation, but he expressly denied making any other statements to Taylor or his girlfriend about "cooperation." Id., at 31, 134. The petitioner, of course, had a vastly different version. He testified that the police had brought his girlfriend into the room and told him, in her presence, that he was facing 10 years to life in prison, but that if he cooperated they might be able to arrange a suspended sentence or probation. Upon hearing that remark, the petitioner's girlfriend became upset and began to cry, at which point the police left the petitioner alone with his friends. Id., at 52. As we noted above, the police expressly denied making any such statements. More importantly, upon comparing the two versions, it becomes clear that in an effort to support its holding, the Court has parsed through the petitioner's story and plucked those tidbits that the police did not expressly contradict. This method of setting forth the facts of a case on appellate review hardly comports with the rule that an appellate court must adopt any reasonable view of the evidence that supports the trial court's ruling. Since there is nothing unreasonable about the police account of the meeting between the petitioner and his friends, that version is the one we must accept on review. At the hearing, Detective Hicks testified that after Taylor asked to speak with his friends, the police left them alone together. There is no suggestion, other than the petitioner's discredited version of the meeting, that the police said anything to the petitioner's girlfriend, or that she became upset. Thus, the Court errs in stating that the petitioner's girlfriend became upset because of statements made by the police, and in intimating that the police created a coercive atmosphere in which the petitioner could not carefully consider his options and, on the basis of his friends' advice, decide to confess to the robbery. 3 In that confession, the petitioner stated that Charles Martin approached him with guns and a plan to rob Moseley's Grocery. Taylor's role in the robbery was to distract Mr. Moseley by buying some groceries. Just before his accomplices pulled out their guns, Taylor put down the groceries and walked outside to see whether an approaching car was a police car. When he saw that it was not a police car, he began to reenter the store, but stopped when he saw the robbery taking place. Thereafter he fled, met his cofelons at a preassigned place, and took his share of the money. Id., at 128-132. 4 In fact, the State did not seriously contend that the arrest had been based on probable cause. See id., at 8, 10. 5 The holding in Brown was derived from this Court's seminal decision in Wong Sun v. United States, 371 U.S. 471, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963), in which we rejected a "but for" test for determining whether to suppress evidence gathered following a Fourth Amendment violation. "We need not hold that all evidence is 'fruit of the poisonous tree' simply because it would not have come to light but for the illegal actions of the police. Rather, the more apt question in such a case is 'whether, granting establishment of the primary illegality, the evidence to which instant objection is made has been come at by exploitation of that illegality or instead by means sufficiently distinguishable to be purged of the primary taint.' Maguire, Evidence of Guilt, 221 (1959)." Id., at 487-488, 83 S.Ct., at 417. 6 The petitioner confessed some six hours after his arrest. As Justice STEVENS noted in his concurring opinion in Dunaway, the "temporal relationship between the arrest and the confession may be an ambiguous factor," 442 U.S., at 220, 99 S.Ct., at 2260, for a lengthy detention could be used to exploit an illegal arrest at least as easily as a brief detention. In the present case, there seems to be nothing remarkable, one way or the other, about the length of detention. 7 The Court has taken each circumstance out of context and examined it to see whether it alone would be enough to purge the taint of the illegal arrest. The Court's failure to consider the circumstances of this case as a whole may have contributed to its erroneous conclusion.

01

457 U.S. 702 102 S.Ct. 2672 73 L.Ed.2d 327 JACKSONVILLE BULK TERMINALS, INC. et al., Petitioners,v.INTERNATIONAL LONGSHOREMEN'S ASSOCIATION et al. No. 80-1045. Argued Jan. 18, 1982. Decided June 24, 1982. Syllabus After President Carter announced certain trade restrictions with the Soviet Union because of its intervention in Afghanistan, respondent International Longshoremen's Association announced that its members would not handle any cargo bound to, or coming from, the Soviet Union. When an affiliated local union refused to load certain goods (not included in the Presidential embargo) bound for the Soviet Union, petitioners (hereafter collectively referred to as the Employer) brought suit in Federal District Court against respondents, the international union, its officers and agents, and the local union (hereafter collectively referred to as the Union), pursuant to § 301(a) of the Labor Management Relations Act. The Employer alleged that the Union's work stoppage violated the terms of a collective-bargaining agreement which contained a no-strike clause and a provision requiring arbitration of disputes. As requested by the Employer, the court ordered the Union to arbitrate the question whether the work stoppage violated the collective-bargaining agreement, and granted a preliminary injunction pending arbitration. The court reasoned that the political motivation behind the work stoppage rendered inapplicable § 4 of the Norris-LaGuardia Act, which prohibits injunctions against strikes "in any case involving or growing out of any labor dispute." The Court of Appeals affirmed the District Court's order insofar as it required arbitration, but disagreed with the conclusion that the Norris-LaGuardia Act was not applicable. Held : 1. The Norris-LaGuardia Act applies to this case, which involves a "labor dispute" even though the work stoppage was politically motivated. Pp. 709-720. (a) The plain language of the Act—prohibiting injunctions in "any" labor dispute and defining "labor dispute" to include "any controversy concerning terms or conditions of employment"—does not except labor disputes having their genesis in political protests. Here, the Employer sought injunctive relief as to the dispute over whether the work stoppage violated the no-strike clause of the bargaining agreement, not as to the event that triggered the stoppage. The term "labor dispute" must not be narrowly construed, the critical element in determining whether the Act applies being whether, as here, "the employer-employee relationship [is] the matrix of the controversy." Columbia River Packers Assn. v. Hinton, 315 U.S. 143, 147, 62 S.Ct. 520, 522, 86 L.Ed. 750. The existence of noneconomic motives does not make the Act inapplicable. Pp. 710-715. (b) The legislative history of both the Norris-LaGuardia Act and the 1947 amendments to the National Labor Relations Act indicates that the Norris-LaGuardia Act was intended to apply to politically motivated work stoppages. Pp. 715-719 (c) The Norris-LaGuardia Act's broad prohibitions will not be constricted, except in narrowly defined situations where accommodation of the Act to specific congressional policy is necessary. Pp. 719-720. 2. Nor may the Union's work stoppage here be enjoined, pending arbitration, under the rationale of Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199, and Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022, on the asserted ground that the dispute underlying the stoppage is arbitrable under the collective-bargaining agreement. While Boys Markets recognized an exception to the anti-injunction provisions of the Norris-LaGuardia Act when the employer sought to enforce the union's contractual obligation to arbitrate grievances rather than to strike over them, Buffalo Forge makes it clear that a Boys Markets injunction pending arbitration may not issue unless the dispute underlying the work stoppage is arbitrable. Here the underlying dispute, whether viewed as an expression of the Union's "moral outrage" at Soviet military policy or as an expression of sympathy for the people of Afghanistan, is plainly not arbitrable under the collective-bargaining agreement. Thus the strike may not be enjoined pending the arbitrator's ruling on the legality of the strike under the no-strike clause of the collective-bargaining agreement. Pp. 720-723 5th Cir., 626 F.2d 455, affirmed. Thomas P. Gies, Washington, D. C., for petitioners. Ernest Mathews, Jr., New York City, for respondents. Justice MARSHALL delivered the opinion of the Court. 1 In this case, we consider the power of a federal court to enjoin a politically motivated work stoppage in an action brought by an employer pursuant to § 301(a) of the Labor Management Relations Act (LMRA), 61 Stat. 156, 29 U.S.C. § 185(a), to enforce a union's obligations under a collective-bargaining agreement. We first address whether the broad anti-injunction provisions of the Norris-LaGuardia Act, 47 Stat. 70, 29 U.S.C. § 101 et seq., apply to politically motivated work stoppages. Finding these provisions applicable, we then consider whether the work stoppage may be enjoined under the rationale of Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199 (1970), and Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976), pending an arbitrator's decision on whether the strike violates the collective-bargaining agreement. 2 * On January 4, 1980, President Carter announced that, due to the Soviet Union's intervention in Afghanistan, certain trade with the Soviet Union would be restricted. Superphosphoric acid (SPA), used in agricultural fertilizer, was not included in the Presidential embargo.1 On January 9, 1980, respondent International Longshoremen's Association (ILA) announced that its members would not handle any cargo bound to, or coming from, the Soviet Union or carried on Russian ships.2 In accordance with this resolution, respondent local union, an ILA affiliate, refused to load SPA bound for the Soviet Union aboard three ships that arrived at the shipping terminal operated by petitioner Jacksonville Bulk Terminals, Inc. (JBT), at the Port of Jacksonville, Fla., during the month of January 1980. 3 In response to this work stoppage, petitioners JBT, Hooker Chemical Corp., and Occidental Petroleum Co. (collectively referred to as the Employer)3 brought this action pursuant to § 301(a) of the LMRA, 29 U.S.C. § 185(a), against respondents ILA, its affiliated local union, and its officers and agents (collectively referred to as the Union). The Employer alleged that the Union's work stoppage violated the collective-bargaining agreement between the Union and JBT. The Employer sought to compel arbitration under the agreement, requested a temporary restraining order and a preliminary injunction pending arbitration, and sought damages. 4 The agreement contains both a broad no-strike clause and a provision requiring the resolution of all disputes through a grievance procedure, ending in arbitration.4 The no-strike clause provides: 5 "During the term of this Agreement, . . . the Union agrees there shall not be any strike of any kind or degree whatsoever, . . . for any cause whatsoever; such causes including but not limited to, unfair labor practices by the Employer or violation of this Agreement. The right of employees not to cross a bona fide picket line is recognized by the Employer. . . ." 6 The United States District Court for the Middle District of Florida ordered the Union to process its grievance in accordance with the contractual grievance procedure. The District Court also granted the Employer's request for a preliminary injunction pending arbitration, reasoning that the political motivation behind the work stoppage rendered the Norris-LaGuardia Act's anti-injunction provisions inapplicable. 7 The United States Court of Appeals for the Fifth Circuit affirmed the District Court's order to the extent it required arbitration of the question whether the work stoppage violated the collective-bargaining agreement. New Orleans Steamship Assn. v. General Longshore Workers, 626 F.2d 455 (1980).5 However, the Court of Appeals disagreed with the District Court's conclusion that the provisions of the Norris-LaGuardia Act are inapplicable to politically motivated work stoppages. Relying on Buffalo Forge, the Court of Appeals further held that the Employer was not entitled to an injunction pending arbitration because the underlying dispute was not arbitrable. We granted certiorari, 450 U.S. 1029, 101 S.Ct. 1737, 68 L.Ed.2d 223 (1981), and agree with the Court of Appeals that the provisions of the Norris-LaGuardia Act apply to this case, and that, under Buffalo Forge, an injunction pending arbitration may not issue. II 8 Section 4 of the Norris-LaGuardia Act provides in part: 9 "No court of the United States shall have jurisdiction to issue any restraining order or temporary or permanent injunction in any case involving or growing out of any labor dispute to prohibit any person or persons participating or interested in such dispute . . . from doing, whether singly or in concert, any of the following acts: 10 "(a) Ceasing or refusing to perform any work or to remain in any relation of employment." 47 Stat. 70, 29 U.S.C. § 104. 11 Congress adopted this broad prohibition to remedy the growing tendency of federal courts to enjoin strikes by narrowly construing the Clayton Act's labor exemption from the Sherman Act's prohibition against conspiracies to restrain trade, see 29 U.S.C. § 52. See, e.g., H.R.Rep.No.669, 72d Cong., 1st Sess., 7-8, 10-11 (1932). This Court has consistently given the anti-injunction provisions of the Norris-LaGuardia Act a broad interpretation, recognizing exceptions only in limited situations where necessary to accommodate the Act to specific federal legislation or paramount congressional policy. See, e.g., Boys Markets, Inc. v. Retail Clerks, 398 U.S., at 249-253, 90 S.Ct., at 1591-1593; Railroad Trainmen v. Chicago River & Indiana R. Co., 353 U.S. 30, 39-42, 77 S.Ct. 635, 639-641, 1 L.Ed.2d 622 (1957). 12 The Boys Markets exception, as refined in Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976), is relevant to our decision today. In Boys Markets, this Court re-examined Sinclair Refining Co. v. Atkinson, 370 U.S. 195, 82 S.Ct. 1328, 8 L.Ed.2d 440 (1962), which held that the Norris-LaGuardia Act precludes a federal district court from enjoining a strike in breach of a collective-bargaining agreement, even where that agreement contains provisions for binding arbitration of the grievance concerning which the strike was called. 398 U.S., at 237-238, 90 S.Ct., at 1585. The Court overruled Sinclair and held that, in order to accommodate the anti-injunction provisions of Norris-LaGuardia to the subsequently enacted provisions of § 301(a) and the strong federal policy favoring arbitration, it was essential to recognize an exception to the anti-injunction provisions for cases in which the employer sought to enforce the union's contractual obligation to arbitrate grievances rather than to strike over them. 398 U.S., at 249-253, 90 S.Ct., at 1591-1593.6 13 After Boys Markets, the Courts of Appeals divided on the question whether a strike could be enjoined under the Boys Markets exception to the Norris-LaGuardia Act pending arbitration, when the strike was not over a grievance that the union had agreed to arbitrate.7 In Buffalo Forge, the Court resolved this conflict and held that the Boys Markets exception does not apply when only the question whether the strike violates the no-strike pledge, and not the dispute that precipitated the strike, is arbitrable under the parties' collective-bargaining agreement.8 14 The Employer argues that the Norris-LaGuardia Act does not apply in this case because the political motivation underlying the Union's work stoppage removes this controversy from that Act's definition of a "labor dispute." Alternatively, the Employer argues that this case fits within the exception to that Act recognized in Boys Markets as refined in Buffalo Forge. We review these arguments in turn. III 15 At the outset, we must determine whether this is a "case involving or growing out of any labor dispute" within the meaning of § 4 of the Norris-LaGuardia Act, 29 U.S.C. § 104. Section 13(c) of the Act broadly defines the term "labor dispute" to include "any controversy concerning terms or conditions of employment." 47 Stat. 73, 29 U.S.C. § 113(c).9 16 The Employer argues that the existence of political motives takes this work stoppage controversy outside the broad scope of this definition. This argument, however, has no basis in the plain statutory language of the Norris-LaGuardia Act or in our prior interpretations of that Act. Furthermore, the argument is contradicted by the legislative history of not only the Norris-LaGuardia Act but also the 1947 amendments to the National Labor Relations Act (NLRA). A. 17 An action brought by an employer against the union representing its employees to enforce a no-strike pledge generally involves two controversies. First, there is the "underlying dispute," which is the event or condition that triggers the work stoppage. This dispute may or may not be political, and it may or may not be arbitrable under the parties' collective-bargaining agreement. Second, there is the parties' dispute over whether the no-strike pledge prohibits the work stoppage at issue. This second dispute can always form the basis for federal-court jurisdiction, because § 301(a) gives federal courts jurisdiction over "[s]uits for violation of contracts between an employer and a labor organization." 29 U.S.C. § 185(a). 18 It is beyond cavil that the second form of dispute—whether the collective-bargaining agreement either forbids or permits the union to refuse to perform certain work—is a "controversy concerning the terms or conditions of employment." 29 U.S.C. § 113(c). This § 301 action was brought to resolve just such a controversy. In its complaint, the Employer did not seek to enjoin the intervention of the Soviet Union in Afghanistan, nor did it ask the District Court to decide whether the Union was justified in expressing disapproval of the Soviet Union's actions. Instead, the Employer sought to enjoin the Union's decision not to provide labor, a decision which the Employer believed violated the terms of the collective-bargaining agreement. It is this contract dispute, and not the political dispute, that the arbitrator will resolve, and on which the courts are asked to rule. 19 The language of the Norris-LaGuardia Act does not except labor disputes having their genesis in political protests. Nor is there any basis in the statutory language for the argument that the Act requires that each dispute relevant to the case be a labor dispute. The Act merely requires that the case involve "any" labor dispute. Therefore, the plain terms of § 4(a) and § 13 of the Norris-LaGuardia Act deprive the federal courts of the power to enjoin the Union's work stoppage in this § 301 action, without regard to whether the Union also has a nonlabor dispute with another entity.10 20 The conclusion that this case involves a labor dispute within the meaning of the Norris-LaGuardia Act comports with this Court's consistent interpretation of that Act.11 Our decisions have recognized that the term "labor dispute" must not be narrowly construed because the statutory definition itself is extremely broad and because Congress deliberately included a broad definition to overrule judicial decisions that had unduly restricted the Clayton Act's labor exemption from the antitrust laws. For example, inMarine Cooks & Stewards v. Panama S.S. Co., 362 U.S. 365, 369, 80 S.Ct. 779, 783, 4 L.Ed.2d 797 (1960), the Court observed: 21 "Th[e] Act's language is broad. The language is broad because Congress was intent upon taking the federal courts out of the labor injunction business except in the very limited circumstances left open for federal jurisdiction under the Norris-LaGuardia Act. The history and background that led Congress to take this view have been adverted to in a number of prior opinions of this Court in which we refused to give the Act narrow intepretations that would have restored many labor dispute controversies to the courts" (emphasis added; footnote omitted). 22 The critical element in determining whether the provisions of the Norris-LaGuardia Act apply is whether "the employer-employee relationship [is] the matrix of the controversy." Columbia River Packers Assn., Inc. v. Hinton, 315 U.S. 143, 147, 62 S.Ct. 520, 522, 86 L.Ed. 750 (1942). In this case, the Employer and the Union representing its employees are the disputants, and their dispute concerns the interpretation of the labor contract that defines their relationship.12 Thus, the employer-employee relationship is the matrix of this controversy. 23 Nevertheless, the Employer argues that a "labor dispute" exists only when the Union's action is taken in its own "economic self-interest." The Employer cites Musicians v. Carroll, 391 U.S. 99, 88 S.Ct. 1562, 20 L.Ed.2d 460 (1968), and Columbia River Packers Assn., supra, for this proposition. In these cases, however, the Court addressed the very different question whether the relevant parties were "labor" groups involved in a labor dispute for the purpose of determining whether their actions were exempt from the antitrust laws.13 These cases do not hold that a union's noneconomic motive inevitably takes the dispute out of the Norris-LaGuardia Act, but only that the protections of that Act do not extend to labor organizations when they cease to act as labor groups or when they enter into illegal combinations with nonlabor groups in restraint of trade.14 Here, there is no question that the Union is a labor group, representing its own interests in a dispute with the Employer over the employees' obligation to provide labor. 24 Even in cases where the disputants did not stand in the relationship of employer and employee, this Court has held that the existence of noneconomic motives does not make the Norris-LaGuardia Act inapplicable. For example, in New Negro Alliance v. Sanitary Grocery Co., 303 U.S. 552, 58 S.Ct. 703, 82 L.Ed. 1012 (1938), this Court held that the Norris-LaGuardia Act prohibited an injunction against picketing by members of a civic group, which was aimed at inducing a store to employ Negro employees. In determining that the group and its members were "persons interested in a labor dispute" within the meaning of § 13, the Court found it immaterial that the picketers, who were neither union organizers nor store employees, were not asserting economic interests commonly associated with labor unions—e.g., terms and conditions of employment in the narrower sense of wages, hours, unionization, or betterment of working conditions. Id., at 560, 58 S.Ct., at 706. Although the lower courts found Norris-LaGuardia inapplicable because the picketing was motivated by the group's "political" or "social" goals of improving the position of Negroes generally, and not by the desire to improve specific conditions of employment, this Court reasoned: "The Act does not concern itself with the background or the motives of the dispute." Id., at 561, 58 S.Ct., at 707. B 25 The Employer's argument that the Union's motivation for engaging in a work stoppage determines whether the Norris-LaGuardia Act applies is also contrary to the legislative history of that Act. The Act was enacted in response to federal-court intervention on behalf of employers through the use of injunctive powers against unions and other associations of employees. This intervention had caused the federal judiciary to fall into disrepute among large segments of this Nation's population. See generally S.Rep.No. 163, 72d Cong., 1st Sess. 8, 16-18 (1932); 75 Cong.Rec. 4915 (1932) (remarks of Sen. Wagner). 26 Apart from the procedural unfairness of many labor injunctions, one of the greatest evils associated with them was the use of tort-law doctrines, which often made the lawfulness of a strike depend upon judicial views of social and economic policy. See, e.g., Cox, Current Problems in the Law of Grievance Arbitration, 30 Rocky Mountain L.Rev. 247, 256 (1958). In debating the Act, its supporters repeatedly expressed disapproval of this Court's interpretations of the Clayton Act's labor exemption—interpretations which permitted a federal judge to find the Act inapplicable based on his or her appraisal of the "legitimacy" of the union's objectives.15 See, e.g., 75 Cong.Rec. 4916 (1932) (remarks of Sen. Wagner) (definition of labor dispute expanded to override Duplex Printing Press Co. v. Deering, 254 U.S. 443, 41 S.Ct. 172, 65 L.Ed. 349 (1921) (holding a strike and picketing with the purpose of unionizing a plant not a labor dispute because the objectives were not legitimate and there was no employer-employee relationship between the disputants)); 75 Cong.Rec., at 5487-5488 (remarks of Rep. Celler) (bill brought forth to remedy decisions allowing injunction in Duplex and in Bedford Cut Stone Co. v. Stone Cutters, 274 U.S. 37, 47 S.Ct. 522, 71 L.Ed. 916 (1927) (holding that decision by workers not to work on nonunion goods not a labor dispute)). See also 75 Cong.Rec., at 4686 (remarks of Sen. Hebert) (Committee minority agreed that injunctions should not have issued in Bedford and Duplex). See generally H.R.Rep.No. 669, 72d Cong., 1st Sess., 8, 10-11 (1932). The legislative history is replete with criticisms of the ability of powerful employers to use federal judges as "strike-breaking" agencies; by virtue of their almost unbridled "equitable discretion," federal judges could enter injunctions based on their disapproval of the employees' objectives, or on the theory that these objectives or actions, although lawful if pursued by a single employee, became unlawful when pursued through the "conspiracy" of concerted activity. See,e.g., 75 Cong.Rec., at 4928-4938, 5466-5468, 5478-5481, 5487-5490. 27 Furthermore, the question whether the Norris-LaGuardia Act would apply to politically motivated strikes was brought to the attention of the 72d Congress when it passed the Act. Opponents criticized the definition of "labor dispute" in § 13(c) on the ground that it would cover politically motivated strikes. Representative Beck argued that federal courts should have jurisdiction to enjoin political strikes like those threatened by labor unions in Europe. Id., at 5471-5473 (discussing threatened strike by British unions protesting the cancellation of leases held by Communist Party members, and threatened strikes by Belgian unions protesting a decision to supply military aid to Poland).16 In response, Representative Oliver argued that the federal courts should not have the power to enjoin such strikes. Id., at 5480-5481. Finally, Representative Beck offered an amendment to the Act that would have permitted federal courts to enjoin strikes called for ulterior purposes, including political motives. This amendment was defeated soundly. See id., at 5507. 28 Further support for our conclusion that Congress believed that the Norris-LaGuardia Act applies to work stoppages instituted for political reasons can be found in the legislative history of the 1947 amendments to the NLRA. That history reveals that Congress rejected a proposal to repeal the Norris-LaGuardia Act with respect to one broad category of political strikes.17 The House bill included definitions of various kinds of labor disputes. See H.R. 3020, 80th Cong., 1st Sess., § 2, 1 Legislative History of the LMRA 158, 160 (1947) (Leg.Hist.); H.R.Rep.No. 245, 80th Cong., 1st Sess., 1, 18-19 (1947), 1 Leg.Hist. 292, 309-310. Of relevance here, § 2(13) defined a "sympathy" strike as a strike "called or conducted not by reason of any dispute between the employer and the employees on strike or participating in such concerted interference, but rather by reason of either (A) a dispute involving another employer or other employees of the same employer, or (B) disagreement with some governmental policy." H.R. 3020, § 2(13), 1 Leg.Hist. 168 (emphasis added). Section 12 of the House bill made this kind of strike "unlawful concerted activity," and "it remove[d] the immunities that the present laws confer upon persons who engage in them." H.R.Rep.No. 245, supra, at 23, 1 Leg.Hist. 314. In particular, the Norris-LaGuardia Act would not apply to suits brought by private parties to enjoin such activity, and damages could be recovered. See H.R.Rep.No. 245, supra, at 23-24, 43-44, 1 Leg.Hist. 314-315, 334-335. In explaining these provisions, the House Report stated that strikes "against a policy of national or local government, which the employer cannot change," should be made unlawful, and that "[t]he bill makes inapplicable in such suits the Norris-LaGuardia Act, which heretofore has protected parties to industrial strife from the consequences of their lawlessness." H.R.Rep.No. 245, supra, at 24, 44, 1 Leg.Hist. 315, 335. 29 The Conference Committee accepted the Senate version, which had eliminated these provisions of the House bill.18 The House Manager's statement accompanying the Conference Report explained that its recommendation did not go as far as the House bill, that § 8(b) prohibits jurisdictional strikes and illegal secondary boycotts, and that the Board, not private parties, may petition a district court under § 10(k) or § 10(l ) to enjoin these activities notwithstanding the provisions of the Norris-LaGuardia Act. H.R.Conf.Rep.No.510, 80th Cong., 1st Sess., 36, 42-43, 57, 58-59 (1947), 1 Leg.Hist. 540, 546-547, 561, 562-563, U.S.Code Cong.Serv.1947, p. 1135. In short, Congress declined in 1947 to adopt a broad "political motivation" exception to the Norris-LaGuardia Act for strikes in protest of some governmental policy. Instead, if a strike of this nature takes the form of a secondary boycott prohibited by § 8(b), Congress chose to give the Board, not private parties, the power to petition a federal district court for an injunction. See 29 U.S.C. §§ 160(k), 160(l). Cf. Longshoremen v. Allied International, Inc., 456 U.S. 212, 102 S.Ct. 1656, 72 L.Ed.2d 21 (1982). C 30 This case, brought by the Employer to enforce its collective-bargaining agreement with the Union, involves a "labor dispute" within any common-sense meaning of that term. Were we to ignore this plain interpretation and hold that the political motivation underlying the work stoppage removes this controversy from the prohibitions of the Norris-LaGuardia Act, we would embroil federal judges in the very scrutiny of "legitimate objectives" that Congress intended to prevent when it passed that Act. The applicability not only of § 4, but also of all of the procedural protections embodied in that Act, would turn on a single federal judge's perception of the motivation underlying the concerted activity.19 The Employer's interpretation is simply inconsistent with the need, expressed by Congress when it enacted the Norris-LaGuardia Act, for clear "mileposts for judges to follow." 75 Cong.Rec. 4935 (1932) (remarks of Sen. Bratton). 31 In essence, the Employer asks us to disregard the legislative history of the Act and to distort the definition of a labor dispute in order to reach what it believes to be an "equitable" result. The Employer's real complaint, however, is not with the Union's political objections to the conduct of the Soviet Union, but with what the Employer views as the Union's breach of contract. The Employer's frustration with this alleged breach of contract should not be remedied by characterizing it as other than a labor dispute. We will not adopt by judicial fiat an interpretation that Congress specifically rejected when it enacted the 1947 amendments to the NLRA. See generally n. 17, supra. In the past, we have consistently declined to constrict Norris-LaGuardia's broad prohibitions except in narrowly defined situations where accommodation of that Act to specific congressional policy is necessary. We refuse to deviate from that path today. IV 32 Alternatively, the Employer argues that the Union's work stoppage may be enjoined under the rationale of Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199 (1970), and Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976), because the dispute underlying the work stoppage is arbitrable under the collective-bargaining agreement. In making this argument, the Employer disavows its earlier argument that the underlying dispute is purely political, and asserts that the Union's work stoppage was motivated by a disagreement with the Employer over the management-rights clause in the collective-bargaining agreement. The Solicitor General, in an amicus brief filed on behalf of the United States, agrees with the Employer that the work stoppage may be enjoined pending arbitration. He contends that in addition to the political dispute, disputes concerning both the management-rights clause and the work-conditions clause underlie the work stoppage, and that at least one of these disputes is arguably arbitrable.20 33 We disagree. Buffalo Forge makes it clear that a Boys Markets injunction pending arbitration should not issue unless the dispute underlying the work stoppage is arbitrable. The rationale of Buffalo Forge compels the conclusion that the Union's work stoppage, called to protest the invasion of Afghanistan by the Soviet Union, may not be enjoined pending the arbitrator's decision on whether the work stoppage violates the no-strike clause in the collective-bargaining agreement. The underlying dispute, whether viewed as an expression of the Union's "moral outrage" at Soviet military policy or as an expression of sympathy for the people of Afghanistan, is plainly not arbitrable under the collective-bargaining agreement. 34 The attempts by the Solicitor General and the Employer to characterize the underlying dispute as arbitrable do not withstand analysis. The "underlying" disputes concerning the management-rights clause or the work-conditions clause simply did not trigger the work stoppage. To the contrary, the applicability of these clauses to the dispute, if any, was triggered by the work stoppage itself. Consideration of whether the strike intruded on the management-rights clause or was permitted by the work-conditions clause may inform the arbitrator's ultimate decision on whether the strike violates the no-strike clause. Indeed, the question whether striking over a nonarbitrable issue violates other provisions of the collective-bargaining agreement may itself be an arbitrable dispute. The fact remains, however, that the strike itself was not over an arbitrable dispute and therefore may not be enjoined pending the arbitrator's ruling on the legality of the strike under the collective-bargaining agreement. 35 The weaknesses in the analysis of the Employer and the Solicitor General can perhaps best be demonstrated by applying it to a pure sympathy strike, which clearly cannot be enjoined pending arbitration under the rationale of Buffalo Forge. If this work stoppage were a pure sympathy strike, it could be characterized alternatively as a dispute over the Employer's right to choose to do business with the employer embroiled in a dispute with a sister union, as a dispute over management's right to assign and direct work, or as a dispute over whether requiring the union to handle goods of the employer whose employees are on strike is an unreasonable work condition.21 None of these characterizations, however, alters the fact, essential to the rationale of Buffalo Forge, that the strike was not over an arbitrable issue and therefore did not directly frustrate the arbitration process. 36 The Employer's argument that this work stoppage may be enjoined pending arbitration really reflects a fundamental disagreement with the rationale of Buffalo Forge, and not a belief that this rationale permits an injunction in this case. The Employer apparently disagrees with the Buffalo Forge Court's conclusion that, in agreeing to broad arbitration and no-strike clauses, the parties do not bargain for injunctive relief to restore the status quo pending the arbitrator's decision on the legality of the strike under the collective-bargaining agreement, without regard to what triggered the strike. Instead, they bargain only for specific enforcement of the union's promise to arbitrate the underlying grievance before resorting to a strike. See 428 U.S., at 410-412, 96 S.Ct., at 3148-3149. The Employer also apparently believes that Buffalo Forge frustrates the arbitration process and encourages industrial strife. But see id., at 412, 96 S.Ct., at 3149.22 However, this disagreement with Buffalo Forge only argues for reconsidering that decision.23 It does not justify distorting the rationale of that case beyond recognition in order to reach the result urged by the Employer. V 37 In conclusion, we hold that an employer's § 301 action to enforce the provisions of a collective-bargaining agreement allegedly violated by a union's work stoppage involves a "labor dispute" within the meaning of the Norris-LaGuardia Act, without regard to the motivation underlying the union's decision not to provide labor. Under our decisions in Boys Markets and Buffalo Forge, when the underlying dispute is not arbitrable, the employer may not obtain injunctive relief pending the arbitrator's ruling on the legality of the strike under the collective-bargaining agreement. Accordingly, the decision of the Court of Appeals is 38 Affirmed. 39 Justice O'CONNOR, concurring in the judgment. 40 Based on the legislative history of the Norris-LaGuardia Act, 29 U.S.C. § 101 et seq., and our previous cases interpreting it, e.g., New Negro Alliance v. Sanitary Grocery Co., 303 U.S. 552, 58 S.Ct. 703, 82 L.Ed. 1012 (1938), the Court correctly concludes that this case involves a labor dispute within the meaning of § 4 of the Act, 29 U.S.C. § 104. The Court also correctly determines that under Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976), no injunction may issue pending arbitration because the underlying political dispute is not arbitrable under the collective-bargaining agreement. Unless the Court is willing to overrule Buffalo Forge, the conclusion reached by the Court in this case is inescapable. Therefore, I concur in the judgment. 41 Chief Justice BURGER, with whom Justice POWELL joins, dissenting. 42 * This case in no sense involves or grows out of a labor dispute as that term is defined in § 13(c) of the Norris-LaGuardia Act, 29 U.S.C. § 113(c). See ante, at 709-710, n. 9. Section 13(c) defines a labor dispute as "any controversy concerning terms or conditions of employment. . . ."1 The dispute in this case is a political dispute and has no relation to any controversy concerning terms or conditions of employment. If Congress had intended to bar federal courts from issuing injunctions in political disputes, it could have simply prohibited federal courts from enjoining strikes rather than limiting its prohibition to controversies concerning terms or conditions of employment. Accordingly, I disagree with the Court's conclusion that the Norris-LaGuardia Act bars a federal court from enjoining this politically motivated work stoppage. 43 The International Longshoremen's Association objects to the Soviet Union's invasion of Afghanistan. As a consequence, it announced that it would not handle any cargo bound to, or coming from, the Soviet Union, or any cargo carried on Soviet ships. This case commenced after the union, pursuant to its political position, refused to load superphosphoric acid onto certain ships bound for the Soviet Union. The union has no objection to any terms or conditions of employment; it would have loaded the superphosphoric acid on any non-Soviet ship bound for a destination other than the Soviet Union. No one has suggested that the union's action is actually motivated to obtain concessions concerning employment conditions. The union refused to handle the cargo simply because a foreign country invaded a neighboring country and the union desired to express its opposition to the invasion. Thus the plain meaning of § 13(c) leads to the conclusion that this case does not involve or grow out of a labor dispute because the union members are not seeking to change their terms or conditions of employment. 44 As the Court recognizes, we have held that the test of whether the Norris-LaGuardia Act applies is whether "the employer-employee relationship [is] the matrix of the controversy." Columbia River Packers Assn., Inc. v. Hinton, 315 U.S. 143, 147, 62 S.Ct. 520, 522, 86 L.Ed. 750 (1942); quoted ante, at 712-713. Federal Courts of Appeals have stated that unions are protected by the Norris-LaGuardia Act when they act to advance the economic interests of their members. See, e.g., Brotherhood of Railroad Trainmen v. Atlantic Coast Line R. Co., 362 F.2d 649, 654 (CA5 1966). These cases illustrate the plain meaning of § 13(c)'s definition of labor dispute—the Norris-LaGuardia Act protects union organizational efforts and efforts to improve working conditions. 45 The Court errs gravely in finding that the matrix of this controversy is the union's relationship with the petitioners. The union's dispute with the petitioners merely flows from its decision to demonstrate its opposition to the invasion of Afghanistan. No economic interests of union members are involved; indeed, the union's policy is contrary to its members' economic interests since it reduces the amount of available work.2 Thus, the cases generally explicating the meaning of § 13(c) lend no support to the notion that this case involves a labor dispute. 46 The federal courts have consistently recognized that the Norris-LaGuardia Act does not apply to politically motivated work stoppages concerning subjects over which employers have no control. These courts, in cases which are for all practical purposes indistinguishable from this case—and which often involved the International Longshoremen's Association—properly concluded that the Act only applies to economic disputes.3 This Court has never before held, as it holds here, that the Norris-LaGuardia Act protects strikes resulting from political disputes rather than from labor disputes. Since the meaning of the words of the statute is plain, and since the applicable precedent supports the conclusion that this is not a labor dispute, we ought to conclude that politically motivated strikes are outside the coverage of the Norris-LaGuardia Act.4 47 Finally, the Court argues that a common-sense interpretation of the meaning of the term "labor dispute" supports its conclusion. But the "common-sense" meaning of a term is not controlling when Congress has provided, as it provided in § 13(c), an explicit definition of a labor dispute. "Common sense" and legislative history ought not change the meaning of the unambiguous words of a statute. It is not contended that any act of petitioners to improve the terms or conditions of employment would have persuaded the union to load the ships. Hence there is no labor dispute under the Norris-LaGuardia Act. II 48 This case, together with our recent decision in Longshoremen v. Allied International, Inc., 456 U.S. 212, 102 S.Ct. 1656, 72 L.Ed.2d 21 (1982), illustrates the inherent flaw in the holding in Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976). If the Court cannot give to ordinary words their ordinary meaning and grasp that the dispute in this case is a purely political dispute rather than having any relation to a labor dispute, it should overrule Buffalo Forge. 49 The controversy in Allied International also resulted from the International Longshoremen's Association's protest over the Soviet invasion of Afghanistan. There we held that the union's refusal to unload shipments from the Soviet Union was a secondary boycott prohibited by § 8(b)(4) of the National Labor Relations Act, 29 U.S.C. § 158(b)(4). The union is therefore liable for damages as a result of its refusal to unload the shipments. Yet the Court today holds that the union may not be enjoined from refusing to load cargo onto ships bound for the Soviet Union. 50 This is all the more perplexing because the union entered into an agreement with petitioners which contained an unequivocal no-strike clause: "During the term of this Agreement, . . . the Union agrees there shall not be any strike of any kind or degree whatsoever, . . . for any cause whatsoever." (Emphasis added). Ante, at 706. In Allied International this union was found liable for damages caused to a party with which it had no such agreement. Here, however, despite the existence of the no-strike agreement between petitioners and the union, the Court holds that the union's illegal acts may not be enjoined. 51 To reach this strange result, the Court first decides that this case involves a labor dispute rather than a political dispute, and therefore is within the scope of the NorrisLaGuardia Act. The Court then contradicts itself and concludes that, since the dispute is really a political protest over Soviet aggression, it may not be enjoined under the Buffalo Forge exception to the rule of Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199 (1970), since a federal court cannot resolve the actual dispute. This case, together with Allied International, persuades me that the artificialBuffalo Forge exception should be abolished. Rather than continuing to engage in mechanical and contradictory analyses as to the character of disputes such as this one, we should hold that a federal court may enjoin a strike pending arbitration when the striking union has agreed to a contract with a no-strike clause such as the one agreed to by petitioners and the ILA. That is what we seemed to hold in Boys Markets, and we should not have tinkered with that holding in Buffalo Forge. 52 There is no rational way to reconcile this holding with Allied International. If we must overrule Buffalo Forge to come to a consistent result, we should do so. 53 Justice POWELL, dissenting. 54 The no-strike clause agreed to by the parties in this case could scarcely be more emphatic: "During the term of this Agreement, . . . the Union agrees there shall not be any strike of any kind or degree whatsoever, . . . for any cause whatsoever " (emphasis added). Ante, at 706. Such a clause is one of the most significant provisions in the bargaining agreement. One can fairly assume that the employer gave considerable ground in other areas of the agreement to gain this apparent guarantee that all disagreements would go first to arbitration. Thus, under the plain language of the agreement of the parties, the strike by the respondents should have been enjoined pending arbitration. 55 But in labor law—since this Court's decision in Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 96 S.Ct. 3141, 49 L.Ed.2d 1022 (1976)—plain language agreed to by a union does not bind it. Buffalo Forge is an aberration. It cannot be reconciled with labor law policy of encouraging industrial peace through arbitration. It severely undercuts Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199 (1970). In a word, Buffalo Forge should be overruled. 56 The internal contradictions in today's decision by the Court further illustrate absence of principle in Buffalo Forge's reasoning. The Court argues that now we must divide the dispute in this case into the "underlying" dispute over Soviet policy and the "other" dispute over the scope of the no-strike clause. I consider this method of analysis artificial and unprincipled. On the one hand, the Court must characterize the dispute in this case as a labor dispute—involving the scope of the no-strike clause—to bring the dispute within the scope of the Norris-LaGuardia Act. But on the other hand, Buffalo Forge requires the Court to contradict itself by insisting that the dispute is "really" over Soviet aggression and therefore that the rule of Boy's Market, and the federal policy in support of arbitration, are inapplicable. 57 The Court should not have it both ways. So long as it adheres to the aberrant analysis in Buffalo Forge, I agree with THE CHIEF JUSTICE that the dispute in this case must be viewed as a political dispute outside the scope of the Norris-LaGuardia Act. I therefore join his dissenting opinion. 58 Justice STEVENS, dissenting. 59 For the reasons stated in Part I of THE CHIEF JUSTICE's dissenting opinion in this case, as well as the reasons stated in Part I of my dissenting opinion in Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 415-424, 96 S.Ct. 3141, 3151-3155, 49 L.Ed.2d 1022, I respectfully dissent. 1 On February 25, 1980, the embargo was extended to include SPA along with other products. On April 24, 1981, President Reagan lifted the SPA embargo as part of his decision to remove restrictions on the sale of grain to the Soviets. By telegrams dated April 24, 1981, and June 5, 1981, the International Longshoremen's Association has recommended to its members that they resume handling goods to and from the Soviet Union. Although the work stoppage is no longer in effect, there remains a live controversy over whether the collective-bargaining agreement prohibits politically motivated work stoppages, and the Union may resume such a work stoppage at any time. As a result, this case is not moot. See Buffalo Forge Co. v. Steelworkers, 428 U.S. 397, 403, n. 8, 96 S.Ct. 3141, 3145, n. 8, 49 L.Ed.2d 1022 (1976). 2 The President of the ILA made the following announcement: "In response to overwhelming demands by the rank and file members of the Union, the leadership of the ILA today ordered immediate suspension in handling all Russian ships and all Russian cargoes in ports from Maine to Texas and Puerto Rico where ILA workers are employed. * * * * * "The reason for this action should be apparent in light of international events that have affected relations between the U. S. & Soviet Union. "However, the decision by the Union leadership was made necessary by the demands of the workers. "It is their will to refuse to work Russian vessels and Russian cargoes under present conditions of the world. "People are upset and they refuse to continue the business as usual policy as long as the Russians insist on being international bully boys. It is a decision in which the Union leadership concurs." Brief for Respondents 2, n. 2. 3 JBT is a wholly owned subsidiary of Oxy Chemical Corp., which is a subsidiary of Hooker. Ownership of all these corporations is ultimately vested in Occidental. Hooker Chemical Co. manufactures SPA at a manufacturing facility in Florida. Pursuant to a bilateral trade agreement between Occidental and the Soviet Union, SPA is shipped to the Soviet Union from the JBT facility in Jacksonville. 4 The grievance and arbitration clause provides in relevant part: "Matters under dispute which cannot be promptly settled between the Local and an individual Employer shall . . . be referred . . . to a Port Grievance Committee . . . . In the event this Port Grievance Committee cannot reach an agreement . . . the dispute shall be referred to the Joint Negotiating Committee . . . . * * * * * "A majority decision of this Committee shall be final and binding on both parties and on all Employers signing this Agreement. In the event the Committee is unable to reach a majority decision within 72 hours after meeting to discuss the case, it shall employ a professional arbitrator. . . ." 5 The Union concedes that the question whether the work stoppage violates the no-strike clause is arbitrable. In a consolidated case, the Court of Appeals upheld an injunction issued by the United States District Court for the Eastern District of Louisiana enforcing an arbitrator's decision that the ILA work stoppage violated a collective-bargaining agreement. 626 F.2d, at 469. 6 In Boys Markets, the underlying dispute was clearly subject to the grievance and arbitration procedures of the collective-bargaining agreement, and the strike clearly violated the no-strike clause. 7 See cases cited in Buffalo Forge, 428 U.S., at 404, n. 9, 96 S.Ct., at 3146, n. 9. 8 In Buffalo Forge, the strike at issue was a sympathy strike in support of sister unions negotiating with the employer. The Court reasoned that there was no need to accommodate the policies of the Norris-LaGuardia Act to § 301 and to the federal policy favoring arbitration when a strike is not called over an arbitrable dispute, because such a strike does not directly frustrate the arbitration process by denying or evading the union's promise to arbitrate. 428 U.S., at 407-412, 96 S.Ct., at 3147-3149. 9 Section 13(c) provides: "(c) The term 'labor dispute' includes any controversy concerning terms or conditions of employment, or concerning the association or representation of persons in negotiating, fixing, maintaining, changing, or seeking to arrange terms or conditions of employment, regardless of whether or not the disputants stand in the proximate relation of employer and employee." 10 Of course, there are exceptions to the Act's prohibitions against enjoining work stoppages. See, e.g., Boys Markets, Inc. v. Retail Clerks, 398 U.S. 235, 90 S.Ct. 1583, 26 L.Ed.2d 199 (1970). The employer may obtain an injunction to enforce an arbitrator's decision that the strike violates the collective-bargaining agreement and can recover damages for the violation, pursuant to § 301 of the LMRA, 29 U.S.C. § 185. See, e.g., Buffalo Forge, supra, at 405, 96 S.Ct., at 3146. See also infra, at 718-719, and n. 18, (discussing Board's authority under 29 U.S.C. §§ 160(k), 160(l ), to petition for an injunction upon finding reasonable cause to believe that the strike is an unfair labor practice). 11 The Employer's reliance on Eastex, Inc. v. NLRB, 437 U.S. 556, 98 S.Ct. 2505, 57 L.Ed.2d 428 (1978), to argue that a politically motivated strike is not a labor dispute is misplaced. In Eastex, we addressed whether certain concerted activity was protected under § 7 of the NLRA, 29 U.S.C. § 157, and we recognized that "[t]here may well be types of conduct or speech that are so purely political or so remotely connected to the concerns of employees as employees as to be beyond the protection of [§ 7]." Id., at 570, n. 20, 98 S.Ct., at 2514, n. 20. Although the definition of a "labor dispute" in § 2(9) of the NLRA, 29 U.S.C. § 152(9), is virtually identical to that in § 13(c) of the Norris-LaGuardia Act, 29 U.S.C. § 113(c), and the two provisions have been construed consistently with one another, e.g., United States v. Hutcheson, 312 U.S. 219, 234, n. 4, 61 S.Ct. 463, 467, n. 4, 85 L.Ed. 780 (1941), this similarity does not advance the Employer's argument. Union activity that prompts a "labor dispute" within the meaning of these sections may be protected by § 7, prohibited by § 8(b), 29 U.S.C. § 158(b), or neither protected nor prohibited. The objective of the concerted activity is relevant in determining whether such activity is protected under § 7 or prohibited by § 8(b), but not in determining whether the activity is a "labor dispute" under § 2(9). Moreover, the conclusion that a purely political work stoppage is not protected under § 7 means simply that the employer is not prohibited by § 8(a)(1) of the NLRA, 29 U.S.C. § 158(a)(1), from discharging or disciplining employees for this activity. It hardly establishes that no "labor dispute" existed within the meaning of § 2(9). Similarly, if the employees protested such sanctions under the collective-bargaining agreement, an arbitrator might ultimately conclude that the sanctions were proper, but this would not alter the obvious fact that the matter is a labor dispute. 12 A labor dispute might be present under the facts of this case even in the absence of the dispute over the scope of the no-strike clause. Regardless of the political nature of the Union's objections to handling Soviet-bound cargo, these objections were expressed in a work stoppage by employees against their employer, which focused on particular work assignments. Thus, apart from the collective-bargaining agreement, the employer-employee relationship would be the matrix of the controversy. We need not decide this question, however, because this case does involve a dispute over the interpretation of the parties' collective-bargaining agreement. 13 In Musicians, the Court held that, although orchestra leaders acted as independent contractors with respect to certain "club-date" engagements, the union's involvement with the leaders was not a combination with a nonlabor group in violation of the Sherman Act. In finding that the leaders were a "labor group," and a party to a labor dispute, the Court relied on the " 'presence of a job or wage competition or some other economic interrelationship affecting legitimate union interests between the union members and the independent contractors.' " 391 U.S., at 106, 88 S.Ct., at 1567 (quoting the opinion of the District Court). In Columbia River Packers Assn., the Court found that the union was merely an association of independent fish sellers involved in a controversy with fish buyers over a contract for the sale of fish; they were not employees of the buyers, nor did they seek to be. 315 U.S., at 147, 62 S.Ct., at 522. The Employer's reliance on Bakery Drivers v. Wagshal, 333 U.S. 437, 68 S.Ct. 630, 92 L.Ed. 792 (1948), is similarly misplaced. In that case, the Court held only that a controversy between two businessmen over delivery times or methods of payment does not become a labor dispute merely because a union representative, with or without his employer's consent, sought to obtain payment pursuant to a particular method. Id., at 443-444, 68 S.Ct., at 633. 14 The Employer's economic-motive analysis also leads to the untenable result that strikes in protest of unreasonably unsafe conditions and some sympathy strikes are not "labor disputes." 15 See Duplex Printing Press Co. v. Deering, 254 U.S. 443, 468-469, 41 S.Ct. 172, 177, 65 L.Ed. 349 (1921). See also Bedford Cut Stone Co. v. Stone Cutters, 274 U.S. 37, 54-55, 47 S.Ct. 522, 527-528, 71 L.Ed. 916 (1927). 16 The thrust of this objection was that the Act's definition of a labor dispute "takes no account whatever of the motives and purposes with which a nation-wide strike or boycott can be commenced and prosecuted." 75 Cong.Rec. 5472 (1932) (remarks of Rep. Beck). 17 In relying on this history, we do not argue that congressional rejection of a broad repeal of the Norris-LaGuardia Act precludes accommodation of that Act to the LMRA. See Sinclair Refining Co. v. Atkinson, 370 U.S. 195, 204-210, 82 S.Ct. 1328, 1333-1336, 8 L.Ed.2d 440 (1962). In Boys Markets, Inc. v. Retail Clerks, 398 U.S., at 249, 90 S.Ct., at 1591, this Court put that argument to rest. Rather, we rely on this legislative history because it demonstrates that Congress believed that the Norris-LaGuardia Act did apply to controversies concerning politically motivated work stoppages. Furthermore, in this case, unlike Boys Markets, we are not asked to accommodate the Norris-LaGuardia Act to a specific federal Act or to the strong policy favoring arbitration. 18 The Senate had declined to adopt these provisions of the House bill. The Senate Report explained that it did not want to impair labor's social gains under the Norris-LaGuardia Act and the NLRA of 1935, but instead wanted to remedy "specific types of injustice" or "clear inequities" by "precise and carefully drawn legislation." S.Rep.No. 105, 80th Cong., 1st Sess., 1 (1947), 1 Leg.Hist. 407. Some of the concerted activities listed in § 12 of the House bill were made unfair labor practices, and the National Labor Relations Board, not private parties, could petition a district court for injunctions against certain unfair labor practices. See S.Rep.No. 105, supra, at 35, 40, 1 Leg.Hist. 441, 446 (reciting proposed revisions to NLRA, §§ 8(b), 10(k), 10(l )). 19 This proposed exception does not limit the judge's discretion to consideration of specified external conduct or of provisions in a collective-bargaining agreement, as does the Boys Markets exception. It provides no guidance to judges in dealing with concerted activity arguably designed to achieve both political and labor-related goals. Such mixed-motivation cases are bound to arise. For example, in United States Steel Corp. v. United Mine Workers, 519 F.2d 1236 (CA5 1975), miners picketed another employer for importing coal from South Africa. The Court of Appeals held that the Norris-LaGuardia Act applied, and that the Boys Markets exception was not available, because "the miners' action was not aimed at [their employer] at all, but rather at the national policy of this country's permitting the importation of South African coal." 519 F.2d, at 1247 (footnote omitted). Under the political-motivation exception, even if the miners had picketed because slave labor was employed to mine the imported coal, the Norris-LaGuardia Act might not apply. Minor variations in the facts would endow the courts with, or divest them of, jurisdiction to issue an injunction, and would create difficult line-drawing problems. 20 The management-rights clause provides: "The Management of the Employer's business and the direction of the work force in the operation of the business are exclusively vested in the Employer as functions of Management. Except as specifically provided in the Agreement, all of the rights, powers, and authority Employer had prior to signing of this Agreement are retained by the Employer." The work-conditions clause provides: "Where hardship is claimed by the Union because of unreasonable or burdensome conditions or where work methods or operations materially change in the future, the problem shall first be discussed between the local and Management involved. In the event an agreement cannot be reached, either party may refer the dispute to the Joint Negotiating Committee and, if the matter cannot be resolved by that Committee, either party may then refer the question to an arbitrator in accordance with the procedure set forth in Clause 15(B)." 21 In fact, the employer in Buffalo Forge made just such a claim. In addition to alleging breach of the no-strike clause, it claimed that the strike was caused by "refusal to follow a supervisor's instructions to cross the . . . picket line." Buffalo Forge, 428 U.S., at 401, 96 S.Ct., at 3144. The District Court found that the strike was in sympathy with the sister union and was not over a dispute that the parties were contractually bound to arbitrate. Id., at 402-403, 96 S.Ct., at 3145. On appeal, the employer did not press its argument that the work stoppage was in part a protest over truckdriving assignments. Id., at 403, n. 8, 96 S.Ct., at 3145, n. 8. 22 The Employer argues that industrial strife is encouraged because employers are given the incentive to discharge or discipline the workers for refusing to work, which is likely to precipitate further strikes. According to this argument, the strike, which began over a nonarbitrable dispute, is transformed into a dispute over an arbitrable issue, i.e., the employer's right under the collective-bargaining agreement to discipline these workers, and may be enjoined under the Boys Markets/Buffalo Forge exception. See, e.g., Complete Auto Transit, Inc. v. Reis, 614 F.2d 1110, 1113-1114 (CA6 1980), aff'd on other grounds, 451 U.S. 401, 101 S.Ct. 1836, 68 L.Ed.2d 248 (1981). This Court has not addressed the validity of this "transformation" analysis. See Complete Auto Transit, Inc. v. Reis, 451 U.S., at 405, n. 4, 101 S.Ct., at 1839, n. 4. 23 The Employer has also requested that we reconsider our decision in Buffalo Forge Co. v. Steelworkers. We decline this invitation. 1 Section 13(c) also includes union organizational activity within its definition of labor dispute, but this case clearly does not involve such activity. 2 The Court's reliance on New Negro Alliance v. Sanitary Grocery Co., 303 U.S. 552, 58 S.Ct. 703, 82 L.Ed. 1012 (1938), is misplaced. Ante, at 714-715. The picketers in that case might not have been seeking to better their own personal economic position, but their purpose was to affect the terms and conditions of employment of the picketed store, since their object was to persuade the store to employ Negroes. Section 13(c) explicitly states that the coverage of the Act does not depend on whether "the disputants stand in the proximate relation of employer and employee." Ante, at 710, n. 9. 3 See Khedivial Line, S. A. E. v. Seafarers International Union, 278 F.2d 49, 50-51 (CA2 1960) (politically motivated blacklist of Egyptian ships to retaliate for Egyptian blacklist of American ships that dealt with Israel is not "labor dispute" triggering Norris-LaGuardia); West Gulf Maritime Assn. v. International Longshoremen's Assn., 413 F.Supp. 372 (S.D.Tex.1975), summarily aff'd, 531 F.2d 574 (CA5 1976) (union's refusal, on political grounds, in violation of a no-strike agreement, to load grain on a ship bound for the Soviet Union does not present a "labor dispute"). 4 The excerpts from the legislative history relied upon by the Court fall short of the clear evidence required to overcome the plain language of § 13(c). See, e.g., Bread Political Action Committee v. FEC, 455 U.S. 577, 581, 102 S.Ct. 1235, 1238, 71 L.Ed.2d 432 (1982). In 1947, Congress declined to amend the federal labor laws so that strikes protesting "disagreement with some governmental policy" would not be protected by the Norris-LaGuardia Act. H.R. 3020, 80th Cong., 1st Sess., § 2(13)(B) (1947), 1 Legislative History of the LMRA 168 (1947); ante, at 717. However, the language of the rejected House version of the amendment was quite broad. There are cases in which unions might disagree with governmental policy and properly take collective action protesting it in order to advance the legitimate economic interests of union members if the terms or conditions of their employment would be affected. Congress might have rejected the House version because of fear that its broad reach would render legitimate union activity unprotected. In 1932, Congress rejected an amendment which would have permitted federal courts to enjoin acts "performed or threatened for an unlawful purpose or with an unlawful intent . . . ." 75 Cong.Rec. 5507 (1932); ante, at 716-717. This amendment would have swept more broadly than the plain language of § 13(c) as adopted. Indeed, Representative Beck's amendment could have rendered the Norris-LaGuardia Act a nullity, since federal judges in the 1930's would have been able to enjoin a strike merely by finding it motivated by an "unlawful purpose." Thus the legislative history does not lead to or compel a conclusion in this case contrary to the plain language of § 13(c).

67

457 U.S. 800 102 S.Ct. 2727 73 L.Ed.2d 396 Bryce N. HARLOW and Alexander P. Butterfield, Petitionersv.A. Ernest FITZGERALD. No. 80-945. Argued Nov. 30, 1981. Decided June 24, 1982. Syllabus In respondent's civil damages action in Federal District Court based on his alleged unlawful discharge from employment in the Department of the Air Force, petitioners, White House aides to former President Nixon, were codefendants with him and were claimed to have participated in the same alleged conspiracy to violate respondent's constitutional and statutory rights as was involved in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349. After extensive pretrial discovery, the District Court denied the motions of petitioners and the former President for summary judgment, holding, inter alia, that petitioners were not entitled to absolute immunity from suit. Independently of the former President, petitioners appealed the denial of their immunity defense, but the Court of Appeals dismissed the appeal. Held: 1. Government officials whose special functions or constitutional status requires complete protection from suits for damages—including certain officials of the Executive Branch, such as prosecutors and similar officials, see Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895, and the President, Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349 are entitled to the defense of absolute immunity. However, executive officials in general are usually entitled to only qualified or good-faith immunity. The recognition of a qualified immunity defense for high executives reflects an attempt to balance competing values: not only the importance of a damages remedy to protect the rights of citizens, but also the need to protect officials who are required to exercise discretion and the related public interest in encouraging the vigorous exercise of official authority. Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90. Federal officials seeking absolute immunity from personal liability for unconstitutional conduct must bear the burden of showing that public policy requires an exemption of that scope. Pp. 806-808. 2. Public policy does not require a blanket recognition of absolute immunity for Presidential aides. Cf. Butz, supra. Pp. 808-813. (a) The rationale of Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583—which held the Speech and Debate Clause derivately applicable to the "legislative acts" of a Senator's aide that would have been privileged if performed by the Senator himself—does not mandate "derivative" absolute immunity for the President's chief aides. Under the "functional" approach to immunity law, immunity protection extends no further than its justification warrants. Pp. 809-811. (b) While absolute immunity might be justified for aides entrusted with discretionary authority in such sensitive areas as national security or foreign policy, a "special functions" rationale does not warrant a blanket recognition of absolute immunity for all Presidential aides in the performance of all their duties. To establish entitlement to absolute immunity, a Presidential aide first must show that the responsibilities of his office embraced a function so sensitive as to require a total shield from liability. He then must demonstrate that he was discharging the protected function when performing the act for which liability is asserted. Under the record in this case, neither petitioner has made the requisite showing for absolute immunity. However, the possibility that petitioners, on remand, can satisfy the proper standards is not foreclosed. Pp. 811-813. 3. Petitioners are entitled to application of the qualified immunity standard that permits the defeat of insubstantial claims without resort to trial. Pp. 813-820 (a) The previously recognized "subjective" aspect of qualified or "good faith" immunity—whereby such immunity is not available if the official asserting the defense "took the action with the malicious intention to cause a deprivation of constitutional rights or other injury," Wood v. Strickland, 420 U.S. 308, 322, 95 S.Ct. 992, 1000, 43 L.Ed.2d 214—frequently has proved incompatible with the principle that insubstantial claims should not proceed to trial. Henceforth, government officials performing discretionary functions generally are shielded from liability for civil damages insofar as their conduct does not violate "clearly established" statutory or constitutional rights of which a reasonable person would have known. Pp. 815-819. (b) The case is remanded for the District Court's reconsideration of the question whether respondent's pretrial showings were insufficient to withstand petitioners' motion for summary judgment. Pp. 819-820. Vacated and remanded. Herbert J. Miller, Jr., Washington, D. C., for petitioner Nixon. Elliot L. Richardson, Washington, D. C., for petitioners Harlow and Butterfield. John E. Nolan, Jr., Washington, D. C., for respondent. Justice POWELL delivered the opinion of the Court. 1 The issue in this case is the scope of the immunity available to the senior aides and advisers of the President of the United States in a suit for damages based upon their official acts. 2 * In this suit for civil damages petitioners Bryce Harlow and Alexander Butterfield are alleged to have participated in a conspiracy to violate the constitutional and statutory rights of the respondent A. Ernest Fitzgerald. Respondent avers that petitioners entered the conspiracy in their capacities as senior White House aides to former President Richard M. Nixon. As the alleged conspiracy is the same as that involved in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, the facts need not be repeated in detail. 3 Respondent claims that Harlow joined the conspiracy in his role as the Presidential aide principally responsible for congressional relations.1 At the conclusion of discovery the supporting evidence remained inferential. As evidence of Harlow's conspiratorial activity respondent relies heavily on a series of conversations in which Harlow discussed Fitzgerald's dismissal with Air Force Secretary Robert Seamans.2 The other evidence most supportive of Fitzgerald's claims consists of a recorded conversation in which the President later voiced a tentative recollection that Harlow was "all for canning" Fitzgerald.3 4 Disputing Fitzgerald's contentions, Harlow argues that exhaustive discovery has adduced no direct evidence of his involvement in any wrongful activity.4 He avers that Secretary Seamans advised him that considerations of efficiency required Fitzgerald's removal by a reduction in force, despite anticipated adverse congressional reaction. Harlow asserts he had no reason to believe that a conspiracy existed. He contends that he took all his actions in good faith.5 5 Petitioner Butterfield also is alleged to have entered the conspiracy not later than May 1969. Employed as Deputy Assistant to the President and Deputy Chief of Staff to H. R. Haldeman,6 Butterfield circulated a White House memorandum in that month in which he claimed to have learned that Fitzgerald planned to "blow the whistle" on some "shoddy purchasing practices" by exposing these practices to public view.7 Fitzgerald characterizes this memorandum as evidence that Butterfield had commenced efforts to secure Fitzgerald's retaliatory dismissal. As evidence that Butterfield participated in the conspiracy to conceal his unlawful discharge and prevent his reemployment, Fitzgerald cites communications between Butterfield and Haldeman in December 1969 and January 1970. After the President had promised at a press conference to inquire into Fitzgerald's dismissal, Haldeman solicited Butterfield's recommendations. In a subsequent memorandum emphasizing the importance of "loyalty," Butterfield counseled against offering Fitzgerald another job in the administration at that time.8 6 For his part, Butterfield denies that he was involved in any decision concerning Fitzgerald's employment status until Haldeman sought his advice in December 1969—more than a month after Fitzgerald's termination had been scheduled and announced publicly by the Air Force. Butterfield states that he never communicated his views about Fitzgerald to any official of the Defense Department. He argues generally that nearly eight years of discovery have failed to turn up any evidence that he caused injury to Fitzgerald.9 7 Together with their codefendant Richard Nixon, petitioners Harlow and Butterfield moved for summary judgment on February 12, 1980. In denying the motion the District Court upheld the legal sufficiency of Fitzgerald's Bivens (Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971)) claim under the First Amendment and his "inferred" statutory causes of action under 5 U.S.C. § 7211 (1976 ed., Supp.IV) and 18 U.S.C. § 1505.10 The court found that genuine issues of disputed fact remained for resolution at trial. It also ruled that petitioners were not entitled to absolute immunity. App. to Pet. for Cert. 1a-3a. 8 Independently of former President Nixon, petitioners invoked the collateral order doctrine and appealed the denial of their immunity defense to the Court of Appeals for the District of Columbia Circuit. The Court of Appeals dismissed the appeal without opinion. Id., at 11a-12a. Never having determined the immunity available to the senior aides and advisers of the President of the United States, we granted certiorari. 452 U.S. 959, 101 S.Ct. 3106, 69 L.Ed.2d 969 (1981).11 II 9 As we reiterated today in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, our decisions consistently have held that government officials are entitled to some form of immunity from suits for damages. As recognized at common law, public officers require this protection to shield them from undue interference with their duties and from potentially disabling threats of liability. 10 Our decisions have recognized immunity defenses of two kinds. For officials whose special functions or constitutional status requires complete protection from suit, we have recognized the defense of "absolute immunity." The absolute immunity of legislators, in their legislative functions, see, e.g., Eastland v. United States Servicemen's Fund, 421 U.S. 491, 95 S.Ct. 1813, 44 L.Ed.2d 324 (1975), and of judges, in their judicial functions, see, e.g., Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978), now is well settled. Our decisions also have extended absolute immunity to certain officials of the Executive Branch. These include prosecutors and similar officials, see Butz v. Economou, 438 U.S. 478, 508-512, 98 S.Ct. 2894, 2911-2916, 57 L.Ed.2d 895 (1978), executive officers engaged in adjudicative functions, id., at 513-517, 98 S.Ct., at 2914-2916, and the President of the United States, see Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349. 11 For executive officials in general, however, our cases make plain that qualified immunity represents the norm. In Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974), we acknowledged that high officials require greater protection than those with less complex discretionary responsibilities. Nonetheless, we held that a governor and his aides could receive the requisite protection from qualified or good-faith immunity. Id., at 247-248, 94 S.Ct. at 1691-1692. In Butz v. Economou, supra, we extended the approach of Scheuer to high federal officials of the Executive Branch. Discussing in detail the considerations that also had underlain our decision in Scheuer, we explained that the recognition of a qualified immunity defense for high executives reflected an attempt to balance competing values: not only the importance of a damages remedy to protect the rights of citizens, 438 U.S., at 504-505, 98 S.Ct., at 2909-2910, but also "the need to protect officials who are required to exercise their discretion and the related public interest in encouraging the vigorous exercise of official authority." Id., at 506, 98 S.Ct., at 2910. Without discounting the adverse consequences of denying high officials an absolute immunity from private lawsuits alleging constitutional violations—consequences found sufficient in Spalding v. Vilas, 161 U.S. 483, 16 S.Ct. 631, 40 L.Ed. 780 (1896), and Barr v. Matteo, 360 U.S. 564, 79 S.Ct. 1335, 3 L.Ed.2d 1434 (1959), to warrant extension to such officials of absolute immunity from suits at common law—we emphasized our expectation that insubstantial suits need not proceed to trial: 12 "Insubstantial lawsuits can be quickly terminated by federal courts alert to the possibilities of artful pleading. Unless the complaint states a compensable claim for relief . . ., it should not survive a motion to dismiss. Moreover, the Court recognized in Scheuer that damages suits concerning constitutional violations need not proceed to trial, but can be terminated on a properly supported motion for summary judgment based on the defense of immunity. . . . In responding to such a motion, plaintiffs may not play dog in the manger; and firm application of the Federal Rules of Civil Procedure will ensure that federal officials are not harassed by frivolous lawsuits." 438 U.S., at 507-508, 98 S.Ct., at 2911-2912 (citations omitted). 13 Butz continued to acknowledge that the special functions of some officials might require absolute immunity. But the Court held that "federal officials who seek absolute exemption from personal liability for unconstitutional conduct must bear the burden of showing that public policy requires an exemption of that scope." Id., at 506, 98 S.Ct. at 2910. This we reaffirmed today in Nixon v. Fitzgerald, 457 U.S., at 747, 102 S.Ct., at 2700. III A. 14 Petitioners argue that they are entitled to a blanket protection of absolute immunity as an incident of their offices as Presidential aides. In deciding this claim we do not write on an empty page. In Butz v. Economou, supra, the Secretary of Agriculture—a Cabinet official directly accountable to the President—asserted a defense of absolute official immunity from suit for civil damages. We rejected his claim. In so doing we did not question the power or the importance of the Secretary's office. Nor did we doubt the importance to the President of loyal and efficient subordinates in executing his duties of office. Yet we found these factors, alone, to be insufficient to justify absolute immunity. "[T]he greater power of [high] officials," we reasoned, "affords a greater potential for a regime of lawless conduct." 438 U.S., at 506, 98 S.Ct., at 2910. Damages actions against high officials were therefore "an important means of vindicating constitutional guarantees." Ibid. Moreover, we concluded that it would be "untenable to draw a distinction for purposes of immunity law between suits brought against state officials under [42 U.S.C.] § 1983 and suits brought directly under the Constitution against federal officials." Id., at 504, 98 S.Ct., at 2909. 15 Having decided in Butz that Members of the Cabinet ordinarily enjoy only qualified immunity from suit, we conclude today that it would be equally untenable to hold absolute immunity an incident of the office of every Presidential subordinate based in the White House. Members of the Cabinet are direct subordinates of the President, frequently with greater responsibilities, both to the President and to the Nation, than White House staff. The considerations that supported our decision in Butz apply with equal force to this case. It is no disparagement of the offices held by petitioners to hold that Presidential aides, like Members of the Cabinet, generally are entitled only to a qualified immunity. B 16 In disputing the controlling authority of Butz, petitioners rely on the principles developed in Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972).12 In Gravel we endorsed the view that "it is literally impossible . . . for Members of Congress to perform their legislative tasks without the help of aides and assistants" and that "the day-to-day work of such aides is so critical to the Members' performance that they must be treated as the latter's alter egos. . . ." Id., at 616-617, 92 S.Ct., at 2622-2623. Having done so, we held the Speech and Debate Clause derivatively applicable to the "legislative acts" of a Senator's aide that would have been privileged if performed by the Senator himself. Id., at 621-622, 92 S.Ct., at 2625-2626. 17 Petitioners contend that the rationale of Gravel mandates a similar "derivative" immunity for the chief aides of the President of the United States. Emphasizing that the President must delegate a large measure of authority to execute the duties of his office, they argue that recognition of derivative absolute immunity is made essential by all the considerations that support absolute immunity for the President himself. 18 Petitioners' argument is not without force. Ultimately, however, it sweeps too far. If the President's aides are derivatively immune because they are essential to the functioning of the Presidency, so should the Members of the Cabinet Presidential subordinates some of whose essential roles are acknowledged by the Constitution itself13—be absolutely immune. Yet we implicitly rejected such derivative immunity in Butz.14 Moreover, in general our cases have followed a "functional" approach to immunity law. We have recognized that the judicial, prosecutorial, and legislative functions require absolute immunity. But this protection has extended no further than its justification would warrant. In Gravel, for example, we emphasized that Senators and their aides were absolutely immune only when performing "acts legislative in nature," and not when taking other acts even "in their official capacity." 408 U.S., at 625, 92 S.Ct., at 2627. See Hutchinson v. Proxmire, 443 U.S. 111, 125-133, 99 S.Ct. 2675, 2683-2687, 61 L.Ed.2d 411 (1979). Our cases involving judges15 and prosecutors16 have followed a similar line. The undifferentiated extension of absolute "derivative" immunity to the President's aides therefore could not be reconciled with the "functional" approach that has characterized the immunity decisions of this Court, indeed including Gravel itself.17 C 19 Petitioners also assert an entitlement to immunity based on the "special functions" of White House aides. This form of argument accords with the analytical approach of our cases. For aides entrusted with discretionary authority in such sensitive areas as national security or foreign policy, absolute immunity might well be justified to protect the unhesitating performance of functions vital to the national interest.18 But a "special functions" rationale does not warrant a blanket recognition of absolute immunity for all Presidential aides in the performance of all their duties. This conclusion too follows from our decision in Butz, which establishes that an executive official's claim to absolute immunity must be justified by reference to the public interest in the special functions of his office, not the mere fact of high station.19 20 Butz also identifies the location of the burden of proof. The burden of justifying absolute immunity rests on the official asserting the claim. 438 U.S., at 506, 98 S.Ct., at 2910. We have not of course had occasion to identify how a Presidential aide might carry this burden. But the general requisites are familiar in our cases. In order to establish entitlement to absolute immunity a Presidential aide first must show that the responsibilities of his office embraced a function so sensitive as to require a total shield from liability.20 He then must demonstrate that he was discharging the protected function when performing the act for which liability is asserted.21 21 Applying these standards to the claims advanced by petitioners Harlow and Butterfield, we cannot conclude on the record before us that either has shown that "public policy requires [for any of the functions of his office] an exemption of [absolute] scope." Butz, 438 U.S., at 506, 98 S.Ct., at 2910. Nor, assuming that petitioners did have functions for which absolute immunity would be warranted, could we now conclude that the acts charged in this lawsuit—if taken at all—would lie within the protected area. We do not, however, foreclose the possibility that petitioners, on remand, could satisfy the standards properly applicable to their claims. IV 22 Even if they cannot establish that their official functions require absolute immunity, petitioners assert that public policy at least mandates an application of the qualified immunity standard that would permit the defeat of insubstantial claims without resort to trial. We agree. A. 23 The resolution of immunity questions inherently requires a balance between the evils inevitable in any available alternative. In situations of abuse of office, an action for damages may offer the only realistic avenue for vindication of constitutional guarantees. Butz v. Economou, supra, at 506, 98 S.Ct., at 2910; see Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S., at 410, 91 S.Ct., at 2011 ("For people in Bivens' shoes, it is damages or nothing"). It is this recognition that has required the denial of absolute immunity to most public officers. At the same time, however, it cannot be disputed seriously that claims frequently run against the innocent as well as the guilty—at a cost not only to the defendant officials, but to society as a whole.22 These social costs include the expenses of litigation, the diversion of official energy from pressing public issues, and the deterrence of able citizens from acceptance of public office. Finally, there is the danger that fear of being sued will "dampen the ardor of all but the most resolute, or the most irresponsible [public officials], in the unflinching discharge of their duties." Gregoire v. Biddle, 177 F.2d 579, 581 (CA2 1949), cert. denied, 339 U.S. 949, 70 S.Ct. 803, 94 L.Ed. 1363 (1950). 24 In identifying qualified immunity as the best attainable accommodation of competing values, in Butz, supra, 438 U.S., at 507-508, 98 S.Ct., 2911-2912, as in Scheuer, 416 U.S., at 245-248, 94 S.Ct., at 1691-1692, we relied on the assumption that this standard would permit "[i]nsubstantial lawsuits [to] be quickly terminated." 438 U.S., at 507-508, 98 S.Ct., at 2911-2912; see Hanrahan v. Hampton, 446 U.S. 754, 765, 100 S.Ct. 1987, 1993, 64 L.Ed.2d 670 (1980) (POWELL, J., concurring in part and dissenting in part).23 Yet petitioners advance persuasive arguments that the dismissal of insubstantial lawsuits without trial—a factor presupposed in the balance of competing interests struck by our prior cases—requires an adjustment of the "good faith" standard established by our decisions. B 25 Qualified or "good faith" immunity is an affirmative defense that must be pleaded by a defendant official. Gomez v. Toledo, 446 U.S. 635, 100 S.Ct. 1920, 64 L.Ed.2d 572 (1980).24 Decisions of this Court have established that the "good faith" defense has both an "objective" and a "subjective" aspect. The objective element involves a presumptive knowledge of and respect for "basic, unquestioned constitutional rights." Wood v. Strickland, 420 U.S. 308, 322, 95 S.Ct. 992, 1001, 43 L.Ed.2d 214 (1975). The subjective component refers to "permissible intentions." Ibid. Characteristically the Court has defined these elements by identifying the circumstances in which qualified immunity would not be available. Referring both to the objective and subjective elements, we have held that qualified immunity would be defeated if an official "knew or reasonably should have known that the action he took within his sphere of official responsibility would violate the constitutional rights of the [plaintiff], or if he took the action with the malicious intention to cause a deprivation of constitutional rights or other injury. . . ." Ibid. (emphasis added).25 26 The subjective element of the good-faith defense frequently has proved incompatible with our admonition in Butz that insubstantial claims should not proceed to trial. Rule 56 of the Federal Rules of Civil Procedure provides that disputed questions of fact ordinarily may not be decided on motions for summary judgment.26 And an official's subjective good faith has been considered to be a question of fact that some courts have regarded as inherently requiring resolution by a jury.27 27 In the context of Butz' attempted balancing of competing values, it now is clear that substantial costs attend the litigation of the subjective good faith of government officials. Not only are there the general costs of subjecting officials to the risks of trial—distraction of officials from their governmental duties, inhibition of discretionary action, and deterrence of able people from public service. There are special costs to "subjective" inquiries of this kind. Immunity generally is available only to officials performing discretionary functions. In contrast with the thought processes accompanying "ministerial" tasks, the judgments surrounding discretionary action almost inevitably are influenced by the decisionmaker's experiences, values, and emotions. These variables explain in part why questions of subjective intent so rarely can be decided by summary judgment. Yet they also frame a background in which there often is no clear end to the relevant evidence. Judicial inquiry into subjective motivation therefore may entail broad-ranging discovery and the deposing of numerous persons, including an official's professional colleagues.28 Inquiries of this kind can be peculiarly disruptive of effective government.29 28 Consistently with the balance at which we aimed in Butz, we conclude today that bare allegations of malice should not suffice to subject government officials either to the costs of trial or to the burdens of broad-reaching discovery. We therefore hold that government officials performing discretionary functions generally are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known. See Procunier v. Navarette, 434 U.S. 555, 565, 98 S.Ct. 855, 861, 55 L.Ed.2d 24 (1978); Wood v. Strickland, 420 U.S., at 322, 95 S.Ct., at 1001.30 29 Reliance on the objective reasonableness of an official's conduct, as measured by reference to clearly established law,31 should avoid excessive disruption of government and permit the resolution of many insubstantial claims on summary judgment. On summary judgment, the judge appropriately may determine, not only the currently applicable law, but whether that law was clearly established at the time an action occurred.32 If the law at that time was not clearly established, an official could not reasonably be expected to anticipate subsequent legal developments, nor could he fairly be said to "know" that the law forbade conduct not previously identified as unlawful. Until this threshold immunity question is resolved, discovery should not be allowed. If the law was clearly established, the immunity defense ordinarily should fail, since a reasonably competent public official should know the law governing his conduct. Nevertheless, if the official pleading the defense claims extraordinary circumstances and can prove that he neither knew nor should have known of the relevant legal standard, the defense should be sustained. But again, the defense would turn primarily on objective factors. 30 By defining the limits of qualified immunity essentially in objective terms, we provide no license to lawless conduct. The public interest in deterrence of unlawful conduct and in compensation of victims remains protected by a test that focuses on the objective legal reasonableness of an official's acts. Where an official could be expected to know that certain conduct would violate statutory or constitutional rights, he should be made to hesitate; and a person who suffers injury caused by such conduct may have a cause of action.33 But where an official's duties legitimately require action in which clearly established rights are not implicated, the public interest may be better served by action taken "with independence and without fear of consequences." Pierson v. Ray, 386 U.S. 547, 554, 87 S.Ct. 1213, 1217, 18 L.Ed.2d 288 (1967).34 C 31 In this case petitioners have asked us to hold that the respondent's pretrial showings were insufficient to survive their motion for summary judgment.35 We think it appropriate, however, to remand the case to the District Court for its reconsideration of this issue in light of this opinion.36 The trial court is more familiar with the record so far developed and also is better situated to make any such further findings as may be necessary. V 32 The judgment of the Court of Appeals is vacated, and the case is remanded for further action consistent with this opinion. 33 So ordered. 34 Justice BRENNAN, with whom Justice MARSHALL, and Justice BLACKMUN join, concurring. 35 I agree with the substantive standard announced by the Court today, imposing liability when a public-official defendant "knew or should have known" of the constitutionally violative effect of his actions. Ante, at , 815, 819. This standard would not allow the official who actually knows that he was violating the law to escape liability for his actions, even if he could not "reasonably have been expected" to know what he actually did know. Ante, at 819, n.33. Thus the clever and unusually well-informed violator of constitutional rights will not evade just punishment for his crimes. I also agree that this standard applies "across the board," to all "government officials performing discretionary functions." Ante, at 818. I write separately only to note that given this standard, it seems inescapable to me that some measure of discovery may sometimes be required to determine exactly what a public-official defendant did "know" at the time of his actions. In this respect the issue before us is very similar to that addressed in Herbert v. Lando, 441 U.S. 153, 99 S.Ct. 1635, 60 L.Ed.2d 115 (1979), in which the Court observed that "[t]o erect an impenetrable barrier to the plaintiff's use of such evidence on his side of the case is a matter of some substance, particularly when defendants themselves are prone to assert their goo[d f]aith. . . ." Id., at 170, 99 S.Ct., at 1645. Of course, as the Court has already noted, ante, at 818-819, summary judgment will be readily available to public-official defendants whenever the state of the law was so ambiguous at the time of the alleged violation that it could not have been "known" then, and thus liability could not ensue. In my view, summary judgment will also be readily available whenever the plaintiff cannot prove, as a threshold matter, that a violation of his constitutional rights actually occurred. I see no reason why discovery of defendants' "knowledge" should not be deferred by the trial judge pending decision of any motion of defendants for summary judgment on grounds such as these. Cf. Herbert v. Lando, supra, at 180, n.4, 99 S.Ct., at 1650, n.4 (POWELL, J., concurring). 36 Justice BRENNAN, Justice WHITE, Justice MARSHALL and Justice BLACKMUN, concurring. 37 We join the Court's opinion but, having dissented in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, we disassociate ourselves from any implication in the Court's opinion in the present case that Nixon v. Fitzgerald was correctly decided. 38 Justice REHNQUIST, concurring. 39 At such time as a majority of the Court is willing to re-examine our holding in Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), I shall join in that undertaking with alacrity. But until that time comes, I agree that the Court's opinion in this case properly disposes of the issues presented, and I therefore join it. 40 Chief Justice BURGER, dissenting. 41 The Court today decides in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, what has been taken for granted for 190 years, that it is implicit in the Constitution that a President of the United States has absolute immunity from civil suits arising out of official acts as Chief Executive. I agree fully that absolute immunity for official acts of the President is, like executive privilege, "fundamental to the operation of Government and inextricably rooted in the separation of powers under the Constitution." United States v. Nixon, 418 U.S. 683, 708, 94 S.Ct. 3090, 3107, 41 L.Ed.2d 1039 (1974).1 42 In this case the Court decides that senior aides of the President do not have derivative immunity from the President. I am at a loss, however, to reconcile this conclusion with our holding in Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972). The Court reads Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), as resolving that question; I do not. Butz is clearly distinguishable.2 43 In Gravel we held that it is implicit in the Constitution that aides of Members of Congress have absolute immunity for acts performed for Members in relation to their legislative function. We viewed the aides' immunity as deriving from the Speech or Debate Clause, which provides that "for any Speech or Debate in either House, [Senators and Representatives] shall not be questioned in any other Place." Art. I, § 6, cl. 1 (emphasis added). Read literally, the Clause would, of course, limit absolute immunity only to the Member and only to speech and debate within the Chamber. But we have read much more into this plain language. The Clause says nothing about "legislative acts" outside the Chambers, but we concluded that the Constitution grants absolute immunity for legislative acts not only "in either House" but in committees and conferences and in reports on legislative activities. 44 Nor does the Clause mention immunity for congressional aides. Yet, going far beyond any words found in the Constitution itself, we held that a Member's aides who implement policies and decisions of the Member are entitled to the same absolute immunity as a Member. It is hardly an overstatement to say that we thus avoided a "literalistic approach," Gravel, supra, at 617, 92 S.Ct. at 2623, and instead looked to the structure of the Constitution and the evolution of the function of the Legislative Branch. In short, we drew this immunity for legislative aides from a functional analysis of the legislative process in the context of the Constitution taken as a whole and in light of 20th-century realities. Neither Presidents nor Members of Congress can, as they once did, perform all their constitutional duties personally.3 45 We very properly recognized in Gravel that the central purpose of a Member's absolute immunity would be "diminished and frustrated" if the legislative aides were not also protected by the same broad immunity. Speaking for the Court in Gravel, Justice WHITE agreed with the Court of Appeals that 46 "it is literally impossible, in view of the complexities of the modern legislative process, with Congress almost constantly in session and matters of legislative concern constantly proliferating, for Members of Congress to perform their legislative tasks without the help of aides and assistants; that the day-to-day work of such aides is so critical to the Members' performance that they must be treated as the latter's alter egos; and that if they are not so recognized, the central role of the Speech or Debate Clause—to prevent intimidation of legislators by the Executive and accountability before a possibly hostile judiciary . . .—will inevitably be diminished and frustrated." 408 U.S., at 616-617, 92 S.Ct., at 2622-2623 (emphasis added). 47 I joined in that analysis and continue to agree with it, for without absolute immunity for these "elbow aides," who are indeed "alter egos," a Member could not effectively discharge all of the assigned constitutional functions of a modern legislator. 48 The Court has made this reality a matter of our constitutional jurisprudence. How can we conceivably hold that a President of the United States, who represents a vastly larger constituency than does any Member of Congress, should not have "alter egos" with comparable immunity? To perform the constitutional duties assigned to the Executive would be "literally impossible, in view of the complexities of the modern [Executive] process, . . . without the help of aides and assistants."4 Id., at 616, 92 S.Ct., at 2623. These words reflect the precise analysis of Gravel, and this analysis applies with at least as much force to a President. The primary layer of senior aides of a President—like a Senator's "alter egos" are literally at a President's elbow, with offices a few feet or at most a few hundred feet from his own desk. The President, like a Member of Congress, may see those personal aides many times in one day. They are indeed the President's "arms" and "fingers" to aid in performing his constitutional duty to see "that the laws [are] faithfully executed." Like a Member of Congress, but on a vastly greater scale, the President cannot personally implement a fraction of his own policies and day-to-day decisions.5 49 For some inexplicable reason the Court declines to recognize the realities in the workings of the Office of a President, despite the Court's cogent recognition in Gravel concerning the realities of the workings of 20th-century Members of Congress. Absent equal protection for a President's aides, how will Presidents be free from the risks of "intimidation . . . by [Congress] and accountability before a possibly hostile judiciary?" Gravel, 408 U.S., at 617, 92 S.Ct., at 2623. Under today's holding in this case the functioning of the Presidency will inevitably be "diminished and frustrated." Ibid. 50 Precisely the same public policy considerations on which the Court now relies in Nixon v. Fitzgerald, and that we relied on only recently in Gravel, are fully applicable to senior Presidential aides. The Court's opinion in Nixon v. Fitzgerald correctly points out that if a President were subject to suit, awareness of personal vulnerability to suit "frequently could distract a President from his public duties, to the detriment of not only the President and his office but also the Nation that the Presidency was designed to serve." At 753, 102 S.Ct., at 2703. This same negative incentive will permeate the inner workings of the Office of the President if the Chief Executive's "alter egos" are not protected derivatively from the immunity of the President. In addition, exposure to civil liability for official acts will result in constant judicial questioning, through judicial proceedings and pretrial discovery, into the inner workings of the Presidential Office beyond that necessary to maintain the traditional checks and balances of our constitutional structure.6 51 I challenge the Court and the dissenters in Nixon v. Fitzgerald who join in the instant holding to say that the effectiveness of Presidential aides will not "inevitably be diminished and frustrated," Gravel, supra, 408 U.S., at 617, 92 S.Ct., at 2623, if they must weigh every act and decision in relation to the risks of future lawsuits. The Gravel Court took note of the burdens on congressional aides: the stress of long hours, heavy responsibilities, constant exposure to harassment of the political arena. Is the Court suggesting the stresses are less for Presidential aides? By construing the Constitution to give only qualified immunity to senior Presidential aides we give those key "alter egos" only lawsuits, winable lawsuits perhaps, but lawsuits nonetheless, with stress and effort that will disperse and drain their energies and their purses.7 52 In this Court we witness the new filing of as many as 100 cases a week, many utterly frivolous and even bizarre. Yet the defending party in many of these cases may have spent or become liable for thousands of dollars in litigation expense. Hundreds of thousands of other cases are disposed of without reaching this Court. When we see the myriad irresponsible and frivolous cases regularly filed in American courts, the magnitude of the potential risks attending acceptance of public office emerges. Those potential risks inevitably will be a factor in discouraging able men and women from entering public service. 53 We—judges collectively—have held that the common law provides us with absolute immunity for ourselves with respect to judicial acts, however erroneous or ill-advised. See, e.g., Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978). Are the lowest ranking of 27,000 or more judges, thousands of prosecutors, and thousands of congressional aides—an aggregate of not less than 75,000 in all—entitled to greater protection than two senior aides of a President? 54 Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), does not dictate that senior Presidential aides be given only qualified immunity. Butz held only that a Cabinet officer exercising discretion was not entitled to absolute immunity; we need not abandon that holding. A senior Presidential aide works more intimately with the President on a daily basis than does a Cabinet officer, directly implementing Presidential decisions literally from hour to hour. 55 In his dissent today in Nixon v. Fitzgerald, Justice WHITE states that the "Court now applies the dissenting view in Butz to the Office of the President." 457 U.S., at 764, 102 S.Ct., at 2709. However, this suggests that a President and his Cabinet officers, who serve only "during the pleasure of the President," are on the same plane constitutionally. It wholly fails to distinguish the role of a President or his "elbow aides" from the role of Cabinet officers, who are department heads rather than "alter egos." It would be in no sense inconsistent to hold that a President's personal aides have greater immunity than Cabinet officers. 56 The Court's analysis in Gravel demonstrates that the question of derivative immunity does not and should not depend on a person's rank or position in the hierarchy, but on the function performed by the person and the relationship of that person to the superior. Cabinet officers clearly outrank United States Attorneys, yet qualified immunity is accorded the former and absolute immunity the latter; rank is important only to the extent that the rank determines the function to be performed. The function of senior Presidential aides, as the "alter egos" of the President, is an integral, inseparable part of the function of the President.8 Justice WHITE was clearly correct in Gravel, stating that Members of Congress could not "perform their legislative tasks without the help of aides and assistants; [and] that the day-to-day work of such aides is so critical to the Members' performance that they must be treated as the latter's alter egos . . . ." 408 U.S., at 616-617, 92 S.Ct., at 2623. 57 By ignoring Gravel and engaging in a wooden application of Butz, the Court significantly undermines the functioning of the Office of the President. Under the Court's opinion in Nixon today it is clear that Presidential immunity derives from the Constitution as much as congressional immunity comes from that source. Can there rationally be one rule for congressional aides and another for Presidential aides simply because the initial absolute immunity of each derives from different aspects of the Constitution? I find it inexplicable why the Court makes no effort to demonstrate why the Chief Executive of the Nation should not be assured that senior staff aides will have the same protection as the aides of Members of the House and Senate. 1 Harlow held this position from the beginning of the Nixon administration on January 20, 1969, through November 4, 1969. On the latter date he was designated as Counselor to the President, a position accorded Cabinet status. He served in that capacity until December 9, 1970, when he returned to private life. Harlow later resumed the duties of Counselor for the period from July 1, 1973, through April 14, 1974. Respondent appears to allege that Harlow continued in a conspiracy against him throughout the various changes of official assignment. 2 The record reveals that Secretary Seamans called Harlow in May 1969 to inquire about likely congressional reaction to a draft reorganization plan that would cause Fitzgerald's dismissal. According to Seamans' testimony, "[W]e [the Air Force] didn't ask [Harlow] to pass judgment on the action itself. We just asked him what the impact would be in the relationship with the Congress." App. 153a, 164a-165a (deposition of Robert Seamans). Through an aide Harlow responded that "this was a very sensitive item on the Hill and that it would be [his] recommendation that [the Air Force] not proceed to make such a change at that time." Id., at 152a. But the Air Force persisted. Seamans spoke to Harlow on at least one subsequent occasion during the spring of 1969. The record also establishes that Secretary Seamans called Harlow on November 4, 1969, shortly after the public announcement of Fitzgerald's impending dismissal, and again in December 1969. See id., at 186a. 3 See id., at 284a. (transcript of a recorded conversation between Richard Nixon and Ronald Ziegler, February 26, 1973). In a conversation with the President on January 31, 1973, John Ehrlichman also recalled that Harlow had discussed the Fitzgerald case with the President. See id., at 218a-221a. (transcript of recorded conversation between Richard Nixon and John Ehrlichman, January 31, 1973). In the same conversation the President himself asserted that he had spoken to Harlow about the Fitzgerald matter, see id., at 218a, but the parties continue to dispute whether Mr. Nixon—at the most relevant moments in the discussion—was confusing Fitzgerald's case with that of another dismissed employee. The President explicitly stated at one point that he previously had been confused. See id., at 220a. 4 See Defendants Memorandum of Points and Authorities in Support of Their Motion for Summary Judgment in Civ. No. 74-178 (DC), p. 7 (Feb. 12, 1980). 5 In support of his version of events Harlow relies particularly on the deposition testimony of Air Force Secretary Seamans, who stated that he regarded abolition of Fitzgerald's position as necessary "to improve the efficiency" of the Financial Management Office of the Air Force and that he never received any White House instruction regarding the Fitzgerald case. App. 159a-160a. Harlow also disputes the probative value of Richard Nixon's recorded remark that Harlow had supported Fitzgerald's firing. Harlow emphasizes the tentativeness of the President's statement. To the President's query whether Harlow was "all for canning [Fitzgerald], wasn't he?", White House Press Secretary Ronald Ziegler in fact gave a negative reply: "No, I think Bryce may have been the other way." Id., at 284a. The President did not respond to Ziegler's comment. 6 The record establishes that Butterfield worked from an office immediately adjacent to the oval office. He had almost daily contact with the President until March 1973, when he left the White House to become Administrator of the Federal Aviation Administration. 7 Id., at 274a. Butterfield reported that this information had been referred to the Federal Bureau of Investigation. In the memorandum Butterfield reported that he had received the information "by word of several mouths, but allegedly from a senior AFL-CIO official originally . . . . Evidently, Fitzgerald attended a recent meeting of the National Democratic Coalition and, while there, revealed his intentions to a labor representative who, fortunately for us, was unsympathetic." Ibid. 8 Id., 99a-100a, 180a-181a. This memorandum, quoted in Nixon v. Fitzgerald, 457 U.S., at 735-736, 102 S.Ct., at 2694-2695, was not sent to the Defense Department. 9 See Memorandum in Support of Summary Judgment, supra, at 26. The history of Fitzgerald's litigation is recounted in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349. Butterfield was named as a defendant in the initial civil action filed by Fitzgerald in 1974. Harlow was named for the first time in respondent's second amended complaint of July 5, 1978. 10 The first of these statutes, 5 U.S.C. § 7211 (1976 ed., Supp.IV), provides generally that "[t]he right of employees . . . to . . . furnish information to either House of Congress, or to a committee or Member thereof, may not be interfered with or denied." The second, 18 U.S.C. § 1505, is a criminal statute making it a crime to obstruct congressional testimony. Neither expressly creates a private right to sue for damages. Petitioners argue that the District Court erred in finding that a private cause of action could be inferred under either statute, and that "special factors" present in the context of the federal employer-employee relationship preclude the recognition of respondent's Bivens action under the First Amendment. The legal sufficiency of respondent's asserted causes of action is not, however, a question that we view as properly presented for our decision in the present posture of this case. See n. 36, infra. 11 As in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, our jurisdiction has been challenged on the basis that the District Court's order denying petitioners' claim of absolute immunity was not an appealable final order and that the Court of Appeals' dismissal of petitioners' appeal establishes that this case was never "in" the Court of Appeals within the meaning of 28 U.S.C. § 1254. As the discussion in Nixon establishes our jurisdiction in this case as well, we need not consider those challenges in this opinion. 12 Petitioners also claim support from other cases that have followed Gravel in holding that congressional employees are derivatively entitled to the legislative immunity provided to United States Senators and Representatives under the Speech and Debate Clause. See Eastland v. United States Servicemen's Fund, 421 U.S. 491, 95 S.Ct. 1813, 44 L.Ed.2d 324 (1975); Doe v. McMillan, 412 U.S. 306, 93 S.Ct. 2018, 36 L.Ed.2d 912 (1973). 13 See U.S.Const., Art. II, § 2 ("The President . . . may require the Opinion, in writing, of the principal Officer in each of the executive Departments, upon any Subject relating to the Duties of their respective Offices. . . ."). 14 The Chief Justice, post, at 828, argues that senior Presidential aides work "more intimately with the President on a daily basis than does a Cabinet officer," and that Butz therefore is not controlling. In recent years, however, such men as Henry Kissinger and James Schlesinger have served in both Presidential advisory and Cabinet positions. Kissinger held both posts simultaneously. In our view it is impossible to generalize about the role of "offices" in an individual President's administration without reference to the functions that particular officeholders are assigned by the President. Butz v. Economou cannot be distinguished on this basis. 15 See, e.g., Supreme Court of Virginia v. Consumers Union of the United States, 446 U.S. 719, 731-737, 100 S.Ct. 1967, 1974-1977, 64 L.Ed.2d 641 (1980); Stump v. Sparkman, 435 U.S. 349, 362, 98 S.Ct. 1099, 1107, 55 L.Ed.2d 331 (1978). 438 U.S., at 362, 98 S.Ct., at 1107. 16 In Imbler v. Pachtman, 424 U.S. 409, 430-431, 96 S.Ct. 984, 994-995, 47 L.Ed.2d 128 (1973), this Court reserved the question whether absolute immunity would extend to "those aspects of the prosecutor's responsibility that cast him in the role of an administrator or investigative officer." Since that time the Courts of Appeals generally have ruled that prosecutors do not enjoy absolute immunity for acts taken in those capacities. See, e.g., Mancini v. Lester, 630 F.2d 990, 992 (CA3 1980); Forsyth v. Kleindienst, 599 F.2d 1203, 1213-1214 (CA3 1979). This Court at least implicitly has drawn the same distinction in extending absolute immunity to executive officials when they are engaged in quasi-prosecutorial functions. See Butz v. Economou, 438 U.S., at 515-517, 98 S.Ct., at 2915-2916. 17 Our decision today in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, in no way abrogates this general rule. As we explained in that opinion, the recognition of absolute immunity for all of a President's acts in office derives in principal part from factors unique to his constitutional responsibilities and station. Suits against other officials including Presidential aides—generally do not invoke separation-of-powers considerations to the same extent as suits against the President himself. 18 Cf. United States v. Nixon, 418 U.S. 683, 710-711, 94 S.Ct. 3090, 3108-3109, 41 L.Ed.2d 1039 (1974) ("[C]ourts have traditionally shown the utmost deference to Presidential responsibilities" for foreign policy and military affairs, and claims of privilege in this area would receive a higher degree of deference than invocations of "a President's generalized interest in confidentiality"); Katz v. United States, 389 U.S. 347, 364, 88 S.Ct. 507, 518, 19 L.Ed.2d 576 (1967) (WHITE, J., concurring) ("We should not require the warrant procedure and the magistrate's judgment if the President of the United States or his chief legal officer, the Attorney General, has considered the requirements of national security and authorized electronic surveillance as reasonable") (emphasis added). 19 Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972), points to a similar conclusion. We fairly may assume that some aides are assigned to act as Presidential "alter egos," id., at 616-617, 92 S.Ct., at 2622-2623, in the exercise of functions for which absolute immunity is "essential for the conduct of the public business," Butz, supra, at 507, 98 S.Ct., at 2911. Cf. Gravel, supra, at 620, 92 S.Ct., at 2624 (derivative immunity extends only to acts within the "central role" of the Speech and Debate Clause in permitting free legislative speech and debate). By analogy to Gravel, a derivative claim to Presidential immunity would be strongest in such "central" Presidential domains as foreign policy and national security, in which the President could not discharge his singularly vital mandate without delegating functions nearly as sensitive as his own. 20 Here as elsewhere the relevant judicial inquiries would encompass considerations of public policy, the importance of which should be confirmed either by reference to the common law or, more likely, our constitutional heritage and structure. See Nixon v. Fitzgerald, 457 U.S., at 747-748, 102 S.Ct., at 2700-2701. 21 The need for such an inquiry is implicit in Butz v. Economou, supra, at 508-517, 98 S.Ct., at 2911-2916; see Imbler v. Pachtman, supra, at 430-431, 96 S.Ct., at 994-995. Cases involving immunity under the Speech and Debate Clause have inquired explicitly into whether particular acts and activities qualified for the protection of the Clause. See, e.g., Hutchinson v. Proxmire, 443 U.S. 111, 99 S.Ct. 2675, 61 L.Ed.2d 411 (1979); Doe v. McMillan, 412 U.S. 306, 93 S.Ct. 2018, 36 L.Ed.2d 912 (1973); Gravel v. United States, supra. 22 See generally Schuck, Suing Our Servants: The Court, Congress, and the Liability of Public Officials for Damages, 1980 S.Ct.Rev. 281, 324-327. 23 The importance of this consideration hardly needs emphasis. This Court has noted the risk imposed upon political officials who must defend their actions and motives before a jury. See Lake Country Estates, Inc. v. Tahoe Regional Planning Agency, 440 U.S. 391, 405, 99 S.Ct. 1171, 1179, 59 L.Ed.2d 401 (1979); Tenney v. Brandhove, 341 U.S. 367, 377-378, 71 S.Ct. 783, 788-789, 95 L.Ed.2d 1019 (1951). As the Court observed in Tenney: "In times of political passion, dishonest or vindictive motives are readily attributed . . . and as readily believed." Id., at 378, 71 S.Ct., at 789. 24 Although Gomez presented the question in the context of an action under 42 U.S.C. § 1983, the Court's analysis indicates that "immunity" must also be pleaded as a defense in actions under the Constitution and laws of the United States. See 446 U.S., at 640, 100 S.Ct., at 1924. Gomez did not decide which party bore the burden of proof on the issue of good faith. Id., at 642, 100 S.Ct., 1924 (Rehnquist, J., concurring). 25 In Wood the Court explicitly limited its holding to the circumstances in which a school board member, "in the specific context of school discipline," 420 U.S., at 322, 95 S.Ct., 1001, would be stripped of claimed immunity in an action under § 1983. Subsequent cases, however, have quoted the Wood formulation as a general statement of the qualified immunity standard. See, e.g., Procunier v. Navarette, 434 U.S. 555, 562-563, 566, 98 S.Ct. 855, 859-860, 862, 55 L.Ed.2d 24 (1978), quoted in Baker v. McCollan, 443 U.S. 137, 139, 99 S.Ct. 2689, 2692, 61 L.Ed.2d 433 (1979). 26 Rule 56(c) states that summary judgment "shall be rendered forthwith if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." In determining whether summary judgment is proper, a court ordinarily must look at the record in the light most favorable to the party opposing the motion, drawing all inferences most favorable to that party. E.g., Poller v. Columbia Broadcasting System, Inc., 368 U.S. 464, 473, 82 S.Ct. 486, 491, 7 L.Ed.2d 458 (1962). 27 E.g., Landrum v. Moats, 576 F.2d 1320, 1329 (CA8 1978); Duchesne v. Sugarman, 566 F.2d 817, 832-833 (CA2 1977); cf. Hutchinson v. Proxmire, 443 U.S., at 120 n.9, 99 S.Ct., at 2680, n.9 (questioning whether the existence of "actual malice," as an issue of fact, may properly be decided on summary judgment in a suit alleging libel of a public figure). 28 In suits against a President's closest aides, discovery of this kind frequently could implicate separation-of-powers concerns. As the Court recognized in United States v. Nixon, 418 U.S., at 708, 94 S.Ct., at 3107: "A President and those who assist him must be free to explore alternatives in the process of shaping policies and making decisions and to do so in a way many would be unwilling to express except privately. These are the considerations justifying a presumptive privilege for Presidential communications. The privilege is fundamental to the operation of Government and inextricably rooted in the separation of powers under the Constitution." 29 As Judge Gesell observed in his concurring opinion in Halperin v. Kissinger, 196 U.S.App.D.C. 285, 307, 606 F.2d 1192, 1214 (1979), aff'd in pertinent part by an equally divided Court, 452 U.S. 713, 101 S.Ct. 3132, 69 L.Ed.2d 367 (1981): "We should not close our eyes to the fact that with increasing frequency in this jurisdiction and throughout the country plaintiffs are filing suits seeking damage awards against high government officials in their personal capacities based on alleged constitutional torts. Each such suit almost invariably results in these officials and their colleagues being subjected to extensive discovery into traditionally protected areas, such as their deliberations preparatory to the formulation of government policy and their intimate thought processes and communications at the presidential and cabinet levels. Such discover [sic ] is wide-ranging, time-consuming, and not without considerable cost to the officials involved. It is not difficult for ingenious plaintiff's counsel to create a material issue of fact on some element of the immunity defense where subtle questions of constitutional law and a decisionmaker's mental processes are involved. A sentence from a casual document or a difference in recollection with regard to a particular policy conversation held long ago would usually, under the normal summary judgment standards, be sufficient [to force a trial]. . . . The effect of this development upon the willingness of individuals to serve their country is obvious." 30 This case involves no issue concerning the elements of the immunity available to state officials sued for constitutional violations under 42 U.S.C. § 1983. We have found previously, however, that it would be "untenable to draw a distinction for purposes of immunity law between suits brought against state officials under § 1983 and suits brought directly under the Constitution against federal officials." Butz v. Economou, 438 U.S., at 504, 98 S.Ct., at 2909. Our decision in no way diminishes the absolute immunity currently available to officials whose functions have been held to require a protection of this scope. 31 This case involves no claim that Congress has expressed its intent to impose "no fault" tort liability on high federal officials for violations of particular statutes or the Constitution. 32 As in Procunier v. Navarette, 434 U.S., at 565, 98 S.Ct., at 861, we need not define here the circumstances under which "the state of the law" should be "evaluated by reference to the opinions of this Court, of the Courts of Appeals, or of the local District Court." 33 Cf. Procunier v. Navarette, supra, 434 U.S., at 565, 98 S.Ct., at 861, quoting Wood v. Strickland, 420 U.S., at 322, 95 S.Ct., at 1001 ("Because they could not reasonably have been expected to be aware of a constitutional right that had not yet been declared, petitioners did not act with such disregard for the established law that their conduct 'cannot reasonably be characterized as being in good faith' "). 34 We emphasize that our decision applies only to suits for civil damages arising from actions within the scope of an official's duties and in "objective" good faith. We express no view as to the conditions in which injunctive or declaratory relief might be available. 35 In Butz, we admonished that "insubstantial" suits against high public officials should not be allowed to proceed to trial. 438 U.S., at 507, 98 S.Ct., at 2911. See Schuck, supra n.22, at 324-327. We reiterate this admonition. Insubstantial lawsuits undermine the effectiveness of government as contemplated by our constitutional structure, and "firm application of the Federal Rules of Civil Procedure" is fully warranted in such cases. 438 U.S., at 508, 98 S.Ct., at 2911. 36 Petitioners also have urged us, prior to the remand, to rule on the legal sufficiency of respondent's "implied" causes of action under 5 U.S.C. § 7211 (1976 ed., Supp.IV) and 18 U.S.C. § 1505 and his Bivens claim under the First Amendment. We do not view petitioners' argument on the statutory question as insubstantial. Cf. Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 4 U.S. 353, 377-378, 102 S.Ct. 1825, 1838-1839, 72 L.Ed.2d 182 (1982) (controlling question in implication of statutory causes of action is whether Congress affirmatively intended to create a damages remedy); Middlesex County Sewerage Auth. v. National Sea Clammers Assn., 453 U.S. 1, 101 S.Ct. 2615, 69 L.Ed.2d 435 (1981) (same); Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 638-639, 101 S.Ct. 2061, 2065-2066, 68 L.Ed.2d 500 (1981) (same). Nor is the Bivens question. Cf. Bush v. Lucas, 647 F.2d 573, 576 (CA5 1981) (holding that the "unique relationship between the Federal Government and its civil service employees is a special consideration which counsels hesitation in inferring a Bivens remedy"). As in Nixon v. Fitzgerald, 457 U.S. 731, 102 S.Ct. 2690, 73 L.Ed.2d 349, however, we took jurisdiction of the case only to resolve the immunity question under the collateral order doctrine. We therefore think it appropriate to leave these questions for fuller consideration by the District Court and, if necessary, by the Court of Appeals. 1 As I noted in Nixon v. Fitzgerald, Presidential immunity for official acts while in office has never been seriously questioned until very recently. At 758, n.1, 102 S.Ct., at 2706, n.1. (BURGER, C.J., concurring). 2 If indeed there is an irreconcilable conflict between Gravel, and Butz, the Court has an obligation to try to harmonize its holdings—or at least tender a reasonable explanation. The Court has done neither. 3 A Senator's allotment for staff varies significantly, but can range from as few as 17 to over 70 persons, in addition to committee staff aides who perform important legislative functions for Members. S.Doc.No. 97-19, pp. 27-106 (1981). House Members have roughly 18 to 26 assistants at any one time, in addition to committee staff aides. H.R.Doc.No. 97-113, pp. 28-174 (1981). 4 In the early years of the Republic, Members of Congress and Presidents performed their duties without staffs of aides and assistants. Washington and Jefferson spent much of their time on their plantations. Congress did not even appropriate funds for a Presidential clerk until 1857. Lincoln opened his own mail, Cleveland answered the phone at the White House, and Wilson regularly typed his own speeches. S. Wayne, The Legislative Presidency 30 (1978). Whatever may have been the situation beginning under Washington, Adams, and Jefferson, we know today that the Presidency functions with a staff that exercises a wide spectrum of authority and discretion and directly assists the President in carrying out constitutional duties. 5 Justice WHITE's dissent in Nixon v. Fitzgerald today expresses great concern that a President may "cause serious injury to any number of citizens even though he knows his conduct violates a statute . . . ." At 764, 102 S.Ct., at 2709. What the dissent wholly overlooks, however, is the plain fact that the absolute immunity does not protect a President for acts outside the constitutional function of a President. 6 The same remedies for checks on Presidential abuse also will check abuses by the comparatively small group of senior aides who act as "alter egos" of the President. The aides serve at the pleasure of the President and thus may be removed by the President. Congressional and public scrutiny maintain a constant and pervasive check on abuses, and such aides may be prosecuted criminally. See Nixon, 457 U.S., at 757, 102 S.Ct., at 2705-2706. However, a criminal prosecution cannot be commenced absent careful consideration by a grand jury at the request of a prosecutor; the same check is not present with respect to the commencement of civil suits in which advocates are subject to no realistic accountability. 7 The Executive Branch may as a matter of grace supply some legal assistance. The Department of Justice has a longstanding policy of representing federal officers in civil suits involving conduct performed within the scope of their employment. In addition, the Department provides for retention of private legal counsel when necessary. See Senate Subcommittee on Administrative Practice and Procedure of the Committee on the Judiciary, Justice Department Retention of Private Legal Counsel to Represent Federal Employees in Civil Lawsuits, 95th Cong., 2d Sess. (Comm.Print 1978). The Congress frequently pays the expenses of defending its Members even as to acts wholly outside the legislative function. 8 This Court had no trouble reconciling Gravel with Kilbourn v. Thompson, 103 U.S. 168, 26 L.Ed. 377 (1881). In Kilbourn the Sergeant-at-Arms of the House of Representatives was held not to share the absolute immunity enjoyed by the Members of Congress who ordered that officer to act.

78

457 U.S. 731 102 S.Ct. 2690 73 L.Ed.2d 349 Richard NIXON, Petitionerv.A. Ernest FITZGERALD. No. 79-1738. Argued Nov. 30, 1981. Decided June 24, 1982. Syllabus During the waning months of the Presidency of Lyndon B. Johnson in 1968, respondent, a management analyst with the Department of the Air Force, testified before a congressional Subcommittee about cost overruns and unexpected technical difficulties concerning the development of a particular airplane. In January 1970, during the Presidency of petitioner Richard M. Nixon, respondent was dismissed from his job during a departmental reorganization and reduction in force, in which his job was eliminated. Respondent complained to the Civil Service Commission, alleging that his separation represented unlawful retaliation for his congressional testimony. The Commission rejected this claim, but concluded that respondent's dismissal offended applicable regulations because it was motivated by "reasons purely personal to" respondent. Respondent thereafter filed suit for damages in Federal District Court against various Defense Department officials and White House aides allegedly responsible for his dismissal. An amended complaint later named petitioner as a defendant. After earlier judicial rulings and extensive pretrial discovery, only three defendants were involved: petitioner and two White House aides (petitioners in Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396). Denying the defendants' motion for summary judgment, the court held that respondent had stated triable causes of action under two federal statutes and the First Amendment, and that petitioner was not entitled to claim absolute Presidential immunity. Petitioner took a collateral appeal of the immunity decision to the Court of Appeals, which dismissed summarily. Held: 1. This Court has jurisdiction to determine the immunity question. Pp. 741-744. (a) The case was "in" the Court of Appeals for purposes of 28 U.S.C. § 1254, which authorizes this Court's review of "[c]ases in" the courts of appeals. The Court of Appeals here dismissed the appeal for lack of jurisdiction. However, petitioner's appeal to the Court of Appeals falls within the "collateral order" doctrine of Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528, as raising a "serious and unsettled question" of law. Although the Court of Appeals had previously ruled in another case that the President was not entitled to absolute immunity, this Court had never so held. Pp. 741-743. (b) Nor was the controversy mooted by an agreement to liquidate damages entered into between the parties after the petition for certiorari was filed and respondent had entered his opposition. Under the terms of the agreement, petitioner paid respondent $142,000; respondent agreed to accept liquidated damages of $28,000 if this Court ruled that petitioner was not entitled to absolute immunity; and no further payments would be made if the decision upheld petitioner's immunity claim. The limited agreement left both parties with a considerable financial stake in the resolution of the question presented in this Court. Cf. Havens Realty Corp. v. Coleman, 455 U.S. 363, 102 S.Ct. 1114, 71 L.Ed.2d 214. Pp. 743-744. 2. Petitioner, as a former President of the United States, is entitled to absolute immunity from damages liability predicated on his official acts. Pp. 744-758 (a) Although there is no blanket recognition of absolute immunity for all federal executive officials from liability for civil damages resulting from constitutional violations, certain officials—such as judges and prosecutors—because of the special nature of their responsibilities, require absolute exemption from liability. Cf. Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895. Determination of the immunity of particular officials is guided by the Constitution, federal statutes, history, and public policy. Pp. 744-748 (b) The President's absolute immunity is a functionally mandated incident of his unique office, rooted in the constitutional tradition of the separation of powers and supported by the Nation's history. Because of the singular importance of the President's duties, diversion of his energies by concern with private lawsuits would raise unique risks to the effective functioning of government. While the separation-of-powers doctrine does not bar every exercise of jurisdiction over the President, a court, before exercising jurisdiction, must balance the constitutional weight of the interest to be served against the dangers of intrusion on the authority and functions of the Executive Branch. The exercise of jurisdiction is not warranted in the case of merely private suits for damages based on a President's official acts. Pp. 748-754. (c) The President's absolute immunity extends to all acts within the "outer perimeter" of his duties of office. Pp.755-757 (d) A rule of absolute immunity for the President does not leave the Nation without sufficient protection against his misconduct. There remains the constitutional remedy of impeachment, as well as the deterrent effects of constant scrutiny by the press and vigilant oversight by Congress. Other incentives to avoid misconduct may include a desire to earn reelection, the need to maintain prestige as an element of Presidential influence, and a President's traditional concern for his historical stature. Pp. 757-758. Reversed and remanded. Herbert J. Miller, Jr., Washington, D. C., for petitioner Nixon. Elliot L. Richardson, Washington, D. C., for petitioners Harlow and Butterfield. John E. Nolan, Jr., Washington, D. C., for respondent. Justice POWELL delivered the opinion of the Court. 1 The plaintiff in this lawsuit seeks relief in civil damages from a former President of the United States. The claim rests on actions allegedly taken in the former President's official capacity during his tenure in office. The issue before us is the scope of the immunity possessed by the President of the United States. 2 * In January 1970 the respondent A. Ernest Fitzgerald lost his job as a management analyst with the Department of the Air Force. Fitzgerald's dismissal occurred in the context of a departmental reorganization and reduction in force, in which his job was eliminated. In announcing the reorganization, the Air Force characterized the action as taken to promote economy and efficiency in the Armed Forces. 3 Respondent's discharge attracted unusual attention in Congress and in the press. Fitzgerald had attained national prominence approximately one year earlier, during the waning months of the Presidency of Lyndon B. Johnson. On November 13, 1968, Fitzgerald appeared before the Subcommittee on Economy in Government of the Joint Economic Committee of the United States Congress. To the evident embarrassment of his superiors in the Department of Defense, Fitzgerald testified that cost-overruns on the C-5A transport plane could approximate $2 billion.1 He also revealed that unexpected technical difficulties had arisen during the development of the aircraft. 4 Concerned that Fitzgerald might have suffered retaliation for his congressional testimony, the Subcommittee on Economy in Government convened public hearings on Fitzgerald's dismissal.2 The press reported those hearings prominently, as it had the earlier announcement that his job was being eliminated by the Department of Defense. At a news conference on December 8, 1969, President Richard Nixon was queried about Fitzgerald's impending separation from Government service.3 The President responded by promising to look into the matter.4 Shortly after the news conference the petitioner asked White House Chief of Staff H. R. Haldeman to arrange for Fitzgerald's assignment to another job within the administration.5 It also appears that the President suggested to Budget Director Robert Mayo that Fitzgerald might be offered a position in the Bureau of the Budget.6 5 Fitzgerald's proposed reassignment encountered resistance within the administration.7 In an internal memorandum of January 20, 1970, White House aide Alexander Butterfield reported to Haldeman that " 'Fitzgerald is no doubt a top-notch cost expert, but he must be given very low marks in loyalty; and after all, loyalty is the name of the game.' "8 Butterfield therefore recommended that " '[W]e should let him bleed, for a while at least.' "9 There is no evidence of White House efforts to reemploy Fitzgerald subsequent to the Butterfield memorandum. 6 Absent any offer of alternative federal employment, Fitzgerald complained to the Civil Service Commission. In a letter of January 20, 1970, he alleged that his separation represented unlawful retaliation for his truthful testimony before a congressional Committee.10 The Commission convened a closed hearing on Fitzgerald's allegations on May 4, 1971. Fitzgerald, however, preferred to present his grievances in public. After he had brought suit and won an injunction, Fitzgerald v. Hampton, 152 U.S.App.D.C. 1, 467 F.2d 755 (1972), public hearings commenced on January 26, 1973. The hearings again generated publicity, much of it devoted to the testimony of Air Force Secretary Robert Seamans. Although he denied that Fitzgerald had lost his position in retaliation for congressional testimony, Seamans testified that he had received "some advice" from the White House before Fitzgerald's job was abolished.11 But the Secretary declined to be more specific. He responded to several questions by invoking "executive privilege."12 7 At a news conference on January 31, 1973, the President was asked about Mr. Seamans' testimony. Mr. Nixon took the opportunity to assume personal responsibility for Fitzgerald's dismissal: 8 "I was totally aware that Mr. Fitzgerald would be fired or discharged or asked to resign. I approved it and Mr. Seamans must have been talking to someone who had discussed the matter with me. No, this was not a case of some person down the line deciding he should go. It was a decision that was submitted to me. I made it and I stick by it."13 9 A day later, however, the White House press office issued a retraction of the President's statement. According to a press spokesman, the President had confused Fitzgerald with another former executive employee. On behalf of the President, the spokesman asserted that Mr. Nixon had not had "put before him the decision regarding Mr. Fitzgerald."14 10 After hearing over 4,000 pages of testimony, the Chief Examiner for the Civil Service Commission issued his decision in the Fitzgerald case on September 18, 1973. Decision on the Appeal of A. Ernest Fitzgerald, as reprinted in App. 60. The Examiner held that Fitzgerald's dismissal had offended applicable civil service regulations. Id., at 86a-87a.15 The Examiner based this conclusion on a finding that the departmental reorganization in which Fitzgerald lost his job, though purportedly implemented as an economy measure, was in fact motivated by "reasons purely personal to" respondent. Id., at 86a. As this was an impermissible basis for a reduction in force,16 the Examiner recommended Fitzgerald's reappointment to his old position or to a job of comparable authority.17 The Examiner, however, explicitly distinguished this narrow conclusion from a suggested finding that Fitzgerald had suffered retaliation for his testimony to Congress. As found by the Commission, "the evidence of record does not support [Fitzgerald's] allegation that his position was abolished and that he was separated . . . in retaliation for his having revealed the C-5A cost overrun in testimony before the Proxmire Committee on November 13, 1968." Id., at 81a. 11 Following the Commission's decision, Fitzgerald filed a suit for damages in the United States District Court. In it he raised essentially the same claims presented to the Civil Service Commission.18 As defendants he named eight officials of the Defense Department, White House aide Alexander Butterfield, and "one or More" unnamed "White House Aides" styled only as "John Does." 12 The District Court dismissed the action under the District of Columbia's 3-year statute of limitations, Fitzgerald v. Seamans, 384 F.Supp. 688 (DDC 1974), and the Court of Appeals affirmed as to all but one defendant, White House aide Alexander Butterfield, Fitzgerald v. Seamans, 180 U.S.App.D.C. 75, 553 F.2d 220 (1977). The Court of Appeals reasoned that Fitzgerald had no reason to suspect White House involvement in his dismissal at least until 1973. In that year, reasonable grounds for suspicion had arisen, most notably through publication of the internal White House memorandum in which Butterfield had recommended that Fitzgerald at least should be made to "bleed for a while" before being offered another job in the administration. Id., at 80, 84, 553 F.2d, at 225, 229. Holding that concealment of illegal activity would toll the statute of limitations, the Court of Appeals remanded the action against Butterfield for further proceedings in the District Court. 13 Following the remand and extensive discovery thereafter, Fitzgerald filed a second amended complaint in the District Court on July 5, 1978. It was in this amended complaint—more than eight years after he had complained of his discharge to the Civil Service Commission—that Fitzgerald first named the petitioner Nixon as a party defendant.19 Also included as defendants were White House aide Bryce Harlow and other officials of the Nixon administration. Additional discovery ensued. By March 1980, only three defendants remained: the petitioner Richard Nixon and White House aides Harlow and Butterfield. Denying a motion for summary judgment, the District Court ruled that the action must proceed to trial. Its order of March 26 held that Fitzgerald had stated triable causes of action under two federal statutes and the First Amendment to the Constitution.20 The court also ruled that petitioner was not entitled to claim absolute Presidential immunity. 14 Petitioner took a collateral appeal of the immunity decision to the Court of Appeals for the District of Columbia Circuit. The Court of Appeals dismissed summarily. It apparently did so on the ground that its recent decision in Halperin v. Kissinger, 196 U.S.App.D.C. 285, 606 F.2d 1192 (1979), aff'd in pertinent part by an equally divided Court, 452 U.S. 713, 101 S.Ct. 3132, 69 L.Ed.2d 367 (1981), had rejected this claimed immunity defense. 15 As this Court has not ruled on the scope of immunity available to a President of the United States, we granted certiorari to decide this important issue. 452 U.S. 959, 101 S.Ct. 3106, 69 L.Ed.2d 969 (1981). II 16 Before addressing the merits of this case, we must consider two challenges to our jurisdiction. In his opposition to the petition for certiorari, respondent argued that this Court is without jurisdiction to review the nonfinal order in which the District Court rejected petitioner's claim to absolute immunity.21 We also must consider an argument that an agreement between the parties has mooted the controversy. 17 Petitioner invokes the jurisdiction of this Court under 28 U.S.C. § 1254, a statute that invests us with authority to review "[c]ases in" the courts of appeals.22 When the petitioner in this case sought review of an interlocutory order denying his claim to absolute immunity, the Court of Appeals dismissed the appeal for lack of jurisdiction. Emphasizing the "jurisdictional" basis for the Court of Appeals' decision, respondent argued that the District Court's order was not an appealable "case" properly "in" the Court of Appeals within the meaning of § 1254. We do not agree. 18 Under the "collateral order" doctrine of Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949), a small class of interlocutory orders are immediately appealable to the courts of appeals. As defined by Cohen, this class embraces orders that "conclusively determine the disputed question, resolve an important issue completely separate from the merits of the action, and [are] effectively unreviewable on appeal from a final judgment." Coopers & Lybrand v. Livesay, 437 U.S. 463, 468, 98 S.Ct. 2454, 2458, 57 L.Ed.2d 351 (1978); see Cohen, supra, 337 U.S., at 546-547, 69 S.Ct., at 1225-26. As an additional requirement, Cohen established that a collateral appeal of an interlocutory order must "presen[t] a serious and unsettled question." 337 U.S., at 547, 69 S.Ct., at 1226. At least twice before this Court has held that orders denying claims of absolute immunity are appealable under the Cohen criteria. See Helstoski v. Meanor, 442 U.S. 500, 99 S.Ct. 2445, 61 L.Ed.2d 30 (1979) (claim of immunity under the Speech and Debate Clause); Abney v. United States, 431 U.S. 651, 97 S.Ct. 2034, 52 L.Ed.2d 651 (1977) (claim of immunity under Double Jeopardy Clause). In previous cases the Court of Appeals for the District of Columbia Circuit also has treated orders denying absolute immunity as appealable under Cohen. See Briggs v. Goodwin, 186 U.S.App.D.C. 179, 227-229, 569 F.2d 10, 58-60 (1977) (Wilkey, J., dissenting on the appealability issue); McSurely v. McClellan, 172 U.S.App.D.C. 364, 372, 521 F.2d 1024, 1032 (1975), aff'd in pertinent part en banc, 180 U.S.App.D.C. 101, 107-108, n. 18, 553 F.2d 1277, 1283-1284, n. 18 (1976), cert. dism'd sub nom. McAdams v. McSurely, 438 U.S. 189, 98 S.Ct. 3116, 57 L.Ed.2d 704 (1978). 19 In "dismissing" the appeal in this case, the Court of Appeals appears to have reasoned that petitioner's appeal lay outside the Cohen doctrine because it raised no "serious and unsettled question" of law. This argument was pressed by the respondent, who asked the Court of Appeals to dismiss on the basis of that court's "controlling" decision in Halperin v. Kissinger, supra. 20 Under the circumstances of this case, we cannot agree that petitioner's interlocutory appeal failed to raise a "serious and unsettled" question. Although the Court of Appeals had ruled in Halperin v. Kissinger that the President was not entitled to absolute immunity, this Court never had so held. And a petition for certiorari in Halperin was pending in this Court at the time petitioner's appeal was dismissed. In light of the special solicitude due to claims alleging a threatened breach of essential Presidential prerogatives under the separation of powers, see United States v. Nixon, 418 U.S. 683, 691-692, 94 S.Ct. 3090, 3099-3100, 41 L.Ed.2d 1039 (1974), we conclude that petitioner did present a "serious and unsettled" and therefore appealable question to the Court of Appeals. It follows that the case was "in" the Court of Appeals under § 1254 and properly within our certiorari jurisdiction.23 B 21 Shortly after petitioner had filed his petition for certiorari in this Court and respondent had entered his opposition, the parties reached an agreement to liquidate damages.24 Under its terms the petitioner Nixon paid the respondent Fitzgerald a sum of $142,000. In consideration, Fitzgerald agreed to accept liquidated damages of $28,000 in the event of a ruling by this Court that petitioner was not entitled to absolute immunity. In case of a decision upholding petitioner's immunity claim, no further payments would be made. 22 The limited agreement between the parties left both petitioner and respondent with a considerable financial stake in the resolution of the question presented in this Court. As we recently concluded in a case involving a similar contract: "Given respondents' continued active pursuit of monetary relief, this case remains 'definite and concrete, touching the legal relations of parties having adverse legal interests.' " Havens Realty Corp. v. Coleman, 455 U.S. 363, 371, 102 S.Ct. 1114, 1120, 71 L.Ed.2d 214 (1982), quoting Aetna Life Ins. Co. v. Haworth, 300 U.S. 227, 240-241, 57 S.Ct. 461, 463-64, 81 L.Ed. 617 (1937). III A. 23 This Court consistently has recognized that government officials are entitled to some form of immunity from suits for civil damages. In Spalding v. Vilas, 161 U.S. 483, 16 S.Ct. 631, 40 L.Ed. 780 (1896), the Court considered the immunity available to the Postmaster General in a suit for damages based upon his official acts. Drawing upon principles of immunity developed in English cases at common law, the Court concluded that "[t]he interests of the people" required a grant of absolute immunity to public officers. Id., at 498, 16 S.Ct., at 637. In the absence of immunity, the Court reasoned, executive officials would hesitate to exercise their discretion in a way "injuriously affect[ing] the claims of particular individuals," id., at 499, 16 S.Ct., at 637, even when the public interest required bold and unhesitating action. Considerations of "public policy and convenience" therefore compelled a judicial recognition of immunity from suits arising from official acts. 24 "In exercising the functions of his office, the head of an Executive Department, keeping within the limits of his authority, should not be under an apprehension that the motives that control his official conduct may, at any time, become the subject of inquiry in a civil suit for damages. It would seriously cripple the proper and effective administration of public affairs as entrusted to the executive branch of the government, if he were subjected to any such restraint." Id., at 498, 16 S.Ct., at 637. 25 Decisions subsequent to Spalding have extended the defense of immunity to actions besides those at common law. In Tenney v. Brandhove, 341 U.S. 367, 71 S.Ct. 783, 95 L.Ed. 1019 (1951), the Court considered whether the passage of 42 U.S.C. § 1983, which made no express provision for immunity for any official, had abrogated the privilege accorded to state legislators at common law. Tenney held that it had not. Examining § 1983 in light of the "presuppositions of our political history" and our heritage of legislative freedom, the Court found it incredible "that Congress . . . would impinge on a tradition so well grounded in history and reason" without some indication of intent more explicit than the general language of the statute. Id., at 376, 71 S.Ct., at 788. Similarly, the decision in Pierson v. Ray, 386 U.S. 547, 87 S.Ct. 1213, 18 L.Ed.2d 288 (1967), involving a § 1983 suit against a state judge, recognized the continued validity of the absolute immunity of judges for acts within the judicial role. This was a doctrine " 'not for the protection or benefit of a malicious or corrupt judge, but for the benefit of the public, whose interest it is that the judges should be at liberty to exercise their functions with independence and without fear of consequences.' " Id., at 554, 87 S.Ct., at 1218, quoting Scott v. Stansfield, L.R. 3 Ex. 220, 223 (1868). See Bradley v. Fisher, 13 Wall. 335, 20 L.Ed. 646 (1872). The Court in Pierson also held that police officers are entitled to a qualified immunity protecting them from suit when their official acts are performed in "good faith." 386 U.S. at 557, 87 S.Ct., at 1219. 26 In Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974), the Court considered the immunity available to state executive officials in a § 1983 suit alleging the violation of constitutional rights. In that case we rejected the officials' claim to absolute immunity under the doctrine of Spalding v. Vilas, finding instead that state executive officials possessed a "good faith" immunity from § 1983 suits alleging constitutional violations. Balancing the purposes of § 1983 against the imperatives of public policy, the Court held that "in varying scope, a qualified immunity is available to officers of the executive branch of government, the variation being dependent upon the scope of discretion and responsibilities of the office and all the circumstances as they reasonably appeared at the time of the action on which liability is sought to be based." 416 U.S., at 247, 94 S.Ct., at 1692. 27 As construed by subsequent cases, Scheuer established a two-tiered division of immunity defenses in § 1983 suits. To most executive officers Scheuer accorded qualified immunity. For them the scope of the defense varied in proportion to the nature of their official functions and the range of decisions that conceivably might be taken in "good faith." This "functional" approach also defined a second tier, however, at which the especially sensitive duties of certain officials—notably judges and prosecutors—required the continued recognition of absolute immunity. See, e.g., Imbler v. Pachtman, 424 U.S. 409, 96 S.Ct. 984, 47 L.Ed.2d 128 (1976) (state prosecutors possess absolute immunity with respect to the initiation and pursuit of prosecutions); Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978) (state judge possesses absolute immunity for all judicial acts). 28 This approach was reviewed in detail in Butz v. Econo mou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), when we considered for the first time the kind of immunity possessed by federal executive officials who are sued for constitutional violations.25 In Butz the Court rejected an argument, based on decisions involving federal officials charged with common-law torts, that all high federal officials have a right to absolute immunity from constitutional damages actions. Concluding that a blanket recognition of absolute immunity would be anomalous in light of the qualified immunity standard applied to state executive officials, id., at 504, 98 S.Ct., at 2909, we held that federal officials generally have the same qualified immunity possessed by state officials in cases under § 1983. In so doing we reaffirmed our holdings that some officials, notably judges and prosecutors, "because of the special nature of their responsibilities," id., at 511, 98 S.Ct., at 2913, "require a full exemption from liability." Id., at 508, 98 S.Ct., at 2911. InButz itself we upheld a claim of absolute immunity for administrative officials engaged in functions analogous to those of judges and prosecutors. Ibid. We also left open the question whether other federal officials could show that "public policy requires an exemption of that scope." Id., at 506, 98 S.Ct., at 2910. B 29 Our decisions concerning the immunity of government officials from civil damages liability have been guided by the Constitution, federal statutes, and history. Additionally, at least in the absence of explicit constitutional or congressional guidance, our immunity decisions have been informed by the common law. See Butz v. Economou, supra, at 508, 98 S.Ct., at 2911; Imbler v. Pachtman, supra, at 421, 96 S.Ct., at 990. This Court necessarily also has weighed concerns of public policy, especially as illuminated by our history and the structure of our government. See, e.g., Butz v. Economou, supra, at 508, 98 S.Ct., at 2911; Imbler v. Pachtman, supra, at 421, 96 S.Ct., at 990; Spalding v. Vilas, 161 U.S., at 498, 16 S.Ct., at 637.26 30 This case now presents the claim that the President of the United States is shielded by absolute immunity from civil damages liability. In the case of the President the inquiries into history and policy, though mandated independently by our cases, tend to converge. Because the Presidency did not exist through most of the development of common law, any historical analysis must draw its evidence primarily from our constitutional heritage and structure. Historical inquiry thus merges almost at its inception with the kind of "public policy" analysis appropriately undertaken by a federal court. This inquiry involves policies and principles that may be considered implicit in the nature of the President's office in a system structured to achieve effective government under a constitutionally mandated separation of powers. IV 31 Here a former President asserts his immunity from civil damages claims of two kinds. He stands named as a defendant in a direct action under the Constitution and in two statutory actions under federal laws of general applicability. In neither case has Congress taken express legislative action to subject the President to civil liability for his official acts.27 32 Applying the principles of our cases to claims of this kind, we hold that petitioner, as a former President of the United States, is entitled to absolute immunity from damages liability predicated on his official acts. We consider this immunity a functionally mandated incident of the President's unique office, rooted in the constitutional tradition of the separation of powers and supported by our history. Justice Story's analysis remains persuasive: 33 "There are . . . incidental powers, belonging to the executive department, which are necessarily implied from the nature of the functions, which are confided to it. Among these, must necessarily be included the power to perform them . . . . The president cannot, therefore, be liable to arrest, imprisonment, or detention, while he is in the discharge of the duties of his office; and for this purpose his person must be deemed, in civil cases at least, to possess an official inviolability." 3 J. Story, Commentaries on the Constitution of the United States § 1563, pp. 418-419 (1st ed. 1833). A. 34 The President occupies a unique position in the constitutional scheme. Article II, § 1, of the Constitution provides that "[t]he executive Power shall be vested in a President of the United States. . . ." This grant of authority establishes the President as the chief constitutional officer of the Executive Branch, entrusted with supervisory and policy responsibilities of utmost discretion and sensitivity. These include the enforcement of federal law—it is the President who is charged constitutionally to "take Care that the Laws be faithfully executed";28 the conduct of foreign affairs—a realm in which the Court has recognized that "[i]t would be intolerable that courts, without the relevant information, should review and perhaps nullify actions of the Executive taken on information properly held secret";29 and management of the Executive Branch—a task for which "imperative reasons requir[e] an unrestricted power [in the President] to remove the most important of his subordinates in their most important duties."30 35 In arguing that the President is entitled only to qualified immunity, the respondent relies on cases in which we have recognized immunity of this scope for governors and cabinet officers. E.g., Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978); Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974). We find these cases to be inapposite. The President's unique status under the Constitution distinguishes him from other executive officials.31 Because of the singular importance of the President's duties, diversion of his energies by concern with private lawsuits would raise unique risks to the effective functioning of government. As is the case with prosecutors and judges— for whom absolute immunity now is established—a President must concern himself with matters likely to "arouse the most intense feelings." Pierson v. Ray, 386 U.S., at 554, 87 S.Ct., at 1218. Yet, as our decisions have recognized, it is in precisely such cases that there exists the greatest public interest in providing an official "the maximum ability to deal fearlessly and impartially with" the duties of his office. Ferri v. Ackerman, 444 U.S. 193, 203, 100 S.Ct. 402, 408, 62 L.Ed.2d 355 (1979). This concern is compelling where the officeholder must make the most sensitive and far-reaching decisions entrusted to any official under our constitutional system.32 Nor can the sheer prominence of the President's office be ignored. In view of the visibility of his office and the effect of his actions on countless people, the President would be an easily identifiable target for suits for civil damages.33 Cognizance of this personal vulnerability frequently could distract a President from his public duties, to the detriment of not only the President and his office but also the Nation that the Presidency was designed to serve. B 36 Courts traditionally have recognized the President's constitutional responsibilities and status as factors counseling judicial deference and restraint.34 For example, while courts generally have looked to the common law to determine the scope of an official's evidentiary privilege,35 we have recognized that the Presidential privilege is "rooted in the separation of powers under the Constitution." United States v. Nixon, 418 U.S., at 708, 94 S.Ct., at 3107. It is settled law that the separation-of-powers doctrine does not bar every exercise of jurisdiction over the President of the United States. See, e.g., United States v. Nixon, supra; United States v. Burr, 25 F.Cas. 187, 191, 196 (No.14,694) (CC Va.1807); cf. Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 72 S.Ct. 863, 96 L.Ed. 1153 (1952).36 But our cases also have established that a court, before exercising jurisdiction, must balance the constitutional weight of the interest to be served against the dangers of intrusion on the authority and functions of the Executive Branch. See Nixon v. Administrator of General Services, 433 U.S. 425, 443, 97 S.Ct. 2777, 2790, 53 L.Ed.2d 867 (1977); United States v. Nixon, supra, at 703-713, 94 S.Ct., at 3105-3110. When judicial action is needed to serve broad public interests—as when the Court acts, not in derogation of the separation of powers, but to maintain their proper balance, cf. Youngstown Sheet & Tube Co. v. Sawyer, supra, or to vindicate the public interest in an ongoing criminal prosecution, see United States v. Nixon, supra —the exercise of jurisdiction has been held warranted. In the case of this merely private suit for damages based on a President's official acts, we hold it is not.37 C 37 In defining the scope of an official's absolute privilege, this Court has recognized that the sphere of protected action must be related closely to the immunity's justifying purposes. Frequently our decisions have held that an official's absolute immunity should extend only to acts in performance of particular functions of his office. See Butz v. Economou, 438 U.S., at 508-517, 98 S.Ct., at 2911-2916; cf. Imbler v. Pachtman, 424 U.S., at 430-431, 96 S.Ct., at 994-995. But the Court also has refused to draw functional lines finer than history and reason would support. See, e.g., Spalding v. Vilas, 161 U.S., at 498, 16 S.Ct., at 637 (privilege extends to all matters "committed by law to [an official's] control or supervision"); Barr v. Matteo, 360 U.S. 564, 575, 79 S.Ct. 1335, 1341, 3 L.Ed.2d 1434 (1959) (fact "that the action here taken was within the outer perimeter of petitioner's line of duty is enough to render the privilege applicable . . ."); Stump v. Sparkman, 435 U.S., at 363 and n. 12, 98 S.Ct., at 1108 and n. 12 (judicial privilege applies even to acts occurring outside "the normal attributes of a judicial proceeding"). In view of the special nature of the President's constitutional office and functions, we think it appropriate to recognize absolute Presidential immunity from damages liability for acts within the "outer perimeter" of his official responsibility. 38 Under the Constitution and laws of the United States the President has discretionary responsibilities in a broad variety of areas, many of them highly sensitive. In many cases it would be difficult to determine which of the President's innumerable "functions" encompassed a particular action. In this case, for example, respondent argues that he was dismissed in retaliation for his testimony to Congress—a violation of 5 U.S.C. § 7211 (1976 ed., Supp.IV) and 18 U.S.C. § 1505. The Air Force, however, has claimed that the underlying reorganization was undertaken to promote efficiency. Assuming that petitioner Nixon ordered the reorganization in which respondent lost his job, an inquiry into the President's motives could not be avoided under the kind of "functional" theory asserted both by respondent and the dissent. Inquiries of this kind could be highly intrusive. 39 Here respondent argues that petitioner Nixon would have acted outside the outer perimeter of his duties by ordering the discharge of an employee who was lawfully entitled to retain his job in the absence of " 'such cause as will promote the efficiency of the service.' " Brief for Respondent 39, citing 5 U.S.C. § 7512(a). Because Congress has granted this legislative protection, respondent argues, no federal official could, within the outer perimeter of his duties of office, cause Fitzgerald to be dismissed without satisfying this standard in prescribed statutory proceedings. 40 This construction would subject the President to trial on virtually every allegation that an action was unlawful, or was taken for a forbidden purpose. Adoption of this construction thus would deprive absolute immunity of its intended effect. It clearly is within the President's constitutional and statutory authority to prescribe the manner in which the Secretary will conduct the business of the Air Force. See 10 U.S.C. § 8012(b). Because this mandate of office must include the authority to prescribe reorganizations and reductions in force, we conclude that petitioner's alleged wrongful acts lay well within the outer perimeter of his authority. V 41 A rule of absolute immunity for the President will not leave the Nation without sufficient protection against misconduct on the part of the Chief Executive.38 There remains the constitutional remedy of impeachment.39 In addition, there are formal and informal checks on Presidential action that do not apply with equal force to other executive officials. The President is subjected to constant scrutiny by the press. Vigilant oversight by Congress also may serve to deter Presidential abuses of office, as well as to make credible the threat of impeachment.40 Other incentives to avoid misconduct may include a desire to earn reelection, the need to maintain prestige as an element of Presidential influence, and a President's traditional concern for his historical stature. 42 The existence of alternative remedies and deterrents establishes that absolute immunity will not place the President "above the law."41 For the President, as for judges and prosecutors, absolute immunity merely precludes a particular private remedy for alleged misconduct in order to advance compelling public ends. VI 43 For the reasons stated in this opinion, the decision of the Court of Appeals is reversed, and the case is remanded for action consistent with this opinion. 44 So ordered. 45 Chief Justice BURGER, concurring. 46 I join the Court's opinion, but I write separately to underscore that the Presidential immunity derives from and is mandated by the constitutional doctrine of separation of powers. Indeed, it has been taken for granted for nearly two centuries.1 In reaching this conclusion we do well to bear in mind that the focus must not be simply on the matter of judging individual conduct in a fact-bound setting; rather, in those familiar terms of John Marshall, it is a Constitution we are expounding. Constitutional adjudication often bears unpalatable fruit. But the needs of a system of government sometimes must outweigh the right of individuals to collect damages. 47 It strains the meaning of the words used to say this places a President "above the law." United States v. Nixon, 418 U.S. 683, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974). The dissents are wide of the mark to the extent that they imply that the Court today recognizes sweeping immunity for a President for all acts. The Court does no such thing. The immunity is limited to civil damages claims. Moreover, a President, like Members of Congress, judges, prosecutors, or congressional aides—all having absolute immunity—are not immune for acts outside official duties.2 Ante, at 753-755. Even the broad immunity of the Speech and Debate Clause has its limits.3 48 In his dissenting opinion, Justice WHITE confuses "judicial process" in the subpoena sense with a civil damages suit. Post, at 778, n. 23. He quotes language from United States v. Nixon, supra, at 706, 94 S.Ct., at 3106, as though that language has some relevance to the matter of immunity from civil damages: 49 "[N]either the doctrine of separation of powers, nor the need for confidentiality . . . without more, can sustain an absolute, unqualified Presidential privilege of immunity from judicial process under all circumstances." Post, at 782. (Emphasis added.) 50 First, it is important to remember that the context of that language is a criminal prosecution. Second, the "judicial process" referred to was, as in United States v. Burr, 25 F.Cas. 30, No.14,692d (CC Va 1807) (Marshall, C.J., sitting at trial as Circuit Justice), a subpoena to the President to produce relevant evidence in a criminal prosecution. No issue of damages immunity was involved either in Burr v. United States v. Nixon. In short, the quoted language has no bearing whatever on a civil action for damages. It is one thing to say that a President must produce evidence relevant to a criminal case, as in Burr and United States v. Nixon, and quite another to say a President can be held for civil damages for dismissing a federal employee. If the dismissal is wrongful the employee can be reinstated with backpay, as was done here. See n. 5, infra. 51 The immunity of a President from civil suits is not simply a doctrine derived from this Court's interpretation of common law or public policy. Absolute immunity for a President for acts within the official duties of the Chief Executive is either to be found in the constitutional separation of powers or it does not exist. The Court today holds that the Constitution mandates such immunity and I agree. 52 The essential purpose of the separation of powers is to allow for independent functioning of each coequal branch of government within its assigned sphere of responsibility, free from risk of control, interference, or intimidation by other branches. United States v. Nixon, supra; Gravel v. United States, 408 U.S. 606, 617, 92 S.Ct. 2614, 2623, 33 L.Ed.2d 583 (1972). Even prior to the adoption of our Constitution, as well as after, judicial review of legislative action was recognized in some instances as necessary to maintain the proper checks and balances. Den on the Dem. of Bayard & Wife v. Singleton, 3 N.C. 42 (1787); Cases of the Judges of the Court of Appeals, 8 Va. 135 (1788). Cf. Marbury v. Madison, 1 Cranch 137, 2 L.Ed. 60 (1803). However, the Judiciary always must be hesitant to probe into the elements of Presidential decisionmaking, just as other branches should be hesitant to probe into judicial decision-making. Such judicial intervention is not to be tolerated absent imperative constitutional necessity. United States v. Nixon, supra, at 709-716, 94 S.Ct., at 3108-3111.4 The Court's opinion correctly observes that judicial intrusion through private damages actions improperly impinges on and hence interferes with the independence that is imperative to the functioning of the office of a President. 53 Exposing a President to civil damages actions for official acts within the scope of the Executive authority would inevitably subject Presidential actions to undue judicial scrutiny as well as subject the President to harassment. The enormous range and impact of Presidential decisions—far beyond that of any one Member of Congress—inescapably means that many persons will consider themselves aggrieved by such acts. Absent absolute immunity, every person who feels aggrieved would be free to bring a suit for damages, and each suit—especially those that proceed on the merits would involve some judicial questioning of Presidential acts, including the reasons for the decision, how it was arrived at, the information on which it was based, and who supplied the information. Such scrutiny of day-to-day decisions of the Executive Branch would be bound to occur if civil damages actions were made available to private individuals. Although the individual who claims wrongful conduct may indeed have sustained some injury, the need to prevent large-scale invasion of the Executive function by the Judiciary far outweighs the need to vindicate the private claims. We have decided that in a similar sense Members of both Houses of Congress—and their aides—must be totally free from judicial scrutiny for legislative acts; the public interest, in other words, outweighs the need for private redress of one claiming injury from legislative acts of a Member or aide of a Member.5 The Court's concern (and the even more emphatic concerns expressed by Justice WHITE's dissent) over "unremedied wrongs" to citizens by a President seem odd when one compares the potential for "wrongs" which thousands of congressional aides, prosecutors, and judges can theoretically inflict—with absolute immunity—on the same citizens for whom this concern is expressed. See n. 2, supra. 54 Judicial intervention would also inevitably inhibit the processes of Executive Branch decisionmaking and impede the functioning of the Office of the President. The need to defend damages suits would have the serious effect of diverting the attention of a President from his executive duties since defending a lawsuit today—even a lawsuit ultimately found to be frivolous often requires significant expenditures of time and money, as many former public officials have learned to their sorrow. This very case graphically illustrates the point. When litigation processes are not tightly controlled—and often they are not—they can be and are used as mechanisms of extortion. Ultimate vindication on the merits does not repair the damage.6 55 I fully agree that the constitutional concept of separation of independent coequal powers dictates that a President be immune from civil damages actions based on acts within the scope of Executive authority while in office.7 Far from placing a President above the law, the Court's holding places a President on essentially the same footing with judges and other officials whose absolute immunity we have recognized. 56 Justice WHITE, with whom Justice BRENNAN, Justice MARSHALL, and Justice BLACKMUN join, dissenting. 57 The four dissenting Members of the Court in Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), argued that all federal officials are entitled to absolute immunity from suit for any action they take in connection with their official duties. That immunity would extend even to actions taken with express knowledge that the conduct was clearly contrary to the controlling statute or clearly violative of the Constitution. Fortunately, the majority of the Court rejected that approach: We held that although public officials perform certain functions that entitle them to absolute immunity, the immunity attaches to particular functions—not to particular offices. Officials performing functions for which immunity is not absolute enjoy qualified immunity; they are liable in damages only if their conduct violated well-established law and if they should have realized that their conduct was illegal. 58 The Court now applies the dissenting view in Butz to the Office of the President: A President, acting within the outer boundaries of what Presidents normally do, may, without liability, deliberately cause serious injury to any number of citizens even though he knows his conduct violates a statute or tramples on the constitutional rights of those who are injured. Even if the President in this case ordered Fitzgerald fired by means of a trumped-up reduction in force, knowing that such a discharge was contrary to the civil service laws, he would be absolutely immune from suit. By the same token, if a President, without following the statutory procedures which he knows apply to himself as well as to other federal officials, orders his subordinates to wiretap or break into a home for the purpose of installing a listening device, and the officers comply with his request, the President would be absolutely immune from suit. He would be immune regardless of the damage he inflicts, regardless of how violative of the statute and of the Constitution he knew his conduct to be, and regardless of his purpose.1 59 The Court intimates that its decision is grounded in the Constitution. If that is the case, Congress cannot provide a remedy against Presidential misconduct and the criminal laws of the United States are wholly inapplicable to the President. I find this approach completely unacceptable. I do not agree that if the Office of President is to operate effectively, the holder of that Office must be permitted, without fear of liability and regardless of the function he is performing, deliberately to inflict injury on others by conduct that he knows violates the law. 60 We have not taken such a scatter-gun approach in other cases. Butz held that absolute immunity did not attach to the office held by a member of the President's Cabinet but only to those specific functions performed by that officer for which absolute immunity is clearly essential. Members of Congress are absolutely immune under the Speech or Debate Clause of the Constitution, but the immunity extends only to their legislative acts. We have never held that in order for legislative work to be done, it is necessary to immunize all of the tasks that legislators must perform. Constitutional immunity does not extend to those many things that Senators and Representatives regularly and necessarily do that are not legislative acts. Members of Congress, for example, repeatedly importune the executive branch and administrative agencies outside hearing rooms and legislative halls, but they are not immune if in connection with such activity they deliberately violate the law. United States v. Brewster, 408 U.S. 501, 92 S.Ct. 2531, 33 L.Ed.2d 507 (1972), for example, makes this clear. Neither is a Member of Congress or his aide immune from damages suits if in order to secure information deemed relevant to a legislative investigation, he breaks into a house and carries away records. Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972). Judges are absolutely immune from liability for damages, but only when performing a judicial function, and even then they are subject to criminal liability. See Dennis v. Sparks, 449 U.S. 24, 31, 101 S.Ct. 183, 188, 66 L.Ed.2d 185 (1980); O'Shea v. Littleton, 414 U.S. 488, 503, 94 S.Ct. 669, 679-80, 38 L.Ed.2d 674 (1974). The absolute immunity of prosecutors is likewise limited to the prosecutorial function. A prosecutor who directs that an investigation be carried out in a way that is patently illegal is not immune. 61 In Marbury v. Madison, 1 Cranch 137, 1675 (1803), the Court, speaking through The Chief Justice, observed that while there were "important political powers" committed to the President for the performance of which neither he nor his appointees were accountable in court, "the question, whether the legality of an act of the head of a department be examinable in a court of justice or not, must always depend on the nature of that act." 1 Cranch, 137, 165 (1803). The Court nevertheless refuses to follow this course with respect to the President. It makes no effort to distinguish categories of Presidential conduct that should be absolutely immune from other categories of conduct that should not qualify for that level of immunity. The Court instead concludes that whatever the President does and however contrary to law he knows his conduct to be, he may, without fear of liability, injure federal employees or any other person within or without the Government. 62 Attaching absolute immunity to the Office of the President, rather than to particular activities that the President might perform, places the President above the law. It is a reversion to the old notion that the King can do no wrong. Until now, this concept had survived in this country only in the form of sovereign immunity. That doctrine forecloses suit against the Government itself and against Government officials, but only when the suit against the latter actually seeks relief against the sovereign. Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 687, 69 S.Ct. 1457, 1460, 93 L.Ed. 1628 (1949). Suit against an officer, however, may be maintained where it seeks specific relief against him for conduct contrary to his statutory authority or to the Constitution. Id., at 698, 69 S.Ct., at 1465. Now, however, the Court clothes the Office of the President with sovereign immunity, placing it beyond the law.2 63 In Marbury v. Madison, supra, 1 Cranch at 163, The Chief Justice, speaking for the Court, observed: "The government of the United States has been emphatically termed a government of laws, and not of men. It will certainly cease to deserve this high appellation, if the laws furnish no remedy for the violation of a vested legal right." Until now, the Court has consistently adhered to this proposition. In Scheuer v. Rhodes, 416 U.S. 232, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974), a unanimous Court held that the Governor of a State was entitled only to a qualified immunity. We reached this position, even though we recognized that 64 "[i]n the case of higher officers of the executive branch . . . the inquiry is far more complex since the range of decisions and choices—whether the formulation of policy, of legislation, of budgets, or of day-to-day decisions—is virtually infinite. . . . In short, since the options which a chief executive and his principal subordinates must consider are far broader and far more subtle than those made by officials with less responsibility, the range of discretion must be comparably broad." Id., at 246-247, 94 S.Ct., at 1691-92. 65 As Justice BRENNAN observed in McGautha v. California, 402 U.S. 183, 252-253, 91 S.Ct. 1454, 1489-90, 28 L.Ed.2d 711 (1971) (dissenting opinion): "The principle that our Government shall be of laws and not of men is so strongly woven into our constitutional fabric that it has found recognition in not just one but several provisions of the Constitution" (footnote omitted). And as THE CHIEF JUSTICE said in Complete Auto Transit, Inc. v. Reis, 451 U.S. 401, 429, 101 S.Ct. 1836, 1851, 68 L.Ed.2d 248 (1981) (dissenting opinion): 66 "Accountability of each individual for individual conduct lies at the core of all law—indeed, of all organized societies. The trend to eliminate or modify sovereign immunity is not an unrelated development; we have moved away from 'The King can do no wrong.' This principle of individual accountability is fundamental if the structure of an organized society is not to be eroded to anarchy and impotence, and it remains essential in civil as well as criminal justice." 67 Unfortunately, the Court now abandons basic principles that have been powerful guides to decision. It is particularly unfortunate since the judgment in this case has few, if any, indicia of a judicial decision; it is almost wholly a policy choice, a choice that is without substantial support and that in all events is ambiguous in its reach and import. 68 We have previously stated that "the law of privilege as a defense to damages actions against officers of Government has 'in large part been of judicial making.' " Butz v. Economou, 438 U.S., at 501-502, 98 S.Ct., at 2908-2909, quoting Barr v. Matteo, 360 U.S. 564, 569, 79 S.Ct. 1335, 1338, 3 L.Ed.2d 1434 (1959). But this does not mean that the Court has simply "enacted" its own view of the best public policy. No doubt judicial convictions about public policy—whether and what kind of immunity is necessary or wise—have played a part, but the courts have been guided and constrained by common-law tradition, the relevant statutory background, and our constitutional structure and history. Our cases dealing with the immunity of Members of Congress are constructions of the Speech or Debate Clause and are guided by the history of such privileges at common law. The decisions dealing with the immunity of state officers involve the question of whether and to what extent Congress intended to abolish the common-law privileges by providing a remedy in the predecessor of 42 U.S.C. § 1983 for constitutional violations by state officials. Our decisions respecting immunity for federal officials—including absolute immunity for judges, prosecutors, and those officials doing similar work—also in large part reflect common-law views, as well as judicial conclusions as to what privileges are necessary if particular functions are to be performed in the public interest. 69 Unfortunately, there is little of this approach in the Court's decision today. The Court casually, but candidly, abandons the functional approach to immunity that has run through all of our decisions. Ante, at 2705. Indeed, the majority turns this rule on its head by declaring that because the functions of the President's office are so varied and diverse and some of them so profoundly important, the office is unique and must be clothed with officewide, absolute immunity. This is policy, not law, and in my view, very poor policy. 70 * In declaring the President to be absolutely immune from suit for any deliberate and knowing violation of the Constitution or of a federal statute, the Court asserts that the immunity is "rooted in the constitutional tradition of the separation of powers and supported by our history."3 Ante, at 749. The decision thus has all the earmarks of a constitutional pronouncement—absolute immunity for the President's office is mandated by the Constitution. Although the Court appears to disclaim this, ante at 748-749, n. 27, it is difficult to read the opinion coherently as standing for any narrower proposition: Attempts to subject the President to liability either by Congress through a statutory action or by the courts through a Bivens (Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971)) proceeding would violate the separation of powers.4 Such a generalized absolute immunity cannot be sustained when examined in the traditional manner and in light of the traditional judicial sources. 71 The petitioner and the United States, as amicus,5 rely principally on two arguments to support the claim of absolute immunity for the President from civil liability: absolute immunity is an "incidental power" of the Presidency, historically recognized as implicit in the Constitution, and absolute immunity is required by the separation-of-powers doctrine. I will address each of these contentions. 72 The Speech or Debate Clause, Art. I, § 6, guarantees absolute immunity to Members of Congress; nowhere, however, does the Constitution directly address the issue of Presidential immunity.6 Petitioner nevertheless argues that the debates at the Constitutional Convention and the early history of constitutional interpretation demonstrate an implicit assumption of absolute Presidential immunity. In support of this position, petitioner relies primarily on three separate items: First, preratification remarks made during the discussion of Presidential impeachment at the Convention and in The Federalist; second, remarks made during the meeting of the first Senate; and third, the views of Justice Story. 73 The debate at the Convention on whether or not the President should be impeachable did touch on the potential dangers of subjecting the President to the control of another branch, the Legislature.7 Gouverneur Morris, for example, complained of the potential for dependency and argued that "[the President] can do no criminal act without Coadjutors who may be punished. In case he should be re-elected, that will be sufficient proof of his innocence."8 Colonel Mason responded to this by asking if "any man [shall] be above Justice" and argued that this was least appropriate for the man "who can commit the most extensive injustice."9 Madison agreed that "it [is] indispensable that some provision should be made for defending the Community against the incapacity, negligence or perfidy of the chief Magistrate."10 Pinckney responded on the other side, believing that if granted the power, the Legislature would hold impeachment "as a rod over the Executive and by that means effectually destroy his independence."11 74 Petitioner concludes from this that the delegates meant impeachment to be the exclusive means of holding the President personally responsible for his misdeeds, outside of electoral politics. This conclusion, however, is hardly supported by the debate. Although some of the delegates expressed concern over limiting Presidential independence, the delegates voted 8 to 2 in favor of impeachment. Whatever the fear of subjecting the President to the power of another branch, it was not sufficient, or at least not sufficiently shared, to insulate the President from political liability in the impeachment process. 75 Moreover, the Convention debate did not focus on wrongs the President might commit against individuals, but rather on whether there should be a method of holding him accountable for what might be termed wrongs against the state.12 Thus, examples of the abuses that concerned delegates were betrayal, oppression, and bribery; the delegates feared that the alternative to an impeachment mechanism would be "tumults & insurrections" by the people in response to suchabuses. 2 Farrand 67. The only conclusions that can be drawn from this debate are that the independence of the Executive was not understood to require a total lack of accountability to the other branches and that there was no general desire to insulate the President from the consequences of his improper acts.13 76 Much the same can be said in response to petitioner's reliance on The Federalist No. 77. In that essay, Hamilton asked whether the Presidency combines "the requisites to safety in the republican sense—a due dependence on the people—a due responsibility." The Federalist No. 77, p. 520 (J. Cooke ed. 1961). He answered that the constitutional plan met this test because it subjected the President to both the electoral process and the possibility of impeachment, including subsequent criminal prosecution. Petitioner concludes from this that these were intended to be the exclusive means of restraining Presidential abuses. This, by no means follows. Hamilton was concerned in The Federalist No. 77, as were the delegates at the Convention, with the larger political abuses—"wrongs against the state"—that a President might commit. He did not consider what legal means might be available for redress of individualized grievances.14 77 That omission should not be taken to imply exclusion in these circumstances is well illustrated by comparing some of the remarks made in the state ratifying conventions with Hamilton's discussion in No. 77. In the North Carolina ratifying convention, for example, there was a discussion of the adequacy of the impeachment mechanism for holding executive officers accountable for their misdeeds. Governor Johnson defended the constitutional plan by distinguishing three legal mechanisms of accountability: 78 "If an officer commits an offence against an individual, he is amenable to the courts of law. If he commits crimes against the state, he may be indicted and punished. Impeachment only extends to high crimes and misdemeanors in a public office. It is a mode of trial pointed out for great misdemeanors against the public."15 79 Governor Johnson surely did not contemplate that the availability of an impeachment mechanism necessarily implied the exclusion of other forms of legal accountability; rather, the method of accountability was to be a function of the character of the wrong. Mr. Maclaine, another delegate to the North Carolina Convention, clearly believed that the courts would remain open to individual citizens seeking redress from injuries caused by Presidential acts: 80 "The President is the superior officer, who is to see the laws put in execution. He is amenable for any maladministration in his office. Were it possible to suppose that the President should give wrong instructions to his deputies, whereby the citizens would be distressed, they would have redress in the ordinary courts of common law."16 81 A similar distinction between different possible forms of Presidential accountability was drawn by Mr. Wilson at the Pennsylvania ratifying convention: 82 "[The President] is placed high, and is possessed of power far from being contemptible; yet not a single privilege is annexed to his character; far from being above the laws, he is amenable to them in his private character as a citizen, and in his public character by impeachment."17 83 There is no more reason to respect the views of Hamilton than those of Wilson: both were members of the Constitutional Convention; both were instrumental in securing the ratification of the Constitution. But more importantly, there is simply no express contradiction in their statements. Petitioner relies on an inference drawn from silence to create this contradiction. The surrounding history simply does not support this inference. 84 The second piece of historical evidence cited by petitioner is an exchange at the first meeting of the Senate, involving Vice President Adams and Senators Ellsworth and Maclay. The debate started over whether or not the words "the President" should be included at the beginning of federal writs, similar to the manner in which English writs ran in the King's name. Senator Maclay thought that this would improperly combine the executive and judicial branches. This, in turn, led to a discussion of the proper relation between the two. Senator Ellsworth and Vice President Adams defended the proposition that 85 "the President, personally, was not subject to any process whatever; could have no action, whatever, brought against him; was above the power of all judges, justices, &c. For [that] would . . . put it in the power of a common justice to exercise any authority over him, and stop the whole machine of government."18 86 In their view the impeachment process was the exclusive form of process available against the President. Senator Maclay ardently opposed this view and put the case of a President committing "murder in the street." In his view, in such a case neither impeachment nor resurrection were the exclusive means of holding the President to the law; rather, there was "loyal justice." Senator Maclay, who recorded the exchange, concludes his notes with the remark that none of this "is worth minuting, but it shows clearly how amazingly fond of the old leaven many people are."19 In his view, Senator Ellsworth and his supporters had not fully comprehended the difference in the political position of the American President and that of the British Monarch. Again, nothing more can be concluded from this than that the proper scope of Presidential accountability, including the question whether the President should be subject to judicial process, was no clearer then than it is now. 87 The final item cited by petitioner clearly supports his position, but is of such late date that it contributes little to understanding the original intent. In his Commentaries on the Constitution, published in 1833, Justice Story described the "incidental powers" of the President: 88 "Among these must necessarily be included the power to perform [his functions] without any obstruction or impediment whatsoever. The President cannot, therefore, be liable to arrest, imprisonment, or detention, while he is in the discharge of the duties of his office; and for this purpose his person must be deemed, in civil cases at least, to possess an official inviolability. In the exercise of his political powers he is to use his own discretion, and he is accountable only to his country and to his own conscience. His decision in relation to these powers is subject to no control, and his discretion, when exercised, is conclusive."20 89 While Justice Story may have been firmly committed to this view in 1833, Senator Pinckney, a delegate to the Convention, was as firmly committed to the opposite view in 1800.21 90 Senator Pinckney, arguing on the floor of the Senate, contrasted the privileges extended to Members of Congress by the Constitution with the lack of any such privileges extended to the President.22 He argued that this was a deliberate choice of the delegates to the Convention, who "well knew how oppressively the power of undefined privileges had been exercised in Great Britain, and were determined no such authority should ever be exercised here." 10 Annals of Cong. 72 (1800). Therefore, "[n]o privilege of this kind was intended for your Executive, nor any except that . . . for your Legislature." Id., at 74.23 91 In previous immunity cases the Court has emphasized the importance of the immunity afforded the particular government official at common law. See Imbler v. Pachtman, 424 U.S. 409, 421, 96 S.Ct. 984, 47 L.Ed.2d 128 (1976). Clearly this sort of analysis is not possible when dealing with an office, the Presidency, that did not exist at common law. To the extent that historical inquiry is appropriate in this context, it is constitutional history, not common law, that is relevant. From the history discussed above, however, all that can be concluded is that absolute immunity from civil liability for the President finds no support in constitutional text or history, or in the explanations of the earliest commentators. This is too weak a ground to support a declaration by this Court that the President is absolutely immune from civil liability, regardless of the source of liability or the injury for which redress is sought. This much the majority implicitly concedes since history and text, traditional sources of judicial argument, merit only a footnote in the Court's opinion. Ante, at 750-752, n. 31. B 92 No bright line can be drawn between arguments for absolute immunity based on the constitutional principle of separation of powers and arguments based on what the Court refers to as "public policy." This necessarily follows from the Court's functional interpretation of the separation-of-powers doctrine: 93 "[I]n determining whether the Act disrupts the proper balance between the coordinate branches, the proper inquiry focuses on the extent to which it prevents the Executive Branch from accomplishing its constitutionally assigned functions." Nixon v. Administrator of General Services, 433 U.S. 425, 443, 97 S.Ct. 2777, 2790, 53 L.Ed.2d 867 (1977). 94 See also United States v. Nixon, 418 U.S. 683, 706-707, 94 S.Ct. 3090, 3106-07, 41 L.Ed.2d 1039 (1974); Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 635, 72 S.Ct. 863, 870, 96 L.Ed. 1153 (1952) (Jackson, J., concurring). Petitioner argues that public policy favors absolute immunity because absent such immunity the President's ability to execute his constitutionally mandated obligations will be impaired. The convergence of these two lines of argument is superficially apparent from the very fact that in both instances the approach of the Court has been characterized as a "functional" analysis. 95 The difference is only one of degree. While absolute immunity might maximize executive efficiency and therefore be a worthwhile policy, lack of such immunity may not so disrupt the functioning of the Presidency as to violate the separation-of-powers doctrine. Insofar as liability in this case is of congressional origin, petitioner must demonstrate that subjecting the President to a private damages action will prevent him from "accomplishing [his] constitutionally assigned functions." Insofar as liability is based on a Bivens action, perhaps a lower standard of functional disruption is appropriate. Petitioner has surely not met the former burden; I do not believe that he has met the latter standard either. 96 Taken at face value, the Court's position that as a matter of constitutional law the President is absolutely immune should mean that he is immune not only from damages actions but also from suits for injunctive relief, criminal prosecutions and, indeed, from any kind of judicial process. But there is no contention that the President is immune from criminal prosecution in the courts under the criminal laws enacted by Congress or by the States for that matter. Nor would such a claim be credible. The Constitution itself provides that impeachment shall not bar "Indictment, Trial, Judgment and Punishment, according to Law." Art. I, § 3, cl. 7. Similarly, our cases indicate that immunity from damages actions carries no protection from criminal prosecution. Supra, at 765-766. 97 Neither can there be a serious claim that the separation-of-powers doctrine insulates Presidential action from judicial review or insulates the President from judicial process. No argument is made here that the President, whatever his liability for money damages, is not subject to the courts' injunctive powers. See, e.g., Youngstown Sheet & Tube Co., supra; Korematsu v. United States, 323 U.S. 214, 65 S.Ct. 193, 89 L.Ed. 194 (1944); Panama Refining Co. v. Ryan, 293 U.S. 388, 55 S.Ct. 241, 79 L.Ed. 446 (1935). Petitioner's attempt to draw an analogy to the Speech or Debate Clause, Brief for Petitioner 45, one purpose of which is "to prevent . . . accountability before a possibly hostile judiciary," Gravel v. United States, 408 U.S., at 617, 92 S.Ct., at 2623, breaks down at just this point. While the Speech or Debate Clause guarantees that "for any Speech or Debate" Congressmen "shall not be questioned in any other Place," and, thus, assures that Congressmen, in their official capacity, shall not be the subject of the courts' injunctive powers, no such protection is afforded the Executive. Indeed, as the cases cited above indicate, it is the rule, not the exception, that executive actions—including those taken at the immediate direction of the President—are subject to judicial review.24 Regardless of the possibility of money damages against the President, then, the constitutionality of the President's actions or their legality under the applicable statutes can and will be subject to review. Indeed, in this very case, respondent Fitzgerald's dismissal was set aside by the Civil Service Commission as contrary to the applicable regulations issued pursuant to authority granted by Congress. 98 Nor can private damages actions be distinguished on the ground that such claims would involve the President personally in the litigation in a way not necessitated by suits seeking declaratory or injunctive relief against certain Presidential actions. The President has been held to be subject to judicial process at least since 1807. United States v. Burr, 25 F.Cas. 30 (No. 14,692d) (CC Va.1807) (Marshall, C.J., sitting as Circuit Justice). Burr "squarely ruled that a subpoena may be directed to the President." Nixon v. Sirica, 159 U.S.App.D.C. 58, 67, 487 F.2d 700, 709 (1973). Chief Justice Marshall flatly rejected any suggestion that all judicial process, in and of itself, constitutes an unwarranted interference in the Presidency: "The guard, furnished to this high officer, to protect him from being harassed by vexatious and unnecessary subpoenas, is to be looked for in the conduct of a court after those subpoenas have issued; not in any circumstance which is to precede their being issued." 25 F.Cas., at 34 (emphasis added). 99 This position was recently rearticulated by the Court in United States v. Nixon, 418 U.S., at 706, 94 S.Ct., at 3106: 100 "[N]either the doctrine of separation of powers, nor the need for confidentiality . . . without more, can sustain an absolute, unqualified Presidential privilege of immunity from judicial process under all circumstances." 101 These two lines of cases establish, then, that neither subjecting Presidential actions to a judicial determination of their constitutionality, nor subjecting the President to judicial process violates the separation-of-powers doctrine. Similarly, neither has been held to be sufficiently intrusive to justify a judicially declared rule of immunity. With respect to intrusion by the judicial process itself on executive functions, subjecting the President to private claims for money damages involves no more than this. If there is a separation-of-powers problem here, it must be found in the nature of the remedy and not in the process involved. 102 We said in Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), that "it is not unfair to hold liable the official who knows or should know he is acting outside the law, and . . . insisting on an awareness of clearly established constitutional limits will not unduly interfere with the exercise of official judgment." Id., at 506-507, 98 S.Ct., at 2910-2911. Today's decision in Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396, makes clear that the President, were he subject to civil liability, could be held liable only for an action that he knew, or as an objective matter should have known, was illegal and a clear abuse of his authority and power. In such circumstances, the question that must be answered is who should bear the cost of the resulting injury—the wrongdoer or the victim. 103 The principle that should guide the Court in deciding this question was stated long ago by Chief Justice Marshall: "The very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury." Marbury v. Madison, 1 Cranch, at 163, 2 L.Ed. 60. Much more recently, the Court considered the role of a damages remedy in the performance of the courts' traditional function of enforcing federally guaranteed rights: "Historically, damages have been regarded as the ordinary remedy for an invasion of personal interests in liberty." Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S., at 395, 91 S.Ct., at 2004.25 To the extent that the Court denies an otherwise appropriate remedy, it denies the victim the right to be made whole and, therefore, denies him "the protection of the laws."26 104 That the President should have the same remedial obligations toward those whom he injures as any other federal officer is not a surprising proposition. The fairness of the remedial principle the Court has so far followed—that the wrongdoer, not the victim, should ordinarily bear the costs of the injury—has been found to be outweighed only in instances where potential liability is "thought to injure the governmental decisionmaking process." Imbler v. Pachtman, 424 U.S., at 437, 96 S.Ct., at 998 (WHITE, J., concurring in judgment). The argument for immunity is that the possibility of a damages action will, or at least should, have an effect on the performance of official responsibilities. That effect should be to deter unconstitutional, or otherwise illegal, behavior. This may, however, lead officers to be more careful and "less vigorous" in the performance of their duties. Caution, of course, is not always a virtue and undue caution is to be avoided. 105 The possibility of liability may, in some circumstances, distract officials from the performance of their duties and influence the performance of those duties in ways adverse to the public interest. But when this "public policy" argument in favor of absolute immunity is cast in these broad terms, it applies to all officers, both state and federal: All officers should perform their responsibilities without regard to those personal interests threatened by the possibility of a lawsuit. See Imbler, supra, at 436, 96 S.Ct. at 998 (WHITE, J., concurring in judgment).27 Inevitably, this reduces the public policy argument to nothing more than an expression of judicial inclination as to which officers should be encouraged to perform their functions with "vigor," although with less care.28 106 The Court's response, until today, to this problem has been to apply the argument to individual functions, not offices, and to evaluate the effect of liability on governmental decisionmaking within that function, in light of the substantive ends that are to be encouraged or discouraged. In this case, therefore, the Court should examine the functions implicated by the causes of action at issue here and the effect of potential liability on the performance of those functions. II 107 The functional approach to the separation-of-powers doctrine and the Court's more recent immunity decisions29 converge on the following principle: The scope of immunity is determined by function, not office. The wholesale claim that the President is entitled to absolute immunity in all of his actions stands on no firmer ground than did the claim that all Presidential communications are entitled to an absolute privilege, which was rejected in favor of a functional analysis, by a unanimous Court in United States v. Nixon, 418 U.S. 683, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974). Therefore, whatever may be true of the necessity of such a broad immunity in certain areas of executive responsibility,30 the only question that must be answered here is whether the dismissal of employees falls within a constitutionally assigned executive function, the performance of which would be substantially impaired by the possibility of a private action for damages. I believe it does not. 108 Respondent has so far proceeded in this action on the basis of three separate causes of action: two federal statutes—5 U.S.C. § 7211 (1976 ed., Supp.IV) and 18 U.S.C. § 1505—and the First Amendment. At this point in the litigation, the availability of these causes of action is not before us. Assuming the correctness of the lower court's determination that the two federal statutes create a private right of action, I find the suggestion that the President is immune from those causes of action to be unconvincing. The attempt to found such immunity upon a separation-of-powers argument is particularly unconvincing. 109 The first of these statutes, 5 U.S.C. § 7211 (1976 ed., Supp.IV), states that "[t]he right of employees . . . to . . . furnish information to either House of Congress, or to a committee or Member thereof, may not be interfered with or denied." The second, 18 U.S.C. § 1505, makes it a crime to obstruct congressional testimony. It does not take much insight to see that at least one purpose of these statutes is to assure congressional access to information in the possession of the Executive Branch, which Congress believes it requires in order to carry out its responsibilities.31 Insofar as these statutes implicate a separation-of-powers argument, I would think it to be just the opposite of that suggested by petitioner and accepted by the majority. In enacting these statutes, Congress sought to preserve its own constitutionally mandated functions in the face of a recalcitrant Executive.32 Thus, the separation-of-powers problem addressed by these statutes was first of all Presidential behavior that intruded upon, or burdened, Congress' performance of its own constitutional responsibilities. It is no response to this to say that such a cause of action would disrupt the President in the furtherance of his responsibilities. That approach ignores the separation-of-powers problem that lies behind the congressional action; it assumes that Presidential functions are to be valued over congressional functions. 110 The argument that Congress, by providing a damages action under these statutes (as is assumed in this case), has adopted an unconstitutional means of furthering its ends, must rest on the premise that Presidential control of executive employment decisions is a constitutionally assigned Presidential function with which Congress may not significantly interfere. This is a frivolous contention. In United States v. Perkins, 116 U.S. 483, 485, 6 S.Ct. 449, 450, 29 L.Ed. 700 (1886), this Court held that "when Congress, by law, vests the appointment of inferior officers in the heads of Departments it may limit and restrict the power of removal as it deems best for the public interest." Whatever the rule may be with respect to high officers, see Humphrey's Executor v. United States, 295 U.S. 602, 55 S.Ct. 869, 79 L.Ed. 1611 (1935), with respect to those who fill traditional bureaucratic positions, restrictions on executive authority are the rule and not the exception.33 This case itself demonstrates the severe statutory restraints under which the President operates in this area. 111 Fitzgerald was a civil service employee working in the Office of the Secretary of the Air Force. Although his position was such as to fall within the "excepted" service, which would ordinarily mean that civil service rules and regulations applicable to removals would not protect him, 5 CFR § 6.4 (1982), his status as a veteran entitled him to special protections. Veterans are entitled to certain civil service benefits afforded to "preference eligibles." 5 U.S.C. § 2108 (1976 ed. and Supp.IV). These benefits include that set forth in 5 U.S.C. § 7513(a) (1976 ed., and Supp.IV): "[A]n agency may take [adverse action] against an employee only for such cause as will promote the efficiency of the service." Similarly, his veteran status entitled Fitzgerald to the protection of the reduction-in-force procedures established by civil service regulation. 5 U.S.C. §§ 3501, 3502 (1976 ed., and Supp.IV). It was precisely those procedures that the Chief Examiner for the Civil Service Commission found had been violated, in his 1973 recommendation that respondent be reappointed to his old position or to a job of comparable authority. 112 This brief review is enough to illustrate my point: Personnel decisions of the sort involved in this case are emphatically not a constitutionally assigned Presidential function that will tolerate no interference by either of the other two branches of Government. More important than this "quantitative" analysis of the degree of intrusion in Presidential decision-making permitted in this area, however, is the "qualitative" analysis suggested in Part I-B above. 113 Absolute immunity is appropriate when the threat of liability may bias the decisionmaker in ways that are adverse to the public interest. But as the various regulations and statutes protecting civil servants from arbitrary executive action illustrate, this is an area in which the public interest is demonstrably on the side of encouraging less "vigor" and more "caution" on the part of decisionmakers. That is, the very steps that Congress has taken to assure that executive employees will be able freely to testify in Congress and to assure that they will not be subject to arbitrary adverse actions indicate that those policy arguments that have elsewhere justified absolute immunity are not applicable here. Absolute immunity would be nothing more than a judicial declaration of policy that directly contradicts the policy of protecting civil servants reflected in the statutes and regulations. 114 If respondent could, in fact, have proceeded on his two statutory claims, the Bivens action would be superfluous. Respondent may not collect damages twice, and the same injuries are put forward by respondent as the basis for both the statutory and constitutional claims. As we have said before, "were Congress to create equally effective alternative remedies, the need for damages relief [directly under the Constitution] might be obviated." Davis v. Passman, 442 U.S. 228, 248, 99 S.Ct. 2264, 2278, 60 L.Ed.2d 846 (1979). Nevertheless, because the majority decides that the President is absolutely immune from a Bivens action as well, I shall express my disagreement with that conclusion. 115 In Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971), we held that individuals who have suffered a compensable injury through a violation of the rights guaranteed them by the Fourth Amendment may invoke the general federal-question jurisdiction of the federal courts in a suit for damages. That conclusion rested on two principles: First, " '[t]he very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws,' " id., at 397, 91 S.Ct., at 2005, quoting Marbury v. Madison, 1 Cranch, at 163, 2 L.Ed. 60; second, "[h]istorically, damages have been regarded as the ordinary remedy for an invasion of personal interests in liberty." 403 U.S., at 395, 91 S.Ct., at 2004. In Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978), we rejected the argument of the Federal Government that federal officers, including Cabinet officers, are absolutely immune from civil liability for such constitutional violations—a position that we recognized would substantially undercut our conclusion in Bivens. We held there that although the performance of certain limited functions will be protected by the shield of absolute immunity, the general rule is that federal officers, like state officers, have only a qualified immunity. Finally, in Davis v. Passman, supra, we held that a Congressman could be held liable for damages in a Bivens-type suit brought in federal court alleging a violation of individual rights guaranteed the plaintiff by the Due Process Clause. In my view, these cases have largely settled the issues raised by the Bivens problem here. 116 These cases established the following principles. First, it is not the exclusive prerogative of the Legislative Branch to create a federal cause of action for a constitutional violation. In the absence of adequate legislatively prescribed remedies, the general federal-question jurisdiction of the federal courts permits the courts to create remedies, both legal and equitable, appropriate to the character of the injury. Second, exercise of this "judicial" function does not create a separation-of-powers problem: We have held both executive and legislative officers subject to this judicially created cause of action and in each instance we have rejected separation-of-powers arguments. Holding federal officers liable for damages for constitutional injuries no more violates separation-of-powers principles than does imposing equitable remedies under the traditional function of judicial review. Third, federal officials will generally have a "qualified immunity" from such suits; absolute immunity will be extended to certain functions only on the basis of a showing that exposure to liability is inconsistent with the proper performance of the official's duties and responsibilities. Finally, Congress retains the power to restrict exposure to liability, and the policy judgments implicit in this decision should properly be made by Congress. 117 The majority fails to recognize the force of what the Court has already done in this area. Under the above principles, the President could not claim that there are no circumstances under which he would be subject to a Bivens-type action for violating respondent's constitutional rights. Rather, he must assert that the absence of absolute immunity will substantially impair his ability to carry out particular functions that are his constitutional responsibility. For the reasons I have presented above, I do not believe that this argument can be successfully made under the circumstances of this case. 118 It is, of course, theoretically possible that the President should be held to be absolutely immune because each of the functions for which he has constitutional responsibility would be substantially impaired by the possibility of civil liability. I do not think this argument is valid for the simple reason that the function involved here does not have this character. On which side of the line other Presidential functions would fall need not be decided in this case. 119 The majority opinion suggests a variant of this argument. It argues, not that every Presidential function has this character, but that distinguishing the particular functions involved in any given case would be "difficult." Ante, at 756.34 Even if this were true, it would not necessarily follow that the President is entitled to absolute immunity: That would still depend on whether, in those unclear instances, it is likely to be the case that one of the functions implicated deserves the protection of absolute immunity. In this particular case, I see no such function.35 120 I do not believe that subjecting the President to a Bivens action would create separation-of-powers problems or "public policy" problems different from those involved in subjecting the President to a statutory cause of action.36 Relying upon the history and text of the Constitution, as well as the analytic method of our prior cases, I conclude that these problems are not sufficient to justify absolute immunity for the President in general, nor under the circumstances of this case in particular. III 121 Because of the importance of this case, it is appropriate to examine the reasoning of the majority opinion. 122 The opinion suffers from serious ambiguity even with respect to the most fundamental point: How broad is the immunity granted the President? The opinion suggests that its scope is limited by the fact that under none of the asserted causes of action "has Congress taken express legislative action to subject the President to civil liability for his official acts." Ante, at 748. We are never told, however, how or why congressional action could make a difference. It is not apparent that any of the propositions relied upon by the majority to immunize the President would not apply equally to such a statutory cause of action; nor does the majority indicate what new principles would operate to undercut those propositions. 123 In the end, the majority seems to overcome its initial hesitation, for it announces that "[w]e consider [absolute] immunity a functionally mandated incident of the President's unique office, rooted in the constitutional tradition of the separation of powers and supported by our history," ante, at 749. See also ante, at 757 ("A rule of absolute immunity for the President will not leave the Nation without sufficient protection against misconduct on the part of the Chief Executive").37 While the majority opinion recognizes that "[i]t is settled law that the separation-of-powers doctrine does not bar every exercise of jurisdiction over the President of the United States," it bases its conclusion, at least in part, on a suggestion that there is a special jurisprudence of the Presidency. Ante, at 2704.38 124 But in United States v. Nixon, 418 U.S. 683, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974), we upheld the power of a Federal District Court to issue a subpoenaduces tecum against the President. In other cases we have enjoined executive officials from carrying out Presidential directives. See e.g., Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 72 S.Ct. 863, 96 L.Ed. 1153 (1952). Not until this case has there ever been a suggestion that the mere formalism of the name appearing on the complaint was more important in resolving separation-of-powers problems than the substantive character of the judicial intrusion upon executive functions. 125 The majority suggests that the separation-of-powers doctrine permits exercising jurisdiction over the President only in those instances where "judicial action is needed to serve broad public interests—as when the Court acts, not in derogation of the separation of powers, but to maintain their proper balance." Ante, at 754. Without explanation, the majority contends that a "merely private suit for damages" does not serve this function. Ibid. 126 The suggestion that enforcement of the rule of law—i.e., subjecting the President to rules of general applicability—does not further the separation of powers, but rather is in derogation of this purpose, is bizarre. At stake in a suit of this sort, to the extent that it is based upon a statutorily created cause of action, is the ability of Congress to assert legal restraints upon the Executive and of the courts to perform their function of providing redress for legal harm. Regardless of what the Court might think of the merits of Mr. Fitzgerald's claim, the idea that pursuit of legal redress offends the doctrine of separation of powers is a frivolous contention passing as legal argument. 127 Similarly, the majority implies that the assertion of a constitutional cause of action—the whole point of which is to assure that an officer does not transgress the constitutional limits on his authority—may offend separation-of-powers concerns. This is surely a perverse approach to the Constitution: Whatever the arguments in favor of absolute immunity may be, it is untenable to argue that subjecting the President to constitutional restrictions will undercut his "unique" role in our system of government. It cannot be seriously argued that the President must be placed beyond the law and beyond judicial enforcement of constitutional restraints upon executive officers in order to implement the principle of separation of powers. 128 Focusing on the actual arguments the majority offers for its holding of absolute immunity for the President, one finds surprisingly little. As I read the relevant section of the Court's opinion, I find just three contentions from which the majority draws this conclusion. Each of them is little more than a makeweight; together they hardly suffice to justify the wholesale disregard of our traditional approach to immunity questions. 129 First, the majority informs us that the President occupies a "unique position in the constitutional scheme," including responsibilities for the administration of justice, foreign affairs, and management of the Executive Branch. Ante, at 749-750. True as this may be, it says nothing about why a "unique" rule of immunity should apply to the President. The President's unique role may indeed encompass functions for which he is entitled to a claim of absolute immunity. It does not follow from that, however, that he is entitled to absolute immunity either in general or in this case in particular. 130 For some reason, the majority believes that this uniqueness of the President shifts the burden to respondent to prove that a rule of absolute immunity does not apply. The respondent has failed in this effort, the Court suggests, because the President's uniqueness makes "inapposite" any analogy to our cases dealing with other executive officers. Ante, at 750. Even if this were true, it would not follow that the President is entitled to absolute immunity; it would only mean that a particular argument is out of place. But the fact is that it is not true. There is nothing in the President's unique role that makes the arguments used in those other cases inappropriate. 131 Second, the majority contends that because the President's "visibility" makes him particularly vulnerable to suits for civil damages, ante, at 753, a rule of absolute immunity is required. The force of this argument is surely undercut by the majority's admission that "there is no historical record of numerous suits against the President." Ante, at 753, n. 33. Even granting that a Bivens cause of action did not become available until 1971, in the 11 years since then there have been only a handful of suits. Many of these are frivolous and dealt with in a routine manner by the courts and the Justice Department. There is no reason to think that, in the future, the protection afforded by summary judgment procedures would not be adequate to protect the President, as they currently protect other executive officers from unfounded litigation. Indeed, given the decision today in Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396 there is even more reason to believe that frivolous claims will not intrude upon the President's time. Even if judicial procedures were found not to be sufficient, Congress remains free to address this problem if and when it develops. 132 Finally, the Court suggests that potential liability "frequently could distract a President from his public duties." Ante, at 753. Unless one assumes that the President himself makes the countless high-level executive decisions required in the administration of government, this rule will not do much to insulate such decisions from the threat of liability. The logic of the proposition cannot be limited to the President; its extension, however, has been uniformly rejected by this Court. See Butz v. Economou, 438 U.S. 478, 98 S.Ct. 2984, 57 L.Ed.2d 985 (1978); Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396. Furthermore, in no instance have we previously held legal accountability in itself to be an unjustifiable cost. The availability of the courts to vindicate constitutional and statutory wrongs has been perceived and protected as one of the virtues of our system of delegated and limited powers. As I argued in Part I, our concern in fashioning absolute immunity rules has been that liability may pervert the decisionmaking process in a particular function by undercutting the values we expect to guide those decisions. Except for the empty generality that the President should have " 'the maximum ability to deal fearlessly and impartially with' the duties of his office,' " ante, at 752, the majority nowhere suggests a particular, disadvantageous effect on a specific Presidential function. The caution that comes from requiring reasonable choices in areas that may intrude on individuals' legally protected rights has never before been counted as a cost. IV 133 The majority may be correct in its conclusion that "[a] rule of absolute immunity . . . will not leave the Nation without sufficient protection against misconduct on the part of the Chief Executive." Ante, at 757. Such a rule will, however, leave Mr. Fitzgerald without an adequate remedy for the harms that he may have suffered. More importantly, it will leave future plaintiffs without a remedy, regardless of the substantiality of their claims. The remedies in which the Court finds comfort were never designed to afford relief for individual harms. Rather, they were designed as political safety valves. Politics and history, however, are not the domain of the courts; the courts exist to assure each individual that he, as an individual, has enforceable rights that he may pursue to achieve a peaceful redress of his legitimate grievances. 134 I find it ironic, as well as tragic, that the Court would so casually discard its own role of assuring "the right of every individual to claim the protection of the laws," Marbury v. Madison, 1 Cranch, at 163, 2 L.Ed. 60, in the name of protecting the principle of separation of powers. Accordingly, I dissent. 135 Justice BLACKMUN, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 136 I join Justice WHITE's dissent. For me, the Court leaves unanswered his unanswerable argument that no man, not even the President of the United States, is absolutely and fully above the law. See United States v. Lee, 106 U.S. 196, 220, 1 S.Ct. 240, 260, 27 L.Ed. 171 (1882),1 and Marbury v. Madison, 1 Cranch 137, 163, 2 L.Ed. 60 (1803).2 Until today, I had thought this principle was the foundation of our national jurisprudence. It now appears that it is not. 137 Nor can I understand the Court's holding that the absolute immunity of the President is compelled by separation-of-powers concerns, when the Court at the same time expressly leaves open, ante, at 748, and n. 27, the possibility that the President nevertheless may be fully subject to congressionally created forms of liability. These two concepts, it seems to me, cannot coexist. 138 I also write separately to express my unalleviated concern about the parties' settlement agreement, the key details of which were not disclosed to the Court by counsel until the veritable "last minute," and even then, only because the Halperins' motion to intervene had directed the Court's attention to them. See ante, at 743-744, n. 24. The Court makes only passing mention of this agreement in Part II-B of its opinion. 139 For me, the case in effect was settled before argument by petitioner's payment of $142,000 to respondent. A much smaller sum of $28,000 was left riding on an outcome favorable to respondent, with nothing at all to be paid if petitioner prevailed, as indeed he now does. The parties publicly stated that the amount of any payment would depend upon subsequent proceedings in the District Court; in fact, the parties essentially had agreed that, regardless of this Court's ruling, no further proceedings of substance would occur in the District Court. Surely, had the details of this agreement been known at the time the petition for certiorari came before the Court, certiorari would have been denied. I cannot escape the feeling that this long-undisclosed agreement comes close to being a wager on the outcome of the case, with all of the implications that entails. 140 Havens Realty Corp. v. Coleman, 455 U.S. 363, 102 S.Ct. 1114, 71 L.Ed.2d 214 (1982), most recently—and, it now appears, most conveniently—decided, affords less than comfortable support for retaining the case.3 The pertinent question here is not whether the case is moot, but whether this is the kind of case or controversy over which we should exercise our power of discretionary review. Cf. United States v. Johnson, 319 U.S. 302, 63 S.Ct. 1075, 87 L.Ed. 1413 (1943). 141 Apprised of all developments, I therefore would have dismissed the writ as having been improvidently granted. The Court, it seems to me, brushes by this factor in order to resolve an issue of profound consequence that otherwise would not be here. Lacking support for such a dismissal, however, I join the dissent. 1 See Economics of Military Procurement: Hearings before the Subcommittee on Economy in Government of the Joint Economic Committee, 90th Cong., 2d Sess., pt. I, pp. 199-201 (1968-1969). It is not disputed that officials in the Department of Defense were both embarrassed and angered by Fitzgerald's testimony. Within less than two months of respondent's congressional appearance, staff had prepared a memorandum for the outgoing Secretary of the Air Force, Harold Brown, listing three ways in which Fitzgerald might be removed from his position. See App. 209a-211a (memorandum of John Lang to Harold Brown, Jan. 6, 1969). Among these was a "reduction in force"—the means by which Fitzgerald ultimately was removed by Brown's successor in office under the new Nixon administration. The reduction in force was announced publicly on November 4, 1969, and Fitzgerald accordingly was separated from the Air Force upon the elimination of his job on January 5, 1970. 2 See The Dismissal of A. Ernest Fitzgerald by the Department of Defense: Hearings before the Subcommittee on Economy in Government of the Joint Economic Committee, 91st Cong., 1st Sess. (1969). Some 60 Members of Congress also signed a letter to the President protesting the "firing of this dedicated public servant" as a "punitive action." Id., at 115-116. 3 A briefing memorandum on the Fitzgerald matter had been prepared by White House staff in anticipation of a possible inquiry at the forthcoming press conference. Authored by aide Patrick Buchanan, it advanced the view that the Air Force was "firing . . . a good public servant." App. 269a (memorandum of Patrick Buchanan to Richard Nixon, Dec. 5, 1969). The memorandum suggested that the President order Fitzgerald's retention by the Defense Department. 4 Id., at 228a. 5 See id., at 109a-112a (deposition of H. R. Haldeman); id., at 137a-141a (deposition of petitioner Richard Nixon). Haldeman's deposed testimony was based on his handwritten notes of December 12, 1969. Id., at 275a. 6 See id., at 126a (deposition of Robert Mayo); id., at 141a (deposition of Richard Nixon). 7 Both Mayo and his deputy, James Schlesinger, appear to have resisted at least partly due to a suspicion that Fitzgerald lacked institutional loyalty to executive policies and that he spoke too freely in communications with friends on Capitol Hill. Both also stated that high-level positions were presently unavailable within the Bureau of the Budget. See id., at 126a (deposition of Robert Mayo); id., at 146a-147a (deposition of James Schlesinger). 8 Quoted in Decision on the Appeal of A. Ernest Fitzgerald (Sept. 18, 1973) (CSC Decision), reprinted in App. 60a, 84a. (Page citations to the CSC Decision refer to the cited page in the Joint Appendix.) 9 Id., at 85a. The memorandum added that " '[W]e owe "first choice on Fitzgerald" to [Senator] Proxmire and others who tried so hard to make him a hero [for exposing the cost overruns].' " Suspicion of Fitzgerald's assumed loyalty toward Senator Proxmire was widely shared in the White House and in the Defense Department. According to the CSC Decision, supra: "While Mr. Fitzgerald has denied that he was 'Senator Proxmier's [sic] boy in the Air Force', and he may honestly believe it, we find this statement difficult to accept. It is evident that the top officials in the Air Force, without specifically saying so, considered him to be just that. . . . We also note that upon leaving the Air Force Mr. Fitzgerald was employed as a consultant by the Proxmire Committee and that Senator Proxmire appeared at the Commission hearing as a character witness for [Fitzgerald]." App. 83a. 10 Id., at 61a. 11 See id., at 83a-84a. 12 See ibid. 13 Id., at 185a. A few hours after the press conference, Mr. Nixon repeated privately to Presidential aide Charles Colson that he had ordered Fitzgerald's firing. Id., at 214a-215a (recorded conversation of Jan. 31, 1973). 14 Id., at 196a (transcription of statement of White House press secretary Ronald Ziegler, Feb. 1, 1973). In a conversation with aide John Ehrlichman, following his conversation with Charles Colson, see n. 13 supra, the President again had claimed responsibility for Fitzgerald's dismissal. When Ehrlichman corrected him on several details, however, the President concluded that he was "thinkin' of another case." Id., at 218a (recorded conversation of Jan. 31, 1973). See id., at 220a. It was after this conversation that the retraction was ordered. 15 Fitzgerald's position in the Air Force was in the "excepted service" and therefore not covered by civil service rules and regulations for the competitive service. Fitzgerald v. Hampton, 152 U.S.App.D.C. 1, 4, 467 F.2d 755, 758 (1972); see CSC Decision, App. 63a-64a. In Hampton, however, the court held that Fitzgerald's employment nonetheless was under "legislative protection," since he was a "preference eligible" veteran entitled to various statutory protections under the Veterans' Preference Act. See 152 U.S.App.D.C., at 4-14, 467 F.2d, at 758-768. Among these were the benefits of the reduction-in-force procedures established by civil service regulation. See id., at 4, 467 F.2d, at 758. 16 The Examiner found that Fitzgerald in fact was dismissed because of his superiors' dissatisfaction with his job performance. App. 86a-87a. Their attitude was evidenced by "statements that he was not a 'team player' and 'not on the Air Force team.' " Id., at 83a. Without deciding whether this would have been an adequate basis for an "adverse action" against Fitzgerald as an "inadequate or unsatisfactory employee," id., at 86a, the Examiner held that the Commission's adverse action procedures, current version codified at 5 CFR pt. 752 (1982), implicitly forbade the Air Force to employ a "reduction in force" as a means of dismissing respondent for reasons "personal to" him. App. 87a. 17 The Commission also ordered that Fitzgerald should receive backpay. Id., at 87a-88a. Following the Commission's order, respondent was offered a new position with the Defense Department, but not one that he regarded as equivalent to his former employment. Fitzgerald accordingly filed an enforcement action in the District Court. This litigation ultimately culminated in a settlement agreement. Under its terms the United States Air Force agreed to reassign Fitzgerald to his former position as Management Systems Deputy to the Assistant Secretary of the Air Force, effective June 21, 1982. See Settlement Agreement in Fitzgerald v. Hampton et al., Civ. No. 76-1486 (DC June 15, 1982). 18 The complaint alleged a continuing conspiracy to deprive him of his job, to deny him reemployment, and to besmirch his reputation. Fitzgerald alleged that the conspiracy had continued through the Commission hearings and remained in existence at the initiation of the lawsuit. See Fitzgerald v. Seamans, 384 F.Supp. 688, 690-692 (DC 1974). 19 The general allegations of the complaint remained essentially unchanged. In averring Nixon's participation in the alleged conspiracy against him, the complaint quoted petitioner's press conference statement that he was "totally aware" of and in fact "approved" Fitzgerald's dismissal. Second Amended Complaint in Fitzgerald v. Butterfield, Civ. No. 74-178 (D.C.), p. 6. 20 See App. to Pet. for Cert. 1a-2a. The District Court held that respondent was entitled to "infer" a cause of action under 5 U.S.C. § 7211 (1976 ed., Supp.IV) and 18 U.S.C. § 1505. Neither expressly confers a private right to sue for relief in damages. The first, 5 U.S.C. § 7211 (1976 ed., Supp.IV), provides generally that "[T]he right of employees . . . to . . . furnish information to either House of Congress, or to a committee or Member thereof, may not be interfered with or denied." The second, 18 U.S.C. § 1505, is a criminal statute making it a crime to obstruct congressional testimony. The correctness of the decision that a cause of action could be "implied" under these statutes is not currently before us. As explained infra, this case is here under the "collateral order" doctrine, for review of the District Court's denial of petitioner's motion to dismiss on the ground that he enjoyed absolute immunity from civil suit. The District Court also held that respondent had stated a claim under the common law of the District of Columbia, but respondent subsequently abandoned his common-law cause of action. See Supplemental Brief in Opposition 2. 21 See Brief in Opposition 2. Although Fitzgerald has not continued to urge this argument, the challenge was jurisdictional, and we therefore address it. 22 The statute provides in pertinent part: "Cases in the courts of appeals may be reviewed by the Supreme Court by the following methods: "(1) By writ of certiorari granted upon the petition of any party to any civil or criminal case, before or after rendition of judgment or decree . . . ." 23 There can be no serious doubt concerning our power to review a court of appeals' decision to dismiss for lack of jurisdiction—a power we have exercised routinely. See, e.g., Gardner v. Westinghouse Broadcasting Co., 437 U.S. 478, 98 S.Ct. 2451, 57 L.Ed.2d 364 (1978). If we lacked authority to do so, decisions to dismiss for want of jurisdiction would be insulated entirely from review by this Court. Nor, now that we have taken jurisdiction of the case, need we remand to the Court of Appeals for a decision on the merits. The immunity question is a pure issue of law, appropriate for our immediate resolution. Especially in light of the Court of Appeals' now-binding decision of the issue presented, concerns of judicial economy fully warrant our decision of the important question presented. 24 Respondent filed a copy of this agreement with the Clerk of this Court on August 24, 1981, as an appendix to his brief in opposition to a motion of Morton, Ina, David, Mark, and Gary Halperin to intervene and for other relief. On June 10, 1980, prior to the Court's action on the petition for certiorari, counsel to the parties had advised the Court that their clients had reached an agreement to liquidate damages, but that there remained a live controversy. Counsel did not include a copy of the agreement in their initial submission. 25 Spalding v. Vilas, 161 U.S. 483, 16 S.Ct. 631, 40 L.Ed. 780 (1896), was distinguished on the ground that the suit against the Postmaster General had asserted a common-law—and not a constitutional—cause of action. See Butz v. Economou, 438 U.S., at 493-495, 98 S.Ct., at 2904-2905. 26 Although the Court in Butz v. Economou, supra, 438 U.S. at 508, 98 S.Ct. at 2911, described the requisite inquiry as one of "public policy," the focus of inquiry more accurately may be viewed in terms of the "inherent" or "structural" assumptions of our scheme of government. 27 In the present case we therefore are presented only with "implied" causes of action, and we need not address directly the immunity question as it would arise if Congress expressly had created a damages action against the President of the United States. This approach accords with this Court's settled policy of avoiding unnecessary decision of constitutional issues. Reviewing this case under the "collateral order" doctrine, see supra, at 742, we assume for purposes of this opinion that private causes of action may be inferred both under the First Amendment and the two statutes on which respondent relies. But it does not follow that we must—in considering a Bivens (Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971)) remedy or interpreting a statute in light of the immunity doctrine—assume that the cause of action runs against the President of the United States. Cf. Tenney v. Brandhove, 341 U.S. 367, 376, 71 S.Ct. 783, 788, 95 L.Ed. 1019 (1951) (construing § 1983 in light of the immunity doctrine, the Court could not accept "that Congress . . . would impinge on a tradition [of legislative immunity] so well grounded in history and reason by covert inclusion in the general language before us," and therefore would not address issues that would arise if Congress had undertaken to deprive state legislators of absolute immunity). Consequently, our holding today need only be that the President is absolutely immune from civil damages liability for his official acts in the absence of explicit affirmative action by Congress. We decide only this constitutional issue, which is necessary to disposition of the case before us. 28 U.S.Const., Art. II, § 3. 29 Chicago & Southern Air Lines, Inc. v. Waterman S.S. Corp., 333 U.S. 103, 111, 68 S.Ct. 431, 436, 92 L.Ed. 568 (1948). 30 Myers v. United States, 272 U.S. 52, 134-135, 47 S.Ct. 21, 31-32, 71 L.Ed. 160 (1926). 31 Noting that the Speech and Debate Clause provides a textual basis for congressional immunity, respondent argues that the Framers must be assumed to have rejected any similar grant of executive immunity. This argument is unpersuasive. First, a specific textual basis has not been considered a prerequisite to the recognition of immunity. No provision expressly confers judicial immunity. Yet the immunity of judges is well settled. See, e.g., Bradley v. Fisher, 13 Wall. 335, 20 L.Ed. 646 (1872); Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978). Second, this Court already has established that absolute immunity may be extended to certain officials of the Executive Branch. Butz v. Economou, 438 U.S., at 511-512, 98 S.Ct., at 2913-14; see Imbler v. Pachtman, 424 U.S. 409, 96 S.Ct. 984, 47 L.Ed.2d 128 (1976) (extending immunity to prosecutorial officials within the Executive Branch). Third, there is historical evidence from which it may be inferred that the Framers assumed the President's immunity from damages liability. At the Constitutional Convention several delegates expressed concern that subjecting the President even to impeachment would impair his capacity to perform his duties of office. See 2 M. Farrand, Records of the Federal Convention of 1787, p. 64 (1911) (remarks of Gouverneur Morris); id., at 66 (remarks of Charles Pinckney). The delegates of course did agree to an Impeachment Clause. But nothing in their debates suggests an expectation that the President would be subjected to the distraction of suits by disappointed private citizens. And Senator Maclay has recorded the views of Senator Ellsworth and Vice President John Adams—both delegates to the Convention—that "the President, personally, was not the subject to any process whatever . . . . For [that] would . . . put it in the power of a common justice to exercise any authority over him and stop the whole machine of Government." Journal of William Maclay 167 (E. Maclay ed. 1890). Justice Story, writing in 1833, held it implicit in the separation of powers that the President must be permitted to discharge his duties undistracted by private lawsuits. 3 J. Story, Commentaries on the Constitution of the United States § 1563, pp. 418-419 (1st ed. 1833) (quoted supra, at 749). Thomas Jefferson also argued that the President was not intended to be subject to judicial process. When Chief Justice Marshall held in United States v. Burr, 25 F.Cas. 30 (No. 14,692d) (CC Va.1807), that a subpoena duces tecum can be issued to a President, Jefferson protested strongly, and stated his broader view of the proper relationship between the Judiciary and the President: "The leading principle of our Constitution is the independence of the Legislature, executive and judiciary of each other, and none are more jealous of this than the judiciary. But would the executive be independent of the judiciary, if he were subject to the commands of the latter, & to imprisonment for disobedience; if the several courts could bandy him from pillar to post, keep him constantly trudging from north to south & east to west, and withdraw him entirely from his constitutional duties? The intention of the Constitution, that each branch should be independent of the others, is further manifested by the means it has furnished to each, to protect itself from enterprises of force attempted on them by the others, and to none has it given more effectual or diversified means than to the executive." 10 The Works of Thomas Jefferson 404 n. (P. Ford ed. 1905) (quoting a letter from President Jefferson to a prosecutor at the Burr trial) (emphasis in the original). See also 5 D. Malone, Jefferson and His Time: Jefferson the President 320-325 (1974). In light of the fragmentary character of the most important materials reflecting the Framers' intent, we do think that the most compelling arguments arise from the Constitution's separation of powers and the Judiciary's historic understanding of that doctrine. See text supra. But our primary reliance on constitutional structure and judicial precedent should not be misunderstood. The best historical evidence clearly supports the Presidential immunity we have upheld. Justice WHITE's dissent cites some other materials, including ambiguous comments made at state ratifying conventions and the remarks of a single publicist. But historical evidence must be weighed as well as cited. When the weight of evidence is considered, we think we must place our reliance on the contemporary understanding of John Adams, Thomas Jefferson, and Oliver Ellsworth. Other powerful support derives from the actual history of private lawsuits against the President. Prior to the litigation explosion commencing with this Court's 1971 Bivens decision, fewer than a handful of damages actions ever were filed against the President. None appears to have proceeded to judgment on the merits. 32 Among the most persuasive reasons supporting official immunity is the prospect that damages liability may render an official unduly cautious in the discharge of his official duties. As Judge Learned Hand wrote in Gregoire v. Biddle, 177 F.2d 579, 581 (CA2 1949), cert. denied, 339 U.S. 949, 70 S.Ct. 803, 94 L.Ed. 1363 (1950), "[t]he justification for . . . [denying recovery] is that it is impossible to know whether the claim is well founded until the case has been tried, and to submit all officials, the innocent as well as the guilty, to the burden of a trial and to the inevitable danger of its outcome, would dampen the ardor of all but the most resolute . . . ." 33 These dangers are significant even though there is no historical record of numerous suits against the President, since a right to sue federal officials for damages for constitutional violations was not even recognized until Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971). 34 This tradition can be traced far back into our constitutional history. See, e.g., Mississippi v. Johnson, 4 Wall. 475, 501, 18 L.Ed. 437 (1866) ("[W]e are fully satisfied that this court has no jurisdiction of a bill to enjoin the President in the performance of his official duties; and that no such bill ought to be received by us"); Kendall v. United States, 12 Pet. 524, 610, 9 L.Ed. 1181 (1838) ("The executive power is vested in a President; and as far as his powers are derived from the constitution, he is beyond the reach of any other department, except in the mode prescribed by the constitution through the impeaching power"). 35 See United States v. Reynolds, 345 U.S. 1, 6-7, 73 S.Ct. 528, 531-32, 97 L.Ed. 727 (1953) (Secretary of the Air Force); Carl Zeiss Stiftung v. V. E. B. Carl Zeiss, Jena, 40 F.R.D. 318, 323-324 (DC 1966), aff'd sub nom. V.E.B. Carl Zeiss, Jena v. Clark, 128 U.S.App.D.C. 10, 384 F.2d 979, cert. denied, 389 U.S. 952, 88 S.Ct. 334, 19 L.Ed.2d 361 (1967) (Department of Justice officials). 36 Although the President was not a party, the Court enjoined the Secretary of Commerce from executing a direct Presidential order. See 343 U.S., at 583, 72 S.Ct., at 865. 37 The Court has recognized before that there is a lesser public interest in actions for civil damages than, for example, in criminal prosecutions. See United States v. Gillock, 445 U.S. 360, 371-373, 100 S.Ct. 1185, 1192-1193, 63 L.Ed.2d 454 (1980); cf. United States v. Nixon, 418 U.S., at 711-712, and n. 19, 94 S.Ct., at 3109-3110 and n. 19 (basing holding on special importance of evidence in a criminal trial and distinguishing civil actions as raising different questions not presented for decision). It never has been denied that absolute immunity may impose a regrettable cost on individuals whose rights have been violated. But, contrary to the suggestion of Justice WHITE's dissent, it is not true that our jurisprudence ordinarily supplies a remedy in civil damages for every legal wrong. The dissent's objections on this ground would weigh equally against absolute immunity for any official. Yet the dissent makes no attack on the absolute immunity recognized for judges and prosecutors. Our implied-rights-of-action cases identify another area of the law in which there is not a damages remedy for every legal wrong. These cases establish that victims of statutory crimes ordinarily may not sue in federal court in the absence of expressed congressional intent to provide a damages remedy. See, e.g., Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 102 S.Ct. 1825, 72 L.Ed.2d 182 (1982); Middlesex County Sewerage Auth. v. National Sea Clammers Assn., 453 U.S. 1, 101 S.Ct. 2615, 69 L.Ed.2d 435 (1981); California v. Sierra Club, 451 U.S. 287, 101 S.Ct. 1775, 68 L.Ed.2d 101 (1981). Justice WHITE does not refer to the jurisprudence of implied rights of action. Moreover, the dissent undertakes no discussion of cases in the Bivens line in which this Court has suggested that there would be no damages relief in circumstances "counseling hesitation" by the judiciary. See Bivens v. Six Unknown Fed. Narcotics Agents, supra, at 396, 91 S.Ct., at 2004; Carlson v. Green, 446 U.S. 14, 19, 100 S.Ct. 1468, 1472, 64 L.Ed.2d 15 (1980) (in direct constitutional actions against officials with "independent status in our constitutional scheme . . . judicially created remedies . . . might be inappropriate"). Even the case on which Justice WHITE places principal reliance, Marbury v. Madison, 1 Cranch 137, 2 L.Ed. 60 (1803), provides dubious support at best. The dissent cites Marbury for the proposition that "[T]he very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury." Id., 1 Cranch at 163. Yet Marbury does not establish that the individual's protection must come in the form of a particular remedy. Marbury, it should be remembered, lost his case in the Supreme Court. The Court turned him away with the suggestion that he should have gone elsewhere with his claim. In this case it was clear at least that Fitzgerald was entitled to seek a remedy before the Civil Service Commission—a remedy of which he availed himself. See supra, at 736-739, and n. 17. 38 The presence of alternative remedies has played an important role in our previous decisions in the area of official immunity. E.g., Imbler v. Pachtman, 424 U.S., at 428-429, 96 S.Ct., at 994 ("We emphasize that the immunity of prosecutors from liability in suits under § 1983 does not leave the public powerless to deter misconduct or to punish that which occurs"). 39 The same remedy plays a central role with respect to the misconduct of federal judges, who also possess absolute immunity. See Kaufman, Chilling Judicial Independence, 88 Yale L.J. 681, 690-706 (1979). Congressmen may be removed from office by a vote of their colleagues. U.S.Const., Art. I, § 5, cl. 2. 40 Prior to petitioner Nixon's resignation from office, the House Judiciary Committee had convened impeachment hearings. See generally Report of the Committee on the Judiciary of the House of Representatives: Impeachment of Richard M. Nixon, President of the United States, H.R.Rep.No. 93-1305 (1974). 41 The dissenting opinions argue that our decision places the President "above the law." This contention is rhetorically chilling but wholly unjustified. The remedy of impeachment demonstrates that the President remains accountable under law for his misdeeds in office. This case involves only a damages remedy. Although the President is not liable in civil damages for official misbehavior, that does not lift him "above" the law. The dissents do not suggest that a judge is "above" the law when he enters a judgment for which he cannot be held answerable in civil damages; or a prosecutor is above the law when he files an indictment; or a Congressman is above the law when he engages in legislative speech or debate. It is simply error to characterize an official as "above the law" because a particular remedy is not available against him. 1 Presidential immunity for official acts while in office has never been seriously questioned until very recently. Ante, at 750-752, n. 31. I can find only one instance in which, prior to our decision in Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971), a citizen sued a former President for acts committed while in office. A suit against Thomas Jefferson was dismissed for being improperly brought in Virginia, thus precluding the necessity of reaching any immunity issue. Livingston v. Jefferson, 15 F.Cas. No. 8,411 p. 660 (No. 8,411) (CC Va.1811). 2 In their "parade of horribles" and lamentations, the dissents also wholly fail to acknowledge why the same perils they fear are not present in the absolute immunity the law has long recognized for numerous other officials. At least 75,000 public officers have absolute immunity from civil damages suits for acts within the scope of their official functions. The dissenting opinions manifest an astonishing blind side in pointing to that old reliable that "no man is above the law." The Court has had no difficulty expanding the absolute immunity of Members of Congress, and in granting derivative absolute immunity to numerous aides of Members. Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972). We have since recognized absolute immunity for judges, Stump v. Sparkman, 435 U.S. 349, 98 S.Ct. 1099, 55 L.Ed.2d 331 (1978), and for prosecutors, Imbler v. Pachtman, 424 U.S. 409, 96 S.Ct. 984, 47 L.Ed.2d 128 (1976), yet the Constitution provides no hint that either judges, prosecutors, or congressional aides should be so protected. Absolute immunity for judges and prosecutors is seen to derive from the common law and public policy, which recognize the need to protect judges and prosecutors from harassment. The potential danger to the citizenry from the malice of thousands of prosecutors and judges is at once more pervasive and less open to constant, public scrutiny than the actions of a President. 3 In United States v. Brewster, 408 U.S. 501, 92 S.Ct. 2531, 33 L.Ed.2d 507 (1972), we held that the Speech and Debate Clause does not prohibit prosecution of a Senator for accepting a bribe designed to influence his legislative acts. 4 Justice WHITE suggests that prior to today, Presidents, prosecutors, judges, congressional aides, and other officials "could have been held liable for the kind of claim put forward by Fitzgerald—a personnel decision allegedly made for unlawful reasons." Post, at 767, n. 2 (emphasis added). But the law does not permit a plaintiff to recite "magic" words in pleadings and have the incantation operate to make these immunities vanish. Justice WHITE errs fundamentally in treating all of the above officials as if the scope of their authority were identical. The authority of a President as head of the Executive Branch of our Government—a wholly unique office—is far broader than that of any other official. As the Court notes, a President has authority in the course of personnel changes in an executive department to make personnel decisions. If the decision is wrong, statutory remedies are provided. See n. 5, infra. This is not to say that, in a given case, it would not be appropriate to raise the question whether an official—even a President—had acted within the scope of the official's constitutional and statutory duties. The doctrine of absolute immunity does not extend beyond such actions. 5 Gravel v. United States, 408 U.S. 606, 92 S.Ct. 2614, 33 L.Ed.2d 583 (1972). The Federal Tort Claims Act of 1946 reflects this policy distinction; in it Congress waived sovereign immunity for certain damages claims, but pointedly excepted any "discretionary function or duty . . . whether or not the discretion involved be abused." 28 U.S.C. § 2680(a). Under the Act, damage resulting from discretionary governmental action is not subject to compensation. See, e.g., Dalehite v. United States, 346 U.S. 15, 73 S.Ct. 956, 97 L.Ed. 1427 (1953). For uncompensated injuries Congress may in its discretion provide separate nonjudicial remedies such as private bills. In this case Fitzgerald received substantial relief through the route provided by Congress: the Civil Service Commission ordered him reinstated with backpay. App. 87a-88a. Similarly situated persons are therefore not without an adequate remedy. But see post, at 797 (WHITE, J., dissenting). In addition, respondent Fitzgerald has also received a settlement of $142,000. It can hardly be said he has had no remedy. 6 Judge Learned Hand described his feelings: "After now some dozen years of experience I must say that as a litigant I should dread a lawsuit beyond almost anything else short of sickness and death." 3 Lectures on Legal Topics, Association of the Bar of the City of New York 105 (1926). 7 The Court suggests that "we need not address directly" whether Congress could create a damages action against a President. Ante, at 748, n. 27. However, the Court's holding, in my view, effectively resolves that issue; once it is established that the Constitution confers absolute immunity, as the Court holds today, legislative action cannot alter that result. Nothing in the Court's opinion is to be read as suggesting that a constitutional holding of this Court can be legislatively overruled or modified. Marbury v. Madison, 1 Cranch 137, 2 L.Ed. 60 (1803). 1 This, of course, is not simply a hypothetical example. See Halperin v. Kissinger, 196 U.S.App.D.C. 285, 606 F.2d 1192 (1979), aff'd by an equally divided Court, 452 U.S. 713, 101 S.Ct. 3132, 69 L.Ed.2d 367 (1981). 2 It is ironic that this decision should come out at the time of the tenth anniversary of the Watergate affair. Even the popular press has drawn from that affair an insight into the character of the American constitutional system that is bound to be profoundly shaken by today's decision: "The important lesson that Watergate established is that no President is above the law. It is a banality, a cliche, but it is a point on which many Americans . . . seem confused." 119 Time, No. 24, p. 28 (June 14, 1982). A majority of the Court shares this confusion. The majority vigorously protests this characterization of its position, ante, at 758, n. 41, arguing that the President remains subject to law in the form of impeachment proceedings. But the abandonment of the rule of law here is not in the result reached, but in the manner of reaching it. The majority fails to apply to the President those principles which we have consistently used to determine the scope and credibility of an absolute immunity defense. It does this because of some preconceived notion of the inapplicability of general rules of law to the President. Similarly, THE CHIEF JUSTICE, like the majority, misses the point in his wholly unconvincing contentions that the Court today does no more than extend to the President the same sort of immunity that we have recognized with respect to Members of Congress, judges, prosecutors, and legislative aides. In none of our previous cases have we extended absolute immunity to all actions "within the scope of the official's constitutional and statutory duties." Concurring opinion of THE CHIEF JUSTICE, ante, at 761, n. 4. Indeed, under the immunity doctrine as it existed prior to today's decision, each of these officials could have been held liable for the kind of claim put forward by Fitzgerald—a personnel decision allegedly made for unlawful reasons. Although such a decision falls within the scope of an official's duties, it does not fall within the judicial, legislative, or prosecutorial functions to which absolute immunity attaches. THE CHIEF JUSTICE's failure to grasp the difference between the functional approach to absolute immunity that we have previously adopted and the nature of today's decision accounts for his misunderstanding of this dissent. 3 Although the majority opinion initially claims that its conclusion is based substantially on "our history," historical analysis in fact plays virtually no part in the analysis that follows. 4 On this point, I am in agreement with the concurring opinion of THE CHIEF JUSTICE. 5 The Solicitor General relies entirely upon the brief filed by his office for petitioners in Kissinger v. Halperin, 452 U.S. 713, 101 S.Ct. 3122, 69 L.Ed.2d 367 (1981). 6 In fact, insofar as the Constitution addresses the issue of Presidential liability, its approach is very different from that taken in the Speech or Debate Clause. The possibility of impeachment assures that the President can be held accountable to the other branches of Government for his actions; the Constitution further states that impeachment does not bar criminal prosecution. 7 The debate is recorded in 2 M. Farrand, Records of the Federal Convention of 1787, pp. 64-69 (1911) (hereinafter Farrand). 8 Id., at 64. 9 Id., at 65. 10 Ibid. 11 Id., at 66. 12 In The Federalist No. 65, p, 439 (J. Cooke ed. 1961), Alexander Hamilton described impeachable offenses as follows: "They are of a nature which may with peculiar propriety be denominated POLITICAL, as they relate chiefly to injuries done immediately to the society itself." 13 The majority's use of the historical record is in line with its other arguments: It puts the burden on respondent to demonstrate no Presidential immunity, rather than on petitioner to prove the appropriateness of this defense. Thus, while noting that the doubts of some of the Framers were not sufficient to prevent the adoption of the Impeachment Clause, the majority nevertheless states that "nothing in [the] debates suggests an expectation that the President would be subjected to [civil damages actions]." Ante, at 751, n. 31. Of course, nothing in the debates suggests an expectation that the President would not be liable in civil suits for damages either. Nevertheless, the debates are one element that the majority cites to support its conclusion that "[t]he best historical evidence clearly supports the Presidential immunity we have upheld." Ante, at 752, n. 31. 14 Other commentary on the proposed Constitution did, however, consider the subject of Presidential immunity. In fact, the subject was discussed in the first major defense of the Constitution published in the United States. In his essays on the Constitution, published in the Independent Gazetteer in September 1787, Tench Coxe included the following statement in his description of the limited power of the proposed Office of the President: "His person is not so much protected as that of a member of the House of Representatives; for he may be proceeded against like any other man in the ordinary course of law." Quoted in 2 The Documentary History of the Ratification of the Constitution 141 (1976) (emphasis in original). 15 4 J. Elliot, Debates on the Federal Constitution 48 (1876 ed.). 16 Id., at 47. 17 2 id., at 480. 18 W. Maclay, Sketches of Debate in the First Senate of the United States in 1789-1791, p. 152 (1969 reprint). 19 Ibid. 20 2 J. Story, Commentaries on the Constitution of the United States § 1569, p. 372 (4th ed. 1873). 21 It is not possible to determine whether this is the same Pinckney that Madison recorded as Pinkney, who objected at the Convention to granting a power of impeachment to the Legislature. Two Charles Pinckneys attended the Convention. Both were from South Carolina. See 3 Farrand 559. 22 Senator Pinckney's comments are recorded at 10 Annals of Cong. 69-83 (1800). Petitioner contends that these remarks are not relevant because they concerned only the authority of Congress to inquire into the origin of an allegedly libelous newspaper article. Reply Brief for Petitioner 7. Although this was the occasion for the remarks, Pinckney did discuss the immunity of Members of Congress as a privilege embodied in the Speech or Debate Clause: "[O]ur Constitution supposes no man . . . to be infallible, but considers them all as mere men, and subject to all the passions, and frailties, and crimes, that men generally are, and accordingly provides for the trial of such as ought to be tried, and leaves the members of the Legislature, for their proceedings, to be amenable to their constituents and to public opinion . . . ." 10 Annals of Cong. 71 (1800). This, then, was one of the privileges of Congress that he was contrasting with those extended (or not extended) to the President. 23 The majority cites one additional piece of historical evidence, a letter by President Jefferson, which it contends demonstrates that Jefferson believed that "the President was not intended to be subject to judicial process. Ante, at 751, n. 31. Thomas Jefferson's views on the relation of the President to the judicial process are, however, not quite so clear as the majority suggests. Jefferson took a variety of positions on the proper relation of Executive and Judicial authority, at different points in his career. It would be surprising if President Jefferson had not argued strongly for such immunity from judicial process, particularly in a confrontation with Chief Justice Marshall. Jefferson's views on this issue before he became President would be of a good deal more significance. In this regard, it is significant that in Jefferson's second and third drafts of the Virginia Constitution, which also proposed a separation of the powers of government into three separate branches, he specifically proposed that the Executive be subject to judicial process: "[H]e shall be liable to action, tho' not to personal restraint for private duties or wrongs." 1 Papers of Thomas Jefferson 350, 360 (1950). Also significant is the fact that when Jefferson's followers tried to impeach Justice Chase in 1804-1805, one of the grounds of their attack on him was that he had refused to subpoena President Adams during the trial of Dr. Cooper for sedition. See E. Corwin, The President: Office and Powers 113 (4th ed. 1957). Finally, it is worth noting that even in the middle of the debate over Chief Justice Marshall's power to subpoena the President during the Burr trial, Jefferson looked to a legislative solution of the confrontation: "I hope however that . . . at the ensuing session of the legislature [the Chief Justice] may have means provided for giving to individuals the benefit of the testimony of the [Executive] functionaries in proper cases." 10 The Works of Thomas Jefferson 407 n. (P. Ford ed. 1905) (quoting a letter from President Jefferson to George Hay, United States District Attorney for Virginia). 24 The Solicitor General, in fact, argues that the possibility of judicial review of Presidential actions supports the claim of absolute immunity: Judicial review "serves to contain and correct the unauthorized exercise of the President's powers," making private damages actions unnecessary in order to achieve the same end. Brief for Petitioners in Kissinger v. Halperin, O.T.1980, No. 79-880, p. 31. See n. 5, supra. 25 See also Justice Harlan's discussion of the appropriateness of the damages remedy in order to redress the violation of certain constitutional rights. Bivens v. Six Unknown Fed. Narcotics Agents, 403 U.S., at 407-410, 91 S.Ct., at 2010-2012 (concurring in judgment). 26 Contrary to the suggestion of the majority, ante, at 754-755, n. 37, I do not suggest that there must always be a remedy in civil damages for every legal wrong or that Marbury v. Madison stands for this proposition. Marbury does, however, suggest the importance of the private interests at stake within the broader perspective of a political system based on the rule of law. The functional approach to immunity questions, which we have previously followed but which the majority today discards, represented an appropriate reconciliation of the conflicting interests at stake. 27 The Court has never held that the "public policy" conclusions it reaches as to the appropriateness of absolute immunity in particular instances are not subject to reversal through congressional action. Implicitly, therefore, the Court has already rejected a constitutionally based, separation-of-powers argument for immunity for federal officials. 28 Surely the fact that officers of the court have been the primary beneficiaries of this Court's pronouncements of absolute immunity gives support to this appearance of favoritism. 29 See Supreme Court of Virginia v. Consumers Union of United States, 446 U.S. 719, 100 S.Ct. 1967, 64 L.Ed.2d 641 (1980); Butz v. Economou, 438 U.S. 478, 511, 98 S.Ct. 2894, 2913, 57 L.Ed.2d 895 (1978). 30 I will not speculate on the Presidential functions which may require absolute immunity, but a clear example would be instances in which the President participates in prosecutorial decisions. 31 See, e.g., 48 Cong.Rec. 4653 (1912) ("During my first session of Congress I was desirous of learning the needs of the postal service and inquiring into the conditions of the employees. To my surprise I found that under an Executive order these civil-service employees could not give me any information") (remarks of Rep. Calder); id., at 4656 ("I believe it is high time that Congress should listen to the appeals of these men and provide a way whereby they can properly present a petition to the Members of Congress for a redress of grievances without the fear of losing their official positions") (remarks of Rep. Reilly); id., at 5157 ("I have always requested employees to consult with me on matters affecting their interest and believe that it is my duty to listen to all respectful appeals and complaints") (remarks of Rep. Evans). Indeed, it is for just this reason that petitioners in Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.3d 396, argue that the statutes do not create a private right of action: "5 U.S.C. § 7211 and 18 U.S.C. § 1505 were designed to protect the legislative process, not to give one such as Fitzgerald a right to seek damages." Brief for Petitioners, O.T.1981, No. 80-945, p. 26, n. 11. 32 Indeed, the impetus for passage of what is now 5 U.S.C. § 7211 (1976 ed., Supp.IV) was the imposition of "gag rules" upon testimony of civil servants before congressional committees. See Exec. Order No. 402 (Jan. 25, 1906); Exec. Order No. 1142 (Nov. 26, 1909). 33 Thus, adverse action may generally be taken against civil servants only "for such cause as will promote the efficiency of the service." 5 U.S.C. §§ 7503, 7513, and 7543 (1976 ed., Supp.IV). 34 The majority also seems to believe that by "function" the Court has in the past referred to "subjective purpose." See ante, at 756 ("an inquiry into the President's motives could not be avoided under the . . . 'functional' theory . . ."). I do not read our cases that way. In Stump v. Sparkman, 435 U.S. 349, 362, 98 S.Ct. 1099, 1107, 55 L.Ed.2d 331 (1978), we held that the factors determining whether a judge's act was a "judicial action" entitled to absolute immunity "relate to the nature of the act itself, i.e., whether it is a function normally performed by a judge, and to the expectations of the parties." Neither of these factors required any analysis of the purpose the judge may have had in carrying out the particular action. Similarly in Butz v. Economou, 438 U.S., at 512-516, 98 S.Ct., at 2913-15, when we determined that certain executive functions were entitled to absolute immunity because they shared "enough of the characteristics of the judicial process," we looked to objective qualities and not subjective purpose. 35 The majority seems to suggest that responsibility for governmental reorganizations is one such function. Ante, at 756. I fail to see why this should be so. 36 Although our conclusions differ, the majority opinion reflects a similar view as to the relationship between the two sources of the causes of action in this case: It does not believe it necessary to differentiate in its own analysis between the statutory and constitutional causes of action. 37 THE CHIEF JUSTICE leaves no doubt that he, at least, reads the majority opinion as standing for the broad proposition that the President is absolutely immune under the Constitution: "I write separately to underscore that the Presidential immunity [as spelled out today] derives from and is mandated by the constitutional doctrine of separation of powers." Concurring opinion of THE CHIEF JUSTICE, ante, at 758. Similarly, THE CHIEF JUSTICE dismisses the majority's claim that it has not decided the question of whether Congress could create a damages action against the President: "[T]he Court's holding . . . effectively resolves that issue; once it is established that the Constitution confers absolute immunity, as the Court holds today, legislative action cannot alter that result." Ante, at 763, n. 7. 38 Contrary to the suggestion of the majority, Mississippi v. Johnson, 4 Wall. 475, 18 L.Ed. 437 (1866), carefully reserved the question of whether a court may compel the President himself to perform ministerial executive functions: "We shall limit our inquiry to the question presented by the objection, without expressing any opinion on the broader issues . . . whether, in any case, the President . . . may be required, by the process of this court, to perform a purely ministerial act under a positive law, or may be held amenable, in any case, otherwise than by impeachment for crime." Id., 71 U.S. (4 Wall) at 498. Similarly, Kendall v. United States, 12 Pet. 524, 9 L.Ed. 1181 (1838), also cited by the majority, did not indicate that the President could never be subject to judicial process. In fact, it implied just the contrary in rejecting the argument that the mandamus sought involved an unconstitutional judicial infringement upon the Executive Branch: "The mandamus does not seek to direct or control the postmaster general in the discharge of any official duty, partaking in any respect of an executive character; but to enforce the performance of a mere ministerial act, which neither he nor the President had any authority to deny or control." Id., at 610. 1 "No man in this country is so high that he is above the law. No officer of the law may set that law at defiance with impunity. All the officers of the government, from the highest to the lowest, are creatures of the law, and are bound to obey it." 2 "The very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury. One of the first duties of government is to afford that protection. In Great Britain the king himself is sued in the respectful form of a petition, and he never fails to comply with the judgment of his court." 3 The agreement in Havens was not final until approved by the District Court, 455 U.S., at 370-371, 102 S.Ct., at 1120. In the present case, the parties made their agreement and presented it to the District Court only after the fact. Further, there was no preliminary payment in Havens. Each respondent there was to receive $400 if the Court denied certiorari or affirmed, and nothing if the Court reversed. Here, $142,000 changed hands regardless of the subsequent disposition of the case, with the much smaller sum of $28,000 resting on the Court's ultimate ruling. For me, this is not the kind of case or controversy contemplated by Art. III of the Constitution.

78

457 U.S. 922 102 S.Ct. 2744 73 L.Ed.2d 482 Giles M. LUGAR, Petitioner,v.EDMONDSON OIL COMPANY, INC. and Ronald L. Barbour. No. 80-1730. Argued Dec. 8, 1981. Decided June 25, 1982. Syllabus This case concerns the relationship between the requirement of "state action" to establish a violation of the Fourteenth Amendment, and the requirement of action "under color of state law" to establish a right to recover under 42 U.S.C. § 1983, which provides a remedy for deprivation of constitutional rights when that deprivation takes place "under color of any statute, ordinance, regulation, custom, or usage" of a State. Respondents filed suit in Virginia state court on a debt owed by petitioner, and sought prejudgment attachment of certain of petitioner's property. Pursuant to Virginia law, respondents alleged, in an ex parte petition, a belief that petitioner was disposing of or might dispose of his property in order to defeat his creditors; acting upon that petition, a Clerk of the state court issued a writ of attachment, which was executed by the County Sheriff; a hearing on the propriety of the attachment was later conducted; and 34 days after the levy the trial judge dismissed the attachment for respondents' failure to establish the alleged statutory grounds for attachment. Petitioner then brought this action in Federal District Court under § 1983, alleging that in attaching his property respondents had acted jointly with the State to deprive him of his property without due process of law. The District Court held that the alleged actions of the respondents did not constitute state action as required by the Fourteenth Amendment, and that the complaint therefore did not state a valid claim under § 1983. The Court of Appeals affirmed, but on the basis that the complaint failed to allege conduct under color of state law for purposes of § 1983 because there was neither usurpation or corruption of official power by a private litigant nor a surrender of judicial power to the private litigant in such a way that the independence of the enforcing officer was compromised to a significant degree. Held : 1. Constitutional requirements of due process apply to garnishment and prejudgment attachment procedures whenever state officers act jointly with a private creditor in securing the property in dispute. Sniadach v. Family Finance Corp., 395 U.S. 337, 89 S.Ct. 1820, 23 L.Ed.2d 349. And if the challenged conduct of the creditor constitutes state action as delimited by this Court's prior decisions, then that conduct is also action under color of state law and will support a suit under § 1983. Pp. 926-935. 2. Conduct allegedly causing the deprivation of a constitutional right protected against infringement by a State must be fairly attributable to the State. In determining the question of "fair attribution," (a) the deprivation must be caused by the exercise of some right or privilege created by the State or by a rule of conduct imposed by it or by a person for whom it is responsible, and (b) the party charged with the deprivation must be a person who may fairly be said to be a state actor, either because he is a state official, because he has acted together with or has obtained significant aid from state officials, or because his conduct is otherwise chargeable to the State. Pp. 936-939. 3. Insofar as petitioner alleged only misuse or abuse by respondents of Virginia law, he did not state a cause of action under § 1983, but challenged only private action. Such challenged conduct could not be ascribed to any governmental decision, nor did respondents have the authority of state officials to put the weight of the State behind their private decision. However, insofar as petitioner's complaint challenged the state statute as being procedurally defective under the Due Process Clause, he did present a valid cause of action under § 1983. The statutory scheme obviously is the product of state action, and a private party's joint participation with state officials in the seizure of disputed property is sufficient to characterize that party as a "state actor" for purposes of the Fourteenth Amendment. Respondents were, therefore, acting under color of state law in participating in the deprivation of petitioner's property. Pp. 939-942. 4th Cir., 639 F.2d 1058, affirmed in part, reversed in part, and remanded. Robert L. Morrison, Jr., Danville, Va., for petitioner. James W. Haskins, Martinsville, Va., for respondents. Justice WHITE delivered the opinion of the Court. 1 The Fourteenth Amendment of the Constitution provides in part: 2 "No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws." 3 Because the Amendment is directed at the States, it can be violated only by conduct that may be fairly characterized as "state action." 4 Title 42 U.S.C. § 1983 provides a remedy for deprivations of rights secured by the Constitution and laws of the United States when that deprivation takes place "under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory. . . ."1 This case concerns the relationship between the § 1983 requirement of action under color of state law and the Fourteenth Amendment requirement of state action. 5 * In 1977, petitioner, a lessee-operator of a truckstop in Virginia, was indebted to his supplier, Edmondson Oil Co., Inc. Edmondson sued on the debt in Virginia state court. Ancillary to that action and pursuant to state law, Edmondson sought prejudgment attachment of certain of petitioner's property. Va.Code § 8.01-533 (1977).2 The prejudgment attachment procedure required only that Edmondson allege, in an ex parte petition, a belief that petitioner was disposing of or might dispose of his property in order to defeat his creditors. Acting upon that petition, a Clerk of the state court issued a writ of attachment, which was then executed by the County Sheriff. This effectively sequestered petitioner's property, although it was left in his possession. Pursuant to the statute, a hearing on the propriety of the attachment and levy was later conducted. Thirty-four days after the levy, a state trial judge ordered the attachment dismissed because Edmondson had failed to establish the statutory grounds for attachment alleged in the petition.3 6 Petitioner subsequently brought this action under 42 U.S.C. § 1983 against Edmondson and its president. His complaint alleged that in attaching his property respondents had acted jointly with the State to deprive him of his property without due process of law. The lower courts construed the complaint as alleging a due process violation both from a misuse of the Virginia procedure and from the statutory procedure itself.4 He sought compensatory and punitive damages for specified financial loss allegedly caused by the improvident attachment. 7 Relying on Flagg Brothers Inc. v. Brooks, 436 U.S. 149, 98 S.Ct. 1729, 56 L.Ed.2d 185 (1978), the District Court held that the alleged actions of the respondents did not constitute state action as required by the Fourteenth Amendment and that the complaint therefore did not state a claim upon which relief could be granted under § 1983. Petitioner appealed; the Court of Appeals for the Fourth Circuit, sitting en banc, affirmed, with three dissenters.5 639 F.2d 1058 (1981). 8 The Court of Appeals rejected the District Court's reliance on Flagg Brothers in finding that the requisite state action was missing in this case. The participation of state officers in executing the levy sufficiently distinguished this case from Flagg Brothers. The Court of Appeals stated the issue as follows: 9 "[W]hether the mere institution by a private litigant of presumptively valid state judicial proceedings, without any prior or subsequent collusion or concerted action by that litigant with the state officials who then proceed with adjudicative, administrative, or executive enforcement of the proceedings, constitutes action under color of state law within contemplation of § 1983." 639 F.2d, at 1061-1062 (footnote omitted). 10 The court distinguished between the acts directly chargeable to respondents and the larger context within which those acts occurred, including the direct levy by state officials on petitioner's property. While the latter no doubt amounted to state action, the former was not so clearly action under color of state law. The court held that a private party acts under color of state law within the meaning of § 1983 only when there is a usurpation or corruption of official power by the private litigant or a surrender of judicial power to the private litigant in such a way that the independence of the enforcing officer has been compromised to a significant degree. Because the court thought none of these elements was present here, the complaint failed to allege conduct under color of state law. 11 Because this construction of the under-color-of-state-law requirement appears to be inconsistent with prior decisions of this Court, we granted certiorari. 452 U.S. 937, 101 S.Ct. 3078, 69 L.Ed.2d 951 (1981). II 12 Although the Court of Appeals correctly perceived the importance of Flagg Brothers to a proper resolution of this case, it misread that case.6 It also failed to give sufficient weight to that line of cases, beginning with Sniadach v. Family Finance Corp., 395 U.S. 337, 89 S.Ct. 1820, 23 L.Ed.2d 349 (1969), in which the Court considered constitutional due process requirements in the context of garnishment actions and prejudgment attachments. See North Georgia Finishing, Inc. v. Di-Chem, Inc., 419 U.S. 601, 95 S.Ct. 719, 42 L.Ed.2d 751 (1975); Mitchell v. W. T. Grant Co., 416 U.S. 600, 94 S.Ct. 1895, 40 L.Ed.2d 406 (1974); Fuentes v. Shevin, 407 U.S. 67, 92 S.Ct. 1983, 32 L.Ed.2d 556 (1972). Each of these cases involved a finding of state action as an implicit predicate of the application of due process standards. Flagg Brothers distinguished them on the ground that in each there was overt, official involvement in the property deprivation; there was no such overt action by a state officer in Flagg Brothers. 436 U.S., at 157, 98 S.Ct., at 1734. Although this case falls on the Sniadach, and not the Flagg Brothers, side of this distinction, the Court of Appeals thought the garnishment and attachment cases to be irrelevant because none but Fuentes arose under 42 U.S.C. § 1983 and because Fuentes was distinguishable.7 It determined that it could ignore all of them because the issue in this case was not whether there was state action, but rather whether respondents acted under color of state law. 13 As we see it, however, the two concepts cannot be so easily disentangled. Whether they are identical or not, the state-action and the under-color-of-state-law requirements are obviously related.8 Indeed, until recently this Court did not distinguish between the two requirements at all. 14 In United States v. Price, 383 U.S. 787, 794, n. 7, 86 S.Ct. 1152, 1157 n.7, 16 L.Ed.2d 267 (1966), we explicitly stated that the requirements were identical: "In cases under § 1983, 'under color' of law has consistently been treated as the same thing as the 'state action' required under the Fourteenth Amendment."9 In support of this proposition the Court cited Smith v. Allwright, 321 U.S. 649, 64 S.Ct. 757, 88 L.Ed. 987 (1944), and Terry v. Adams, 345 U.S. 461, 73 S.Ct. 809, 97 L.Ed. 1152 (1953).10 In both of these cases black voters in Texas challenged their exclusion from party primaries as a violation of the Fifteenth Amendment and sought relief under 8 U.S.C. § 43 (1946 ed.).11 In each case, the Court understood the problem before it to be whether the discriminatory policy of a private political association could be characterized as "state action within the meaning of the Fifteenth Amendment." Smith, supra, 321 U.S., at 664, 64 S.Ct., at 765.12 Having found state action under the Constitution, there was no further inquiry into whether the action of the political associations also met the statutory requirement of action "under color of state law." 15 Similarly, it is clear that in a § 1983 action brought against a state official, the statutory requirement of action "under color of state law" and the "state action" requirement of the Fourteenth Amendment are identical. The Court's conclusion in United States v. Classic, 313 U.S. 299, 326, 61 S.Ct. 1031, 1043, 85 L.Ed. 1368 (1941), that "[m]isuse of power, possessed by virtue of state law and made possible only because the wrongdoer is clothed with the authority of state law, is action taken 'under color of' state law," was founded on the rule announced in Ex parte Virginia, 100 U.S. 339, 346-347, 25 L.Ed. 676 (1880), that the actions of a state officer who exceeds the limits of his authority constitute state action for purposes of the Fourteenth Amendment.13 16 The decision of the Court of Appeals rests on a misreading of Flagg Brothers. In that case the Court distinguished two elements of a § 1983 action: 17 "[Plaintiffs] are first bound to show that they have been deprived of a right 'secured by the Constitution and the laws' of the United States. They must secondly show that Flagg Brothers deprived them of this right acting 'under color of any statute' of the State of New York. It is clear that these two elements denote two separate areas of inquiry. Adickes v. S. H. Kress & Co., 398 U.S. 144, 150 [90 S.Ct. 1598, 1604, 26 L.Ed.2d 142] (1970)." 436 U.S., at 155-156, 98 S.Ct., at 1732-1733. 18 Plaintiffs' case foundered on the first requirement. Because a due process violation was alleged and because the Due Process Clause protects individuals only from governmental and not from private action, plaintiffs had to demonstrate that the sale of their goods was accomplished by state action. The Court concluded that the sale, although authorized by state law, did not amount to state action under the Fourteenth Amendment, and therefore set aside the Court of Appeals' contrary judgment. 19 There was no reason in Flagg Brothers to address the question whether there was action under color of state law. The Court expressly eschewed deciding whether that requirement was satisfied by private action authorized by state law. Id., at 156, 98 S.Ct., at 1733. Although the state-action and under-color-of-state-law requirements are "separate areas of inquiry," Flagg Brothers did not hold nor suggest that state action, if present, might not satisfy the § 1983 requirement of conduct under color of state law. Nevertheless, the Court of Appeals relied on Flagg Brothers to conclude in this case that state action under the Fourteenth Amendment is not necessarily action under color of state law for purposes of § 1983. We do not agree. 20 The two-part approach to a § 1983 cause of action, referred to in Flagg Brothers, was derived from Adickes v. S. H. Kress & Co., 398 U.S. 144, 150, 90 S.Ct. 1598, 1604, 26 L.Ed.2d 142 (1970). Adickes was a § 1983 action brought against a private party, based on a claim of racial discrimination in violation of the Equal Protection Clause of the Fourteenth Amendment. Although stating that the § 1983 plaintiff must show both that he has been deprived "of a right secured by the 'Constitution and laws' of the United States" and that the defendant acted "under color of any statute . . . of any State," ibid., we held that the private party's joint participation with a state official in a conspiracy to discriminate would constitute both "state action essential to show a direct violation of petitioner's Fourteenth Amendment equal protection rights" and action " 'under color' of law for purposes of the statute." Id., at 152, 90 S.Ct., at 1606.14 In support of our conclusion that a private party held to have violated the Fourteenth Amendment "can be liable under § 1983," ibid., we cited that part of United States v. Price, 383 U.S., at 794, n. 7, 86 S.Ct., at 1157, n. 7, in which we had concluded that state action and action under color of state law are the same (quoted above). Adickes provides no support for the Court of Appeals' novel construction of § 1983.15 B 21 The decision of the Court of Appeals is difficult to reconcile with the Court's garnishment and prejudgment attachment cases and with the congressional purpose in enacting § 1983. 22 Beginning with Sniadach v. Family Finance Corp., 395 U.S. 337, 89 S.Ct. 1820, 23 L.Ed.2d 349 (1969), the Court has consistently held that constitutional requirements of due process apply to garnishment and prejudgment attachment procedures whenever officers of the State act jointly with a creditor in securing the property in dispute. Sniadach and North Georgia Finishing, Inc. v. Di-Chem, Inc., 419 U.S. 601, 95 S.Ct. 719, 42 L.Ed.2d 751 (1975), involved state-created garnishment procedures; Mitchell v. W. T. Grant Co., 416 U.S. 600, 94 S.Ct. 1895, 40 L.Ed.2d 406 (1974), involved execution of a vendor's lien to secure disputed property. In each of these cases state agents aided the creditor in securing the disputed property; but in each case the federal issue arose in litigation between creditor and debtor in the state courts and no state official was named as a party. Nevertheless, in each case the Court entertained and adjudicated the defendant-debtor's claim that the procedure under which the private creditor secured the disputed property violated federal constitutional standards of due process. Necessary to that conclusion is the holding that private use of the challenged state procedures with the help of state officials constitutes state action for purposes of the Fourteenth Amendment. 23 Fuentes v. Shevin, 407 U.S. 67, 92 S.Ct. 1983, 32 L.Ed.2d 556 (1972), was a § 1983 action brought against both a private creditor and the State Attorney General. The plaintiff sought declaratory and injunctive relief, on due process grounds, from continued enforcement of state statutes authorizing prejudgment replevin. The plaintiff prevailed; if the Court of Appeals were correct in this case, there would have been no § 1983 cause of action against the private parties. Yet they remained parties, and judgment ran against them in this Court.16 24 If a defendant debtor in state-court debt collection proceedings can successfully challenge, on federal due process grounds, the plaintiff creditor's resort to the procedures authorized by a state statute, it is difficult to understand why that same behavior by the state-court plaintiff should not provide a cause of action under § 1983. If the creditor-plaintiff violates the debtor-defendant's due process rights by seizing his property in accordance with statutory procedures, there is little or no reason to deny to the latter a cause of action under the federal statute, § 1983, designed to provide judicial redress for just such constitutional violations. 25 To read the "under color of any statute" language of the Act in such a way as to impose a limit on those Fourteenth Amendment violations that may be redressed by the § 1983 cause of action would be wholly inconsistent with the purpose of § 1 of the Civil Rights Act of 1871, 17 Stat. 13, from which § 1983 is derived. The Act was passed "for the express purpose of 'enforc[ing] the Provisions of the Fourteenth Amendment.' " Lynch v. Household Finance Corp., 405 U.S. 538, 545, 92 S.Ct. 1113, 1118, 31 L.Ed.2d 424 (1972). The history of the Act is replete with statements indicating that Congress thought it was creating a remedy as broad as the protection that the Fourteenth Amendment affords the individual. Perhaps the most direct statement of this was that of Senator Edmunds, the manager of the bill in the Senate: "[Section 1 is] so very simple and really reenact[s] the Constitution." Cong.Globe, 42d Cong., 1st Sess., 569 (1871). Representative Bingham similarly stated that the bill's purpose was "the enforcement . . . of the Constitution on behalf of every individual citizen of the Republic . . . to the extent of the rights guarantied to him by the Constitution." Id., App. 81.17 26 In sum, the line drawn by the Court of Appeals is inconsistent with our prior cases and would substantially undercut the congressional purpose in providing the § 1983 cause of action. If the challenged conduct of respondents constitutes state action as delimited by our prior decisions, then that conduct was also action under color of state law and will support a suit under § 1983.18 III 27 As a matter of substantive constitutional law the state-action requirement reflects judicial recognition of the fact that "most rights secured by the Constitution are protected only against infringement by governments," Flagg Brothers, 436 U.S., at 156, 98 S.Ct., at 1733. As the Court said in Jackson v. Metropolitan Edison Co., 419 U.S. 345, 349, 95 S.Ct. 449, 453, 42 L.Ed.2d 477 (1974): 28 "In 1883, this Court in the Civil Rights Cases, 109 U.S. 3 [3 S.Ct. 18, 27 L.Ed. 835], affirmed the essential dichotomy set forth in [the Fourteenth] Amendment between deprivation by the State, subject to scrutiny under its provisions, and private conduct, 'however discriminatory or wrongful,' against which the Fourteenth Amendment offers no shield." 29 Careful adherence to the "state action" requirement preserves an area of individual freedom by limiting the reach of federal law and federal judicial power. It also avoids imposing on the State, its agencies or officials, responsibility for conduct for which they cannot fairly be blamed. A major consequence is to require the courts to respect the limits of their own power as directed against state governments and private interests. Whether this is good or bad policy, it is a fundamental fact of our political order. 30 Our cases have accordingly insisted that the conduct allegedly causing the deprivation of a federal right be fairly attributable to the State. These cases reflect a two-part approach to this question of "fair attribution." First, the deprivation must be caused by the exercise of some right or privilege created by the State or by a rule of conduct imposed by the state or by a person for whom the State is responsible. In Sniadach, Fuentes, W. T. Grant, and North Georgia, for example, a state statute provided the right to garnish or to obtain prejudgment attachment, as well as the procedure by which the rights could be exercised. Second, the party charged with the deprivation must be a person who may fairly be said to be a state actor. This may be because he is a state official, because he has acted together with or has obtained significant aid from state officials, or because his conduct is otherwise chargeable to the State. Without a limit such as this, private parties could face constitutional litigation whenever they seek to rely on some state rule governing their interactions with the community surrounding them. 31 Although related, these two principles are not the same. They collapse into each other when the claim of a constitutional deprivation is directed against a party whose official character is such as to lend the weight of the State to his decisions. See Monroe v. Pape, 365 U.S. 167, 172, 81 S.Ct. 473, 476, 5 L.Ed.2d 492 (1961). The two principles diverge when the constitutional claim is directed against a party without such apparent authority, i.e., against a private party. The difference between the two inquiries is well illustrated by comparing Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972), with Flagg Brothers, supra. 32 In Moose Lodge, the Court held that the discriminatory practices of the appellant did not violate the Equal Protection Clause because those practices did not constitute "state action." The Court focused primarily on the question of whether the admittedly discriminatory policy could in any way be ascribed to a governmental decision.19 The inquiry, therefore, looked to those policies adopted by the State that were applied to appellant. The Court concluded as follows: 33 "We therefore hold, that with the exception hereafter noted, the operation of the regulatory scheme enforced by the Pennsylvania Liquor Control Board does not sufficiently implicate the State in the discriminatory guest policies of Moose Lodge to . . . make the latter 'state action' within the ambit of the Equal Protection Clause of the Fourteenth Amendment." 407 U.S., at 177, 92 S.Ct., at 1973. 34 In other words, the decision to discriminate could not be ascribed to any governmental decision; those governmental decisions that did affect Moose Lodge were unconnected with its discriminatory policies.20 35 Flagg Brothers focused on the other component of the state-action principle. In that case, the warehouseman proceeded under New York Uniform Commercial Code, § 7-210, and the debtor challenged the constitutionality of that provision on the grounds that it violated the Due Process and Equal Protection Clauses of the Fourteenth Amendment. Undoubtedly the State was responsible for the statute. The response of the Court, however, focused not on the terms of the statute but on the character of the defendant to the § 1983 suit: Action by a private party pursuant to this statute, without something more, was not sufficient to justify a characterization of that party as a "state actor." The Court suggested that that "something more" which would convert the private party into a state actor might vary with the circumstances of the case. This was simply a recognition that the Court has articulated a number of different factors or tests in different contexts: e.g., the "public function" test, see Terry v. Adams, 345 U.S. 461, 73 S.Ct. 809, 97 L.Ed. 1152 (1953); Marsh v. Alabama, 326 U.S. 501, 66 S.Ct. 276, 90 L.Ed. 265 (1946); the "state compulsion" test, see Adickes v. S. H. Kress & Co., 398 U.S. at 170, 90 S.Ct. at 1615; the "nexus" test, see Jackson v. Metropolitan Edison Co., 419 U.S. 345, 95 S.Ct. 449, 42 L.Ed.2d 477 (1974); Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961); and in the case of prejudgment attachments, a "joint action test," Flagg Brothers, 436 U.S., at 157, 98 S.Ct., at 1734.21 Whether these different tests are actually different in operation or simply different ways of characterizing the necessarily fact-bound inquiry that confronts the Court in such a situation need not be resolved here. See Burton, supra, 365 U.S., at 722, 81 S.Ct., at 860 ("Only by sifting facts and weighing circumstances can the nonobvious involvement of the State in private conduct be attributed its true significance"). IV 36 Turning to this case, the first question is whether the claimed deprivation has resulted from the exercise of a right or privilege having its source in state authority. The second question is whether, under the facts of this case, respondents, who are private parties, may be appropriately characterized as "state actors." Both the District Court and the Court of Appeals noted the ambiguous scope of petitioner's contentions: "There has been considerable confusion throughout the litigation on the question whether Lugar's ultimate claim of unconstitutional deprivation was directed at the Virginia statute itself or only at its erroneous application to him." 639 F.2d, at 1060, n. 1. Both courts held that resolution of this ambiguity was not necessary to their disposition of the case: both resolved it, in any case, in favor of the view that petitioner was attacking the constitutionality of the statute as well as its misapplication. In our view, resolution of this issue is essential to the proper disposition of the case. 37 Petitioner presented three counts in his complaint. Count three was a pendent claim based on state tort law; counts one and two claimed violations of the Due Process Clause. Count two alleged that the deprivation of property resulted from respondents' "malicious, wanton, willful, opressive [sic], [and] unlawful acts." By "unlawful," petitioner apparently meant "unlawful under state law." To say this, however, is to say that the conduct of which petitioner complained could not be ascribed to any governmental decision; rather, respondents were acting contrary to the relevant policy articulated by the State. Nor did they have the authority of state officials to put the weight of the State behind their private decision, i.e., this case does not fall within the abuse of authority doctrine recognized in Monroe v. Pape, 365 U.S. 167, 81 S.Ct. 473, 5 L.Ed.2d 492 (1961). That respondents invoked the statute without the grounds to do so could in no way be attributed to a state rule or a state decision. Count two, therefore, does not state a cause of action under § 1983 but challenges only private action. 38 Count one is a different matter. That count describes the procedures followed by respondents in obtaining the prejudgment attachment as well as the fact that the state court subsequently ordered the attachment dismissed because respondents had not met their burden under state law. Petitioner then summarily states that this sequence of events deprived him of his property without due process. Although it is not clear whether petitioner is referring to the state-created procedure or the misuse of that procedure by respondents, we agree with the lower courts that the better reading of the complaint is that petitioner challenges the state statute as procedurally defective under the Fourteenth Amendment.22 39 While private misuse of a state statute does not describe conduct that can be attributed to the State, the procedural scheme created by the statute obviously is the product of state action. This is subject to constitutional restraints and properly may be addressed in a § 1983 action, if the second element of the state-action requirement is met as well. 40 As is clear from the discussion in Part II, we have consistently held that a private party's joint participation with state officials in the seizure of disputed property is sufficient to characterize that party as a "state actor" for purposes of the Fourteenth Amendment. The rule in these cases is the same as that articulated in Adickes v. S. H. Kress & Co., supra, 398 U.S., at 152, 90 S.Ct., at 1605-1606, in the context of an equal protection deprivation: 41 " 'Private persons, jointly engaged with state officials in the prohibited action, are acting "under color" of law for purposes of the statute. To act "under color" of law does not require that the accused be an officer of the State. It is enough that he is a willful participant in joint activity with the State or its agents,' " quoting United States v. Price, 383 U.S., at 794, 86 S.Ct., at 1157. 42 The Court of Appeals erred in holding that in this context "joint participation" required something more than invoking the aid of state officials to take advantage of state-created attachment procedures. That holding is contrary to the conclusions we have reached as to the applicability of due process standards to such procedures. Whatever may be true in other contexts, this is sufficient when the State has created a system whereby state officials will attach property on the ex parte application of one party to a private dispute. 43 In summary, petitioner was deprived of his property through state action; respondents were, therefore, acting under color of state law in participating in that deprivation. Petitioner did present a valid cause of action under § 1983 insofar as he challenged the constitutionality of the Virginia statute; he did not insofar as he alleged only misuse or abuse of the statute.23 44 The judgment is reversed in part and affirmed in part, and the case is remanded for further proceedings consistent with this opinion. 45 So ordered. Chief Justice BURGER, dissenting. 46 Whether we are dealing with suits under § 1983 or suits brought pursuant to the Fourteenth Amendment, in my view the inquiry is the same: is the claimed infringement of a federal right fairly attributable to the State. Rendell-Baker v. Kohn, 457 U.S. 830, 838, 102 S.Ct. 2764, 2770, 73 L.Ed.2d 418. Applying this standard, it cannot be said that the actions of the named respondents are fairly attributable to the State.* Respondents did no more than invoke a presumptively valid state prejudgment attachment procedure available to all. Relying on a dubious "but for" analysis, the Court erroneously concludes that the subsequent procedural steps taken by the State in attaching a putative debtor's property in some way transforms respondents' acts into actions of the State. This case is no different from the situation in which a private party commences a lawsuit and secures injunctive relief which, even if temporary, may cause significant injury to the defendant. Invoking a judicial process, of course, implicates the State and its officers but does not transform essentially private conduct into actions of the State. Dennis v. Sparks, 449 U.S. 24, 101 S.Ct. 183, 66 L.Ed.2d 185 (1980). Similarly, one who practices a trade or profession, drives an automobile, or builds a house under a state license is not engaging in acts fairly attributable to the state. In both Dennis and the instant case petitioner's remedy lies in private suits for damages such as malicious prosecution. The Court's opinion expands the reach of the statute beyond anything intended by Congress. It may well be a consequence of too casually falling into a semantical trap because of the figurative use of the term "color of state law." Justice POWELL, with whom Justice REHNQUIST and Justice O'CONNOR join, dissenting. 47 Today's decision is a disquieting example of how expansive judicial decisionmaking can ensnare a person who had every reason to believe he was acting in strict accordance with law. The case began nearly five years ago as the outgrowth of a simple suit on a debt in a Virginia state court. Respondent—a small wholesale oil dealer in Southside, Va.—brought suit against petitioner Lugar, a truckstop owner who had failed to pay a debt.1 The suit was to collect this indebtedness. Fearful that petitioner might dissipate his assets before the debt was collected, respondent also filed a petition in state court seeking sequestration of certain of Lugar's assets. He did so under a Virginia statute, traceable at least to 1819, that permits creditors to seek prejudgment attachment of property in the possession of debtors.2 No court had questioned the validity of the statute, and it remains presumptively valid. The Clerk of the state court duly issued a writ of attachment, and the County Sheriff then executed it. There is no allegation that respondent conspired with the state officials to deny petitioner the fair protection of state or federal law. 48 Respondent ultimately prevailed in his lawsuit. The petitioner Lugar was ordered by a court to pay his debt. A state court did find, however, that Lugar's assets should not have been attached prior to a judgment on the underlying action. 49 Following this decision Lugar instituted legal action in the United States District Court for the Western District of Virginia. Suing under a federal statute, 42 U.S.C. § 1983, Lugar alleged that the respondent—by filing a petition in state court—had acted "under color of law" and had caused the deprivation of constitutional rights under the Fourteenth Amendment—an Amendment that does not create rights enforceable against private citizens, such as one would have assumed respondent to be, but only against the States. Rendell-Baker v. Kohn, 457 U.S. 830, 837, 102 S.Ct. 2764, 2769, 73 L.Ed.2d 418; Flagg Bros., Inc. v. Brooks, 436 U.S. 149, 156, 98 S.Ct. 1729, 1733, 56 L.Ed.2d 185 (1978); Shelley v. Kraemer, 334 U.S. 1, 13, 68 S.Ct. 836, 842, 92 L.Ed. 1161 (1948); Civil Rights Cases, 109 U.S. 3, 11, 3 S.Ct. 18, 21, 27 L.Ed. 835 (1883).3 Both the District Court and the Court of Appeals agreed that petitioner had no cause of action under § 1983. They sensibly found that respondent could not be held responsible for any deprivation of constitutional rights and that the suit did not belong in federal court. 50 This Court today reverses the judgment of those lower courts. It holds that respondent, a private citizen who did no more than commence a legal action of a kind traditionally initiated by private parties, thereby engaged in "state action." This decision is as unprecedented as it is implausible. It is plainly unjust to the respondent, and the Court makes no argument to the contrary. Respondent, who was represented by counsel, could have had no notion that his filing of a petition in state court, in the effort to secure payment of a private debt, made him a "state actor" liable in damages for allegedly unconstitutional action by the Commonwealth of Virginia. Nor is the Court's analysis consistent with the mode of inquiry prescribed by our cases. On the contrary, the Court undermines fundamental distinctions between the common-sense categories of state and private conduct and between the legal concepts of "state action" and private action "under color of law." 51 * The plain language of 42 U.S.C. § 1983 establishes that a plaintiff must satisfy two jurisdictional requisites to state an actionable claim. First, he must allege the violation of a right "secured by the Constitution and laws" of the United States. Because "most rights secured by the Constitution are protected only against infringement by governments," Flagg Bros., Inc. v. Brooks, 436 U.S., at 156, 98 S.Ct., at 1733, this requirement compels an inquiry into the presence of state action. Second, a § 1983 plaintiff must show that the alleged deprivation was caused by a person acting "under color" of law. In Flagg Bros., this Court affirmed that "these two elements denote two separate areas of inquiry." Id., at 155-156, 98 S.Ct., at 1733. See Adickes v. S. H. Kress & Co., 398 U.S. 144, 152, 90 S.Ct. 1598, 1605, 26 L.Ed.2d 142 (1970). 52 This case demonstrates why separate inquiries are required. Here it is not disputed that the Virginia Sheriff and Clerk of Court, the state officials who sequestered petitioner's property in the manner provided by Virginia law, engaged in state action. Yet the petitioner, while alleging constitutional injury from this action by state officials, did not sue the State or its agents. In these circumstances the Court of Appeals correctly stated that the relevant inquiry was the second identified in Flagg Bros.: whether the respondent, a private citizen whose only action was to invoke a presumptively valid state attachment process, had acted under color of state law in "causing" the State to deprive petitioner of alleged constitutional rights.4 Consistently with past decisions of this Court, the Court of Appeals concluded that respondent's private conduct had not occurred under color of law. 53 Rejecting the reasoning of the Court of Appeals, the Court opinion inexplicably conflates the two inquiries mandated by Flagg Bros. Ignoring that this case involves two sets of actions—one by respondent, who merely filed a suit and accompanying sequestration petition; another by the state officials, who issued the writ and executed the lien—it wrongly frames the question before the Court, not as whether the private respondent acted under color of law in filing the petition, but as "whether . . . respondents, who are private parties, may be appropriately characterized as 'state actors.' " Ante, at 939. It then concludes that they may, on the theory that a private party who invokes "the aid of state officials to take advantage of state-created attachment procedures" is a "joint participant" with the State and therefore a "state actor." "The rule," the Court asserts, is as follows: 54 "Private persons, jointly engaged with state officials in the prohibited action, are acting 'under color' of law for purposes of the statute. To act 'under color' of law does not require that the accused be an officer of the State. It is enough that he is a willful participant in a joint activity with the State or its agents." Ante, at 941, quoting Adickes v. S. H. Kress & Co., supra, at 152, 90 S.Ct., at 1605, in turn quoting United States v. Price, 383 U.S. 787, 794, 86 S.Ct. 1152, 1156, 16 L.Ed.2d 267 (1966). 55 There are at least two fallacies in the Court's conclusion. First, as is apparent from the quotation, our cases have not established that private "joint participants" with state officials themselves necessarily become state actors. Where private citizens interact with state officials in the pursuit of merely private ends, the appropriate inquiry generally is whether the private parties have acted "under color of law." Second, even when the inquiry is whether an action occurred under color of law, our cases make clear that the "joint participation" standard is not satisfied when a private citizen does no more than invoke a presumptively valid judicial process in pursuit only of legitimate private ends. II 56 As this Court recognized in Monroe v. Pape, 365 U.S. 167, 172, 81 S.Ct. 473, 476, 5 L.Ed.2d 492 (1961), the historic purpose of § 1983 was to prevent state officials from using the cloak of their authority under state law to violate rights protected against state infringement by the Fourteenth Amendment.5 The Court accordingly is correct that an important inquiry in a § 1983 suit against a private party is whether there is an allegation of wrongful "conduct that can be attributed to the State." Ante, at 941. This is the first question referred to in Flagg Bros. But there still remains the second Flagg Bros. question: whether this state action fairly can be attributed to the respondent, whose only action was to invoke a presumptively valid attachment statute. This question, unasked by the Court, reveals the fallacy of its conclusion that respondent may be held accountable for the attachment of property because he was a "state actor."6 From the occurrence of state action taken by the Sheriff who sequestered petitioner's property it does not follow that respondent became a "state actor" simply because the Sheriff was. This Court, until today, has never endorsed this non sequitur. 57 It of course is true that respondent's private action was followed by state action, and that the private and the state actions were not unconnected. But "[t]hat the State responds to [private] actions by [taking action of its own] does not render it responsible for those [private] actions." Blum v. Yaretsky, 457 U.S. 991, 1005, 102 S.Ct. 2777, 2786, 73 L.Ed.2d 534. See Flagg Bros., 436 U.S., at 164-165, 98 S.Ct., at 1737-1738; Jackson v. Metropolitan Edison Co., 419 U.S. 345, 357, 95 S.Ct. 449, 456-457, 42 L.Ed.2d 477 (1974). And where the State is not responsible for a private decision to behave in a certain way, the private action generally cannot be considered "state action" within the meaning of our cases. See, e.g., Blum v. Yaretsky, 457 U.S., at 1004-1005, 102 S.Ct., at 2786; Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 172-173, 92 S.Ct. 1965, 1971-1972, 32 L.Ed.2d 627 (1972). As in Jackson v. Metropolitan Edison Co., supra, "[r]espondent's exercise of the choice allowed by state law where the initiative comes from it and not from the State, does not make its action in doing so 'state action' for purposes of the Fourteenth Amendment." 419 U.S., at 357, 95 S.Ct., at 457 (footnote omitted). 58 This Court of course has held that private parties are amenable to suit under § 1983 when "jointly engaged" with state officials in the violation of constitutional rights. See Adickes v. S. H. Kress & Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970).7 Yet the Court, in advancing its "joint participation" theory, does not cite a single case in which a private decision to invoke a presumptively valid state legal process has been held to constitute state action. Even the quotation on which the Court principally relies for its statement of the applicable "rule," ante, at 941, does not refer to state action. Rather, it states explicitly that "[p]rivate persons, jointly engaged with state officials in the prohibited action, are acting 'under color' of law for purposes of the statute." 59 As illustrated by this quotation, our cases have recognized a distinction between "state action" and private action under "color of law." This distinction is sound in principle. It also is consistent with and supportive of the distinction between "private" conduct and government action that is subject to the procedural limitations of the Due Process Clause of the Fourteenth Amendment. As the Court itself notes: "Careful adherence to the 'state action' requirement preserves an area of individual freedom by limiting the reach of federal law and federal judicial power. It also avoids imposing on the State, its agencies or officials, responsibility for conduct for which they cannot fairly be blamed." Ante, at 936. 60 A "color of law" inquiry acknowledges that private individuals, engaged in unlawful joint behavior with state officials, may be personally responsible for wrongs that they cause to occur. But it does not confuse private actors with the State—the fallacy of the analysis adopted today by the Court. In this case involving the private action of the respondent in petitioning the state courts of Virginia, the appropriate inquiry as to respondent's liability is not whether he was a state actor, but whether he acted under color of law. It is to this question that I therefore turn. III 61 Contrary to the position of the Court, our cases do not establish that a private party's mere invocation of state legal procedures constitutes "joint participation" or "conspiracy" with state officials satisfying the § 1983 requirement of action under color of law. In Dennis v. Sparks, 449 U.S. 24, 101 S.Ct. 183, 66 L.Ed.2d 185 (1980), we held that private parties acted under color of law when corruptly conspiring with a state judge in a joint scheme to defraud. In so holding, however, we explicitly stated that "merely resorting to the courts and being on the winning side of a lawsuit does not make a party a co-conspirator or a joint actor with the judge." Id., at 28, 101 S.Ct., at 186. This conclusion is reinforced by our more recent decision in Polk County v. Dodson, 454 U.S. 312, 102 S.Ct. 445, 70 L.Ed.2d 509 (1981). As we held to be true with respect to the defense of a criminal defendant, invocation of state legal process is "essentially a private function . . . for which state office and authority are not needed." Id., at 319, 102 S.Ct., at 450. These recent decisions make clear that independent, private decisions made in the context of litigation cannot be said to occur under color of law.8 The Court nevertheless advances two principal grounds for its holding to the contrary. 62 * The Court argues that petitioner's action under § 1983 is supported by cases in which this Court has applied due process standards to state garnishment and prejudgment attachment procedures. The Court relies specifically on Sniadach v. Family Finance Corp., 395 U.S. 337, 89 S.Ct. 1820, 23 L.Ed.2d 349 (1969); Fuentes v. Shevin, 407 U.S. 67, 92 S.Ct. 1983, 32 L.Ed.2d 556 (1972); Mitchell v. W. T. Grant Co., 416 U.S. 600, 94 S.Ct. 1895, 40 L.Ed.2d 406 (1974); and North Georgia Finishing, Inc. v. Di-Chem, Inc., 419 U.S. 601, 95 S.Ct. 719, 42 L.Ed.2d 751 (1975). According to the Court, these cases establish that a private party acts "under color" of law when seeking the attachment of property under an unconstitutional state statute.9 In fact, a careful reading demonstrates that they provide no authority for this proposition. 63 Of the cases cited by the Court, Sniadach, Mitchell, and Di-Chem all involved attacks on the validity of state attachment or garnishment statutes. None of the cases alleged that the private creditor was a joint actor with the State, and none involved a claim for damages against the creditor. Each case involved a state suit, not a federal action under § 1983. It therefore was unnecessary in any of these cases for this Court to consider whether the creditor, by virtue of instituting the attachment or garnishment, became a state actor or acted under color of state law. There is not one word in any of these cases that so characterizes the private creditor.10 In Fuentes v. Shevin, the Court did consider a § 1983 action against a private creditor as well as the State Attorney General.11 Again, however, the only question before this Court was the validity of a state statute. No claim was made that the creditor was a joint actor with the State or had acted under color of law. No damages were sought from the creditor. Again, there was no occasion for this Court to consider the status under § 1983 of the private party, and there is not a word in the opinion that discusses this. As with Sniadach, Mitchell, andDi-Chem, Fuentes thus fails to establish that a private party's mere invocation of state attachment or garnishment procedures represents action under color of law—even in a case in which those procedures are subsequently held to be unconstitutional. B 64 In addition to relying on cases involving the constitutionality of state attachment and garnishment statutes, the Court advances a "joint participation" theory based on Adickes v. S. H. Kress Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970). In Adickes the plaintiff sued a private restaurant under § 1983, alleging a conspiracy between the restaurant and local police to deprive her of the right to equal treatment in a place of public accommodation. Id., at 152, 153, 90 S.Ct., at 1605-1606. Reversing the decision below, this Court upheld the cause of action. It found that the private defendant, in "conspiring" with local police to obtain official enforcement of a state custom of racial segregation, engaged in a " 'joint activity with the State or its agents' " and therefore acted under color of law within the meaning of § 1983. Id., at 152, 90 S.Ct., at 1606 (quoting United States v. Price, 383 U.S., at 794, 86 S.Ct., at 1156). 65 Contrary to the suggestion of the Court, however, Justice Harlan's Court opinion in Adickes did not purport to define the term "under color of law." Attending closely to the facts presented, the Court observed that "[w]hatever else may also be necessary to show that a person has acted 'under color of [a] statute' for purposes of § 1983, . . . we think it essential that he act with the knowledge of and pursuant to that statute." 398 U.S., at 162, n. 23, 90 S.Ct., at 1611, n. 23 (emphasis added). As indicated by this choice of language, the Court clearly seems to have contemplated some limiting principle. A citizen summoning the police to enforce the law ordinarily would not be considered to have engaged in a "conspiracy." Nor, presumably, would such a citizen be characterized as acting under color of law and thereby risking amenability to suit for constitutional violations that subsequently might occur. Surely there is nothing in Adickes to indicate that the Court would have found action under color of law in cases of this kind. 66 Although Adickes is distinguishable from these hypotheticals, the current case is not. The conduct in Adickes occurred in 1964, 10 years after Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), and after the decade of publicized litigation that followed in its wake. In view of the intense national focus on issues of racial discrimination, it is virtually inconceivable that a private citizen then could have acted in the innocent belief that the state law and customs involved in Adickes still were presumptively valid. As Justice Harlan wrote, "[f]ew principles of law are more firmly stitched into our constitutional fabric than the proposition that a State must not discriminate against a person because of his race or the race of his companions, or in any way act to compel or encourage segregation." 398 U.S., at 151-152, 90 S.Ct., at 1605. Construed as resting on this basis, Adickes establishes that a private party acts under color of law when he conspires with state officials to secure the application of a state law so plainly unconstitutional as to enjoy no presumption of validity. In such a context, the private party could be characterized as hiding behind the authority of law and as engaging in "joint participation" with the State in the deprivation of constitutional rights.12 Here, however, petitioner has alleged no conspiracy. Nor has he even alleged that respondent was invoking the aid of a law he should have known to be constitutionally invalid.13 Finally, there is no allegation that respondent's decision to invoke legal process was in any way compelled by the law or custom of the State in which he lived. In this context Adickes simply is inapposite. 67 Today's decision therefore is as unprecedented as it is unjust.14 1 Title 42 U.S.C. § 1983, at the time in question, provided in full: "Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured in an action at law, suit in equity, or other proper proceeding for redress." 2 At the time of the attachment in question, this section was codified as Va.Code § 8-519 (1973). 3 The principal action then proceeded to the entry of judgment on the debt in favor of Edmondson and some of petitioner's property was sold in execution of the judgment. 4 In his answer to respondents' motion to dismiss on abstention grounds petitioner stated that "[n]o question of the constitutional validity of the State statutes is made." Plaintiff's Memorandum in Opposition to Motion to Dismiss 3. The District Court responded to this as follows: "[D]espite plaintiff's protests to the contrary . . . the complaint can only be read as challenging the constitutionality of Virginia's attachment statute." App. to Pet. for Cert. 38. The Court of Appeals agreed. 639 F.2d 1058, 1060, n. 1 (CA4 1981). 5 The case was originally argued before a three-judge panel. The Court of Appeals, however, acting sua sponte, set the matter for a rehearing en banc. 6 Justice POWELL suggests that our opinion is not "consistent with the mode of inquiry prescribed by our cases." Post, at 946. We believe the situation to be just the opposite. We rely precisely upon the ground that the majority itself put forth in Flagg Brothers to distinguish that case from the earlier prejudgment attachment cases: "This total absence of overt official involvement plainly distinguishes this case from earlier decisions imposing procedural restrictions on creditors' remedies." 436 U.S., at 157, 98 S.Ct., at 1734. Justice POWELL at no point mentions this aspect of the Flagg Brothers decision. The method of inquiry we adopt is that suggested by Adickes v. S. H. Kress & Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970), and seemingly approved in Flagg Brothers : Joint action with a state official to accomplish a prejudgment deprivation of a constitutionally protected property interest will support a § 1983 claim against a private party. 7 The Court of Appeals held Fuentes v. Shevin not to be relevant because the defendants in that case included the State Attorney General, as well as the private creditor. In the court's view, the presence of a state official made the "private party defendant . . . merely a nominal party to the action for injunctive relief." 639 F.2d, at 1068, n. 22. Judge Butzner, in dissent, found Fuentes to be directly controlling. 8 The Court of Appeals itself recognized this when it stated that in two of three basic patterns of § 1983 litigation—that in which the defendant is a public official and that in which he is a private party—there is no distinction between state action and action under color of state law. Only when there is joint action by private parties and state officials, the court stated, could a distinction arise between these two requirements. 9 We also stated that if an indictment "allege[s] conduct on the part of the 'private' defendants which constitutes 'state action,' [it alleges] action 'under color' of law within [18 U.S.C.] § 242." 383 U.S., at 794, n. 7, 86 S.Ct., at 1157 n. 7. In Monroe v. Pape, 365 U.S. 167, 185, 81 S.Ct. 473, 483, 5 L.Ed.2d 492 (1961), the Court held that "under color of law" has the same meaning in 18 U.S.C. § 242 as it does in § 1983. 10 Besides these two Supreme Court cases, the Court cited a number of lower court cases in support of the proposition that the constitutional concept of state action satisfies the statutory requirement of action under color of state law. Simkins v. Moses H. Cone Memorial Hospital, 323 F.2d 959 (CA4 1963); Smith v. Holiday Inns, 336 F.2d 630 (CA6 1964); Hampton v. City of Jacksonville, 304 F.2d 320 (CA5 1962); Boman v. Birmingham Transit Co., 280 F.2d 531 (CA5 1960); Kerr v. Enoch Pratt Free Library, 149 F.2d 212 (CA4 1945). Each of these cases involved litigation between private parties in which the plaintiffs alleged unconstitutional discrimination. In each case, the only inquiry was whether the private-party defendant met the state-action requirement of the Fourteenth Amendment. Once that requirement was met, the courts granted the relief sought. 11 Title 8 U.S.C. § 43 (1946 ed.) was reclassified as 42 U.S.C. § 1983 in 1952. 12 There was no opinion for the Court in Terry v. Adams. All three opinions in support of the reversal of the lower court decision pose the question as to whether the action of the private political association in question, the Jaybird Democratic Association, constituted state action for purposes of the Fifteenth Amendment. None suggests that a Fifteenth Amendment violation by the private association might not support a cause of action because of a failure to prove action under color of state law. 13 United States v. Classic did not involve § 1983 directly; rather, it interpreted 18 U.S.C. § 242 (then 18 U.S.C. § 52 (1940 ed.)), which is the criminal counterpart of 42 U.S.C. § 1983. See n. 9, supra, on the relationship between 18 U.S.C. § 242 and 42 U.S.C. § 1983. 14 The Adickes opinion contained the following statement, 398 U.S., at 162, n. 23, 90 S.Ct., at 1611 n. 23. "Whatever else may also be necessary to show that a person has acted 'under color of [a] statute, for purposes of § 1983, . . . we think it essential that he act with the knowledge of and pursuant to that statute." This statement obviously was meant neither to establish the definition of action under color of state law, nor to establish a distinction between this statutory requirement and the constitutional standard of state action. The statement was made in response to an argument that the discrimination by the private party was pursuant to the state trespass statute and that this would satisfy the requirements of § 1983. The Court rejected this because there had been no factual showing that the defendants had acted with knowledge of, or pursuant to, this statute. It was in this context, that this statement was made. Justice BRENNAN, writing separately, did suggest in Adickes that "when a private party acts alone, more must be shown . . . to establish that he acts 'under color of' a state statute or other authority than is needed to show that his action constitutes state action." Id., at 210, 90 S.Ct., at 1630 (footnote omitted). Even in his view, however, when a private party acts in conjunction with a state official, whatever satisfies the state-action requirement of the Fourteenth Amendment satisfies the under-color-of-state-law requirement of the statute. Justice BRENNAN's position rested, at least in part, on a much less strict standard of what would constitute "state action" in the area of racial discrimination than that adopted by the majority. In any case, the position he articulated there has never been adopted by the Court. 15 Justice POWELL's discussion of Adickes confuses the two counts of the complaint in that case. There was a conspiracy count which alleged that respondent—a private party—and a police officer had conspired "(1) 'to deprive [petitioner] of her right to enjoy equal treatment and service in a place of public accommodation'; and (2) to cause her arrest 'on the false charge of vagrancy.' " Id., at 149-150, 90 S.Ct., at 1604. It was with respect to this count, which did not allege any unconstitutional statute or custom, that the Court held that joint action of the private party and the police officer was sufficient to support a § 1983 suit against that party. The other count of her complaint was a substantive count in which she alleged that the private act of discrimination was pursuant to a "custom of the community to segregate the races in public eating places." Here the Court did not rely on any "joint action" theory, but held that "petitioner would show an abridgment of her equal protection right, if she proves that Kress refused her service because of a state-enforced custom." Id., at 171, 173, 90 S.Ct., at 1615, 1616. Justice POWELL is wrong when he summarizes Adickes as holding that "a private party acts under color of law when he conspires with state officials to secure the application of a state law so plainly unconstitutional as to enjoy no presumption of validity." Post, at 954-955. This is to confuse the conspiracy and the substantive counts at issue in Adickes. Unless one argues that the state vagrancy law was unconstitutional—an argument no one made in Adickes—the joint action count of Adickes did not involve a state law, whether "plainly unconstitutional" or not. 16 We thus find incomprehensible Justice POWELL's statement that we cite no cases in which a private decision to invoke a presumptively valid state legal process has been held to be state action. Post, at 950. Likewise, his discussion of these cases, post, at 952-953, steadfastly ignores the predicate for the holding in each case that the debtor could challenge the constitutional adequacy of the private creditor's seizure of his property. That predicate was necessarily the principle that a private party's invocation of a seemingly valid prejudgment remedy statute, coupled with the aid of a state official, satisfies the state-action requirement of the Fourteenth Amendment and warrants relief against the private party. 17 In fact, throughout the congressional debate over the 1871 Act, the bill was officially described as a bill "to enforce the provisions of the fourteenth amendment to the Constitution of the United States, and for other purposes." See also, e.g., remarks of Senator Trumbull in describing the purpose of the House in passing the Act: "[A]s the bill passed the House of Representatives, it was understood by the members of that body to go no further than to protect persons in the rights which were guarantied to them by the Constitution and laws of the United States," Cong.Globe, 42d Cong., 1st Sess., 579 (1871); and remarks of Representative Shellabarger on the relationship between § 1 of the bill and the Fourteenth Amendment, id., App. 68. 18 Our conclusion in this case is not inconsistent with the statement in Flagg Brothers that "these two elements [state action and action under color of state law] denote two separate areas of inquiry." 436 U.S., at 155-156, 98 S.Ct., at 1732-1733. First, although we hold that conduct satisfying the state-action requirement of the Fourteenth Amendment satisfies the statutory requirement of action under color of state law, it does not follow from that that all conduct that satisfies the under-color-of-state-law requirement would satisfy the Fourteenth Amendment requirement of state action. If action under color of state law means nothing more than that the individual act "with the knowledge of and pursuant to that statute," Adickes v. S. H. Kress & Co., 398 U.S., at 162, n.23, 90 S.Ct., at 1611 n.23, then clearly under Flagg Brothers that would not, in itself, satisfy the state-action requirement of the Fourteenth Amendment. Second, although we hold in this case that the under-color-of-state-law requirement does not add anything not already included within the state-action requirement of the Fourteenth Amendment, § 1983 is applicable to other constitutional provisions and statutory provisions that contain no state-action requirement. Where such a federal right is at issue, the statutory concept of action under color of state law would be a distinct element of the case not satisfied implicitly by a finding of a violation of the particular federal right. Nor is our decision today inconsistent with Polk County v. Dodson, 454 U.S. 312, 102 S.Ct. 445, 70 L.Ed.2d 509 (1981). In Polk County, we held that a public defender's actions, when performing a lawyer's traditional functions as counsel in a state criminal proceeding, would not support a § 1983 suit. Although we analyzed the public defender's conduct in light of the requirement of action "under color of state law," we specifically stated that it was not necessary in that case to consider whether that requirement was identical to the "state action" requirement of the Fourteenth Amendment: "Although this Court has sometimes treated the questions as if they were identical, see United States v. Price, 383 U.S. 787, 794, and n.7, 86 S.Ct. 1152, 1157, and n.7, 16 L.Ed.2d 267 (1966), we need not consider their relationship in order to decide this case." Id., at 322, n.12, 102 S.Ct., at 451, n.12. We concluded there that a public defender, although a state employee, in the day-to-day defense of his client, acts under canons of professional ethics in a role adversarial to the State. Accordingly, although state employment is generally sufficient to render the defendant a state actor under our analysis, infra, at 937, it was "peculiarly difficult" to detect any action of the State in the circumstances of that case. 454 U.S., at 320, 102 S.Ct., at 451. In Polk County, we also rejected respondent's claims against governmental agencies because he "failed to allege any policy that arguably violated his rights under the Sixth, Eighth, or Fourteenth Amendments." Id., at 326, 102 S.Ct., at 454. Because respondent failed to challenge any rule of conduct or decision for which the State was responsible, his allegations would not support a claim of state action under the analysis proposed below. Infra, at 937. Thus, our decision today does not suggest a different outcome in Polk County. 19 There are elements of the other state-action inquiry in the opinion as well. This is found primarily in the effort to distinguish the relationship of Moose Lodge and the State from that between the State and the restaurant considered in Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961). See 407 U.S., at 175, 92 S.Ct., at 1972. 20 The "one exception" further illustrates this point. The Court enjoined enforcement of a state rule requiring Moose Lodge to comply with its own constitution and bylaws insofar as they contained racially discriminatory provisions. State enforcement of this rule, either judicially or administratively, would, under the circumstances, amount to a governmental decision to adopt a racially discriminatory policy. 21 Contrary to the suggestion of Justice POWELL's dissent, we do not hold today that "a private party's mere invocation of state legal procedures constitutes 'joint participation' or 'conspiracy' with state officials satisfying the § 1983 requirement of action under color of law." Post, at 951. The holding today, as the above analysis makes clear, is limited to the particular context of prejudgment attachment. 22 This confusion in the nature of petitioner's allegations continued in oral argument in this Court. Although at various times counsel for petitioner seemed to deny that petitioner challenged the constitutionality of the statute, see, e.g., Tr. of Oral Arg. at 11, he also stated that "[t]he claim is that the action as taken, even if it were just line by line in accordance with Virginia law—whether or not they did it right, the claim is that it was in violation of Lugar's constitutional rights." Id., at 19. 23 Justice POWELL is concerned that private individuals who innocently make use of seemingly valid state laws would be responsible, if the law is subsequently held to be unconstitutional, for the consequences of their actions. In our view, however, this problem should be dealt with not by changing the character of the cause of action but by establishing an affirmative defense. A similar concern is at least partially responsible for the availability of a good-faith defense, or qualified immunity, to state officials. We need not reach the question of the availability of such a defense to private individuals at this juncture. What we said in Adickes, 398 U.S., at 174, n.44, 90 S.Ct., at 1617, n.44, when confronted with this question is just as applicable today: "We intimate no views concerning the relief that might be appropriate if a violation is shown. The parties have not briefed these remedial issues, and if a violation is proved they are best explored in the first instance below in light of the new record that will be developed on remand. Nor do we mean to determine at this juncture whether there are any defenses available to defendants in § 1983 actions like the one at hand. Cf. Pierson v. Ray, 386 U.S. 547 [87 S.Ct. 1213, 18 L.Ed.2d 288] (1967)" (citations omitted). * The pleadings in this case amply demonstrate that the challenged conduct was directed solely at respondents' acts. The unlawful actions alleged were that respondents made "conclusory allegations," App. 5, respondents lacked a "factual basis" for attachment, id., at 10, and respondents lacked "good cause to believe facts which would support" attachment. Id., at 19. There is no allegation of collusion or conspiracy with state actors. 1 The state action, filed in the name of the Edmondson Oil Co., alleged that Lugar owed $41,983 for products and merchandise previously delivered. App. 22. In the present suit Lugar has named as defendants both the Edmondson Oil Co. and its president, Ronald Barbour. As the respondent Barbour is the sole stockholder of Edmondson Oil Co., id., at 2, and appears to have directed all its actions in this litigation, see id., at 26, I refer throughout to Barbour as if he were the sole respondent. 2 See Va.Code § 8.01-533 et seq. (1977). At the time of the attachment in this case, the applicable provisions were Va.Code § 8-519 et seq. (1973). The Virginia attachment provisions have remained essentially in their present form despite numerous recodifications since 1819. See Va.Code § 8-519 et seq. (1950); Va.Code § 6378 et seq. (1919); Va.Code § 2959 et seq. (1887); Va.Code, ch. 151 (1849); Va.Code, ch. 123 (1819). 3 Title 42 U.S.C. § 1983, at the time in question, provided: "Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured in an action at law, suit in equity, or other proper proceeding for redress." 4 Judge Phillips' excellent opinion for the en banc Court of Appeals correctly defined the question presented as "whether the mere institution by a private litigant of presumptively valid state judicial proceedings, without any prior or subsequent collusion or concerted action by that litigant with the state officials who then proceed with adjudicative, administrative, or executive enforcement of the proceedings, constitutes action under color of state law within the contemplation of § 1983." 639 F.2d 1058, 1061-1062 (CA4 1981) (footnote omitted). 5 State officials acting in their official capacities, even if in abuse of their lawful authority, generally are held to act "under color" of law. E.g., Monroe v. Pape, 365 U.S., at 171-172, 81 S.Ct., at 475-476; Ex parte Virginia, 100 U.S. 339, 346-347, 25 L.Ed. 676 (1880). This is because such officials are "clothed with the authority" of state law, which gives them power to perpetrate the very wrongs that Congress intended § 1983 to prevent. United States v. Classic, 313 U.S. 299, 326, 61 S.Ct. 1031, 1043, 85 L.Ed. 1368 (1941); Ex parte Virginia, supra, 100 U.S. at 346-347, 25 L.Ed. 676. Cf. Polk County v. Dodson, 454 U.S. 312, 102 S.Ct. 445, 70 L.Ed.2d 509 (1981) (a public defender, representing an indigent client in a criminal proceeding, performs a function for which the authority of his state office is not needed, and therefore does not act under color of state law when engaged in a defense attorney's traditionally private roles). 6 The Court, ante, at 928, quotes United States v. Price, 383 U.S. 787, 794, n.7, 86 S.Ct. 1152, 1156, n. 7 (1966), as establishing that "[i]n cases under § 1983, 'under color' of law has consistently been treated as the same thing as the 'state action' required under the Fourteenth Amendment." In Price, however, the same conduct by the same actors constituted both "state action" and the action "under color" of law. See 383 U.S., at 794, n. 7, 86 S.Ct., at 1156, n. 7 (if an indictment alleges "conduct on the part of the 'private' defendants which constitutes 'state action,' [it also alleges] action 'under color of law'. . ."). The situation in this case is quite different. The present case involves "state action" by the Sheriff—action that also was "under color of law" under Price. But the real question here is whether the conduct of the private respondent constituted either state action or action under color of law. The Price quotation plainly does not resolve this question. And the cases cited in Price, on which the Court also relies, are similarly inapposite. 7 In Adickes the term "jointly engaged" appears to have been used specifically to connote engagement in a "conspiracy." See 398 U.S., at 152-153, 90 S.Ct., at 1605-1606. 8 The Court avers that its holding "is limited to the particular context of prejudgment attachment." Ante, at 939, n.21. However welcome, this limitation lacks a principled basis. It is unclear why a private party engages in state action when filing papers seeking an attachment of property, but not when seeking other relief (e.g., an injunction), or when summoning police to investigate a suspected crime. 9 At one stage in the litigation the respondent averred that his lawsuit raised "[n]o question of the constitutional validity of the State statutes." Plaintiff's Memorandum in Opposition to Motion to Dismiss 3. The District Court nevertheless concluded that "the complaint can only be read as challenging the constitutionality of Virginia's attachment statute." App. to Pet. for Cert. 38. The Court of Appeals agreed. 639 F.2d, at 1060, and n.1. 10 The Court finds support for its contrary view only by reading these cases as implicitly embracing the same fallacy as the Court does today. In Sniadach, Mitchell, and Di-Chem—as in this case—there was no question that state action had occurred. There, as here, some official of the State—an undisputed state actor—had undertaken either to attach property or garnish wages. For the Court, the occurrence of state action by these state officials ipso facto establishes that the private plaintiffs also must have been viewed as state actors. Given the presence of state action by the state officials, however, there was no need to inquire whether the private parties also were state actors. It is plain from the opinions that the Court did not do so. Nor, in cases arising in state court, was there any need to consider whether the private defendants had acted under color of law within the meaning of § 1983. 11 Fuentes was consolidated with a case involving similar facts, Epps v. Cortese, 326 F.Supp. 127 (E.D.Pa.1971). 12 Arguing that the patent unconstitutionality of racial discrimination was irrelevant to the "conspiracy" count in Adickes, the Court charges that this discussion confuses the conspiracy and the substantive causes of action. Ante, at 932, n.15. The Court's view is difficult to understand. In Adickes the private defendant allegedly conspired with the police to "deprive plaintiff of her right to enjoy equal treatment and service in a place of public accommodation," 398 U.S., at 150, n.5, 90 S.Ct., at 1604, n.5, and apparently to cause her discriminatory and legally baseless arrest under a vagrancy statute. Because the vagrancy statute was not challenged as invalid on its face, the Court concludes that the "joint action" or "conspiracy" count "did not involve a state law, whether 'plainly unconstitutional' or not." Ante, at 932, n.15. This conclusion is simply wrong. In the first place, the alleged "conspiracy" included an agreement to enforce a state law requiring racial segregation in restaurants. This law plainly was unconstitutional. Further, even the vagrancy statute certainly would have been unconstitutional as applied to enforce racial segregation. Presumably it was for these reasons that the Court agreed that the private defendant had "conspir[ed]" with the local police. 398 U.S., at 152, 90 S.Ct., at 1605. Adickes is entirely a different case from the one at bar. 13 At least one scholarly commentator has stated a cautious conclusion that the Virginia attachment provisions would satisfy the standards established by this Court's recent due process decisions. See Brabham, Sniadach Through Di-Chem and Backwards: An Analysis of Virginia's Attachment and Detinue Statutes, 12 U.Rich.L.Rev. 157, 195-199 (1977). The correctness of this conclusion is not of course an issue in the present posture of the case, nor is it directly relevant to the case's proper resolution. 14 The Court suggests that respondent may be entitled to claim good-faith immunity from this suit for civil damages. Ante, at 942, and n.23. This is a positive suggestion with which I agree. A holding of immunity will mitigate the ultimate cost of this litigation. It would not, however, convert the Court's holding into a just one. This case already has been in litigation for nearly five years. It will now be remanded for further proceedings. Respondent, solely because he undertook to assert rights authorized by a presumptively valid state statute, will have been subjected to the expense, distractions, and hazards of a protracted litigation.

12

457 U.S. 853 102 S.Ct. 2799 73 L.Ed.2d 435 BOARD OF EDUCATION, ISLAND TREES UNION FREE SCHOOL DISTRICT NO. 26 et al., Petitioners,v.Steven A. PICO, by his next friend Frances Pico et al. No. 80-2043. Argued March 2, 1982. Decided June 25, 1982. Syllabus Petitioner Board of Education, rejecting recommendations of a committee of parents and school staff that it had appointed, ordered that certain books, which the Board characterized as "anti-American, anti-Christian, anti-Sem[i]tic, and just plain filthy," be removed from high school and junior high school libraries. Respondent students then brought this action for declaratory and injunctive relief under 42 U.S.C. § 1983 against the Board and petitioner Board members, alleging that the Board's actions had denied respondents their rights under the First Amendment. The District Court granted summary judgment in petitioners' favor. The Court of Appeals reversed and remanded for a trial on the merits of respondents' allegations. Held : The judgment is affirmed. 638 F.2d 404, affirmed. Justice BRENNAN, joined by Justice MARSHALL and Justice STEVENS, concluded: 1 1. The First Amendment imposes limitations upon a local school board's exercise of its discretion to remove books from high school and junior high school libraries. Pp. 863-872. 2 (a) Local school boards have broad discretion in the management of school affairs, but such discretion must be exercised in a manner that comports with the transcendent imperatives of the First Amendment. Students do not "shed their constitutional rights to freedom of speech or expression at the schoolhouse gate," Tinker v. Des Moines School Dist., 393 U.S. 503, 506, 89 S.Ct. 733, 736, 21 L.Ed.2d 731, and such rights may be directly and sharply implicated by the removal of books from the shelves of a school library. While students' First Amendment rights must be construed "in light of the special characteristics of the school environment," ibid., the special characteristics of the school library make that environment especially appropriate for the recognition of such rights. Pp. 863-869. 3 (b) While petitioners might rightfully claim absolute discretion in matters of curriculum by reliance upon their duty to inculcate community values in schools, petitioners' reliance upon that duty is misplaced where they attempt to extend their claim of absolute discretion beyond the compulsory environment of the classroom into the school library and the regime of voluntary inquiry that there holds sway. P. 869. 4 (c) Petitioners possess significant discretion to determine the content of their school libraries, but that discretion may not be exercised in a narrowly partisan or political manner. Whether petitioners' removal of books from the libraries denied respondents their First Amendment rights depends upon the motivation behind petitioners' actions. Local school boards may not remove books from school libraries simply because they dislike the ideas contained in those books and seek by their removal to "prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion." West Virginia Board of Education v. Barnette, 319 U.S. 624, 642, 63 S.Ct. 1178, 1187, 87 L.Ed. 1628. If such an intention was the decisive factor in petitioners' decision, then petitioners have exercised their discretion in violation of the Constitution. Pp. 869-872 5 2. The evidentiary materials before the District Court must be construed favorably to respondents, given the procedural posture of this case. When so construed, those evidentiary materials raise a genuine issue of material fact as to whether petitioners exceeded constitutional limitations in exercising their discretion to remove the books at issue from their school libraries. Respondents' allegations, and some of the evidentiary materials before the District Court, also fail to exclude the possibility that petitioners' removal procedures were highly irregular and ad hoc—the antithesis of those procedures that might tend to allay suspicions regarding petitioners' motivation. Pp. 872-875. 6 Justice BLACKMUN concluded that a proper balance between the limited constitutional restriction imposed on school officials by the First Amendment and the broad state authority to regulate education, would be struck by holding that school officials may not remove books from school libraries for the purpose of restricting access to the political ideas or social perspectives discussed in the books, when that action is motivated simply by the officials' disapproval of the ideas involved. Pp. 879-882. 7 Justice WHITE, while agreeing that there should be a trial to resolve the factual issues, concluded that there is no necessity at this point for discussing the extent to which the First Amendment limits the school board's discretion to remove books from the school libraries. Pp. 883-884. 8 George W. Lipp, Jr., Babylon, N. Y., for petitioners. 9 Alan H. Levine, New York Civil Liberties Union, New York City, for respondents. 10 Justice BRENNAN announced the judgment of the Court and delivered an opinion, in which Justice MARSHALL and Justice STEVENS joined, and in which Justice BLACKMUN joined except for Part II-A-(1). 11 The principal question presented is whether the First Amendment1 imposes limitations upon the exercise by a local school board of its discretion to remove library books from high school and junior high school libraries. 12 * Petitioners are the Board of Education of the Island Trees Union Free School District No. 26, in New York, and Richard Ahrens, Frank Martin, Christina Fasulo, Patrick Hughes, Richard Melchers, Richard Michaels, and Louis Nessim. When this suit was brought, Ahrens was the President of the Board, Martin was the Vice President, and the remaining petitioners were Board members. The Board is a state agency charged with responsibility for the operation and administration of the public schools within the Island Trees School District, including the Island Trees High School and Island Trees Memorial Junior High School. Respondents are Steven Pico, Jacqueline Gold, Glenn Yarris, Russell Rieger, and Paul Sochinski. When this suit was brought, Pico, Gold, Yarris, and Rieger were students at the High School, and Sochinski was a student at the Junior High School. 13 In September 1975, petitioners Ahrens, Martin, and Hughes attended a conference sponsored by Parents of New York United (PONYU), a politically conservative organization of parents concerned about education legislation in the State of New York. At the conference these petitioners obtained lists of books described by Ahrens as "objectionable," App. 22, and by Martin as "improper fare for school students," id., at 101.2 It was later determined that the High School library contained nine of the listed books, and that another listed book was in the Junior High School library.3 In February 1976, at a meeting with the Superintendent of Schools and the Principals of the High School and Junior High School, the Board gave an "unofficial direction" that the listed books be removed from the library shelves and delivered to the Board's offices, so that Board members could read them.4 When this directive was carried out, it became publicized, and the Board issued a press release justifying its action. It characterized the removed books as "anti-American, anti-Christian, anti-Sem[i]tic, and just plain filthy," and concluded that "[i]t is our duty, our moral obligation, to protect the children in our schools from this moral danger as surely as from physical and medical dangers." 474 F.Supp. 387, 390 (EDNY 1979). 14 A short time later, the Board appointed a "Book Review Committee," consisting of four Island Trees parents and four members of the Island Trees schools staff, to read the listed books and to recommend to the Board whether the books should be retained, taking into account the books' "educational suitability," "good taste," "relevance," and "appropriateness to age and grade level." In July, the Committee made its final report to the Board, recommending that five of the listed books be retained5 and that two others be removed from the school libraries.6 As for the remaining four books, the Committee could not agree on two,7 took no position on one,8 and recommended that the last book be made available to students only with parental approval.9 The Board substantially rejected the Committee's report later that month, deciding that only one book should be returned to the High School library without restriction,10 that another should be made available subject to parental approval,11 but that the remaining nine books should "be removed from elementary and secondary libraries and [from] use in the curriculum." Id., at 391.12 The Board gave no reasons for rejecting the recommendations of the Committee that it had appointed. 15 Respondents reacted to the Board's decision by bringing the present action under 42 U.S.C. § 1983 in the United States District Court for the Eastern District of New York. They alleged that petitioners had 16 "ordered the removal of the books from school libraries and proscribed their use in the curriculum because particular passages in the books offended their social, political and moral tastes and not because the books, taken as a whole, were lacking in educational value." App. 4. 17 Respondents claimed that the Board's actions denied them their rights under the First Amendment. They asked the court for a declaration that the Board's actions were unconstitutional, and for preliminary and permanent injunctive relief ordering the Board to return the nine books to the school libraries and to refrain from interfering with the use of those books in the schools' curricula. Id., at 5-6. 18 The District Court granted summary judgment in favor of petitioners. 474 F.Supp. 387 (1979). In the court's view, "the parties substantially agree[d] about the motivation behind the board's actions," id., at 391—namely, that 19 "the board acted not on religious principles but on its conservative educational philosophy, and on its belief that the nine books removed from the school library and curriculum were irrelevant, vulgar, immoral, and in bad taste, making them educationally unsuitable for the district's junior and senior high school students." Id., at 392. 20 With this factual premise as its background, the court rejected respondents' contention that their First Amendment rights had been infringed by the Board's actions. Noting that statutes, history, and precedent had vested local school boards with a broad discretion to formulate educational policy,13 the court concluded that it should not intervene in " 'the daily operations of school systems' " unless " 'basic constitutional values' " were " 'sharply implicate[d],' "14 and determined that the conditions for such intervention did not exist in the present case. Acknowledging that the "removal [of the books] . . . clearly was content-based," the court nevertheless found no constitutional violation of the requisite magnitude: 21 "The board has restricted access only to certain books which the board believed to be, in essence, vulgar. While removal of such books from a school library may . . . reflect a misguided educational philosophy, it does not constitute a sharp and direct infringement of any first amendment right." Id., at 397. 22 A three-judge panel of the United States Court of Appeals for the Second Circuit reversed the judgment of the District Court, and remanded the action for a trial on respondents' allegations. 638 F.2d 404 (1980). Each judge on the panel filed a separate opinion. Delivering the judgment of the court, Judge Sifton treated the case as involving "an unusual and irregular intervention in the school libraries' operations by persons not routinely concerned with such matters," and concluded that petitioners were obliged to demonstrate a reasonable basis for interfering with respondents' First Amendment rights. Id., at 414-415. He then determined that, at least at the summary judgment stage, petitioners had not offered sufficient justification for their action,15 and concluded that respondents "should have . . . been offered an opportunity to persuade a finder of fact that the ostensible justifications for [petitioners'] actions . . . were simply pretexts for the suppression of free speech." Id., at 417.16 Judge Newman concurred in the result. Id., at 432-438. He viewed the case as turning on the contested factual issue of whether petitioners' removal decision was motivated by a justifiable desire to remove books containing vulgarities and sexual explicitness, or rather by an impermissible desire to suppress ideas. Id., at 436-437.17 We granted certiorari, 454 U.S. 891, 102 S.Ct. 385, 70 L.Ed.2d 205 (1981). II 23 We emphasize at the outset the limited nature of the substantive question presented by the case before us. Our precedents have long recognized certain constitutional limits upon the power of the State to control even the curriculum and classroom. For example, Meyer v. Nebraska, 262 U.S. 390, 43 S.Ct. 625, 67 L.Ed. 1042 (1923), struck down a state law that forbade the teaching of modern foreign languages in public and private schools, and Epperson v. Arkansas, 393 U.S. 97, 89 S.Ct. 266, 21 L.Ed.2d 228 (1968), declared unconstitutional a state law that prohibited the teaching of the Darwinian theory of evolution in any state-supported school. But the current action does not require us to re-enter this difficult terrain, which Meyer and Epperson traversed without apparent misgiving. For as this case is presented to us, it does not involve textbooks, or indeed any books that Island Trees students would be required to read.18 Respondents do not seek in this Court to impose limitations upon their school Board's discretion to prescribe the curricula of the Island Trees schools. On the contrary, the only books at issue in this case are library books, books that by their nature are optional rather than required reading. Our adjudication of the present case thus does not intrude into the classroom, or into the compulsory courses taught there. Furthermore, even as to library books, the action before us does not involve the acquisition of books. Respondents have not sought to compel their school Board to add to the school library shelves any books that students desire to read. Rather, the only action challenged in this case is the removal from school libraries of books originally placed there by the school authorities, or without objection from them. 24 The substantive question before us is still further constrained by the procedural posture of this case. Petitioners were granted summary judgment by the District Court. The Court of Appeals reversed that judgment, and remanded the action for a trial on the merits of respondents' claims. We can reverse the judgment of the Court of Appeals, and grant petitioners' request for reinstatement of the summary judgment in their favor, only if we determine that "there is no genuine issue as to any material fact," and that petitioners are "entitled to a judgment as a matter of law." Fed.Rule Civ.Proc. 56(c). In making our determination, any doubt as to the existence of a genuine issue of material fact must be resolved against petitioners as the moving party. Adickes v. S.H. Kress & Co., 398 U.S. 144, 157-159, 90 S.Ct. 1598, 1608-1609, 26 L.Ed.2d 142 (1970). Furthermore, "[o]n summary judgment the inferences to be drawn from the underlying facts contained in [the affidavits, attached exhibits, and depositions submitted below] must be viewed in the light most favorable to the party opposing the motion." United States v. Diebold, Inc., 369 U.S. 654, 655, 82 S.Ct. 993, 994, 8 L.Ed.2d 176 (1962). 25 In sum, the issue before us in this case is a narrow one, both substantively and procedurally. It may best be restated as two distinct questions. First, does the First Amendment impose any limitations upon the discretion of petitioners to remove library books from the Island Trees High School and Junior High School? Second, if so, do the affidavits and other evidentiary materials before the District Court, construed most favorably to respondents, raise a genuine issue of fact whether petitioners might have exceeded those limitations? If we answer either of these questions in the negative, then we must reverse the judgment of the Court of Appeals and reinstate the District Court's summary judgment for petitioners. If we answer both questions in the affirmative, then we must affirm the judgment below. We examine these questions in turn. A. 26 (1) 27 The Court has long recognized that local school boards have broad discretion in the management of school affairs. See, e.g., Meyer v. Nebraska, supra, at 402, 43 S.Ct., at 627; Pierce v. Society of Sisters, 268 U.S. 510, 534, 45 S.Ct. 571, 573, 69 L.Ed. 1070 (1925). Epperson v. Arkan sas, supra, 393 U.S., at 104, 89 S.Ct., at 270, reaffirmed that, by and large, "public education in our Nation is committed to the control of state and local authorities," and that federal courts should not ordinarily "intervene in the resolution of conflicts which arise in the daily operation of school systems." Tinker v. Des Moines School Dist., 393 U.S. 503, 507, 89 S.Ct. 733, 736, 21 L.Ed.2d 731 (1969), noted that we have "repeatedly emphasized . . . the comprehensive authority of the States and of school officials . . . to prescribe and control conduct in the schools." We have also acknowledged that public schools are vitally important "in the preparation of individuals for participation as citizens," and as vehicles for "inculcating fundamental values necessary to the maintenance of a democratic political system." Ambach v. Norwick, 441 U.S. 68, 76-77, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979). We are therefore in full agreement with petitioners that local school boards must be permitted "to establish and apply their curriculum in such a way as to transmit community values," and that "there is a legitimate and substantial community interest in promoting respect for authority and traditional values be they social, moral, or political." Brief for Petitioners 10.19 28 At the same time, however, we have necessarily recognized that the discretion of the States and local school boards in matters of education must be exercised in a manner that comports with the transcendent imperatives of the First Amendment. In West Virginia Board of Education v. Barnette, 319 U.S. 624, 63 S.Ct. 1178, 87 L.Ed. 1628 (1943), we held that under the First Amendment a student in a public school could not be compelled to salute the flag. We reasoned: 29 "Boards of Education . . . have, of course, important, delicate, and highly discretionary functions, but none that they may not perform within the limits of the Bill of Rights. That they are educating the young for citizenship is reason for scrupulous protection of Constitutional freedoms of the individual, if we are not to strangle the free mind at its source and teach youth to discount important principles of our government as mere platitudes." Id., at 637, 63 S.Ct. at 1185. 30 Later cases have consistently followed this rationale. Thus Epperson v. Arkansas, invalidated a State's anti-evolution statute as violative of the Establishment Clause, and reaffirmed the duty of federal courts "to apply the First Amendment's mandate in our educational system where essential to safeguard the fundamental values of freedom of speech and inquiry." 393 U.S., at 104, 89 S.Ct., at 270. And Tinker v. Des Moines School Dist., supra, held that a local school board had infringed the free speech rights of high school and junior high school students by suspending them from school for wearing black armbands in class as a protest against the Government's policy in Vietnam; we stated there that the "comprehensive authority . . . of school officials" must be exercised "consistent with fundamental constitutional safeguards." 393 U.S., at 507, 89 S.Ct., at 736. In sum, students do not "shed their constitutional rights to freedom of speech or expression at the schoolhouse gate," id., at 506, 89 S.Ct., at 736, and therefore local school boards must discharge their "important, delicate, and highly discretionary functions" within the limits and constraints of the First Amendment. 31 The nature of students' First Amendment rights in the context of this case requires further examination. West Virginia Board of Education v. Barnette, supra, is instructive. There the Court held that students' liberty of conscience could not be infringed in the name of "national unity" or "patriotism." 319 U.S., at 640-641, 63 S.Ct., at 1186. We explained that 32 "the action of the local authorities in compelling the flag salute and pledge transcends constitutional limitations on their power and invades the sphere of intellect and spirit which it is the purpose of the First Amendment to our Constitution to reserve from all official control." Id., at 642, 63 S.Ct., at 1187. 33 Similarly, Tinker v. Des Moines School Dist., supra, held that students' rights to freedom of expression of their political views could not be abridged by reliance upon an "undifferentiated fear or apprehension of disturbance" arising from such expression: 34 "Any departure from absolute regimentation may cause trouble. Any variation from the majority's opinion may inspire fear. Any word spoken, in class, in the lunchroom, or on the campus, that deviates from the views of another person may start an argument or cause a disturbance. But our Constitution says we must take this risk, Terminiello v. Chicago, 337 U.S. 1 [69 S.Ct. 894, 93 L.Ed. 1131] (1949); and our history says that it is this sort of hazardous freedom—this kind of openness—that is the basis of our national strength and of the independence and vigor of Americans who grow up and live in this . . . often disputatious society." 393 U.S., at 508-509, 89 S.Ct., at 737. 35 In short, "First Amendment rights, applied in light of the special characteristics of the school environment, are available to . . . students." Id., at 506, 89 S.Ct., at 736. 36 Of course, courts should not "intervene in the resolution of conflicts which arise in the daily operation of school systems" unless "basic constitutional values" are "directly and sharply implicate[d]" in those conflicts. Epperson v. Arkansas, 393 U.S., at 104, 89 S.Ct., at 270. But we think that the First Amendment rights of students may be directly and sharply implicated by the removal of books from the shelves of a school library. Our precedents have focused "not only on the role of the First Amendment in fostering individual self-expression but also on its role in affording the public access to discussion, debate, and the dissemination of information and ideas." First National Bank of Boston v. Bellotti, 435 U.S. 765, 783, 98 S.Ct. 1407, 1419, 55 L.Ed.2d 707 (1978). And we have recognized that "the State may not, consistently with the spirit of the First Amendment, contract the spectrum of available knowledge." Griswold v. Connecticut, 381 U.S. 479, 482, 85 S.Ct. 1678, 1680, 14 L.Ed.2d 510 (1965). In keeping with this principle, we have held that in a variety of contexts "the Constitution protects the right to receive information and ideas." Stanley v. Georgia, 394 U.S. 557, 564, 89 S.Ct. 1243, 1247, 22 L.Ed.2d 542 (1969); see Kleindienst v. Mandel, 408 U.S. 753, 762-763, 92 S.Ct. 2576, 2581, 33 L.Ed.2d 683 (1972) (citing cases). This right is an inherent corollary of the rights of free speech and press that are explicitly guaranteed by the Constitution, in two senses. First, the right to receive ideas follows ineluctably from the sender's First Amendment right to send them: "The right of freedom of speech and press . . . embraces the right to distribute literature, and necessarily protects the right to receive it." Martin v. Struthers, 319 U.S. 141, 143, 63 S.Ct. 862, 863, 87 L.Ed. 1313 (1943) (citation omitted). "The dissemination of ideas can accomplish nothing if otherwise willing addressees are not free to receive and consider them. It would be a barren marketplace of ideas that had only sellers and no buyers." Lamont v. Postmaster General, 381 U.S. 301, 308, 85 S.Ct. 1493, 1497, 14 L.Ed.2d 398 (1965) (BRENNAN, J., concurring). 37 More importantly, the right to receive ideas is a necessary predicate to the recipient's meaningful exercise of his own rights of speech, press, and political freedom. Madison admonished us: 38 "A popular Government, without popular information, or the means of acquiring it, is but a Prologue to a Farce or a Tragedy; or, perhaps both. Knowledge will forever govern ignorance: And a people who mean to be their own Governors, must arm themselves with the power which knowledge gives." 9 Writings of James Madison 103 (G. Hunt ed. 1910).20 39 As we recognized in Tinker, students too are beneficiaries of this principle: 40 "In our system, students may not be regarded as closed-circuit recipients of only that which the State chooses to communicate. . . . [S]chool officials cannot suppress 'expressions of feeling with which they do not wish to contend.' " 393 U.S., at 511, 89 S.Ct., at 739 (quoting Burnside v. Byars, 363 F.2d 744, 749 (CA5 1966)). 41 In sum, just as access to ideas makes it possible for citizens generally to exercise their rights of free speech and press in a meaningful manner, such access prepares students for active and effective participation in the pluralistic, often contentious society in which they will soon be adult members. Of course all First Amendment rights accorded to students must be construed "in light of the special characteristics of the school environment." Tinker v. Des Moines School Dist., 393 U.S., at 506, 89 S.Ct., at 736. But the special characteristics of the school library make that environment especially appropriate for the recognition of the First Amendment rights of students. 42 A school library, no less than any other public library, is "a place dedicated to quiet, to knowledge, and to beauty." Brown v. Louisiana, 383 U.S. 131, 142, 86 S.Ct. 719, 724, 15 L.Ed.2d 637 (1966) (opinion of Fortas, J.). Keyishian v. Board of Regents, 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967), observed that " 'students must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding.' "21 The school library is the principal locus of such freedom. As one District Court has well put it, in the school library 43 "a student can literally explore the unknown, and discover areas of interest and thought not covered by the prescribed curriculum. . . . Th[e] student learns that a library is a place to test or expand upon ideas presented to him, in or out of the classroom." Right to Read Defense Committee v. School Committee, 454 F.Supp. 703, 715 (Mass.1978). 44 Petitioners emphasize the inculcative function of secondary education, and argue that they must be allowed unfettered discretion to "transmit community values" through the Island Trees schools. But that sweeping claim overlooks the unique role of the school library. It appears from the record that use of the Island Trees school libraries is completely voluntary on the part of students. Their selection of books from these libraries is entirely a matter of free choice; the libraries afford them an opportunity at self-education and individual enrichment that is wholly optional. Petitioners might well defend their claim of absolute discretion in matters of curriculum by reliance upon their duty to inculcate community values. But we think that petitioners' reliance upon that duty is misplaced where, as here, they attempt to extend their claim of absolute discretion beyond the compulsory environment of the classroom, into the school library and the regime of voluntary inquiry that there holds sway. 45 (2) 46 In rejecting petitioners' claim of absolute discretion to remove books from their school libraries, we do not deny that local school boards have a substantial legitimate role to play in the determination of school library content. We thus must turn to the question of the extent to which the First Amendment places limitations upon the discretion of petitioners to remove books from their libraries. In this inquiry we enjoy the guidance of several precedents. West Virginia Board of Education v. Barnette, stated: 47 "If there is any fixed star in our constitutional constellation, it is that no official, high or petty, can prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion . . . . If there are any circumstances which permit an exception, they do not now occur to us." 319 U.S., at 642, 63 S.Ct., at 1187. 48 This doctrine has been reaffirmed in later cases involving education. For example, Keyishian v. Board of Regents, supra, 385 U.S., at 603, 87 S.Ct., at 683, noted that "the First Amendment . . . does not tolerate laws that cast a pall of orthodoxy over the classroom"; see also Epperson v. Arkansas, 393 U.S., at 104-105, 89 S.Ct., at 270. And Mt. Healthy City Board of Ed. v. Doyle, 429 U.S. 274, 97 S.Ct. 568, 50 L.Ed.2d 471 (1977), recognized First Amendment limitations upon the discretion of a local school board to refuse to rehire a nontenured teacher. The school board in Mt. Healthy had declined to renew respondent Doyle's employment contract, in part because he had exercised his First Amendment rights. Although Doyle did not have tenure, and thus "could have been discharged for no reason whatever," Mt. Healthy held that he could "nonetheless establish a claim to reinstatement if the decision not to rehire him was made by reason of his exercise of constitutionally protected First Amendment freedoms." Id., at 283-284, 97 S.Ct., at 574. We held further that once Doyle had shown "that his conduct was constitutionally protected, and that this conduct was a 'substantial factor' . . . in the Board's decision not to rehire him," the school board was obliged to show "by a preponderance of the evidence that it would have reached the same decision as to respondent's reemployment even in the absence of the protected conduct." Id., at 287, 97 S.Ct., at 576. 49 With respect to the present case, the message of these precedents is clear. Petitioners rightly possess significant discretion to determine the content of their school libraries. But that discretion may not be exercised in a narrowly partisan or political manner. If a Democratic school board, motivated by party affiliation, ordered the removal of all books written by or in favor of Republicans, few would doubt that the order violated the constitutional rights of the students denied access to those books. The same conclusion would surely apply if an all-white school board, motivated by racial animus, decided to remove all books authored by blacks or advocating racial equality and integration. Our Constitution does not permit the official suppression of ideas. Thus whether petitioners' removal of books from their school libraries denied respondents their First Amendment rights depends upon the motivation behind petitioners' actions. If petitioners intended by their removal decision to deny respondents access to ideas with which petitioners disagreed, and if this intent was the decisive factor in petitioners' decision,22 then petitioners have exercised their discretion in violation of the Constitution. To permit such intentions to control official actions would be to encourage the precise sort of officially prescribed orthodoxy unequivocally condemned in Barnette. On the other hand, respondents implicitly concede that an unconstitutional motivation would not be demonstrated if it were shown that petitioners had decided to remove the books at issue because those books were pervasively vulgar. Tr. of Oral Arg. 36. And again, respondents concede that if it were demonstrated that the removal decision was based solely upon the "educational suitability" of the books in question, then their removal would be "perfectly permissible." Id., at 53. In other words, in respondents' view such motivations, if decisive of petitioners' actions, would not carry the danger of an official suppression of ideas, and thus would not violate respondents' First Amendment rights. 50 As noted earlier, nothing in our decision today affects in any way the discretion of a local school board to choose books to add to the libraries of their schools. Because we are concerned in this case with the suppression of ideas, our holding today affects only the discretion to remove books. In brief, we hold that local school boards may not remove books from school library shelves simply because they dislike the ideas contained in those books and seek by their removal to "prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion." West Virginia Board of Education v. Barnette, 319 U.S., at 642, 63 S.Ct., at 1187. Such purposes stand inescapably condemned by our precedents. B 51 We now turn to the remaining question presented by this case: Do the evidentiary materials that were before the District Court, when construed most favorably to respondents, raise a genuine issue of material fact whether petitioners exceeded constitutional limitations in exercising their discretion to remove the books from the school libraries? We conclude that the materials do raise such a question, which forecloses summary judgment in favor of petitioners. 52 Before the District Court, respondents claimed that petitioners' decision to remove the books "was based on [their] personal values, morals and tastes." App. 139. Respondents also claimed that petitioners objected to the books in part because excerpts from them were "anti-American." Id., at 140. The accuracy of these claims was partially conceded by petitioners,23 and petitioners' own affidavits lent further support to respondents' claims.24 In addition, the record developed in the District Court shows that when petitioners offered their first public explanation for the removal of the books, they relied in part on the assertion that the removed books were "anti-American," and "offensive to . . . Americans in general." 474 F.Supp., at 390.25 Furthermore, while the Book Review Committee appointed by petitioners was instructed to make its recommendations based upon criteria that appear on their face to be permissible—the books' "educational suitability," "good taste," "relevance," and "appropriateness to age and grade level," App. 67—the Committee's recommendations that five of the books be retained and that only two be removed were essentially rejected by petitioners, without any statement of reasons for doing so. Finally, while petitioners originally defended their removal decision with the explanation that "these books contain obscenities, blasphemies, brutality, and perversion beyond description," 474 F.Supp., at 390, one of the books, A Reader for Writers, was removed even though it contained no such language. 638 F.2d, at 428, n. 6 (Mansfield, J., dissenting). 53 Standing alone, this evidence respecting the substantive motivations behind petitioners' removal decision would not be decisive. This would be a very different case if the record demonstrated that petitioners had employed established, regular, and facially unbiased procedures for the review of controversial materials. But the actual record in the case before us suggests the exact opposite. Petitioners' removal procedures were vigorously challenged below by respondents, and the evidence on this issue sheds further light on the issue of petitioners' motivations.26 Respondents alleged that in making their removal decision petitioners ignored "the advice of literary experts," the views of "librarians and teachers within the Island Trees School system," the advice of the Superintendent of Schools, and the guidance of publications that rate books for junior and senior high school students. App. 128-129. Respondents also claimed that petitioners' decision was based solely on the fact that the books were named on the PONYU list received by petitioners Ahrens, Martin, and Hughes, and that petitioners "did not undertake an independent review of other books in the [school] libraries." Id., at 129-130. Evidence before the District Court lends support to these claims. The record shows that immediately after petitioners first ordered the books removed from the library shelves, the Superintendent of Schools reminded them that "we already have a policy . . . designed expressly to handle such problems," and recommended that the removal decision be approached through this established channel. See n. 4, supra. But the Board disregarded the Superintendent's advice, and instead resorted to the extraordinary procedure of appointing a Book Review Committee—the advice of which was later rejected without explanation. In sum, respondents' allegations and some of the evidentiary materials presented below do not rule out the possibility that petitioners' removal procedures were highly irregular and ad hoc—the antithesis of those procedures that might tend to allay suspicions regarding petitioners' motivations. 54 Construing these claims, affidavit statements, and other evidentiary materials in a manner favorable to respondents, we cannot conclude that petitioners were "entitled to a judgment as a matter of law." The evidence plainly does not foreclose the possibility that petitioners' decision to remove the books rested decisively upon disagreement with constitutionally protected ideas in those books, or upon a desire on petitioners' part to impose upon the students of the Island Trees High School and Junior High School a political orthodoxy to which petitioners and their constituents adhered. Of course, some of the evidence before the District Court might lead a finder of fact to accept petitioners' claim that their removal decision was based upon constitutionally valid concerns. But that evidence at most creates a genuine issue of material fact on the critical question of the credibility of petitioners' justifications for their decision: On that issue, it simply cannot be said that there is no genuine issue as to any material fact. 55 The mandate shall issue forthwith. 56 Affirmed. 57 Justice BLACKMUN, concurring in part and concurring in the judgment. 58 While I agree with much in today's plurality opinion, and while I accept the standard laid down by the plurality to guide proceedings on remand, I write separately because I have a somewhat different perspective on the nature of the First Amendment right involved. 59 * To my mind, this case presents a particularly complex problem because it involves two competing principles of constitutional stature. On the one hand, as the dissenting opinions demonstrate, and as we all can agree, the Court has acknowledged the importance of the public schools "in the preparation of individuals for participation as citizens, and in the preservation of the values on which our society rests." Ambach v. Norwick, 441 U.S. 68, 76, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979). See, also, ante, at 863-864 (plurality opinion). Because of the essential socializing function of schools, local education officials may attempt "to promote civic virtues," Ambach v. Norwick, 441 U.S., at 80, 99 S.Ct., at 1596, and to "awake[n] the child to cultural values." Brown v. Board of Education, 347 U.S. 483, 493, 74 S.Ct. 686, 691, 98 L.Ed. 873 (1954). Indeed, the Constitution presupposes the existence of an informed citizenry prepared to participate in governmental affairs, and these democratic principles obviously are constitutionally incorporated into the structure of our government. It therefore seems entirely appropriate that the State use "public schools [to] . . . inculcat[e] fundamental values necessary to the maintenance of a democratic political system." Ambach v. Norwick, 441 U.S., at 77, 99 S.Ct., at 1594. 60 On the other hand, as the plurality demonstrates, it is beyond dispute that schools and school boards must operate within the confines of the First Amendment. In a variety of academic settings the Court therefore has acknowledged the force of the principle that schools, like other enterprises operated by the State, may not be run in such a manner as to "prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion." West Virginia Board of Education v. Barnette, 319 U.S. 624, 642, 63 S.Ct. 1178, 1187, 87 L.Ed. 1628 (1943). While none of these cases define the limits of a school board's authority to choose a curriculum and academic materials, they are based on the general proposition that "state-operated schools may not be enclaves of totalitarianism. . . . In our system, students may not be regarded as closed-circuit recipients of only that which the State chooses to communicate." Tinker v. Des Moines School Dist., 393 U.S. 503, 511, 89 S.Ct. 733, 739, 21 L.Ed.2d 731 (1969). 61 The Court in Tinker thus rejected the view that "a State might so conduct its schools as to 'foster a homogeneous people.' " Id., at 511, 89 S.Ct., at 739, quoting Meyer v. Nebraska, 262 U.S. 390, 402, 43 S.Ct. 625, 627, 67 L.Ed. 1042 (1923). Similarly, Keyishian v. Board of Regents, 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967)—a case that involved the State's attempt to remove "subversives" from academic positions at its universities, but that addressed itself more broadly to public education in general—held that "[t]he classroom is peculiarly the 'marketplace of ideas' "; the First Amendment therefore "does not tolerate laws that cast a pall of orthodoxy over the classroom." Id., at 603, 87 S.Ct., at 683. And Barnette is most clearly applicable here: its holding was based squarely on the view that "[f]ree public education, if faithful to the ideal of secular instruction and political neutrality, will not be partisan or enemy of any class, creed, party, or faction." 319 U.S., at 637, 63 S.Ct., at 1185. The Court therefore made it clear that imposition of "ideological discipline" was not a proper undertaking for school authorities. Ibid. 62 In combination with more generally applicable First Amendment rules, most particularly the central proscription of content-based regulations of speech, see Police Department of Chicago v. Mosley, 408 U.S. 92, 92 S.Ct. 2286, 33 L.Ed.2d 212 (1972), the cases outlined above yield a general principle: the State may not suppress exposure to ideas—for the sole purpose of suppressing exposure to those ideas—absent sufficiently compelling reasons. Because the school board must perform all its functions "within the limits of the Bill of Rights," Barnette, 319 U.S., at 637, 63 S.Ct., at 1185, this principle necessarily applies in at least a limited way to public education. Surely this is true in an extreme case: as the plurality notes, it is difficult to see how a school board, consistent with the First Amendment, could refuse for political reasons to buy books written by Democrats or by Negroes, or books that are "anti-American" in the broadest sense of that term. Indeed, Justice REHNQUIST appears "cheerfully [to] concede" this point. Post, at 907 (dissenting opinion). 63 In my view, then, the principle involved here is both narrower and more basic than the "right to receive information" identified by the plurality. I do not suggest that the State has any affirmative obligation to provide students with information or ideas, something that may well be associated with a "right to receive." See post, at 887 (BURGER, C.J., dissenting); post, at 915-918 (REHNQUIST, J., dissenting). And I do not believe, as the plurality suggests, that the right at issue here is somehow associated with the peculiar nature of the school library, see ante, at 868-869; if schools may be used to inculcate ideas, surely libraries may play a role in that process.1 Instead, I suggest that certain forms of state discrimination between ideas are improper. In particular, our precedents command the conclusion that the State may not act to deny access to an idea simply because state officials disapprove of that idea for partisan or political reasons.2 64 Certainly, the unique environment of the school places substantial limits on the extent to which official decisions may be restrained by First Amendment values. But that environment also makes it particularly important that some limits be imposed. The school is designed to, and inevitably will, inculcate ways of thought and outlooks; if educators intentionally may eliminate all diversity of thought, the school will "strangle the free mind at its source and teach youth to discount important principles of our government as mere platitudes." Barnette, 319 U.S., at 637, 63 S.Ct., at 1185. As I see it, then, the question in this case is how to make the delicate accommodation between the limited constitutional restriction that I think is imposed by the First Amendment, and the necessarily broad state authority to regulate education. In starker terms, we must reconcile the schools' "inculcative" function with the First Amendment's bar on "prescriptions of orthodoxy." II 65 In my view, we strike a proper balance here by holding that school officials may not remove books for the purpose of restricting access to the political ideas or social perspectives discussed in them, when that action is motivated simply by the officials' disapproval of the ideas involved. It does not seem radical to suggest that state action calculated to suppress novel ideas or concepts is fundamentally antithetical to the values of the First Amendment. At a minimum, allowing a school board to engage in such conduct hardly teaches children to respect the diversity of ideas that is fundamental to the American system. In this context, then, the school board must "be able to show that its action was caused by something more than a mere desire to avoid the discomfort and unpleasantness that always accompany an unpopular viewpoint," Tinker v. Des Moines School Dist., 393 U.S., at 509, 89 S.Ct., at 737, and that the board had something in mind in addition to the suppression of partisan or political views it did not share. 66 As I view it, this is a narrow principle. School officials must be able to choose one book over another, without outside interference, when the first book is deemed more relevant to the curriculum, or better written, or when one of a host of other politically neutral reasons is present. These decisions obviously will not implicate First Amendment values. And even absent space or financial limitations, First Amendment principles would allow a school board to refuse to make a book available to students because it contains offensive language, cf. FCC v. Pacifica Foundation, 438 U.S. 726, 757, 98 S.Ct. 3026, 3044, 57 L.Ed.2d 1073 (1978) (POWELL, J., concurring), or because it is psychologically or intellectually inappropriate for the age group, or even, perhaps, because the ideas it advances are "manifestly inimical to the public welfare." Pierce v. Society of Sisters, 268 U.S. 510, 534, 45 S.Ct. 571, 573, 69 L.Ed. 1070 (1925). And, of course, school officials may choose one book over another because they believe that one subject is more important, or is more deserving of emphasis. 67 As is evident from this discussion, I do not share Justice REHNQUIST's view that the notion of "suppression of ideas" is not a useful analytical concept. See post, at 918-920 (dissenting opinion). Indeed, Justice REHNQUIST's discussion itself demonstrates that "access to ideas" has been given meaningful application in a variety of contexts. See post., at 910-920, 914 ("[e]ducation consists of the selective presentation and explanation of ideas"). And I believe that tying the First Amendment right to the purposeful suppression of ideas makes the concept more manageable than Justice REHNQUIST acknowledges. Most people would recognize that refusing to allow discussion of current events in Latin class is a policy designed to "inculcate" Latin, not to suppress ideas. Similarly, removing a learned treatise criticizing American foreign policy from an elementary school library because the students would not understand it is an action unrelated to the purpose of suppressing ideas. In my view, however, removing the same treatise because it is "anti-American" raises a far more difficult issue. 68 It is not a sufficient answer to this problem that a State operates a school in its role as "educator," rather than its role as "sovereign," see post, at 908-910 (REHNQUIST, J., dissenting), for the First Amendment has application to all the State's activities. While the State may act as "property owner" when it prevents certain types of expressive activity from taking place on public lands, for example, see post, at 908-909, few would suggest that the State may base such restrictions on the content of the speaker's message, or may take its action for the purpose of suppressing access to the ideas involved. See Police Department of Chicago v. Mosley, 408 U.S., at 96, 92 S.Ct., at 2290. And while it is not clear to me from Justice REHNQUIST's discussion whether a State operates its public libraries in its "role as sovereign," surely difficult constitutional problems would arise if a State chose to exclude "anti-American" books from its public libraries—even if those books remained available at local bookstores. 69 Concededly, a tension exists between the properly inculcative purposes of public education and any limitation on the school board's absolute discretion to choose academic materials. But that tension demonstrates only that the problem here is a difficult one, not that the problem should be resolved by choosing one principle over another. As the Court has recognized, school officials must have the authority to make educationally appropriate choices in designing a curriculum: "the State may 'require teaching by instruction and study of all in our history and in the structure and organization of our government, including the guaranties of civil liberty, which tend to inspire patriotism and love of country.' " Barnette, 319 U.S., at 631, 63 S.Ct., at 1181, quoting Minersville School District v. Gobitis, 310 U.S. 586, 604, 60 S.Ct. 1010, 1017, 84 L.Ed. 1375 (1940) (Stone, J., dissenting). Thus school officials may seek to instill certain values "by persuasion and example," 319 U.S., at 640, 63 S.Ct., at 1186, or by choice of emphasis. That sort of positive educational action, however, is the converse of an intentional attempt to shield students from certain ideas that officials find politically distasteful. Arguing that the majority in the community rejects the ideas involved, see post, at 889, 891-892 (BURGER, C.J., dissenting), does not refute this principle: "The very purpose of a Bill of Rights was to withdraw certain subjects from the vicissitudes of political controversy, to place them beyond the reach of majorities and officials . . . ." Barnette, 319 U.S., at 638, 63 S.Ct., at 1185. 70 As THE CHIEF JUSTICE notes, the principle involved here may be difficult to apply in an individual case. See post, at 889 (dissenting opinion). But on a record as sparse as the one before us, the plurality can hardly be faulted for failing to explore every possible ramification of its decision. And while the absence of a record "underscore[s] the views of those of us who originally felt that the cas[e] should not be taken," Ferguson v. Moore-McCormack Lines, Inc., 352 U.S. 521, 559, 77 S.Ct. 457, 478, 1 L.Ed.2d 511 (1957) (opinion of Harlan, J.) the case is here, and must be decided. 71 Because I believe that the plurality has derived a standard similar to the one compelled by my analysis, I join all but Part II-A(1) of the plurality opinion. 72 Justice WHITE, concurring in the judgment. 73 The District Court found that the books were removed from the school library because the school board believed them "to be, in essence, vulgar." 474 F.Supp. 387, 397 (EDNY 1979). Both Court of Appeals judges in the majority concluded, however, that there was a material issue of fact that precluded summary judgment sought by petitioners. The unresolved factual issue, as I understand it, is the reason or reasons underlying the school board's removal of the books. I am not inclined to disagree with the Court of Appeals on such a fact-bound issue and hence concur in the judgment of affirmance. Presumably this will result in a trial and the making of a full record and findings on the critical issues. 74 The plurality seems compelled to go further and issue a dissertation on the extent to which the First Amendment limits the discretion of the school board to remove books from the school library. I see no necessity for doing so at this point. When findings of fact and conclusions of law are made by the District Court, that may end the case. If, for example, the District Court concludes after a trial that the books were removed for their vulgarity, there may be no appeal. In any event, if there is an appeal, if there is dissatisfaction with the subsequent Court of Appeals' judgment, and if certiorari is sought and granted, there will be time enough to address the First Amendment issues that may then be presented. 75 I thus prefer the course taken by the Court in Kennedy v. Silas Mason Co., 334 U.S. 249, 68 S.Ct. 1031, 92 L.Ed. 1347 (1948), a suit involving overtime compensation under the Fair Labor Standards Act. Summary judgment had been granted by the District Court and affirmed by the Court of Appeals. This Court reversed, holding that summary judgment was improvidently granted, and remanded for trial so that a proper record could be made. The Court expressly abjured issuing its advice on the legal issues involved. Writing for the Court, Justice Jackson stated: 76 "We consider it the part of good judicial administration to withhold decision of the ultimate questions involved in this case until this or another record shall present a more solid basis of findings based on litigation or on a comprehensive statement of agreed facts. While we might be able, on the present record, to reach a conclusion that would decide the case, it might well be found later to be lacking in the thoroughness that should precede judgment of this importance and which it is the purpose of the judicial process to provide. 77 "Without intimating any conclusion on the merits, we vacate the judgments below and remand the case to the District Court for reconsideration and amplification of the record in the light of this opinion and of present contentions." Id., at 257, 68 S.Ct., at 1034. 78 We took a similar course in a unanimous per curiam opinion in Dombrowski v. Eastland, 387 U.S. 82, 87 S.Ct. 1425, 18 L.Ed.2d 577 (1967). There we overturned a summary judgment since it was necessary to resolve a factual dispute about collaboration between one of the respondents and a state legislative committee. We remanded, saying: "In the absence of the factual refinement which can occur only as a result of trial, we need not and, indeed, could not express judgment as to the legal consequences of such collaboration, if it occurred." Id., at 84, 87 S.Ct., at 1427. 79 The Silas Mason case turned on issues of statutory construction. It is even more important that we take a similar course in cases like Dombrowski, which involved Speech or Debate Clause immunity, and in this one, which poses difficult First Amendment issues in a largely uncharted field. We should not decide constitutional questions until it is necessary to do so, or at least until there is better reason to address them than are evident here. I therefore concur in the judgment of affirmance. 80 Chief Justice BURGER, with whom Justice POWELL, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 81 The First Amendment, as with other parts of the Constitution, must deal with new problems in a changing world. In an attempt to deal with a problem in an area traditionally left to the states, a plurality of the Court, in a lavish expansion going beyond any prior holding under the First Amendment, expresses its view that a school board's decision concerning what books are to be in the school library is subject to federal-court review.1 Were this to become the law, this Court would come perilously close to becoming a "super censor" of school board library decisions. Stripped to its essentials, the issue comes down to two important propositions: first, whether local schools are to be administered by elected school boards, or by federal judges and teenage pupils; and second, whether the values of morality, good taste, and relevance to education are valid reasons for school board decisions concerning the contents of a school library. In an attempt to place this case within the protection of the First Amendment, the plurality suggests a new "right" that, when shorn of the plurality's rhetoric, allows this Court to impose its own views about what books must be made available to students.2 82 * A. 83 I agree with the fundamental proposition that "students do not 'shed their constitutional rights to freedom of speech or expression at the schoolhouse gate.' " Ante, at 865. For example, the Court has held that a school board cannot compel a student to participate in a flag salute ceremony, West Virginia Bd. of Education v. Barnette, 319 U.S. 624, 63 S.Ct. 1178, 87 L.Ed. 1628 (1943), or prohibit a student from expressing certain views, so long as that expression does not disrupt the educational process. Tinker v. Des Moines School Dist., 393 U.S. 503, 89 S.Ct. 733, 21 L.Ed.2d 731 (1969). Here, however, no restraints of any kind are placed on the students. They are free to read the books in question, which are available at public libraries and bookstores; they are free to discuss them in the classroom or elsewhere. Despite this absence of any direct external control on the students' ability to express themselves, the plurality suggests that there is a new First Amendment "entitlement" to have access to particular books in a school library. 84 The plurality cites Meyer v. Nebraska, 262 U.S. 390, 43 S.Ct. 625, 67 L.Ed. 1042 (1923), which struck down a state law that restricted the teaching of modern foreign languages in public and private schools, and Epperson v. Arkansas, 393 U.S. 97, 89 S.Ct. 266, 21 L.Ed.2d 228 (1968), which declared unconstitutional under the Establishment Clause a law banning the teaching of Darwinian evolution, to establish the validity of federal-court interference with the functioning of schools. The plurality finds it unnecessary "to re-enter this difficult terrain," ante, at 861, yet in the next breath relies on these very cases and others to establish the previously unheard of "right" of access to particular books in the public school library.3 The apparent underlying basis of the plurality's view seems to be that students have an enforceable "right" to receive the information and ideas that are contained in junior and senior high school library books. Ante, at 866. This "right" purportedly follows "ineluctably" from the sender's First Amendment right to freedom of speech and as a "necessary predicate" to the recipient's meaningful exercise of his own rights of speech, press, and political freedom. Ante, at 866-867. No such right, however, has previously been recognized. 85 It is true that where there is a willing distributor of materials, the government may not impose unreasonable obstacles to dissemination by the third party. Virginia Pharmacy Board v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976). And where the speaker desires to express certain ideas, the government may not impose unreasonable restraints. Tinker v. Des Moines School Dist., supra. It does not follow, however, that a school board must affirmatively aid the speaker in his communication with the recipient. In short the plurality suggests today that if a writer has something to say, the government through its schools must be the courier. None of the cases cited by the plurality establish this broad-based proposition. 86 First, the plurality argues that the right to receive ideas is derived in part from the sender's First Amendment rights to send them. Yet we have previously held that a sender's rights are not absolute. Rowan v. Post Office Dept., 397 U.S. 728, 90 S.Ct. 1484, 25 L.Ed.2d 736 (1970).4 Never before today has the Court indicated that the government has an obligation to aid a speaker or author in reaching an audience. 87 Second, the plurality concludes that "the right to receive ideas is a necessary predicate to the recipient's meaningful exercise of his own rights of speech, press, and political freedom." Ante, at 867 (emphasis in original). However, the "right to receive information and ideas," Stanley v. Georgia, 394 U.S. 557, 564, 89 S.Ct. 1243, 1247, 22 L.Ed.2d 542 (1969), cited ante, at 867, does not carry with it the concomitant right to have those ideas affirmatively provided at a particular place by the government. The plurality cites James Madison to emphasize the importance of having an informed citizenry. Ibid. We all agree with Madison, of course, that knowledge is necessary for effective government. Madison's view, however, does not establish a right to have particular books retained on the school library shelves if the school board decides that they are inappropriate or irrelevant to the school's mission. Indeed, if the need to have an informed citizenry creates a "right," why is the government not also required to provide ready access to a variety of information? This same need would support a constitutional "right" of the people to have public libraries as part of a new constitutional "right" to continuing adult education. 88 The plurality also cites Tinker, supra, to establish that the recipient's right to free speech encompasses a right to have particular books retained on the school library shelf. Ante, at 868. But the cited passage of Tinker notes only that school officials may not prohibit a student from expressing his or her view on a subject unless that expression interferes with the legitimate operations of the school. The government does not "contract the spectrum of available knowledge." Griswold v. Connecticut, 381 U.S. 479, 482, 85 S.Ct. 1678, 1680, 14 L.Ed.2d 510 (1965), cited ante, at 866, by choosing not to retain certain books on the school library shelf; it simply chooses not to be the conduit for that particular information. In short, even assuming the desirability of the policy expressed by the plurality, there is not a hint in the First Amendment, or in any holding of this Court, of a "right" to have the government provide continuing access to certain books. B 89 Whatever role the government might play as a conduit of information, schools in particular ought not be made a slavish courier of the material of third parties. The plurality pays homage to the ancient verity that in the administration of the public schools " 'there is a legitimate and substantial community interest in promoting respect for authority and traditional values be they social, moral, or political.' " Ante, at 864. If, as we have held, schools may legitimately be used as vehicles for "inculcating fundamental values necessary to the maintenance of a democratic political system," Ambach v. Norwick, 441 U.S. 68, 77, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979), school authorities must have broad discretion to fulfill that obligation. Presumably all activity within a primary or secondary school involves the conveyance of information and at least an implied approval of the worth of that information. How are "fundamental values" to be inculcated except by having school boards make content-based decisions about the appropriateness of retaining materials in the school library and curriculum. In order to fulfill its function, an elected school board must express its views on the subjects which are taught to its students. In doing so those elected officials express the views of their community; they may err, of course, and the voters may remove them. It is a startling erosion of the very idea of democratic government to have this Court arrogate to itself the power the plurality asserts today. 90 The plurality concludes that under the Constitution school boards cannot choose to retain or dispense with books if their discretion is exercised in a "narrowly partisan or political manner." Ante, at 870. The plurality concedes that permissible factors are whether the books are "pervasively vulgar," ante, at 871, or educationally unsuitable. Ibid. "Educational suitability," however, is a standardless phrase. This conclusion will undoubtedly be drawn in many—if not most—instances because of the decisionmaker's content-based judgment that the ideas contained in the book or the idea expressed from the author's method of communication are inappropriate for teenage pupils. 91 The plurality also tells us that a book may be removed from a school library if it is "pervasively vulgar." But why must the vulgarity be "pervasive" to be offensive? Vulgarity might be concentrated in a single poem or a single chapter or a single page, yet still be inappropriate. Or a school board might reasonably conclude that even "random" vulgarity is inappropriate for teenage school students. A school board might also reasonably conclude that the school board's retention of such books gives those volumes an implicit endorsement. Cf. FCC v. Pacifica Foundation, 438 U.S. 726, 98 S.Ct. 3026, 57 L.Ed.2d 1073 (1978). 92 Further, there is no guidance whatsoever as to what constitutes "political" factors. This Court has previously recognized that public education involves an area of broad public policy and " 'go[es] to the heart of representative government.' " Ambach v. Norwick, supra, at 74, 99 S.Ct., at 1593. As such, virtually all educational decisions necessarily involve "political" determinations. 93 What the plurality views as valid reasons for removing a book at their core involve partisan judgments. Ultimately the federal courts will be the judge of whether the motivation for book removal was "valid" or "reasonable." Undoubtedly the validity of many book removals will ultimately turn on a judge's evaluation of the books. Discretion must be used, and the appropriate body to exercise that discretion is the local elected school board, not judges.5 94 We can all agree that as a matter of educational policy students should have wide access to information and ideas. But the people elect school boards, who in turn select administrators, who select the teachers, and these are the individuals best able to determine the substance of that policy. The plurality fails to recognize the fact that local control of education involves democracy in a microcosm. In most public schools in the United States the parents have a large voice in running the school.6 Through participation in the election of school board members, the parents influence, if not control, the direction of their children's education. A school board is not a giant bureaucracy far removed from accountability for its actions; it is truly "of the people and by the people." A school board reflects its constituency in a very real sense and thus could not long exercise unchecked discretion in its choice to acquire or remove books. If the parents disagree with the educational decisions of the school board, they can take steps to remove the board members from office. Finally, even if parents and students cannot convince the school board that book removal is inappropriate, they have alternative sources to the same end. Books may be acquired from bookstores, public libraries, or other alternative sources unconnected with the unique environment of the local public schools.7 II 95 No amount of "limiting" language could rein in the sweeping "right" the plurality would create. The plurality distinguishes library books from textbooks because library books "by their nature are optional rather than required reading." Ante, at 862. It is not clear, however, why this distinction requires greater scrutiny before "optional" reading materials may be removed. It would appear that required reading and textbooks have a greater likelihood of imposing a " 'pall of orthodoxy' " over the educational process than do optional reading. Ante, at 870. In essence, the plurality's view transforms the availability of this "optional" reading into a "right" to have this "optional" reading maintained at the demand of teenagers. 96 The plurality also limits the new right by finding it applicable only to the removal of books once acquired. Yet if the First Amendment commands that certain books cannot be removed, does it not equally require that the same books be acquired? Why does the coincidence of timing become the basis of a constitutional holding? According to the plurality, the evil to be avoided is the "official suppression of ideas." Ante, at 871. It does not follow that the decision to remove a book is less "official suppression" than the decision not to acquire a book desired by someone.8 Similarly, a decision to eliminate certain material from the curriculum, history for example, would carry an equal—probably greater—prospect of "official suppression." Would the decision be subject to our review? III 97 Through use of bits and pieces of prior opinions unrelated to the issue of this case, the plurality demeans our function of constitutional adjudication. Today the plurality suggests that the Constitution distinguishes between school libraries and school classrooms, between removing unwanted books and acquiring books. Even more extreme, the plurality concludes that the Constitution requires school boards to justify to its teenage pupils the decision to remove a particular book from a school library. I categorically reject this notion that the Constitution dictates that judges, rather than parents, teachers, and local school boards, must determine how the standards of morality and vulgarity are to be treated in the classroom. 98 Justice POWELL, dissenting. 99 The plurality opinion today rejects a basic concept of public school education in our country: that the States and locally elected school boards should have the responsibility for determining the educational policy of the public schools. After today's decision any junior high school student, by instituting a suit against a school board or teacher, may invite a judge to overrule an educational decision by the official body designated by the people to operate the schools. 100 * School boards are uniquely local and democratic institutions. Unlike the governing bodies of cities and counties, school boards have only one responsibility: the education of the youth of our country during their most formative and impressionable years. Apart from health, no subject is closer to the hearts of parents than their children's education during those years. For these reasons, the governance of elementary and secondary education traditionally has been placed in the hands of a local board, responsible locally to the parents and citizens of school districts. Through parent-teacher associations (PTA's), and even less formal arrangements that vary with schools, parents are informed and often may influence decisions of the board. Frequently, parents know the teachers and visit classes. It is fair to say that no single agency of government at any level is closer to the people whom it serves than the typical school board. 101 I therefore view today's decision with genuine dismay. Whatever the final outcome of this suit and suits like it, the resolution of educational policy decisions through litigation, and the exposure of school board members to liability for such decisions, can be expected to corrode the school board's authority and effectiveness. As is evident from the generality of the plurality's "standard" for judicial review, the decision as to the educational worth of a book is a highly subjective one. Judges rarely are as competent as school authorities to make this decision; nor are judges responsive to the parents and people of the school district.1 102 The new constitutional right, announced by the plurality, is described as a "right to receive ideas" in a school. Ante, at 867. As the dissenting opinions of THE CHIEF JUSTICE and Justice REHNQUIST so powerfully demonstrate, however, this newfound right finds no support in the First Amendment precedents of this Court. And even apart from the inappropriateness of judicial oversight of educational policy, the new constitutional right is framed in terms that approach a meaningless generalization. It affords little guidance to courts, if they—as the plurality now authorizes them—are to oversee the inculcation of ideas. The plurality does announce the following standard: A school board's "discretion may not be exercised in a narrowly partisan or political manner." Ante, at 870. But this is a standardless standard that affords no more than subjective guidance to school boards, their counsel, and to courts that now will be required to decide whether a particular decision was made in a "narrowly partisan or political manner." Even the "chancellor's foot" standard in ancient equity jurisdiction was never this fuzzy. 103 As Justice REHNQUIST tellingly observes, how does one limit on a principled basis—today's new constitutional right? If a 14-year-old child may challenge a school board's decision to remove a book from the library, upon what theory is a court to prevent a like challenge to a school board's decision not to purchase that identical book? And at the even more "sensitive" level of "receiving ideas," does today's decision entitle student oversight of which courses may be added or removed from the curriculum, or even of what a particular teacher elects to teach or not teach in the classroom? Is not the "right to receive ideas" as much—or indeed even more—implicated in these educational questions?2 II 104 The plurality's reasoning is marked by contradiction. It purports to acknowledge the traditional role of school boards and parents in deciding what should be taught in the schools. It states the truism that the schools are "vitally important 'in the preparation of individuals for participation as citizens,' and as vehicles for 'inculcating fundamental values necessary to the maintenance of a democratic political system.' " Ante, at 864. Yet when a school board, as in this case, takes its responsibilities seriously and seeks to decide what the fundamental values are that should be imparted, the plurality finds a constitutional violation. 105 Just this Term the Court held, in an opinion I joined, that the children of illegal aliens must be permitted to attend the public schools. See Plyler v. Doe, 457 U.S. 202, 102 S.Ct. 2382, 72 L.Ed.2d 786. Quoting from earlier opinions, the Court noted that " 'the public schoo[l is] a most vital civic institution for the preservation of a democratic system of government' " and that the public schools are "the primary vehicle for transmitting 'the values on which our society rests.' " Ante, at 221, 102 S.Ct., at 2397. By denying to illegal aliens the opportunity "to absorb the values and skills upon which our social order rests" the law under review placed a lifelong disability upon these illegal alien children. Ibid. 106 Today the plurality drains much of the content from these apt phrases. A school board's attempt to instill in its students the ideas and values on which a democratic system depends is viewed as an impermissible suppression of other ideas and values on which other systems of government and other societies thrive. Books may not be removed because they are indecent; extol violence, intolerance, and racism; or degrade the dignity of the individual. Human history, not the least that of the 20th century, records the power and political life of these very ideas. But they are not our ideas or values. Although I would leave this educational decision to the duly constituted board, I certainly would not require a school board to promote ideas and values repugnant to a democratic society or to teach such values to children. 107 In different contexts and in different times, the destruction of written materials has been the symbol of despotism and intolerance. But the removal of nine vulgar or racist books from a high school library by a concerned local school board does not raise this specter. For me, today's decision symbolizes a debilitating encroachment upon the institutions of a free people. 108 Attached as an Appendix hereto is Judge Mansfield's summary of excerpts from the books at issue in this case. APPENDIX TO OPINION OF POWELL, J., DISSENTING 109 "The excerpts which led the Board to look into the educational suitability of the books in question are set out (with minor corrections after comparison with the text of the books themselves) below. The pagination and the underlinings are retained from the original report used by the board. In newer editions of some of the books, the quotes appear at different pages. 110 "1) SOUL ON ICE by Eldridge Cleaver PAGE QUOTE 111 157-158 '. . . There are white men who will pay you to fuck their wives. They approach you and say, "How would you like to fuck a white woman?" "What is this?" you ask. "On the up-and-up," he assures you. "It's all right. She's my wife. She needs black rod, is all. She has to have it. It's like a medicine or drug to her. She has to have it. I'll pay you. It's all on the level, no trick involved. Interested?" You go with him and he drives you to their home. The three of you go into the bedroom. There is a certain type who will leave you and his wife alone and tell you to pile her real good. After it is all over, he will pay you and drive you to wherever you want to go. Then there are some who like to peep at you through a keyhole and watch you have his woman, or peep at you through a window, or lie under the bed and listen to the creaking of the bed as you work out. There is another type who likes to masturbate while he stands beside the bed and watches you pile her. There is the type who likes to eat his woman up after you get through piling her. And there is the type who only wants you to pile her for a little while, just long enough to thaw her out and kick her motor over and arouse her to heat, then he wants you to jump off real quick and he will jump onto her and together they can make it from there by themselves.' 112 "2) A HERO AIN'T NOTHING BUT A SANDWICH by Alice Childress PAGE QUOTE 113 10 'Hell, no! Fuck the society.' 114 64-65 'The hell with the junkie, the wino, the capitalist, the welfare checks, the world . . . yeah, and fuck you too!' 115 75-76 'They can have back the spread and curtains, I'm too old for them fuckin bunnies anyway.' 116 "3) THE FIXER by Bernard Malamud PAGE QUOTE 117 52 'What do you think goes on in the wagon at night: Are the drivers on their knees fucking their mothers?' 118 90 'Fuck yourself, said the blinker, etc.' 119 92 'Who else would do anything like that but a motherfucking Zhid?' 120 146 'No more noise out of you or I'll shoot your Jew cock off.' 121 189 'Also there's a lot of fucking in the Old Testament, so how is that religious?' 122 192 'You better go fuck yourself, Bok, said Kogin, I'm onto your Jew tricks.' 215 'Ding-dong giddyap. A Jew's cock's in the devil's hock.' 123 216 'You cocksucker Zhid, I ought make you lick it up off the floor.' 124 "4) GO ASK ALICE by Anonymous PAGE QUOTE 125 31 'I wonder if sex without acid could be so exciting, so wonderful, so indescribable. I always thought it just took a minute, or that it would be like dogs mating.' 126 47 'Chris and I walked into Richie and Ted's apartment to find the bastards stoned and making love to each other . . . low class queer.' 127 81 'shitty, goddamned, pissing, ass, goddamned beJesus, screwing life's, ass, shit. Doris was ten and had humped with who knows how many men in between . . . her current stepfather started having sex with her but good . . . sonofabitch balling her' 128 83 'but now when I face a girl its like facing a boy. I get all excited and turned on. I want to screw with the girl. . . .' 129 84 'I'd rather screw with a guy . . . sometimes I want one of the girls to kiss me. I want her to touch me, to have her sleep under me.' 130 84 'Another day, another blow job . . . If I don't give Big Ass a blow he'll cut off my supply . . . and LittleJacon is yelling, "Mama, Daddy can't come now. He's humping Carla." 131 85 'Shit, goddamn, goddamn prick, son-of-a-bitch, ass, pissed, bastard, goddamn, bullshit 132 94 'I hope you have a nice orgasm with your dog tonight.' 110 'You fucking Miss Polly pure 133 117 'Then he said that all I needed was a good fuck.' 134 146 'It might be great because I'm practically a virgin in the sense that I've never had sex except when I've been stoned. . . .' 135 "5) SLAUGHTERHOUSE FIVE by Kurt Vonnegut, Jr. PAGE QUOTE 136 29 'Get out of the road, you dumb motherfucker.' The last word was still a novelty in the speech of white people in 1944. It was fresh and astonishing to Billy, who had never fucked anybody . . .' 137 32 'You stake a guy out on an anthill in the desert—see? He's facing upward, and you put honey all over his balls and pecker, and you cut off his eyelids so he has to stare at the sun till he dies.' 138 34 'He had a prophylactic kit containing two tough condoms 'For the prevention of disease only!' . . . He had a dirty picture of a woman attempting sexual intercourse with a shetland pony.' 139 94 & 95 'But the Gospels actually taught this: Before you kill somebody, make absolutely sure he isn't well connected . . . The flaw in the Christ stories, said the visitor from outer space, was that Christ who didn't look like much, was actually the son of the Most Powerful Being in the Universe. Readers understood that, so, when they came to the crucifixion, they naturally thought . . . Oh boy—they sure picked the wrong guy to lynch this time! And that thought had a brother: There are right people to lynch. People not well connected. . . . The visitor from outer space made a gift to Earth of a new Gospel. In it, Jesus really WAS a nobody, and a pain in the neck to a lot of people with better connections then he had . . . . So the people amused themselves one day by nailing him to a cross and planting the cross in the ground. There couldn't possibly be any repercussions, the lynchers thought . . . since the new Gospel hammered home again and again what a nobody Jesus was. And then just before the nobody died . . . . The voice of God came crashing down. He told the people that he was adopting the bum as his son . . . God said this: From this moment on, He will punish horribly anybody who torments a bum who has no connections.' 140 99 'They told him that there could be no Earthling babies without male homosexuals. There could be babies without female homosexuals.' 141 120 'Why don't you go fuck yourself? Don't think I haven't tried . . . he was going to have revenge, and that revenge was sweet . . . It's the sweetest thing there is, said Lazzaro. People fuck with me, he said, and Jesus Christ are they ever fucking sorry.' 142 122 'And he'll pull out a gun and shoot his pecker off. The stranger'll let him think a couple of seconds about who Paul Lazzaro is and what life's gonna be like without a pecker. Then he'll shoot him once in the guts and walk away. . . . He died on account of this silly cocksucker here. So I promised him I'd have this silly cocksucker shot after the war.' 143 134 'In my prison cell I sit . . . With my britches full of shit, And my balls are bouncing gently on the floor. And I see the bloody snag when she bit me in the bag . . . Oh, I'll never fuck a Polack any more.' 144 173 'And the peckers of the young men would still be semi-erect, and their muscles would be bulging like cannonballs.' 145 175 'They didn't have hard-ons . . . Everybody else did.' 146 177 'The magazine, which was published for lonesome men to jerk off to.' 147 178 'and one critic said. . . . 'To describe blow-jobs artistically." 148 "6) THE BEST SHORT STORIES BY NEGRO WRITERS Ed. by Langston Hughes PAGE QUOTE 149 176 'like bat's shit and camel piss,' 150 228 'that no-count bitch of a daughter of yours is up there up North making a whore of herself.' 151 237 'they made her get out and stand in front of the headlights of the car and pull down her pants and raise her dress they said that was the only way they could be sure. And you can imagine what they said and what they did—.' 152 303 'You need some pussy. Come on, let's go up to the whore house on the hill.' 153 'Oh, these bastards, these bastards, this God damned Army and the bastards in it. The sons of bitches!' 154 436 'he produced a brown rag doll, looked at her again, then grabbed the doll by its legs and tore it part way up the middle. Then he jammed his finger into the rip between the doll's legs. The other men laughed. . . .' 155 444 'The pimps, hustlers, lesbians, and others trying to misuse me.' 156 462 'But she had straight firm legs and her breasts were small and upright. No doubt if she'd had children her breasts would be hanging like little empty purses.' 157 464 'She first became aware of the warm tense nipples on her breasts. Her hands went up gently to clam them.' 'In profile, his penis hung like a stout tassle. She could even tell that he was circumcised.' 158 406 'Cadillac Bill was busy following Luheaster around, rubbing her stomach and saying, "Magic Stomach, Magic Stomach, bring me a little baby cadillac." 'One of the girls went upstairs with Red Top and stayed for about forty five minutes.' 159 "7) BLACK BOY by Richard Wright PAGE QUOTE 160 70-71 'We black children—seven or eight or nine years of age used to run to the Jew's store and shout: 161 . . . Bloody Christ Killers Never trust a Jew Bloody Christ Killers 162 What won't a Jew do . . . Red, white and blue Your pa was a Jew Your ma a dirty dago 163 What the hell is you?' 164 265 'Crush that nigger's nuts, nigger!' 'Hit that nigger!' 'Aw, fight, you goddam niggers!' 'Sock 'im, in his f-k-g-piece!' 'Make 'im bleed!' 165 "8) LAUGHING BOY by Oliver LaFarge PAGE QUOTE 166 38 'I'll tell you, she is all bad; for two bits she will do the worst thing.' 258-9 'I was frightened when he wanted me to lie with him, but he made me feel all right. He knew all about how to make women forget themselves, that man.' 167 "9) THE NAKED APE by Desmond Morris PAGE QUOTE 168 73-74 'Also, the frontal approach provides the maximum possibility for stimulation of the female's clitoris during the pelvic thrusting of the male. It is true that it will be passively, stimulated by the pulling effect of the male's thrusts, regardless of his body position in relation to the female, but in a face-to-face mating there will in addition be the direct rhythmic pressure of the male's pubic region on to the clitoral area, and this will considerably heighten the stimulation . . .' 'So it seems plausible to consider that face-to-face copulation is basic to our species. There are, of course, a number of variations that do not eliminate the frontal element: male above, female above, side by side, squatting, standing, and so on, but the most efficient and commonly used one is with both partners horizontal, the male above the female. . . .' 169 80 '. . . This broadening of the penis results in the female's external genitals being subjected to much more pulling and pushing during the performance of pelvic thrusts. With each inward thrust of the penis, the clitoral region is pulled downwards and then with each withdrawal, it moves up again. Add to this the rhythmic pressure being exerted on the clitoris region by the pubic region of the frontally copulating male, and you have a repeated massaging of the clitoris that—were she a male—would virtually be masturbatory.' 170 94-99 '. . . If either males or females cannot for some reason obtain sexual access to their opposite numbers, they will find sexual outlets in other ways. They may use other members of their own sex, or they may even use members of other species, or they may masturbate. . . .' 171 "10) READER FOR WRITERS . . ." 172 638 F.2d 404, 419-422, n. 1 (CA2 1980) (Mansfield, J., dissenting). 173 Justice REHNQUIST, with whom THE CHIEF JUSTICE and Justice POWELL join, dissenting. 174 Addressing only those aspects of the constitutional question which must be decided to determine whether or not the District Court was correct in granting summary judgment, I conclude that it was. I agree fully with the views expressed by THE CHIEF JUSTICE, and concur in his opinion. I disagree with Justice BRENNAN's opinion because it is largely hypothetical in character, failing to take account of the facts as admitted by the parties pursuant to local rules of the District Court for the Eastern District of New York, and because it is analytically unsound and internally inconsistent.1 175 * A. 176 Justice BRENNAN's opinion deals far more sparsely with the procedural posture of this case than it does with the constitutional issues which it conceives to arise under the First Amendment. It first launches into a confusing, discursive exegesis on these constitutional issues as applied to junior high school and high school libraries, ante, at 863-872, and only thereafter does it discuss the state of the record before the Court. Ante, at 872-875. Because the record facts should always establish the limits of the Court's constitutional analysis, and are particularly relevant in cases where the trial court has granted summary judgment, I think that Justice BRENNAN's approach violates our "long . . . considered practice not to decide abstract, hypothetical or contingent questions, or to decide any constitutional question in advance of the necessity for its decision." Alabama State Federation of Labor v. McAdory, 325 U.S. 450, 461, 65 S.Ct. 1384, 1389, 89 L.Ed. 1725 (1945). (citations omitted). 177 When Justice BRENNAN finally does address the state of the record, he refers to snippets and excerpts of the relevant facts to explain why a grant of summary judgment was improper. But he totally ignores the effect of Rule 9(g) of the local rules of the District Court, under which the parties set forth their version of the disputed facts in this case.2 Since summary judgment was entered against respondents, they are entitled to have their version of the facts, as embodied in their Rule 9(g) statement, accepted for purposes of our review. Since the parties themselves are presumably the best judges of the extent of the factual dispute between them, however, respondents certainly are not entitled to any more favorable version of the facts than that contained in their own Rule 9(g) statement. Justice BRENNAN's combing through the record of affidavits, school bulletins, and the like for bits and snatches of dispute is therefore entirely beside the point at this stage of the case. 178 Considering only the respondents' description of the factual aspects of petitioners' motivation, Justice BRENNAN's apparent concern that the Board's action may have been a sinister political plot "to suppress ideas" may be laid to rest. The members of the Board, in deciding to remove these books, were undoubtedly influenced by their own "personal values, morals, and tastes,"3 just as any member of a school board is apt to be so influenced in making decisions as to whether a book is educationally suitable. Respondents essentially conceded that some excerpts of the removed books "contained profanities, some were sexually explicit, some were ungrammatical, some were anti-American, and some were offensive to racial, religious or ethnic groups."4 179 Respondents also agreed that, "[a]lthough the books themselves were excluded from use in the schools in any way, [petitioners] have not precluded discussion about the themes of the books or the books themselves." App. 140. Justice BRENNAN's concern with the "suppression of ideas" thus seems entirely unwarranted on this state of the record, and his creation of constitutional rules to cover such eventualities is entirely gratuitous. Though for reasons stated in Part II of this opinion I entirely disagree with Justice BRENNAN's treatment of the constitutional issue, I also disagree with his opinion for the entirely separate reason that it is not remotely tailored to the facts presented by this case. 180 In the course of his discussion, Justice BRENNAN states: 181 "Petitioners rightly possess significant discretion to determine the content of their school libraries. But that discretion may not be exercised in a narrowly partisan or political manner. If a Democratic school board, motivated by party affiliation, ordered the removal of all books written by or in favor of Republicans, few would doubt that the order violated the constitutional rights of the students. . . . The same conclusion would surely apply if an all-white school board, motivated by racial animus, decided to remove all books authored by blacks or advocating racial equality and integration. Our Constitution does not permit the official suppression of ideas." Ante, 870-871 (emphasis in original). 182 I can cheerfully concede all of this, but as in so many other cases the extreme examples are seldom the ones that arise in the real world of constitutional litigation. In this case the facts taken most favorably to respondents suggest that nothing of this sort happened. The nine books removed undoubtedly did contain "ideas," but in the light of the excerpts from them found in the dissenting opinion of Judge Mansfield in the Court of Appeals, it is apparent that eight of them contained demonstrable amounts of vulgarity and profanity, see 638 F.2d 404, 419-422, n. 1 (CA2 1980), and the ninth contained nothing that could be considered partisan or political, see id., at 428, n. 6. As already demonstrated, respondents admitted as much. Petitioners did not, for the reasons stated hereafter, run afoul of the First and Fourteenth Amendments by removing these particular books from the library in the manner in which they did. I would save for another day—feeling quite confident that that day will not arrive—the extreme examples posed in Justice BRENNAN's opinion. B 183 Considerable light is shed on the correct resolution of the constitutional question in this case by examining the role played by petitioners. Had petitioners been the members of a town council, I suppose all would agree that, absent a good deal more than is present in this record, they could not have prohibited the sale of these books by private booksellers within the municipality. But we have also recognized that the government may act in other capacities than as sovereign, and when it does the First Amendment may speak with a different voice: 184 "[I]t cannot be gainsaid that the State has interests as an employer in regulating the speech of its employees that differ significantly from those it possesses in connection with regulation of the speech of the citizenry in general. The problem in any case is to arrive at a balance between the interests of the teacher, as a citizen, in commenting upon matters of concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees." Pickering v. Board of Education, 391 U.S. 563, 568, 88 S.Ct. 1731, 1734, 20 L.Ed.2d 811 (1968). 185 By the same token, expressive conduct which may not be prohibited by the State as sovereign may be proscribed by the State as property owner: "The State, no less than a private owner of property, has power to preserve the property under its control for the use to which it is lawfully dedicated." Adderley v. Florida, 385 U.S. 39, 47, 87 S.Ct. 242, 247, 17 L.Ed.2d 149 (1966) (upholding state prohibition of expressive conduct on certain state property). 186 With these differentiated roles of government in mind, it is helpful to assess the role of government as educator, as compared with the role of government as sovereign. When it acts as an educator, at least at the elementary and secondary school level, the government is engaged in inculcating social values and knowledge in relatively impressionable young people. Obviously there are innumerable decisions to be made as to what courses should be taught, what books should be purchased, or what teachers should be employed. In every one of these areas the members of a school board will act on the basis of their own personal or moral values, will attempt to mirror those of the community, or will abdicate the making of such decisions to so-called "experts."5 In this connection I find myself entirely in agreement with the observation of the Court of Appeals for the Seventh Circuit in Zykan v. Warsaw Community School Corp., 631 F.2d 1300, 1305 (7th Cir. 1980), that it is "permissible and appropriate for local boards to make educational decisions based upon their personal social, political and moral views." In the very course of administering the many-faceted operations of a school district, the mere decision to purchase some books will necessarily preclude the possibility of purchasing others. The decision to teach a particular subject may preclude the possibility of teaching another subject. A decision to replace a teacher because of ineffectiveness may by implication be seen as a disparagement of the subject matter taught. In each of these instances, however, the book or the exposure to the subject matter may be acquired elsewhere. The managers of the school district are not proscribing it as to the citizenry in general, but are simply determining that it will not be included in the curriculum or school library. In short, actions by the government as educator do not raise the same First Amendment concerns as actions by the government as sovereign. II 187 Justice BRENNAN would hold that the First Amendment gives high school and junior high school students a "right to receive ideas" in the school. Ante, at 867. This right is a curious entitlement. It exists only in the library of the school, and only if the idea previously has been acquired by the school in book form. It provides no protection against a school board's decision not to acquire a particular book, even though that decision denies access to ideas as fully as removal of the book from the library, and it prohibits removal of previously acquired books only if the remover "dislike[s] the ideas contained in those books," even though removal for any other reason also denies the students access to the books. Ante, at 871-872 188 But it is not the limitations which Justice BRENNAN places on the right with which I disagree; they simply demonstrate his discomfort with the new doctrine which he fashions out of whole cloth. It is the very existence of a right to receive information, in the junior high school and high school setting, which I find wholly unsupported by our past decisions and inconsistent with the necessarily selective process of elementary and secondary education. A. 189 The right described by Justice BRENNAN has never been recognized in the decisions of this Court and is not supported by their rationale. Justice BRENNAN correctly observes that students do not "shed their constitutional rights to freedom of speech or expression at the schoolhouse gate." Tinker v. Des Moines School District, 393 U.S. 503, 506, 89 S.Ct. 733, 736, 21 L.Ed.2d 731 (1969). But, as this language from Tinker suggests, our past decisions in this area have concerned freedom of speech and expression, not the right of access to particular ideas. We have held that students may not be prevented from symbolically expressing their political views by the wearing of black arm bands, Tinker v. Des Moines School District, supra, and that they may not be forced to participate in the symbolic expression of saluting the flag, West Virginia Board of Education v. Barnette, 319 U.S. 624, 63 S.Ct. 1178, 87 L.Ed. 1628 (1943). But these decisions scarcely control the case before us. Neither the District Court nor the Court of Appeals found that petitioners' removal of books from the school libraries infringed respondents' right to speak or otherwise express themselves. 190 Despite Justice BRENNAN's suggestion to the contrary, this Court has never held that the First Amendment grants junior high school and high school students a right of access to certain information in school. It is true that the Court has recognized a limited version of that right in other settings, and Justice BRENNAN quotes language from five such decisions and one of his own concurring opinions in order to demonstrate the viability of the right-to-receive doctrine. Ante, at 866-867. But not one of these cases concerned or even purported to discuss elementary or secondary educational institutions.6 Justice BRENNAN brushes over this significant omission in First Amendment law by citing Tinker v. Des Moines School District for the proposition that "students too are beneficiaries of this [right-to-receive] principle." Ante, at 868. But Tinker held no such thing. One may readTinker in vain to find any recognition of a First Amendment right to receive information. Tinker, as already mentioned, was based entirely on the students' right to express their political views. 191 Nor does the right-to-receive doctrine recognized in our past decisions apply to schools by analogy. Justice BRENNAN correctly characterizes the right of access to ideas as "an inherent corollary of the rights of free speech and press" which "follows ineluctably from the sender's First Amendment right to send them." Ante, at 867 (emphasis in original). But he then fails to recognize the predicate right to speak from which the students' right to receive must follow. It would be ludicrous, of course, to contend that all authors have a constitutional right to have their books placed in junior high school and high school libraries. And yet without such a right our prior precedents would not recognize the reciprocal right to receive information. Justice BRENNAN disregards this inconsistency with our prior cases and fails to explain the constitutional or logical underpinnings of a right to hear ideas in a place where no speaker has the right to express them. 192 Justice BRENNAN also correctly notes that the reciprocal nature of the right to receive information derives from the fact that it "is a necessary predicate to the recipient's meaningful exercise of his own rights of speech, press, and political freedom." Ibid. (emphasis in original). But the denial of access to ideas inhibits one's own acquisition of knowledge only when that denial is relatively complete. If the denied ideas are readily available from the same source in other accessible locations, the benefits to be gained from exposure to those ideas have not been foreclosed by the State. This fact is inherent in the right-to-receive cases relied on by Justice BRENNAN, every one of which concerned the complete denial of access to the ideas sought.7 Our past decisions are thus unlike this case where the removed books are readily available to students and non-students alike at the corner bookstore or the public library. B 193 There are even greater reasons for rejecting Justice BRENNAN's analysis, however, than the significant fact that we have never adopted it in the past. "The importance of public schools in the preparation of individuals for participation as citizens, and in the preservation of the values on which our society rests, has long been recognized by our decisions." Ambach v. Norwick, 441 U.S. 68, 76, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979). Public schools fulfill the vital role of teaching students the basic skills necessary to function in our society, and of "inculcating fundamental values necessary to the maintenance of a democratic political system." Id., at 77, 99 S.Ct., at 1594. The idea that such students have a right of access, in the school, to information other than that thought by their educators to be necessary is contrary to the very nature of an inculcative education. 194 Education consists of the selective presentation and explanation of ideas. The effective acquisition of knowledge depends upon an orderly exposure to relevant information. Nowhere is this more true than in elementary and secondary schools, where, unlike the broad-ranging inquiry available to university students, the courses taught are those thought most relevant to the young students' individual development. Of necessity, elementary and secondary educators must separate the relevant from the irrelevant, the appropriate from the inappropriate. Determining what information not to present to the students is often as important as identifying relevant material. This winnowing process necessarily leaves much information to be discovered by students at another time or in another place, and is fundamentally inconsistent with any constitutionally required eclecticism in public education. 195 Justice BRENNAN rejects this idea, claiming that it "overlooks the unique role of the school library." Ante, at 869. But the unique role referred to appears to be one of Justice BRENNAN's own creation. No previous decision of this Court attaches unique First Amendment significance to the libraries of elementary and secondary schools. And in his paean of praise to such libraries as the "environment especially appropriate for the recognition of the First Amendment rights of students," ante, at 868, Justice BRENNAN turns to language about public libraries from the three-Justice plurality in Brown v. Louisiana, 383 U.S. 131, 86 S.Ct. 719, 15 L.Ed.2d 637 (1966), and to language about universities and colleges from Keyishian v. Board of Regents, 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967). Ante, at 868. Not only is his authority thus transparently thin, but also, and more importantly, his reasoning misapprehends the function of libraries in our public school system. 196 As already mentioned, elementary and secondary schools are inculcative in nature. The libraries of such schools serve as supplements to this inculcative role. Unlike university or public libraries, elementary and secondary school libraries are not designed for freewheeling inquiry; they are tailored, as the public school curriculum is tailored, to the teaching of basic skills and ideas. Thus, Justice BRENNAN cannot rely upon the nature of school libraries to escape the fact that the First Amendment right to receive information simply has no application to the one public institution which, by its very nature, is a place for the selective conveyance of ideas. 197 After all else is said, however, the most obvious reason that petitioners' removal of the books did not violate respondents' right to receive information is the ready availability of the books elsewhere. Students are not denied books by their removal from a school library. The books may be borrowed from a public library, read at a university library, purchased at a bookstore, or loaned by a friend. The government as educator does not seek to reach beyond the confines of the school. Indeed, following the removal from the school library of the books at issue in this case, the local public library put all nine books on display for public inspection. Their contents were fully accessible to any inquisitive student. C 198 Justice BRENNAN's own discomfort with the idea that students have a right to receive information from their elementary or secondary schools is demonstrated by the artificial limitations which he places upon the right—limitations which are supported neither by logic nor authority and which are inconsistent with the right itself. The attempt to confine the right to the library is one such limitation, the fallacies of which have already been demonstrated. 199 As a second limitation, Justice BRENNAN distinguishes the act of removing a previously acquired book from the act of refusing to acquire the book in the first place: "[N]othing in our decision today affects in any way the discretion of a local school board to choose books to add to the libraries of their schools. [O]ur holding today affects only the discretion to remove books." Ante, at 871-872 (emphasis in original). If Justice BRENNAN truly has found a "right to receive ideas," ante, at 866-867, however, this distinction between acquisition and removal makes little sense. The failure of a library to acquire a book denies access to its contents just as effectively as does the removal of the book from the library's shelf. As a result of either action the book cannot be found in the "principal locus" of freedom discovered by Justice BRENNAN. Ante, at 868. 200 The justification for this limiting distinction is said by Justice BRENNAN to be his concern in this case with "the suppression of ideas." Ante, at 871. Whatever may be the analytical usefulness of this appealing sounding phrase, see Part II-D, infra, the suppression of ideas surely is not the identical twin of the denial of access to information. Not every official act which denies access to an idea can be characterized as a suppression of the idea. Thus unless the "right to receive information" and the prohibition against "suppression of ideas" are each a kind of Mother-Hubbard catch phrase for whatever First Amendment doctrines one wishes to cover, they would not appear to be interchangeable. 201 Justice BRENNAN's reliance on the "suppression of ideas" to justify his distinction between acquisition and removal of books has additional logical pitfalls. Presumably the distinction is based upon the greater visibility and the greater sense of conscious decision thought to be involved in the removal of a book, as opposed to that involved in the refusal to acquire a book. But if "suppression of ideas" is to be the talisman, one would think that a school board's public announcement of its refusal to acquire certain books would have every bit as much impact on public attention as would an equally publicized decision to remove the books. And yet only the latter action would violate the First Amendment under Justice BRENNAN's analysis. 202 The final limitation placed by Justice BRENNAN upon his newly discovered right is a motive requirement: the First Amendment is violated only "[i]f petitioners intended by their removal decision to deny respondents access to ideas with which petitioners disagreed." Ante, at 871 (emphasis in original). But bad motives and good motives alike deny access to the books removed. If Justice BRENNAN truly recognizes a constitutional right to receive information, it is difficult to see why the reason for the denial makes any difference. Of course Justice BRENNAN's view is that intent matters because the First Amendment does not tolerate an officially prescribed orthodoxy. Ante, at 870-871. But this reasoning mixes First Amendment apples and oranges. The right to receive information differs from the right to be free from an officially prescribed orthodoxy. Not every educational denial of access to information casts a pall of orthodoxy over the classroom. 203 It is difficult to tell from Justice BRENNAN's opinion just what motives he would consider constitutionally impermissible. I had thought that the First Amendment proscribes content-based restrictions on the marketplace of ideas. See Widmar v. Vincent, 454 U.S. 263, 269-270, 102 S.Ct. 269, 274, 70 L.Ed.2d 440 (1981). Justice BRENNAN concludes, however, that a removal decision based solely upon the "educational suitability" of a book or upon its perceived vulgarity is " 'perfectly permissible.' " Ante, at 871 (quoting Tr. of Oral Arg. 53). But such determinations are based as much on the content of the book as determinations that the book espouses pernicious political views. 204 Moreover, Justice BRENNAN's motive test is difficult to square with his distinction between acquisition and removal. If a school board's removal of books might be motivated by a desire to promote favored political or religious views, there is no reason that its acquisition policy might not also be so motivated. And yet the "pall of orthodoxy" cast by a carefully executed book-acquisition program apparently would not violate the First Amendment under Justice BRENNAN's view. D 205 Intertwined as a basis for Justice BRENNAN's opinion, along with the "right to receive information," is the statement that "[o]ur Constitution does not permit the official suppression of ideas." Ante, at 871 (emphasis in original). There would be few champions, I suppose, of the idea that our Constitution does permit the official suppression of ideas; my difficulty is not with the admittedly appealing catchiness of the phrase, but with my doubt that it is really a useful analytical tool in solving difficult First Amendment problems. Since the phrase appears in the opinion "out of the blue," without any reference to previous First Amendment decisions of this Court, it would appear that the Court for years has managed to decide First Amendment cases without it. 206 I would think that prior cases decided under established First Amendment doctrine afford adequate guides in this area without resorting to a phrase which seeks to express "a complicated process of constitutional adjudication by a deceptive formula." Kovacs v. Cooper, 336 U.S. 77, 96, 69 S.Ct. 448, 458, 93 L.Ed. 513 (1949) (Frankfurter, J., concurring). A school board which publicly adopts a policy forbidding the criticism of United States foreign policy by any student, any teacher, or any book on the library shelves is indulging in one kind of "suppression of ideas." A school board which adopts a policy that there shall be no discussion of current events in a class for high school sophomores devoted to second-year Latin "suppresses ideas" in quite a different context. A teacher who had a lesson plan consisting of 14 weeks of study of United States history from 1607 to the present time, but who because of a week's illness is forced to forgo the most recent 20 years of American history, may "suppress ideas" in still another way. 207 I think a far more satisfactory basis for addressing these kinds of questions is found in the Court's language in Tinker v. Des Moines School District, where we noted: 208 "[A] particular symbol—black armbands worn to exhibit opposition to this Nation's involvement in Vietnam—was singled out for prohibition. Clearly, the prohibition of expression of one particular opinion, at least without evidence that it is necessary to avoid material and substantial interference with schoolwork or discipline, is not constitutionally permissible." 393 U.S., at 510-511, 89 S.Ct., at 738-739. 209 In the case before us the petitioners may in one sense be said to have "suppressed" the "ideas" of vulgarity and profanity, but that is hardly an apt description of what was done. They ordered the removal of books containing vulgarity and profanity, but they did not attempt to preclude discussion about the themes of the books or the books themselves. App. 140. Such a decision, on respondents' version of the facts in this case, is sufficiently related to "educational suitability" to pass muster under the First Amendment. E 210 The inconsistencies and illogic of the limitations placed by Justice BRENNAN upon his notion of the right to receive ideas in school are not here emphasized in order to suggest that they should be eliminated. They are emphasized because they illustrate that the right itself is misplaced in the elementary and secondary school setting. Likewise, the criticism of Justice BRENNAN's newly found prohibition against the "suppression of ideas" is by no means intended to suggest that the Constitution permits the suppression of ideas; it is rather to suggest that such a vague and imprecise phrase, while perhaps wholly consistent with the First Amendment, is simply too diaphanous to assist careful decision of cases such as this one. 211 I think the Court will far better serve the cause of First Amendment jurisprudence by candidly recognizing that the role of government as sovereign is subject to more stringent limitations than is the role of government as employer, property owner, or educator. It must also be recognized that the government as educator is subject to fewer strictures when operating an elementary and secondary school system than when operating an institution of higher learning. Cf. Tilton v. Richardson, 403 U.S. 672, 685-686, 91 S.Ct. 2091, 2099, 29 L.Ed.2d 790 (1971) (opinion of BURGER, C. J.). With respect to the education of children in elementary and secondary schools, the school board may properly determine in many cases that a particular book, a particular course, or even a particular area of knowledge is not educationally suitable for inclusion within the body of knowledge which the school seeks to impart. Without more, this is not a condemnation of the book or the course; it is only a determination akin to that referred to by the Court in Village of Euclid v. Ambler Realty Co., 272 U.S. 365, 388, 47 S.Ct. 114, 118, 71 L.Ed. 303 (1926): "A nuisance may be merely a right thing in the wrong place,—like a pig in the parlor instead of the barnyard." III 212 Accepting as true respondents' assertion that petitioners acted on the basis of their own "personal values, morals and tastes," App. 139, I find the actions taken in this case hard to distinguish from the myriad choices made by school boards in the routine supervision of elementary and secondary schools. "Courts do not and cannot intervene in the resolution of conflicts which arise in the daily operation of school systems and which do not directly and sharply implicate basic constitutional values." Epperson v. Arkansas, 393 U.S. 97, 104, 89 S.Ct. 266, 270, 21 L.Ed.2d 228 (1968). In this case respondents' rights of free speech and expression were not infringed, and by respondents' own admission no ideas were "suppressed." I would leave to another day the harder cases. 213 Justice O'CONNOR, dissenting. 214 If the school board can set the curriculum, select teachers, and determine initially what books to purchase for the school library, it surely can decide which books to discontinue or remove from the school library so long as it does not also interfere with the right of students to read the material and to discuss it. As Justice REHNQUIST persuasively argues, the plurality's analysis overlooks the fact that in this case the government is acting in its special role as educator. 215 I do not personally agree with the Board's action with respect to some of the books in question here, but it is not the function of the courts to make the decisions that have been properly relegated to the elected members of school boards. It is the school board that must determine educational suitability, and it has done so in this case. I therefore join THE CHIEF JUSTICE's dissent. 1 The Amendment provides in pertinent part that "Congress shall make no law . . . abridging the freedom of speech, or of the press." It applies to the States by virtue of the Fourteenth Amendment. Gitlow v. New York, 268 U.S. 652, 666, 45 S.Ct. 625, 629, 69 L.Ed. 1138 (1925); Grosjean v. American Press Co., 297 U.S. 233, 244, 56 S.Ct. 444, 446, 80 L.Ed. 660 (1936). 2 The District Court noted, however, that petitioners "concede that the books are not obscene." 474 F.Supp. 387, 392 (EDNY 1979). 3 The nine books in the High School library were: Slaughter House Five, by Kurt Vonnegut, Jr.; The Naked Ape, by Desmond Morris; Down These Mean Streets, by Piri Thomas; Best Short Stories of Negro Writers, edited by Langston Hughes; Go Ask Alice, of anonymous authorship; Laughing Boy, by Oliver LaFarge; Black Boy, by Richard Wright; A Hero Ain't Nothin' But A Sandwich, by Alice Childress; and Soul On Ice, by Eldridge Cleaver. The book in the Junior High School library was A Reader for Writers, edited by Jerome Archer. Still another listed book, The Fixer, by Bernard Malamud, was found to be included in the curriculum of a twelfth-grade literature course. 474 F.Supp., at 389 and nn. 2-4. 4 The Superintendent of Schools objected to the Board's informal directive, noting: "[W]e already have a policy . . . designed expressly to handle such problems. It calls for the Superintendent, upon receiving an objection to a book or books, to appoint a committee to study them and make recommendations. I feel it is a good policy—and it is Board policy—and that it should be followed in this instance. Furthermore, I think it can be followed quietly and in such a way as to reduce, perhaps avoid, the public furor which has always attended such issues in the past." App. 44. The Board responded to the Superintendent's objection by repeating its directive "that all copies of the library books in question be removed from the libraries to the Board's office." Id., at 47 (emphasis in original). 5 The Fixer, Laughing Boy, Black Boy, Go Ask Alice, and Best Short Stories by Negro Writers. 474 F.Supp., at 391, nn. 6-7. 6 The Naked Ape and Down These Mean Streets. 474 F.Supp., at 391, n. 8. 7 Soul on Ice and A Hero Ain't Nothin' But A Sandwich. 474 F.Supp., at 391, n. 9. 8 A Reader for Writers. 474 F.Supp., at 391, n. 11. The reason given for this disposition was that all members of the Committee had not been able to read the book. Id., at 391. 9 Slaughter House Five. 474 F.Supp., at 391, n. 10. 10 Laughing Boy. 474 F.Supp., at 391, n. 12. 11 Black Boy. 474 F.Supp., at 391, n. 13. 12 As a result, the nine removed books could not be assigned or suggested to students in connection with school work. Id., at 391. However, teachers were not instructed to refrain from discussing the removed books or the ideas and positions expressed in them. App. 131. 13 474 F.Supp., at 396-397, citing Presidents Council, District 25 v. Community School Board # 25, 457 F.2d 289 (CA2 1972); James v. Board of Education, 461 F.2d 566, 573 (CA2 1972); East Hartford Educational Assn. v. Board of Education, 562 F.2d 838, 856 (CA2 1977) (en banc). 14 474 F.Supp., at 395, quoting Presidents Council, District 25 v. Community School Board # 25, supra, at 291 (in turn quoting Epperson v. Arkansas, 393 U.S. 97, 104, 89 S.Ct. 266, 270, 21 L.Ed.2d 228 (1968)). 15 After criticizing "the criteria for removal" employed by petitioners as "suffer[ing] from excessive generality and overbreadth," and the procedures used by petitioners as "erratic, arbitrary and free-wheeling," Judge Sifton observed that "precision of regulation and sensitivity to First Amendment concerns" were "hardly established" by such procedures. 638 F.2d, at 416. 16 Judge Sifton stated that it could be inferred from the record that petitioners' "political views and personal taste [were] being asserted not in the interests of the children's well-being, but rather for the purpose of establishing those views as the correct and orthodox ones for all purposes in the particular community." Id., at 417. 17 Judge Mansfield dissented, id., at 419-432, based upon a distinctly different reading of the record developed in the District Court. According to Judge Mansfield, "the undisputed evidence of the motivation for the Board's action was the perfectly permissible ground that the books were indecent, in bad taste, and unsuitable for educational purposes." Id., at 430. He also asserted that in reaching its decision "the Board [had] acted carefully, conscientiously and responsibly after according due process to all parties concerned." Id., at 422. Judge Mansfield concluded that "the First Amendment entitles students to reasonable freedom of expression but not to freedom from what some may consider to be excessively moralistic or conservative selection by school authorities of library books to be used as educational tools." Id., at 432. 18 Four of respondents' five causes of action complained of petitioners' "resolutions ordering the removal of certain books from the school libraries of the District and prohibiting the use of those books in the curriculum." App. 5. The District Court concluded that "respect for . . . the school board's substantial control over educational content . . . preclude[s] any finding of a first amendment violation arising out of removal of any of the books from use in the curriculum." 474 F.Supp., at 397. This holding is not at issue here. Respondents' fifth cause of action complained that petitioners' "resolutions prohibiting the use of certain books in the curriculum of schools in the District" had "imposed upon teachers in the District arbitrary and unreasonable restrictions upon their ability to function as teachers in violation of principles of academic freedom." App. 6. The District Court held that respondents had not proved this cause of action: "before such a claim may be sustained there must at least be a real, not an imagined controversy." 474 F.Supp., at 397. Respondents have not sought review of that holding in this Court. 19 Respondents also agree with these propositions. Tr. of Oral Arg. 28, 41. 20 For a modern version of this observation, see A. Meiklejohn, Free Speech and Its Relation to Self-Government 26 (1948): "Just so far as . . . the citizens who are to decide an issue are denied acquaintance with information or opinion or doubt or disbelief or criticism which is relevant to that issue, just so far the result must be ill-considered, ill-balanced planning, for the general good." See also Butler v. Michigan, 352 U.S. 380, 383-384, 77 S.Ct. 524, 525-526, 1 L.Ed.2d 412 (1957); Procunier v. Martinez, 416 U.S. 396, 408-409, 94 S.Ct. 1800, 1808-1809, 40 L.Ed.2d 224 (1974); Houchins v. KQED, Inc., 438 U.S. 1, 30, 98 S.Ct. 2588, 2604, 57 L.Ed.2d 553 (1978) (STEVENS, J., dissenting) ("[T]he First Amendment protects not only the dissemination but also the receipt of information and ideas"); Saxbe v. Washington Post Co., 417 U.S. 843, 862-863, 94 S.Ct. 2811, 2821, 41 L.Ed.2d 514 (1974) (POWELL, J., dissenting) ("[P]ublic debate must not only be unfettered; it must be informed. For that reason this Court has repeatedly stated that First Amendment concerns encompass the receipt of information and ideas as well as the right of free expression"). 21 385 U.S., at 603, 87 S.Ct., at 683, quoting Sweezy v. New Hampshire, 354 U.S. 234, 250, 77 S.Ct. 1203, 1211, 1 L.Ed.2d 1311 (1957) (opinion of Warren, C. J.). 22 By "decisive factor" we mean a "substantial factor" in the absence of which the opposite decision would have been reached. See Mt. Healthy City Board of Ed. v. Doyle, 429 U.S. 274, 287, 97 S.Ct. 568, 576, 50 L.Ed.2d 471 (1977). 23 Petitioners acknowledged that their "evaluation of the suitability of the books was based on [their] personal values, morals, tastes and concepts of educational suitability." App. 142. But they did not accept, and thus apparently denied, respondents' assertion that some excerpts were objected to as "anti-American." Ibid. 24 For example, petitioner Ahrens stated: "I am basically a conservative in my general philosophy and feel that the community I represent as a school board member shares that philosophy. . . . I feel that it is my duty to apply my conservative principles to the decision making process in which I am involved as a board member and I have done so with regard to . . . curriculum formation and content and other educational matters." Id., at 21. "We are representing the community which first elected us and re-elected us and our actions have reflected its intrinsic values and desires." Id., at 27. Petitioners Fasulo, Hughes, Melchers, Michaels, and Nessim made a similar statement that they had "represented the basic values of the community in [their] actions." Id., at 120. 25 When asked to give an example of "anti-Americanism" in the removed books, petitioners Ahrens and Martin both adverted to A Hero Ain't Nothin' But A Sandwich, which notes at one point that George Washington was a slaveholder. See A. Childress, A Hero Ain't Nothin' But A Sandwich 43 (1973); Deposition of Petitioner Ahrens 89; Deposition of Petitioner Martin 20-22. Petitioner Martin stated: "I believe it is anti-American to present one of the nation's heroes, the first President, . . . in such a negative and obviously one-sided life. That is one example of what I would consider anti-American." Deposition of Petitioner Martin 22. 26 We have recognized in numerous precedents that when seeking to distinguish activities unprotected by the First Amendment from other, protected activities, the State must employ "sensitive tools" in order to achieve a precision of regulation that avoids the chilling of protected activities. See, e.g., Speiser v. Randall, 357 U.S. 513, 525-526, 78 S.Ct. 1332, 1341-1342, 2 L.Ed.2d 1460 (1958); NAACP v. Button, 371 U.S. 415, 433, 83 S.Ct. 328, 338, 9 L.Ed.2d 405 (1963); Keyishian v. Board of Regents, 385 U.S. 589, 603-604, 87 S.Ct. 675, 683-684, 17 L.Ed.2d 629 (1967); Blount v. Rizzi, 400 U.S. 410, 417, 91 S.Ct. 423, 428, 27 L.Ed.2d 498 (1971). In the case before us, the presence of such sensitive tools in petitioners' decisionmaking process would naturally indicate a concern on their part for the First Amendment rights of respondents; the absence of such tools might suggest a lack of such concern. See 638 F.2d, at 416-417 (opinion of Sifton, J.). 1 As a practical matter, however, it is difficult to see the First Amendment right that I believe is at work here playing a role in a school's choice of curriculum. The school's finite resources—as well as the limited number of hours in the day require that education officials make sensitive choices between subjects to be offered and competing areas of academic emphasis; subjects generally are excluded simply because school officials have chosen to devote their resources to one rather than to another subject. As is explained below, a choice of this nature does not run afoul of the First Amendment. In any event, the Court has recognized that students' First Amendment rights in most cases must give way if they interfere "with the schools' work or [with] the rights of other students to be secure and to be let alone," Tinker v. Des Moines School Dist., 393 U.S. 503, 508, 89 S.Ct. 733, 737, 21 L.Ed.2d 731 (1969), and such interference will rise to intolerable levels if public participation in the management of the curriculum becomes commonplace. In contrast, library books on a shelf intrude not at all on the daily operation of a school. I also have some doubt that there is a theoretical distinction between removal of a book and failure to acquire a book. But as Judge Newman observed, there is a profound practical and evidentiary distinction between the two actions: "removal, more than failure to acquire, is likely to suggest that an impermissible political motivation may be present. There are many reasons why a book is not acquired, the most obvious being limited resources, but there are few legitimate reasons why a book, once acquired, should be removed from a library not filled to capacity." 638 F.2d 404, 436 (CA2 1980) (Newman, J., concurring in result). 2 In effect, my view presents the obverse of the plurality's analysis: while the plurality focuses on the failure to provide information, I find crucial the State's decision to single out an idea for disapproval and then deny access to it. 1 At the outset, the plurality notes that certain school board members found the books in question "objectionable" and "improper" for junior and senior high school students. What the plurality apparently finds objectionable is that the inquiry as to the challenged books was initially stimulated by what is characterized as "a politically conservative organization of parents concerned about education," which had concluded that the books in question were "improper fare for school students." Ante, at 856. As noted by the District Court, however, and in the plurality opinion, ante, at 859, both parties substantially agreed about the motivation of the school board in removing the books: "[T]he board acted not on religious principles but on its conservative educational philosophy, and on its belief that the nine books removed from the school library and curriculum were irrelevant, vulgar, immoral, and in bad taste, making them educationally unsuitable for the district's junior and senior high school students." 474 F.Supp. 387, 392 (1979). 2 In oral argument counsel advised the Court that of the original plaintiffs, only "[o]ne of them is still in school . . . until this June, and will assumedly graduate in June. There is a potential question of mootness." Tr. of Oral Arg. 4-5 (emphasis added). The sole surviving plaintiff has therefore either recently been graduated from high school or is within days or even hours of graduation. Yet the plurality expresses views on a very important constitutional issue. Fortunately, there is no binding holding of the Court on the critical constitutional issue presented. We do well to remember the admonition of Justice Frankfurter that "the most fundamental principle of constitutional adjudication is not to face constitutional questions but to avoid them, if at all possible." United States v. Lovett, 328 U.S. 303, 320, 66 S.Ct. 1073, 1081, 90 L.Ed. 1252 (1946) (concurring opinion). In the same vein, Justice Stone warned that "the only check upon our own exercise of power is our own sense of self-restraint." United States v. Butler, 297 U.S. 1, 79, 56 S.Ct. 312, 325, 80 L.Ed. 477 (1936) (dissenting opinion). 3 Of course, it is perfectly clear that, unwise as it would be, the board could wholly dispense with the school library, so far as the First Amendment is concerned. 4 In Rowan a unanimous Court upheld the right of a homeowner to direct the local post office to stop delivery of unwanted materials that the householder viewed as "erotically arousing or sexually provocative." 5 Indeed, this case is illustrative of how essentially all decisions concerning the retention of school library books will become the responsibility of federal courts. As noted in n. 1, supra, the parties agreed that the school board in this case acted not on religious principles but "on its belief that the nine books removed from the school library and curriculum were irrelevant, vulgar, immoral, and in bad taste, making them educationally unsuitable for the district's junior and senior high school students." Despite this agreement as to motivation, the case is to be remanded for a determination of whether removal was in violation of the standard adopted by the plurality. The school board's error appears to be that it made its own determination rather than relying on experts. Ante, at 874-875. 6 Epperson v. Arkansas, 393 U.S. 97, 104, 89 S.Ct. 266, 270, 21 L.Ed.2d 228 (1968). There are approximately 15,000 school districts in the country. U.S. Bureau of Census, Statistical Abstract of the United States 297 (102d ed. 1981) (Table 495: Number of Local Governments, by Taxing Power and Type, and Public School Systems—States: 1972 and 1977). See also Diamond, The First Amendment and Public Schools: The Case Against Judicial Intervention, 59 Texas L.Rev. 477, 506-507, n. 130 (1981). 7 Other provisions of the Constitution, such as the Establishment Clause, Epperson v. Arkansas, supra, and the Equal Protection Clause, also limit the discretion of the school board. 8 The formless nature of the "right" found by the plurality in this case is exemplified by this purported distinction. Presumably a school board could, for any reason, choose not to purchase a book for its library. Once it purchases that book, however, it is "locked in" to retaining it on the school shelf until it can justify a reason for its removal. This anomalous result of "book tenure" was pointed out by the District Court in this case. 474 F.Supp., at 395-396. See also Presidents Council, District 25 v. Community School Board No. 25, 457 F.2d 289, 293 (CA2 1972). Under the plurality view, if a school board wants to be assured that it maintains control over the education of its students, every page of every book sought to be acquired must be read before a purchase decision is made. 1 The plurality speaks of the need for "sensitive" decisionmaking, pursuant to "regular" procedures. See ante, at 874, n. 26, 875. One wonders what indeed does this mean. In this case, for example, the board did not act precipitously. It simply did not agree with the recommendations of a committee it had appointed. Would the plurality require—as a constitutional matter that the board delegate unreviewable authority to such a committee? 2 The plurality suggests that the books in a school library derive special protection under the Constitution because the school library is a place in which students exercise unlimited choice. See ante, at 868-869. This suggestion is without support in law or fact. It is contradicted by this very case. The school board in this case does not view the school library as a place in which students pick from an unlimited range of books—some of which may be inappropriate for young people. Rather, the school library is analogous to an assigned reading list within which students may exercise a degree of choice. 1 I also disagree with Justice WHITE's conclusion that he need not decide the constitutional issue presented by this case. That view seems to me inconsistent with the "rule of four"—"that any case warranting consideration in the opinion of [four Justices] of the Court will be taken and disposed of" on the merits, Ferguson v. Moore-McCormack Lines, Inc., 352 U.S. 521, 560, 77 S.Ct. 457, 479, 1 L.Ed.2d 511 (1957) (opinion of Harlan, J.)—which we customarily follow in exercising our certiorari jurisdiction. His concurrence, although not couched in such language, is in effect a single vote to dismiss the writ of certiorari as improvidently granted. Justice Harlan debated this issue with Justice Frankfurter in Ferguson v. Moore-McCormack Lines, supra, and his view ultimately attracted the support of six out of the seven remaining Members of the Court. He stated: "In my opinion due adherence to [the 'rule of four'] requires that once certiorari has been granted a case should be disposed of on the premise that it is properly here, in the absence of considerations appearing which were not manifest or fully apprehended at the time certiorari was granted. In [this case] I am unable to say that such considerations exist, even though I do think that the arguments on the merits underscored the views of those of us who originally felt that the cas[e] should not be taken because [it] involved only issues of fact, and presented nothing of sufficient general importance to warrant this substantial expenditure of the Court's time." Id., at 559, 77 S.Ct., at 478. The case upon which Justice WHITE relies, Kennedy v. Silas Mason Co., 334 U.S. 249, 68 S.Ct. 1031, 92 L.Ed. 1347 (1948), was disposed of in an opinion which commanded the votes of seven of the nine Members of the Court. There could therefore be no question of an infringement of the "rule of four." Certainly any intimation from that case that this Court should not review questions of law in cases where the District Court has granted summary judgment is belied by subsequent decisions too numerous to catalogue. See, e.g., Ernst & Ernst v. Hochfelder, 425 U.S. 185, 96 S.Ct. 1375, 47 L.Ed.2d 668 (1976); Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975); Mills v. Alabama, 384 U.S. 214, 86 S.Ct. 1434, 16 L.Ed.2d 484 (1966). 2 Rule 9(g) of the local rules of the United States District Court for the Eastern District of New York provides: "Upon any motion for summary judgment pursuant to Rule 56 of the Rules of Civil Procedure, there shall be annexed to the notice of motion a separate, short and concise statement of the material facts as to which the moving party contends there is no genuine issue to be tried. "The papers opposing a motion for summary judgment shall include a separate, short and concise statement of the material facts as to which it is contended that there exists a genuine issue to be tried. "All material facts set forth in the statement required to be served by the moving party will be deemed to be admitted unless controverted by the statement required to be served by the opposing party." 3 Paragraph 4 of respondents' Rule 9(g) statement asserts that petitioners' "evaluation of the suitability of the books was based on [their] personal values, morals, and tastes." App. 139. 4 Paragraph 8 of respondents' Rule 9(g) statement reads: "Defendants Ahrens and Martin objected to those excerpts because some contained profanities, some were sexually explicit, some were ungrammatical, some were anti-American, and some were offensive to racial, religious or ethnic groups." App. 140. 5 There are intimations in Justice BRENNAN's opinion that if petitioners had only consulted literary experts, librarians, and teachers their decision might better withstand First Amendment attack. Ante, at 874, and n. 26. These observations seem to me wholly fatuous; surely ideas are no more accessible or no less suppressed if the school board merely ratifies the opinion of some other group rather than following its own opinion. 6 The right of corporations to make expenditures or contributions in order to influence ballot issues was the question presented in First National Bank of Boston v. Bellotti, 435 U.S. 765, 783, 98 S.Ct. 1407, 1419, 55 L.Ed.2d 707 (1978), and the language which Justice BRENNAN quotes from that decision, ante, at 866, was explicitly limited to "the Court's decisions involving corporations in the business of communications or entertainment." 435 U.S., at 783, 98 S.Ct., at 1419. In Kleindienst v. Mandel, 408 U.S. 753, 92 S.Ct. 2576, 33 L.Ed.2d 683 (1972), the Court upheld the power of Congress and the Executive Branch to prevent the entry into this country of a Marxist theoretician who had been invited to lecture at an American university, despite the First Amendment rights of citizens who wished to hear him. Stanley v. Georgia, 394 U.S. 557, 89 S.Ct. 1243, 22 L.Ed.2d 542 (1969), held that the First Amendment prohibits States from making the private possession of obscene material a crime, and Griswold v. Connecticut, 381 U.S. 479, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965), held that the right of privacy prohibits States from forbidding the use of contraceptives. Finally, Martin v. Struthers, 319 U.S. 141, 63 S.Ct. 862, 87 L.Ed. 1313 (1943), held that the First Amendment protects the door-to-door distribution of religious literature. Justice BRENNAN's concurring opinion appears in a case which considered the constitutionality of certain postal statutes. Lamont v. Postmaster General, 381 U.S. 301, 85 S.Ct. 1493, 14 L.Ed.2d 398 (1965). 7 In First National Bank of Boston v. Bellotti, supra, public access to corporate viewpoints on ballot issues not directly affecting the corporations was foreclosed by the Massachusetts law prohibiting corporate expenditures to express such viewpoints. In Kleindienst v. Mandel, supra, the Court noted that the potential recipients of Mandel's ideas were completely deprived of the "particular qualities inherent in sustained, face-to-face debate, discussion and questioning." 408 U.S., at 765, 92 S.Ct., at 2582. The Georgia law in Stanley v. Georgia, supra, criminalized all private possession of obscene material, and the statute in Griswold v. Connecticut, supra, criminalized all use of contraceptive devices or actions encouraging the use of such devices. The ordinance at issue in Martin v. Struthers, supra, forbade all door-to-door distribution of religious literature, while the statute challenged in Lamont v. Postmaster General, supra, required persons receiving Communist propaganda in the mails affirmatively to state their desire to receive such mailings.

23

457 U.S. 991 102 S.Ct. 2777 73 L.Ed.2d 534 Barbara BLUM, Commissioner of the New York State Department of Social Services, et al., Petitionersv.William YARETSKY et al. No. 80-1952. Argued March 24, 1982. Decided June 25, 1982. Syllabus As a participating State in the Medicaid program established by the Social Security Act, New York provides Medicaid assistance to eligible persons who receive care in private nursing homes, which are designated as either "skilled nursing facilities" (SNF's) or "health related facilities" (HRF's), the latter providing less extensive, and generally less expensive, medical care than the former. The nursing homes are directly reimbursed by the State for the reasonable cost of health care services. To obtain Medicaid assistance, an individual must satisfy eligibility standards in terms of income or resources and must seek medically necessary services. As to the latter requirement, federal regulations require each nursing home to establish a utilization review committee (URC) of physicians whose functions include periodically assessing whether each patient is receiving the appropriate level of care, and thus whether the patient's continued stay in the facility is justified. Respondents, who were Medicaid patients in an SNF, instituted a class action in Federal District Court after the nursing home's URC decided that they should be transferred to a lower level of care in an HRF and so notified local officials, and after administrative hearings resulting in affirmance by state officials of the local officials' decision to discontinue benefits unless respondents accepted transfer to an HRF. Respondents alleged, inter alia, that they had not been afforded adequate notice either of the URC decisions and the reasons supporting them or of their right to an administrative hearing to challenge those decisions, as required by the Due Process Clause of the Fourteenth Amendment. Respondents later added claims as to procedural safeguards that should also apply to URC decisions transferring a patient to a higher level of care and to transfers of any kind initiated by the nursing homes themselves or by the patients' attending physicians. Ultimately the court approved a consent judgment establishing procedural rights applicable to URC-initiated transfers to lower levels of care, and ruled in respondents' favor as to transfers to higher levels of care and all transfers initiated by the facility or its agent. The court permanently enjoined petitioner state officials and all SNF's and HRF's in the State from permitting or ordering discharges of class members or their transfers to a different level of care, without prior written notice and an evidentiary hearing. The Court of Appeals affirmed, holding that URC-initiated transfers to a higher level of care and all discharges and transfers by nursing homes or attending physicians involved "state action" for purposes of the Fourteenth Amendment. Held : 1. Respondents have standing to challenge the procedural adequacy of facility-initiated discharges and transfers to lower levels of care. Although respondents were threatened only with URC-initiated transfers to lower levels of care, and although the consent judgment in the District Court halted implementation of such URC decisions, the threat that the nursing homes might determine, independently of the URC decisions, that respondents' continued stay at current levels of care was not medically necessary is not imaginary or speculative but is quite realistic. However, the threat of transfers to higher levels of care is not of sufficient immediacy and reality that respondents presently have standing to seek an adjudication of the procedures attending such transfers. Thus the District Court exceeded its authority under Art. III in adjudicating the procedures governing transfers to higher levels of care. Pp. 999-1002. 2. Respondents failed to establish "state action" in the nursing homes' decisions to discharge or transfer Medicaid patients to lower levels of care and thus failed to prove that petitioners have violated rights secured by the Fourteenth Amendment. Pp. 1002-1012. (a) The mere fact that a private business is subject to state regulation does not by itself convert its action into that of the State for purposes of the Fourteenth Amendment. A State normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement that the choice must in law be deemed to be that of the State. Pp. 1003-1005. (b) The fact that the State responds to the nursing homes' discharge or transfer decisions by adjusting the patients' Medicaid benefits does not render it responsible for those decisions. Moreover, the pertinent statutes and regulations do not constitute affirmative commands by the State for summary discharge or transfer of Medicaid patients who are thought to be inappropriately placed in nursing facilities. The State, by requiring completion by physicians or nursing homes of forms relating to a patient's condition and discharge or transfer decisions, is not responsible for the decisions of the physicians or nursing homes. Those decisions ultimately turn on medical judgments made by private parties according to professional standards that are not established by the State. Similarly, regulations imposing penalties on nursing homes that fail to discharge or transfer patients whose continued stay is inappropriate do not themselves dictate the decision to discharge or transfer in a particular case. And even though the State subsidizes the cost of the facilities, pays the expenses of the patients, and licenses the facilities, the action of the nursing homes is not thereby converted into "state action." Nor do the nursing homes perform a function that has been "traditionally the exclusive prerogative of the State," Jackson v. Metropolitan Edison Co., 419 U.S. 345, 353, 95 S.Ct. 449, 454, 42 L.Ed.2d 477, so as to establish the required nexus between the State and the challenged action. Pp. 1005-1012. 2nd Cir., 629 F.2d 817, reversed. Judith A. Gordon, New York City, for petitioners. John E. Kirklin, New York City, for respondents. Justice REHNQUIST delivered the opinion of the Court. 1 Respondents represent a class of Medicaid patients challenging decisions by the nursing homes in which they reside to discharge or transfer patients without notice or an opportunity for a hearing. The question is whether the State may be held responsible for those decisions so as to subject them to the strictures of the Fourteenth Amendment. 2 * Congress established the Medicaid program in 1965 as Title XIX of the Social Security Act, 42 U.S.C. § 1396 et seq., (1976 ed. and Supp.IV), to provide federal financial assistance to States that choose to reimburse certain medical costs incurred by the poor. As a participating State, New York provides Medicaid assistance to eligible persons who receive care in private nursing homes, which are designated as either "skilled nursing facilities" (SNF's) or "health related facilities" (HRF's).1 The latter provide less extensive, and generally less expensive, medical care than the former.2 Nursing homes chosen by Medicaid patients are directly reimbursed by the State for the reasonable cost of health care services, N.Y.Soc.Serv.Law § 367-a.1 (McKinney Supp.1981). 3 An individual must meet two conditions to obtain Medicaid assistance. He must satisfy eligibility standards defined in terms of income or resources and he must seek medically necessary services. See 42 U.S.C. § 1396. To assure that the latter condition is satisfied,3 federal regulations require each nursing home to establish a utilization review committee (URC) of physicians whose functions include periodically assessing whether each patient is receiving the appropriate level of care, and thus whether the patient's continued stay in the facility is justified.4 42 CFR §§ 456.305, 456.406 (1981). If the URC determines that the patient should be discharged or transferred to a different level of care, either more or less intensive, it must notify the state agency responsible for administering Medicaid assistance.5 42 CFR §§ 456.337(c), 456.437(d) (1981); 10 NYCRR §§ 416.9(f)(2), (3), 421.13(f)(2), (3) (1980). 4 At the time their complaint was filed, respondents Yaretsky and Cuevas were patients in the American Nursing Home, an SNF located in New York City. Both were recipients of assistance under the Medicaid program. In December 1975 the nursing home's URC decided that respondents did not need the care they were receiving and should be transferred to a lower level of care in an HRF. New York City officials, who were then responsible for administering the Medicaid program in the city, were notified of this decision and prepared to reduce or terminate payments to the nursing home for respondents' care. Following administrative hearings, state social service officials affirmed the decision to discontinue benefits unless respondents accepted a transfer to an HRF providing a reduced level of care. 5 Respondents then commenced this suit, acting individually and on behalf of a class of Medicaid-eligible residents of New York nursing homes.6 Named as defendants were the Commissioners of the New York Department of Social Services and the Department of Health. Respondents alleged in part that the defendants had not afforded them adequate notice either of URC decisions and the reasons supporting them or of their right to an administrative hearing to challenge those decisions.7 Respondents maintained that these actions violated their rights under state and federal law and under the Due Process Clause of the Fourteenth Amendment. They sought injunctive relief and damages.8 6 In January 1978 the District Court certified a class9 and issued a preliminary injunction, restraining the defendants from reducing or terminating Medicaid benefits without timely written notice to the patients, provided by state or local officials, of the reasons for the URC decision, the defendants' proposed action, and the patients' right to an evidentiary hearing and continued benefits pending administrative resolution of the claim. App. 100-101, ¶ 2.10 The court's accompanying opinion relied primarily on existing federal and state regulations. Id., at 112-115. 7 In March 1979 the District Court issued a pretrial order that identified a new claim raised by respondents that a panoply of procedural safeguards should apply to URC decisions transferring a patient to a higher, i.e., more intensive, level of medical care, as well as to decisions recommending transfers to a lower level of care. In addition, respondents claimed that such safeguards were required prior to transfers of any kind initiated by the nursing homes themselves or by the patients' attending physicians. Id., at 157, ¶ II(J); 166-167, ¶ II(J). Respondents asserted that all of these transfers deprived patients of interests protected by the Fourteenth Amendment and were the product of "state action." Id., at 167, ¶ II(J).11 8 In October 1979 the District Court approved a consent judgment incorporating the relief previously awarded by the preliminary injunction and establishing additional substantive and procedural rights applicable to URC-initiated transfers to lower levels of care. Id., at 227-239. The consent judgment left several issues of law to be decided by the District Court. The most important, for our purposes, was "whether there is state action and a constitutional right to a pre-transfer evidentiary hearing in a patient transfer to a higher level of care and/or a patient transfer initiated by the facility or its agents." Id., at 234-235, ¶ VIII(A)(1). Ultimately, the District Court answered that question in respondents' favor, although without elaborating its reasons. Id., at 240. The court permanently enjoined petitioners, as well as all SNF's and HRF's in the State, from permitting or ordering the discharge of class members, or their transfer to a different level of care, without providing advance written notice and an evidentiary hearing on "the validity and appropriateness of the proposed action." Id., at 242-243. 9 The Court of Appeals for the Second Circuit affirmed that portion of the District Court's judgment we have described above. 629 F.2d 817 (1980).12 The court held that URC-initiated transfers from a lower level of care to a higher one, and all discharges and transfers initiated by the nursing homes or attending physicians, "involve state action affecting constitutionally protected property and liberty interests." Id., at 820. The court premised its identification of state action on the fact that state authorities "responded" to the challenged transfers by adjusting the patients' Medicaid benefits. Ibid. Citing our opinion in Jackson v. Metropolitan Edison Co., 419 U.S. 345, 351, 95 S.Ct. 449, 453, 42 L.Ed.2d 477 (1974), the court viewed this response as establishing a sufficiently close "nexus" between the State and either the nursing homes or the URC's to justify treating their actions as those of the State itself. 10 We granted certiorari to consider the Court of Appeals' conclusions about the nature of state action. 454 U.S. 815, 102 S.Ct. 90, 70 L.Ed.2d 83 (1981). We now reverse its judgment. II 11 We first address a question raised by petitioners regarding our jurisdiction under Art. III. They contend that respondents, who were threatened with URC-initiated transfers to lower levels of care, are without standing to object either to URC-initiated transfers to higher levels of care or to transfers of any kind initiated by nursing homes or attending physicians. According to petitioners, respondents obtained complete relief in the consent judgment approved by the District Court in October 1979, which afforded substantive and procedural rights to patients who are the subject of URC-initiated transfers to lower levels of care. Since they have not been threatened with transfers of any other kind, they have no standing to object, and the District Court consequently was without Art. III jurisdiction to enter its judgment. 12 It is axiomatic that the judicial power conferred by Art. III may not be exercised unless the plaintiff shows "that he personally has suffered some actual or threatened injury as a result of the putatively illegal conduct of the defendant." Gladstone, Realtors v. Village of Bellwood, 441 U.S. 91, 99, 99 S.Ct. 1601, 1607, 60 L.Ed.2d 66 (1979). It is not enough that the conduct of which the plaintiff complains will injure someone. The complaining party must also show that he is within the class of persons who will be concretely affected. Nor does a plaintiff who has been subject to injurious conduct of one kind possess by virtue of that injury the necessary stake in litigating conduct of another kind, although similar, to which he has not been subject. See Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 166-167, 92 S.Ct. 1965, 1968-69, 32 L.Ed.2d 627 (1972). 13 Respondents appear to recognize these principles, but contend that although the October 1979 consent judgment halted the implementation of adverse URC decisions recommending discharge or transfer to lower levels of care, the URC determinations themselves were left undisturbed. These determinations reflected the judgment of physicians, chosen by the nursing homes, that respondents' continued stay in their facilities was not medically necessary. Consequently, respondents maintain that they are subject to the serious threat that the nursing home administrators will reach similar conclusions and will themselves initiate patient discharges or transfers without adequate notice or hearings. Petitioners belittle this suggestion, noting that the consent judgment permanently enjoined all New York nursing homes, as well as petitioners, from implementing URC transfers to lower levels of care; this injunction bars the nursing homes from adopting the URC decisions as their own. Petitioners concede, however, that the consent judgment permits the nursing homes and respondents' attending physicians to decide independently to initiate transfers. 14 We conclude that the threat of facility-initiated discharges or transfers to lower levels of care is sufficiently substantial that respondents have standing to challenge their procedural adequacy. In reaching this conclusion, we are mindful of "the primary conception that federal judicial power is to be exercised . . . only at the instance of one who is himself immediately harmed, or immediately threatened with harm, by the challenged action." Poe v. Ullman, 367 U.S. 497, 504, 81 S.Ct. 1752, 1756, 6 L.Ed.2d 989 (1961). Of course, "[o]ne does not have to await the consummation of threatened injury to obtain preventive relief." Pennsylvania v. West Virginia, 262 U.S. 553, 593, 43 S.Ct. 658, 664, 67 L.Ed. 1117 (1923), quoted in Babbitt v. Farm Workers, 442 U.S. 289, 298, 99 S.Ct. 2301, 2308, 60 L.Ed.2d 895 (1979). "[T]he question becomes whether any perceived threat to respondents is sufficiently real and immediate to show an existing controversy. . . ." O'Shea v. Littleton, 414 U.S. 488, 496, 94 S.Ct. 669, 676, 38 L.Ed.2d 674 (1974). Even accepting petitioners' characterization of the scope of the permanent injunction embodied in the consent judgment, the nursing homes in which respondents reside remain free to determine independently that respondents' continued stay at current levels of care is not medically necessary. The possibility that they will do so is not "imaginary or speculative." Younger v. Harris, 401 U.S. 37, 42, 91 S.Ct. 746, 749, 27 L.Ed.2d 669 (1971). In light of similar determinations already made by the committee of physicians chosen by the facilities to make such assessments, the threat is quite realistic. See O'Shea v. Littleton, supra, 414 U.S. at 496, 94 S.Ct., at 676 ("past wrongs are evidence bearing on whether there is real and immediate threat of repeated injury"). 15 We cannot conclude, however, that the threat of transfers to higher levels of care, whether initiated by the URC's, the nursing homes, or attending physicians, is "of sufficient immediacy and reality," Golden v. Zwickler, 394 U.S. 103, 108, 89 S.Ct. 956, 959, 22 L.Ed.2d 113 (1969), that respondents have standing to seek an adjudication of the procedures attending such transfers. Nothing in the record available to this Court suggests that any of the individual respondents have been either transferred to more intensive care or threatened with such transfers. It is not inconceivable that respondents will one day confront this eventuality, but assessing the possibility now would "tak[e] us into the area of speculation and conjecture." O'Shea v. Littleton, supra, 414 U.S. at 497, 94 S.Ct., at 676.13 16 Moreover, the conditions under which such transfers occur are sufficiently different from those which respondents do have standing to challenge that any judicial assessment of their procedural adequacy would be wholly gratuitous and advisory. Transfers to higher levels of care are recommended when the patient's medical needs cannot be satisfied by the facility in which he or she currently resides. Although respondents contend that all transfers threaten elderly patients with physical or psychological trauma, one may infer that refusal to accept a transfer to a higher level of care could itself be a decision with potentially traumatic consequences. The same cannot be said of discharges or transfers to less intensive care. In addition, transfers to more intensive care typically result in an increase in Medicaid benefits to match the increased cost of medically necessary care. Respondents' constitutional attack on discharges or transfers to a lower level of care presupposes adeprivation of protected property interests. Finally, since July 1978, petitioners have adhered to a policy permitting Medicaid patients to refuse URC-recommended transfers to higher levels of care without jeopardizing their Medicaid benefits. App. 180, ¶ 56. No similar policy was in force with respect to other transfers until the District Court mandated its adoption. 17 We conclude, therefore, that although respondents have standing to challenge facility-initiated discharges and transfers to lower levels of care, the District Court exceeded its authority in adjudicating the procedures governing transfers to higher levels of care. We turn now to the "state action" question presented by petitioners. III 18 The Fourteenth Amendment of the Constitution provides in part that "[n]o State shall . . . deprive any person of life, liberty, or property without due process of law." Since this Court's decision in the Civil Rights Cases, 109 U.S. 3, 3 S.Ct. 18, 27 L.Ed. 835 (1883), "the principle has become firmly embedded in our constitutional law that the action inhibited by the first section of the Fourteenth Amendment is only such action as may fairly be said to be that of the States." Shelley v. Kraemer, 334 U.S. 1, 13, 68 S.Ct. 836, 842, 92 L.Ed. 1161 (1948). "That Amendment erects no shield against merely private conduct, however discriminatory or wrongful." Ibid. See Jackson v. Metropolitan Edison Co., 419 U.S. 345, 95 S.Ct. 449, 42 L.Ed.2d 477 (1974); Adickes v. S. H. Kress & Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970). 19 Faithful adherence to the "state action" requirement of the Fourteenth Amendment requires careful attention to the gravamen of the plaintiff's complaint. In this case, respondents objected to the involuntary discharge or transfer of Medicaid patients by their nursing homes without certain procedural safeguards.14 They have named as defendants state officials responsible for administering the Medicaid program in New York. These officials are also responsible for regulating nursing homes in the State, including those in which respondents were receiving care. But respondents are not challenging particular state regulations or procedures, and their arguments concede that the decision to discharge or transfer a patient originates not with state officials, but with nursing homes that are privately owned and operated. Their lawsuit, therefore, seeks to hold state officials liable for the actions of private parties, and the injunctive relief they have obtained requires the State to adopt regulations that will prohibit the private conduct of which they complain. 20 This case is obviously different from those cases in which the defendant is a private party and the question is whether his conduct has sufficiently received the imprimatur of the State so as to make it "state" action for purposes of the Fourteenth Amendment. See, e.g., Flagg Bros., Inc. v. Brooks, 436 U.S. 149, 98 S.Ct. 1729, 56 L.Ed.2d 185 (1978); Jackson v. Metropolitan Edison Co., supra; Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972); Adickes v. S. H. Kress & Co., supra. It also differs from other "state action" cases in which the challenged conduct consists of enforcement of state laws or regulations by state officials who are themselves parties in the lawsuit; in such cases the question typically is whether the private motives which triggered the enforcement of those laws can fairly be attributed to the State. See, e.g., Peterson v. City of Greenville, 373 U.S. 244, 83 S.Ct. 1119, 10 L.Ed.2d 323 (1963). But both these types of cases shed light upon the analysis necessary to resolve the present case. 21 First, although it is apparent that nursing homes in New York are extensively regulated, "[t]he mere fact that a business is subject to state regulation does not by itself convert its action into that of the State for purposes of the Fourteenth Amendment." Jackson v. Metropolitan Edison Co., 419 U.S., at 350, 95 S.Ct., at 453. The complaining party must also show that "there is a sufficiently close nexus between the State and the challenged action of the regulated entity so that the action of the latter may be fairly treated as that of the State itself." Id., at 351, 95 S.Ct., at 453. The purpose of this requirement is to assure that constitutional standards are invoked only when it can be said that the State is responsible for the specific conduct of which the plaintiff complains. The importance of this assurance is evident when, as in this case, the complaining party seeks to hold the State liable for the actions of private parties. 22 Second, although the factual setting of each case will be significant, our precedents indicate that a State normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement, either overt or covert, that the choice must in law be deemed to be that of the State. Flagg Bros., Inc. v. Brooks, supra, 436 U.S. at 166, 98 S.Ct., at 1738; Jackson v. Metropolitan Edison Co., supra, 419 U.S. at 357, 95 S.Ct., at 456; Moose Lodge No. 107 v. Irvis, supra, 407 U.S. at 173, 92 S.Ct., at 1971; Adickes v. S. H. Kress & Co., supra, 398 U.S. at 170, 90 S.Ct., at 1615. Mere approval of or acquiescence in the initiatives of a private party is not sufficient to justify holding the State responsible for those initiatives under the terms of the Fourteenth Amendment. See Flagg Bros., supra, 436 U.S. at 164-165, 98 S.Ct., at 1737-38; Jackson v. Metropolitan Edison Co., supra, 419 U.S. at 357, 95 S.Ct., at 456. 23 Third, the required nexus may be present if the private entity has exercised powers that are "traditionally the exclusive prerogative of the State." Jackson v. Metropolitan Edison Co., supra, at 353, 95 S.Ct., at 454; see Flagg Bros., Inc. v. Brooks, supra, 436 U.S. at 157-161, 98 S.Ct., at 1733-36. B 24 Analyzed in the light of these principles, the Court of Appeals' finding of state action cannot stand. The court reasoned that state action was present in the discharge or transfer decisions implemented by the nursing homes because the State responded to those decisions by adjusting the patient's Medicaid benefits. Respondents, however, do not challenge the adjustment of benefits, but the discharge or transfer of patients to lower levels of care without adequate notice or hearings. That the State responds to such actions by adjusting benefits does not render it responsible for those actions. The decisions about which respondents complain are made by physicians and nursing home administrators, all of whom are concededly private parties. There is no suggestion that those decisions were influenced in any degree by the State's obligation to adjust benefits in conformity with changes in the cost of medically necessary care. 25 Respondents do not rest on the Court of Appeals' rationale, however. They argue that the State "affirmatively commands" the summary discharge or transfer of Medicaid patients who are thought to be inappropriately placed in their nursing facilities. Were this characterization accurate, we would have a different question before us. However, our review of the statutes and regulations identified by respondents does not support respondents' characterization of them. 26 As our earlier summary of the Medicaid program explained, a patient must meet two essential conditions in order to obtain financial assistance. He must satisfy eligibility criteria defined in terms of income and resources and he must seek medically necessary services. 42 U.S.C. § 1396. To assure that nursing home services are medically necessary, federal law requires that a physician so certify at the time the Medicaid patient is admitted and periodically thereafter. 42 U.S.C. § 1396b(g)(1) (1976 ed. and Supp.IV). New York requires that the physician complete a "long term care placement form" devised by the Department of Health, called the DMS-1. 10 NYCRR §§ 415.1(a), 420.1(b) (1980). A completed form provides, inter alia, a numerical score corresponding to the physician's assessment of the patient's mental and physical health. As petitioners note, however, the physicians, and not the forms, make the decision about whether the patient's care is medically necessary.15 A physician can authorize a patient's admission to a nursing facility despite a "low" score on the form. See 10 NYCRR §§ 415.1(a)(2), 420.1(b)(2) (1978).16 We cannot say that the State, by requiring completion of a form, is responsible for the physician's decision. 27 In any case, respondents' complaint is about nursing home decisions to discharge or transfer, not to admit, Medicaid patients. But we are not satisfied that the State is responsible for those decisions either.17 The regulations cited by respondents require SNF's and HRF's "to make all efforts possible to transfer patients to the appropriate level of care or home as indicated by the patient's medical condition or needs," 10 NYCRR §§ 416.9(d)(1), 421.13(d)(1) (1980).18 The nursing homes are required to complete patient care assessment forms designed by the State and "provide the receiving facility or provider with a current copy of same at the time of discharge to an alternate level of care facility or home." 10 NYCRR §§ 416.9(d)(4), 421.13(d)(4) (1980). 28 These regulations do not require the nursing homes to rely on the forms in making discharge or transfer decisions, nor do they demonstrate that the State is responsible for the decision to discharge or transfer particular patients. Those decisions ultimately turn on medical judgments made by private parties according to professional standards that are not established by the State.19 This case, therefore, is not unlike Polk County v. Dodson, 454 U.S. 312, 102 S.Ct. 445, 70 L.Ed.2d 509 (1981), in which the question was whether a public defender acts "under color of" state law within the meaning of 42 U.S.C. § 1983 when representing an indigent defendant in a state criminal proceeding.20 Although the public defender was employed by the State and appointed by the State to represent the respondent, we concluded that "[t]his assignment entailed functions and obligations in no way dependent on state authority." Id., at 318, 102 S.Ct., at 449. The decisions made by the public defender in the course of representing his client were framed in accordance with professional canons of ethics, rather than dictated by any rule of conduct imposed by the State. The same is true of nursing home decisions to discharge or transfer particular patients because the care they are receiving is medically inappropriate.21 29 Respondents next point to regulations which, they say, impose a range of penalties on nursing homes that fail to discharge or transfer patients whose continued stay is inappropriate. One regulation excludes from participation in the Medicaid program health care providers who "[f]urnished items or services that are substantially in excess of the beneficiary's needs." 42 CFR § 420.101(a)(2) (1981). The State is also authorized to fine health care providers who violate applicable regulations. 10 NYCRR § 414.18 (1978). As we have previously concluded, however, those regulations themselves do not dictate the decision to discharge or transfer in a particular case. Consequently, penalties imposed for violating the regulations add nothing to respondents' claim of state action. 30 As an alternative position, respondents argue that even if the State does not command the transfers at issue, it reviews and either approves or rejects them on the merits. The regulations cited by respondents will not bear this construction. Although the State requires the nursing homes to complete patient care assessment forms and file them with state Medicaid officials, 10 NYCRR §§ 415.1(a), 420.1(b) (1978), and although federal law requires that state officials review these assessments, 42 CFR §§ 456.271, 456.372 (1981), nothing in the regulations authorizes the officials to approve or disapprove decisions either to retain or discharge particular patients, and petitioners specifically disclaim any such responsibility. Instead, the State is obliged to approve or disapprove continued payment of Medicaid benefits after a change in the patient's need for services. See 42 CFR § 435.916 (1981). Adjustments in benefit levels in response to a decision to discharge or transfer a patient does not constitute approval or enforcement of that decision. As we have already concluded, this degree of involvement is too slim a basis on which to predicate a finding of state action in the decision itself. 31 Finally, respondents advance the rather vague generalization that such a relationship exists between the State and the nursing homes it regulates that the State may be considered a joint participant in the homes' discharge and transfer of Medicaid patients. For this proposition they rely upon Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961). Respondents argue that state subsidization of the operating and capital costs of the facilities, payment of the medical expenses of more than 90% of the patients in the facilities, and the licensing of the facilities by the State, taken together convert the action of the homes into "state" action. But accepting all of these assertions as true, we are nonetheless unable to agree that the State is responsible for the decisions challenged by respondents. As we have previously held, privately owned enterprises providing services that the State would not necessarily provide, even though they are extensively regulated, do not fall within the ambit of Burton. Jackson v. Metropolitan Edison Co., 419 U.S., at 357-358, 95 S.Ct., at 456-57. That programs undertaken by the State result in substantial funding of the activities of a private entity is no more persuasive than the fact of regulation of such an entity in demonstrating that the State is responsible for decisions made by the entity in the course of its business. 32 We are also unable to conclude that the nursing homes perform a function that has been "traditionally the exclusive prerogative of the State." Jackson v. Metropolitan Edison Co., supra, at 353, 95 S.Ct., at 454. Respondents' argument in this regard is premised on their assertion that both the Medicaid statute and the New York Constitution make the State responsible for providing every Medicaid patient with nursing home services. The state constitutional provisions cited by respondents, however, do no more than authorize the legislature to provide funds for the care of the needy. See N.Y.Const., Art. XVII, §§ 1, 3. They do not mandate the provision of any particular care, much less long-term nursing care. Similarly, the Medicaid statute requires that the States provide funding for skilled nursing services as a condition to the receipt of federal moneys. 42 U.S.C. §§ 1396a(a)(13)(B), 1396d(a)(4)(A) (1976 ed. and Supp.IV). It does not require that the States provide the services themselves. Even if respondents' characterization of the State's duties were correct, however, it would not follow that decisions made in the day-to-day administration of a nursing home are the kind of decisions traditionally and exclusively made by the sovereign for and on behalf of the public. Indeed, respondents make no such claim, nor could they. IV 33 We conclude that respondents have failed to establish "state action" in the nursing homes' decisions to discharge or transfer Medicaid patients to lower levels of care.22 Consequently, they have failed to prove that petitioners have violated rights secured by the Fourteenth Amendment. The contrary judgment of the Court of Appeals is accordingly 34 Reversed. 35 Justice WHITE, concurring in the judgments. 36 The issue in Blum v. Yaretsky, No. 80-1952 [457 U.S. 991, 102 S.Ct. 2777, 73 L.Ed.2d 534], is whether a private nursing home's decision to discharge or transfer a Medicaid patient satisfies the state-action requirement of the Fourteenth Amendment. To satisfy this requirement, respondents must show that the transfer or discharge is made on the basis of some rule of decision for which the state is responsible. Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S.Ct. 2744, 2754, 73 L.Ed.2d 482. It is not enough to show that the State takes certain actions in response to this private decision. The rule of decision implicated in the actions at issue here appears to be nothing more than a medical judgment. This is the clear import of the majority's conclusion that the "decisions ultimately turn on medical judgments made by private parties according to professional standards that are not established by the State," at 2788, with which I agree. 37 Similarly, the allegations of the petitioners in Rendell-Baker v. Kohn, 457 U.S. 830, 102 S.Ct. 2764, 73 L.Ed.2d 418, fail to satisfy the state-action requirement. In this case, the question of state-action focuses on an employment decision made by a private school that receives most its funding from public sources and is subject to state regulation in certain respects. For me, the critical factor is the absence of any allegation that the employment decision was itself based upon some rule of conduct or policy put forth by the State. As the majority states, "in contrast to the extensive regulation of the school generally, the various regulators showed relatively little interest in the school's personnel matters." 457 U.S., at 841, 102 S.Ct., at 2772. The employment decision remains, therefore, a private decision not fairly attributable to the state. 38 Accordingly, I concur in the judgments. 39 Justice BRENNAN, with whom Justice MARSHALL joins, dissenting. 40 If the Fourteenth Amendment is to have its intended effect as a restraint on the abuse of state power, courts must be sensitive to the manner in which state power is exercised. In an era of active government intervention to remedy social ills, the true character of the State's involvement in, and coercive influence over, the activities of private parties, often through complex and opaque regulatory frameworks, may not always be apparent. But if the task that the Fourteenth Amendment assigns to the courts is thus rendered more burdensome, the courts' obligation to perform that task faithfully, and consistently with the constitutional purpose, is rendered more, not less, important. 41 In deciding whether "state action"1 is present in the context of a claim brought under 42 U.S.C. § 1983 (1976 ed., Supp.IV), the ultimate determination is simply whether the § 1983 defendant has brought the force of the State to bear against the § 1983 plaintiff in a manner the Fourteenth Amendment was designed to inhibit. Where the defendant is a government employee, this inquiry is relatively straightforward. But in deciding whether "state action" is present in actions performed directly by persons other than government employees, what is required is a realistic and delicate appraisal of the State's involvement in the total context of the action taken. "Only by sifting facts and weighing circumstances can the nonobvious involvement of the State in private conduct be attributed its true significance." Burton v. Wilmington Parking Authority, 365 U.S. 715, 722, 81 S.Ct. 856, 860, 6 L.Ed.2d 45 (1961). See Lugar v. Edmondson Oil Co., 457 U.S. 922, 939-942, 102 S.Ct. 2744, 2755-2756, 73 L.Ed.2d 482.2 The Court today departs from the Burton precept, ignoring the nature of the regulatory framework presented by this case in favor of the recitation of abstract tests and a pigeonhole approach to the question of state action. But however correct the Court's tests may be in the abstract, they are worth nothing if they are not faithfully applied. Bolstered by its own preconception of the decisionmaking process challenged by respondents, and of the relationship between the State, the nursing home operator, and the nursing home resident, the Court subjects the regulatory scheme at issue here to only the most perfunctory examination. The Court thus fails to perceive the decisive involvement of the State in the private conduct challenged by the respondents. 42 * A. 43 The Court's analysis in this case is simple, but it is also demonstrably flawed, for it proceeds upon a premise that is factually unfounded. The Court first describes the decision to transfer a nursing home resident from one level of care to another as involving nothing more than a physician's independent assessment of the appropriate medical treatment required by that resident. Building upon that factual premise, the Court has no difficulty concluding that the State plays no decisive role in the transfer decision: By reducing the resident's benefits to meet the change in treatment prescribed, the State is simply responding to "medical judgments made by private parties according to professional standards that are not established by the State." Ante, at 1008. If this were an accurate characterization of the circumstances of this case, I too would conclude that there was no "state action" in the nursing home's decision to transfer. A doctor who prescribes drugs for a patient on the basis of his independent medical judgment is not rendered a state actor merely because the State may reimburse the patient in different amounts depending upon which drug is prescribed. 44 But the level-of-care decisions at issue in this case, even when characterized as the "independent" decision of the nursing home, see ante, at 1000, have far less to do with the exercise of independent professional judgment than they do with the State's desire to save money. To be sure, standards for implementing the level-of-care scheme established by the Medicaid program are framed with reference to the underlying purpose of that program—to provide needed medical services. And not surprisingly, the State relies on doctors to implement this aspect of its Medicaid program. But the idea of two mutually exclusive levels of care skilled nursing care and intermediate care—embodied in the federal regulatory scheme and implemented by the State, reflects no established medical model of health care. On the contrary, the two levels of long-term institutionalized care enshrined in the Medicaid scheme are legislative constructs, designed to serve governmental cost-containment policies. 45 The fiscal underpinning of the level-of-care determinations at issue here are apparent from the legislative history of the "intermediate care" concept. In 1967, Congress was concerned with the increasing costs of the Medicaid program. Congress' motivation in establishing a program of reimbursement for care in intermediate-care facilities flowed directly from these fiscal concerns. Thus the Senate Finance Committee Report on the Social Security Amendments of 1967, S.Rep.No.744, 90th Cong., 1st Sess., 188 (1967), expressed concern with the fact that only skilled nursing care was available under Medicaid: "[B]ecause of a decided financial advantage to a State under present matching formulas," States tended to classify recipients as in need of " 'skilled nursing home' care." As a consequence, the Report noted, "a strong case exists for introducing another level of care for which vendor payments would be available." Ibid. The result was an amendment to Title XI of the Social Security Act, creating a new treatment track for "categorically needy" medicaid patients, called "intermediate care." As summarized on the Senate floor: 46 "The committee bill would provide for a vendor payment in behalf of persons . . . who are living in facilities which are more than boarding houses but which are less than skilled nursing homes. The rate of Federal sharing for payments for care in those institutions would be at the same rate as for medical assistance under title XIX. Such homes would have to meet safety and sanitation standards comparable to those required for nursing homes in a given state. 47 "This provision should result in a reduction in the cost of title XIX by allowing States to relocate substantial numbers of welfare recipients who are now in skilled nursing homes in lower cost institutions." 113 Cong.Rec. 32599 (1967) (emphasis added). 48 To implement this cost-saving mechanism, the Federal Government has required States participating in the Medicaid Program to establish elaborate systems of periodic "utilization review."3 With respect to patients whose expenses are not reimbursed through Medicaid, these attempts to assign the patient to one of two mutually exclusive "levels of care" would be anomalous. While the criteria used to determine which patients require the services of "skilled-nursing facilities," which require "intermediate care facilities," and which require no long-term institutional care at all, obviously have a medical nexus, those criteria are not geared to the specific needs of particular residents as determined by a physician; the level-of-care determination is not analogous to choosing specific medication or rehabilitative services needed by a nursing home patient. The inherent imprecision of using two broad levels to classify facilities and residents has been noted by the commentators.4 The vigor with which these reviews are performed in the nursing home context, see infra, at 1022-1024, is extraordinarily un medical in character. From a purely medical standpoint, the idea of shifting nursing home residents from a "higher level of care" to a "lower level of care," which almost invariably involves transfer from one facility to another, rarely makes sense. As one commentator has observed: "These transfers eject helpless, disoriented people from the places they have lived for months or even years to facilities, not of their own choosing, that they have never seen before. The evidence is overwhelming that, without extraordinary preparatory efforts that are hardly ever made, any move is harmful for the preponderance of the frail elderly." B. Vladeck, Unloving Care 140 (1980). 49 The arbitrariness of the statutory system of treatment levels is evident from a comparison of the proportion of nursing home residents in skilled nursing facilities (SNF's) and those in intermediate care facilities (ICF's) in different States. A 1973 survey of 32 States revealed that 47.9% of Medicaid patients were in SNF's, 52.1% were in ICF's. But the proportion of SNF and ICF beds varied enormously from State to State. For example, less than 10% of Medicaid recipients receiving long-term institutional care in States such as Louisiana, Maine, Oregon, and Virginia were in SNF's; the number housed in SNF's in New York and Pennsylvania was nearly 80%, and in Florida and Georgia the figure was closer to 90%.5 Quite obviously, the answer to this disparity lies not in medical considerations or judgments, but rather in the varying fiscal policies, and the vigor of enforcement, in the participating States. 50 In New York, the nursing home operator is required to "maintain a discharge planning program to . . . document that the facility has made and is continuing to make all efforts possible to transfer patients to the appropriate level of care or home as indicated by the patient's medical condition or needs." 10 NYCRR § 416.9(d)(1) (emphasis added). See also § 421.13(d)(1) (1980).6 The responsibility the State assigns to nursing home operators to transfer patients to appropriate levels of care is, of course, designed primarily to implement the State's goal of reducing Medicaid costs,7 and the termination or reduction of benefits follows forthwith upon the facility's discharge or transfer of a resident. As the court below noted: "The state has, in essence, delegated a decision to . . . reduce a public assistance recipient's benefits to a 'private' party," 629 F.2d 817, 820 (CA2 1980), by assigning to that private party the responsibility to determine the recipient's need. But we should not rely on that fact alone in evaluating the nexus between the State and the challenged private action. Here the State's involvement clearly extends to supplying the standards to be used in making the delegated decision. B 51 Ignoring the State's fiscal interest in the level-of-care determination, the Court proceeds to a cursory, and misleading, discussion of the State's involvement in the assignment of residents to particular levels of care. In my view, an accurate and realistic appraisal of the procedures actually employed in the State of New York leaves no doubt that not only has the State established the system of treatment levels and utilization review in order to further its own fiscal goals, but that the State prescribes with as much precision as is possible the standards by which individual determinations are to be made. 52 The Court notes that at the time of admission the admitting physician is required to complete a long-term placement form called the DMS-1. 10 NYCRR §§ 415.1(a), 420.1(b) (1978). The Court dismisses the significance of the form by noting blandly that a "completed form provides . . . a numerical score corresponding to the physician's assessment of the patient's mental and physical health," and then commenting: "As petitioners note, . . . the physicians, and not the forms, make the decision about whether the patient's care is medically necessary. A physician can authorize a patient's admission to a nursing facility despite a 'low' score on the form. See 10 NYCRR §§ 415.1(a)(2), 420.1(b)(2) (1978)." Ante, at 1006 (footnote omitted and emphasis added). The Court concludes: "We cannot say that the State, by requiring completion of a form, is responsible for the physician's decision." Ante, at 1006-1007 (emphasis added). A closer look at the regulations at issue suggests that petitioners have been less than candid in their characterization of the admission process and the role of the numerical score. 53 New York's regulations mandate that the nursing home operator shall 54 "admit a patient only on physician's orders and in accordance with the patient assessment criteria and standards as promulgated and published by the department (New York State LongTerm Care Placement Form [DMS-1] and New York State Numerical Standards Master Sheet [DMS-9] ) . . . which shall include, as a minimum: 55 "(1) an assessment, performed prior to admission by or on behalf of the agency or person seeking admission for the patient, of the patient's level of care needs according to the patient assessment criteria and standards promulgated and published by the department." 10 NYCRR § 415.1 (1978) (emphasis added). 56 The details of the DMS-9 Numerical Standards Master Sheet also bear more emphasis than the Court gives them, for that form describes with particularity the patients who are entitled to SNF care, ICF care, or no long-term residential care at all. The DMS-9 provides numerical scores for various resident dysfunctions. For example, if the resident is incontinent with urine often, he receives a score of 20; if seldom, a score of 10; if never, a score of 0. A similar rating is made as to stool incontinence: often, 40; seldom, 20; never, 0. A tabulation is made with respect to "function status." For example, if the resident can walk only with "some help," he receives 35 points; only with "total help," 70 points; if he cannot walk, 105 points. If the resident needs "total help" to dress, he receives 80 points; if "some help" is required, 40 points. Ratings are also made of the patient's "mental status." For example, if the patient is never alert, he receives 40 points; if sometimes alert, 20 points; always alert, 0 points. If his judgment is always impaired, he receives, 30 points; sometimes, 15 points; never, 0 points. And ratings are also set forth for other physical "impairments." For example, if the patient's vision is unimpaired, he receives 0 points; if he has partial sight, 1 point; if he is blind, 2 points. 57 The criterion for admission to a SNF is a DMS-9 "predictor score" of 180. 10 NYCRR § 415.1(a)(2) (1978). For admission to an HRF (health-related facility), the required score is 60. § 420.1(b)(2). Where the admission, or denial of admission, is based on the guidelines set forth in these regulations, there is, of course, no doubt, that the State is directly, and solely, "responsible for the specific conduct of which the plaintiff complains," ante, at 1004 (emphasis omitted), even if it has chosen to authorize a private party to implement that decision.8 58 The Court dismisses the specific state standards for denying admission set forth in the regulations, and tabulated according to the DMS-9, by emphasizing what it perceives as an alternative method for gaining admission to a nursing home. In the Court's view, this alternative route to admission takes the whole scheme outside the realm of state action because it hinges on a "physician's assessment" of what is medically necessary. In characterizing the admission process as the independent assessment of a physician, the Court relies upon, but fails to quote, the following state regulations. The language of those regulations bears noting: 59 "[F]or those patients failing to meet the criteria and standards for admission to the . . . facility [as measured by the DMS-9], a certification signed by a physician member of the transferring facility's utilization review agent or signed by the responsible social services district local medicaid medical director or designee indicating the reason(s) the patient requires [the facility's level of care, is required]." 10 NYCRR §§ 415.1(a)(2) (1978) (emphasis added). See also § 420.1(b)(2). 60 As this provision makes clear, if the potential resident does not qualify under the specific standards of the DMS-1, as tabulated on the DMS-9, the patient can be admitted only on the basis of direct approval by Medicaid officials themselves, or on the basis of a determination by the utilization review agent of the transferring facility—and, of course, such agents are themselves clearly part and parcel of the statutory cost-control process.9 See n. 8, supra. No decision is made on the basis of a medical judgment exercised outside the regulatory framework, by the resident's personal physician acting on the basis of his personal medical judgment. The attending physician's role is, at this stage, limited to "scoring" the patient's condition according to standards set forth by the State on the DMS-9. 61 Yet the State's involvement does not end with the initial certification. Within five days after admission, the matter is again subjected to assessment, this time by the operator of the transferee facility. This time the transferee nursing home operator is required to tabulate the DMS-9 score. If the patient's score is not adequate by the standards of the DMS-9, admission must be denied unless sanctioned by the facility's utilization review agent.10 The utilization review agent of the admitting facility, like that of the transferring facility, operates under a "written utilization control plan, approved by the department [of health]." 10 NYCRR §§ 416.9, 421.13 (1980). And that statutory body has the final say in each instance. There can thus be little doubt that in the vast majority of cases, decisions as to "level of treatment" in the admission process are made according to the State's specified criteria. That some deviation from the most literal application of the State's guidelines is permitted cannot change the character of the State's involvement. Indeed, absent such provision for exceptional cases, the formularized approach embodied in the DMS-9 would be unconscionable. And indeed, even with respect to these exceptional cases, the admissions procedure is administered through bodies whose structure and operations conform to state requirements, and whose decisions follow state guidelines—albeit guidelines somewhat more flexible than the DMS-1, in allowing some "psychosocial" factors to be taken into account. See infra, this page and 1025-1026. 62 The Court dismisses all this by noting that "[w]e cannot say that the State, by requiring completion of a form, is responsible for the physician's decision." Ante, at 1006-1007. The Court then notes that "[i]n any case, respondents' complaint is about nursing home decisions to discharge or transfer, not to admit, Medicaid patients." Ante, at 1007. This is true, of course. But where, one might ask, is the Court's discussion of the frequent utilization reviews that occur after admission? The State's regulations require that the operator shall provide for "continued stay reviews . . . to promote efficient and effective use of available health facilities and services every 30 days for the first 90 days, and every 90 days thereafter, for each nursing home patient." 10 NYCRR § 416.9(b)(1) (1980) (skilled nursing facilities) (emphasis added). See also § 421.13(b)(1) (health-related facilities, every 90 days). 63 The continued stay reviews parallel the admission determination with respect to both the State's procedural and substantive standards.11 Again, the DMS-1 and the DMS-9 channel the medical inquiry and function as the principal determinants of the resident's status, for whenever a resident does not achieve an appropriate score on the DMS-1, as determined by a nonphysician representative of the utilization review agent, the resident's case is directed to a physician member. That physician member does not personally examine the resident, but rather relies on the DMS-1 and other documentary information. See App. 172-173. If the matter is resolved adversely to the resident, only then must the attending physician be notified. The attending physician is allowed to present relevant information, though the final decision remains with the utilization review agent. See 10 NYCRR §§ 416.9(b)(2), 421.13(b)(2) (1980). And again, the State's substantive standards, not independent medical judgment, pervade review determinations. Evaluations are based only on the DMS-1 and DMS-9 tabulation, on a "psycho-social" evaluation respecting the resident's response to transfer and other physical, emotional, and mental characteristics of the patient, on the resident's discharge plan (prepared according to state regulations), and upon "additional criteria and standards . . . which shall have been approved by the department [of health]." 10 NYCRR §§ 416.9(b)(4), (v), 421.13(b)(4)(v) (1980) (emphasis added).12 64 The Court concludes with this assessment of the statutory scheme: 65 "These regulations do not require the nursing homes to rely on the forms in making discharge or transfer decisions, nor do they demonstrate that the State is responsible for the decision to discharge or transfer particular patients. Those decisions ultimately turn on medical judgments made by private parties according to professional standards that are not established by the State." Ante, at 1008. 66 The Court is wrong. As a fair reading of the relevant regulations makes clear, the State (and Federal Government) have created, and administer, the level system as a cost-saving tool of the Medicaid program. The impetus for this active program of review imposed upon the nursing home operator is primarily this fiscal concern. The State has set forth precisely the standards upon which the level-of-care determinations are to be made, and has delegated administration of the program to the nursing home operators, rather than assume the burden of administering the program itself. Thus, not only does the program implement the State's fiscal goals, but, to paraphrase the Court, "[t]hese requirements . . . make the State responsible for actual decisions to discharge or transfer particular patients." See ante, at 1008, n. 18. Where, as here, a private party acts on behalf of the State to implement state policy, his action is state action. II 67 The deficiency in the Court's analysis is dramatized by its inattention to the special characteristics of the nursing home. Quite apart from the State's specific involvement in the transfer decisions at issue in this case, the nature of the nursing home as an institution, sustained by state and federal funds, and pervasively regulated by the State so as to ensure that it is properly implementing the governmental undertaking to provide assistance to the elderly and disabled that is embodied in the Medicaid program, undercuts the Court's sterile approach to the state action inquiry in this case. The private nursing homes of the Nation exist, and profit, at the sufferance of state and federal Medicaid and Medicare agencies. The degree of interdependence between the State and the nursing home is far more pronounced than it was between the State and the private entity in Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961). The State subsidizes practically all of the operating and capital costs of the facility, and pays the medical expenses of more than 90% of its residents. And, in setting reimbursement rates, the State generally affords the nursing homes a profit as well. Even more striking is the fact that the residents of those homes are, by definition, utterly dependent on the State for their support and their placement. For many, the totality of their social network is the nursing home community. Within that environment, the nursing home operator is the immediate authority, the provider of food, clothing, shelter, and health care, and, in every significant respect, the functional equivalent of a State. Cf. Marsh v. Alabama, 326 U.S. 501, 66 S.Ct. 276, 90 L.Ed. 265 (1946). Surely, in this context we must be especially alert to those situations in which the State "has elected to place its power, property and prestige behind" the actions of the nursing home owner. See Burton v. Wilmington Parking Authority, supra, 365 U.S. at 725, 81 S.Ct., at 861. 68 Yet, whatever might be the status of the nursing home operator where the State has simply left the resident in his charge, while paying for the resident's support and care, it is clear that the State has not simply left nursing home patients to the care of nursing home operators. No one would doubt that nursing homes are "pervasively regulated" by State and Federal Governments; virtually every action by the operator is subject to state oversight. But the question at this stage is not whether the procedures set forth in the state and federal regulatory scheme are sufficient to protect the residents' interests. We are confronted with the question preliminary to any Fourteenth Amendment challenge: whether the State has brought its force to bear against the plaintiffs through the office of these private parties. In answering that question we may safely assume that when the State chooses to perform its governmental undertakings through private institutions, and with the aid of private parties, not every action of those private parties is state action. But when the State directs, supports, and encourages those private parties to take specific action, that is state action. 69 We may hypothesize many decisions of nursing home operators that affect patients, but are not attributable to the State.13 But with respect to decisions to transfer patients downward from one level of care to another, if that decision is in any way connected with the statutory review structure set forth above,14 then there is no doubt that the standard for decision, and impetus for the decision, is the responsibility of the State. Indeed, with respect to the level-of-care determination, the State does everything but pay the nursing home operator a fixed salary. Because the State is clearly responsible for the specific conduct of petitioners about which respondents complain, and because this renders petitioners state actors for purposes of the Fourteenth Amendment, I dissent. 1 N.Y.Soc.Serv.Law § 365-a.2(b) (McKinney Supp.1982). Title XIX requires as a condition to the receipt of federal funds that participating States provide financial assistance to eligible persons in need of "skilled nursing facility services." 42 U.S.C. §§ 1396a(a)(13)(B), 1396d(a)(4)(A) (1976 ed. and Supp.IV). Federal assistance is also available to States that choose to reimburse the cost of "intermediate care facility services." § 1396d(a)(15). See §§ 1396d(c), (f). New York regulations refer to facilities that provide the latter type of care as HRF's. 10 NYCRR § 414.1(a) (1981). 2 Compare 10 NYCRR §§ 416.1-416.2 with §§ 421.1-421.2 (1978). The parties have stipulated that Medicaid reimbursement rates for HRF's are generally lower than those for SNF's. See App. 169, ¶ 12. 3 Congress has provided that federal funds supplied to assist in reimbursing nursing home costs will be reduced unless the participating State provides for the periodic review of patient care "to safeguard against unnecessary utilization of such care and services and to assure that payments . . . are not in excess of reasonable charges consistent with efficiency, economy, and quality of care." 42 U.S.C. § 1396a(a)(30). See §§ 1396b(g)(1)(C), 1396b(i)(4), 1395x(k). 4 These committees must be composed of private physicians who are not directly responsible for the patient whose care is being reviewed. 42 CFR §§ 456.306, 456.406 (1981). Under New York law, the committee members may not be employed by the SNF or HRF and may not have a financial interest in any residential care facility. 10 NYCRR §§ 416.9(b)(2), 421.13(b)(2) (1980). 5 If the committee determines that a discharge or transfer is called for, it must afford the patient's attending physician an opportunity to present his views, although the committee's decision ultimately is final. 42 CFR §§ 456.336(f), (h), 456.436(f), (i) (1981). See 10 NYCRR §§ 731.11, 741.14 (1980). 6 The class was defined to include patients "who have been, are or will be threatened or forced to leave their nursing homes and have their Medicaid benefits reduced or terminated as a result of 'Utilization Review' committee findings alleging that they are not eligible for the level of nursing home care they receive." App. 19, ¶ 1. The complaint also named as a plaintiff the New York chapter of the Gray Panthers, an organization that "has among its objectives the development of a health care system for the elderly which provides quality health care to all persons." Id., at 21, ¶ 5. 7 The complaint also alleged that URC transfers to lower levels of care and corresponding reductions in Medicaid benefits were arbitrary and were caused by improperly constituted URC's that acted without adequate written criteria and failed to afford adequate notice either to the patients or their attending physicians. 8 Ten individuals, who are also respondents in this Court, later intervened in the suit. Each intervenor was a resident of either an SNF or an HRF and had been the subject of a URC decision recommending transfer to a lower level of care. The intervenors all were afforded administrative hearings resulting in affirmance of petitioners' decisions to reduce or terminate Medicaid benefits if the intervenors did not follow URC recommendations. 9 The class was defined to include "all persons who are residents in skilled nursing or intermediate care facilities in the State of New York and who, following utilization review recommendations and/or fair hearings, are determined by defendants to be ineligible to receive the level of care at the facilities in which they reside and to be subject to reduction or termination of their Medicaid benefits." Id., at 45. 10 The court also required the defendants to afford class members access to all pertinent case files and medical records. Id., at 101-102. The Court of Appeals for the Second Circuit upheld portions of the injunction challenged by petitioners. Yaretsky v. Blum, 592 F.2d 65 (2nd Cir. 1979). 11 The pretrial order also redefined the class to include "all residents of skilled nursing and health related nursing facilities in New York State who are recipients of Medicaid benefits." App. 151. 12 The court modified the injunction by relieving petitioners of obligations that, in the opinion of federal authorities, would render the State ineligible for Medicaid funding. 629 F.2d, at 822. The court also reversed the District Court's holding that state administrators were precluded by due process or state law from rejecting a hearing officer's recommendation favorable to a patient without reading a verbatim transcript of the hearing and the exhibits. Id., at 822-825. This holding is not before us. 13 Respondents suggest that members of the class they represent have been transferred to higher levels of care as a result of URC decisions. Respondents, however, "must allege and show that they personally have been injured, not that injury has been suffered by other, unidentified members of the class to which they belong and which they purport to represent." Warth v. Seldin, 422 U.S. 490, 502, 95 S.Ct. 2197, 2207, 45 L.Ed.2d 343 (1975). Unless these individuals "can thus demonstrate the requisite case or controversy between themselves personally and [petitioners], 'none may seek relief on behalf of himself or any other member of the class.' O'Shea v. Littleton, 414 U.S. 488, 494, 94 S.Ct. 669, 675, 38 L.Ed.2d 674 (1974)." Ibid. 14 "From the beginning of this lawsuit the respondents' challenge has been to the involuntary discharge or transfer of Medicaid patients from and by their nursing facilities without adequate safeguards. . . . Thus, the claim before this Court is whether state action attaches to a nursing facility's summary discharge or transfer of the patient. . . ." Brief for Respondents 21-22 (emphasis in original). 15 A completed DMS-1 form provides a summary of the patient's medical condition. Five of the eleven questions devoted to this subject require the assignment of numerical values. See 10 NYCRR App. C-1 (1978). A range of numerical values to be used in completing these questions are set forth in a second form, called the DMS-9. See ibid. The dissent's discussion of the DMS-9 suggests that completion of the DMS-1 form is a purely mechanical exercise that does not require the exercise of independent medical judgment. The dissent's discussion is incomplete. The other six questions on the DMS-1 ask the physician such questions as whether the patient requires daily supervision by a registered nurse, whether complications would arise without skilled nursing care, whether a program of therapy is necessary, and if so what kind, whether the patient should be considered for different levels of care, and whether the patient is medically qualified for the level of care he or she is receiving. The physician brings to bear his own medical judgment in answering these questions; their placement on the form would be inexplicable if the numerical scores were dispositive. 16 The dissent belittles this fact by noting that the decision to depart from the form in admitting a patient is made by a physician member of the nursing home's URC, and that such persons are "part and parcel of the statutory cost control process." Post, at 1022. This signifies nothing more than the fact, disputed by no one, that the State requires utilization review in order to reduce unnecessary Medicaid expenditures. It remains true that physician members of the URC's are not employed by the State and, more important, render medical judgments concerning the patient's health needs that the State does not prescribe and for which it is not responsible. We must also emphasize, of course, that we are ultimately concerned with decisions to transfer patients who have already been admitted. Apropos of this relevant issue, the dissent observes, post, at 1023, that once a patient has been admitted, the State requires, as a condition to the disbursement of Medicaid funds, that within five days after admission the nursing home operator assess the patient's status according to standards contained in the DMS-1 and DMS-9 forms. As the dissent is also aware, post, at 1023, n. 10, a physician member of the URC has the power to determine that the patient needs the level of care he is receiving despite an adverse score on the DMS-1. 10 NYCRR §§ 416.9(a)(2)(i), 421.13(a)(2)(i) (1980). That decision, rendered after consultation with the patient's attending physician, is purely a medical judgment for which the State, as before, is not responsible. 17 The dissent condemns us for conducting a "cursory" review of the regulations governing utilization review, post, at 1019, and pointedly asks "where . . . is the Court's discussion of the frequent utilization reviews that occur after admission?" Post, at 1024. The dissent, in its headlong dive into the sea of state regulations, forgets that patient transfers to lower levels of care initiated by utilization review committees are simply not part of this case. As we noted earlier, such transfers were the subject of a consent judgment in October 1979. We are concerned only with transfers initiated by the patients' attending physicians or the nursing home administrators themselves. Therefore, we have focused on regulations that concern decisions which are not the product of URC recommendations. As we explain in the text, those regulations do not demonstrate that the State is responsible for the transfers with which we are concerned. 18 Federal regulations also require SNF's and HRF's to obtain from admitting physicians a plan of discharge for each patient. 42 CFR §§ 456.280(b)(6), 456.380(b)(6) (1981). State regulations require that nursing home staff members assist in the preparation of these plans, which are designed to summarize "the patient's potential for return to the community, for transfer to another more appropriate setting or for achieving or maintaining the best obtainable level of function in the nursing home." 10 NYCRR §§ 416.1(k)(2)(ii), 421.3(b)(2) (1976). These requirements hardly make the State responsible for actual decisions to discharge or transfer particular patients. 19 The dissent characterizes as "factually unfounded," post, at 1014, our conclusion that decisions initiated by nursing homes and physicians to transfer patients to lower levels of care ultimately depend on private judgments about the health needs of the patients. It asserts that different levels of care exist only because of the State's desire to save money, and that the same interest explains the requirement that nursing homes transfer patients who do not need the care they are receiving. Post, at 1014-1019. We do not suggest otherwise. Transfers to lower levels of care are not mandated by the patients' health needs. But they occur only after an assessment of those needs. In other words, although "downward" transfers are made possible and encouraged for efficiency reasons, they can occur only after the decision is made that the patient does not need the care he or she is currently receiving. The State is simply not responsible for that decision, although it clearly responds to it. In concrete terms, therefore, if a particular patient objects to his transfer to a different nursing facility, the "fault" lies not with the State but ultimately with the judgment, made by concededly private parties, that he is receiving expensive care that he does not need. That judgment is a medical one, not a question of accounting. 20 This case, of course, does not involve the "under color of law" requirement of § 1983. Nevertheless, it is clear that the reasoning employed in Polk County is equally applicable to "state action" cases such as this one. 21 Respondents also point to statutes requiring the State periodically to send medical review teams to conduct on-site inspections of all SNF's and HRF's. During these inspections, state employees are required to review the appropriateness of each patient's continued stay in the facility and to report their findings to the nursing home and the agency responsible for administering the Medicaid program in the State. 42 U.S.C. §§ 1396a(a)(26), (31), 1396b(g)(1)(D) (1976 ed. and Supp.IV). See 42 CFR § 456.611 (1981). Petitioners concede that these inspections can result in a discharge or transfer directed by state health officials. As they correctly argue, however, transfers of this kind are not the subject of respondents' complaint and none are presented by the record. 22 As a postscript to their "state action" arguments, respondents suggest that this Court avoid the issue by holding that federal and state statutes and regulations require the procedural safeguards which they seek. The lower courts did not pass on this assertion, and we decline to do so as well. 1 As the Court noted in Lugar v. Edmondson Oil Co., 457 U.S. 922, 926-932, 102 S.Ct. 2744, 2748-2751, 73 L.Ed.2d 482, the state action necessary to support a claimed violation of the Fourteenth Amendment, and the action "under color of law" required by 42 U.S.C. § 1983 (1976 ed., Supp.IV), represent parallel avenues of inquiry in a case claiming a remedy under § 1983 for a violation of the Fourteenth Amendment's Due Process Clause. Of course, the "color of law" inquiry required by § 1983 focuses directly on the question whether the conduct of the particular § 1983 defendant is sufficiently connected with the state action that is present whenever the constitutionality of a state law, regulation, or practice is properly challenged. But this question may just as easily be framed as whether the § 1983 defendant is a "state actor." 2 In Lugar, we addressed a decidedly different question of "state action." In that case, the § 1983 plaintiff sought damages against a private party who had availed himself of an unconstitutional state attachment procedure, and had enlisted the aid of government officials to impair plaintiff's property for his own benefit. We concluded that "a private party's joint participation with state officials in the seizure of disputed property is sufficient to characterize that party as a 'state actor' for purposes of the Fourteenth Amendment." 457 U.S., at 941, 102 S.Ct., at 2756. Here the State affirmatively relies upon and requires private parties to implement specific deprivations of benefits according to standards and procedures that the State has established and enforces for its own benefit. The imprint of state power on the private party's actions would seem in this circumstance to be even more significant. 3 The State must provide for the periodic review of patient care "to safeguard against unnecessary utilization of such care and services and to assure that payments . . . are not in excess of reasonable charges consistent with efficiency, economy, and quality of care." 42 U.S.C. § 1396a(a)(30). See 42 U.S.C. §§ 1395x(k), 1396a(a)(31), 1396b(g)(1)(C) (1976 ed. and Supp.IV); 42 CFR §§ 456.305, 456.406 (1981). There is no need here to dwell on the very detailed federal requirements, except to note that if the State fails to ensure that the physician certifications and utilization review procedures are implemented for each patient in each facility, the State is subject to a loss of Medicaid funds commensurate with the extent of the failure to ensure such utilization review. See 42 U.S.C. §§ 1396b(g), (i)(4) (1976b ed. and Supp.IV); 42 CFR §§ 456.650-456.657 (1981). 4 See, e.g., Bishop, Plough, & Willemain, Nursing Home Levels of Care: Problems and Alternatives, 2 Health Care Financing Rev., No. 2, pp. 33, 36 (1980). 5 See B. Vladeck, Unloving Care 138 (1980). "There is no reason to believe that Medicaid recipients in Georgia or Pennsylvania are ten times as likely to need skilled care as those in Oklahoma or Oregon, but they are ten times as likely to get it, or at least to get something called 'skilled care.' " Id., at 137. 6 If the nursing home fails to assign the patients to the level of care the State deems appropriate, it is subject to sanction. Federal regulations provide that health care providers who furnish "items or services that are substantially in excess of the beneficiary's needs" may be excluded from participating in the program. 42 CFR § 420.101(a)(2) (1981). A nursing home that fails to follow state regulations is also subject to state-imposed daily penalties. See 10 NYCRR § 414.18 (1978). It is also clear that under the federal scheme, the State's responsibility extends to ensuring proper assessment of every resident. See 42 U.S.C. §§ 1396a(26)(A), 1396a(31)(A), 1396b(g)(1)(D) (1976 ed. and Supp.IV). 7 To acknowledge that the active system of utilization review serves a primarily fiscal purpose is not to demean the importance of that purpose, or the extent of overplacement of Medicaid recipients in skilled nursing facilities. That figure has been variously estimated at 10 to 40 percent. See Bishop, Plough, & Willemain, supra, n. 4. 8 The Court mistakes the significance of the DMS-1, and the relevant inquiry, when it attempts to characterize that form as merely an instrument for recording the exercise of an independently exercised medical judgment. See ante, at 1006, n. 15. Of course, a medical background is essential in filling out the forms. But it remains clear that the State's standards are to be applied in making the transfer determination. The Court concludes that the patient assessment standards prescribed by the State may be easily disregarded. But the regulations themselves clearly demonstrate that those standards are not merely precatory. Notably, the regulations specify that "patient assessment standards shall not be applied to residents admitted to the residential health care facility prior to March 1, 1977." 10 NYCRR §§ 416.9(a)(1), 421.13(a)(1) (1980) (emphasis added). See also §§ 416.9(b)(4)(vi), 421.13(b)(4)(vi). If the forms merely recorded the exercise of an independent medical judgment, rather than prescribed the standards upon which that judgment must be exercised, why would it be necessary to exempt certain patients from the inquiry? Indeed, the regulations specifically provide for a different set of standards to be applied to the continued stay review of patients admitted to a facility prior to March 1, 1977. See 10 NYCRR §§ 416.9(b)(4)(vii), 421.13(b)(4)(vii) (1980) ("the standards for residents admitted to the facility prior to March 1, 1977 shall be developed by the utilization review agent and approved by the department"). Again, if the determination were in reality based on an independent medical assessment, it seems inconceivable to me that the State would have any interest in requiring different standards for different patients depending on when the patient had been admitted. 9 Federal regulations require each nursing home to establish a utilization review committee whose functions include review of admission decisions, and the periodic assessment of the resident's condition to determine whether the resident's continued stay in the facility is justified. See 42 CFR §§ 456.301, 456.406 (1981). These review agents, as they are deemed in the New York regulations, are composed of physicians not directly responsible for the patient whose care is being reviewed. §§ 456.306, 456.606. Under New York law, the physicians of the review agent may not have a financial interest in a residential care facility. 10 NYCRR §§ 416.9(b)(2), 421.13(b)(2) (1980). In New York, the review agent generally consists of two or more physicians selected and appointed by the facility. Medicaid provides reimbursement for their services. App. 173. 10 A physician member of the utilization review agent has the power to determine that the patient qualifies for the type of care that the facility offers, even if the patient's score on the DMS-1 is insufficient. 10 NYCRR §§ 416.9(a)(2)(i), 421.13(a)(2)(i) (1980). If that physician member confirms that the patient is not in need of the facility's level of care, he must then notify the patient's attending physician "and afford that physician an opportunity for consultation." § 416.9(a)(2)(ii). But even if the attending physician disagrees with the adverse admission finding of the utilization review agent physician, it is the utilization review agent, not the attending physician, that makes the admission decision. §§ 416.9(a)(2)(iv), 421.13(a)(2)(iv). The utilization review agent must, however, notify "the responsible social services district" of "any adverse admission decision." §§ 416.9(a)(3), 421.13(a)(3). 11 The Court takes issue with our reliance on the nature of continued stay reviews performed by the utilization review agent, noting that "patient transfers to lower levels of care initiated by utilization review committees are simply not part of this case." Ante, at 1007, n. 17. The Court's position with respect to the work of the utilization review committee is schizophrenic at best: The Court expressly relies on its characterization of the review committee's work as representing an independent physician's assessment in reaching its conclusion that the DMS-1 and DMS-9 do not supply the criterion controlling the nursing home operator's decision to admit or retain a patient in the home. Ante, at 1006, see discussion supra, at 1022. In any event, the Court simply misses the point. The nursing home operator is under a continuing duty "to make all efforts possible to transfer patients to the appropriate level of care or home as indicated by the patient's medical condition or needs." 10 NYCRR §§ 416.9(d)(1), 421.13(d)(1) (1980). Whether performed through the utilization review agent, or whether undertaken by the nursing home operator directly, transfers premised on the "patient's medical condition or needs" are to be made with reference to the State's definition of "need." 12 If it is finally determined by the utilization review agent that the patient should be assigned to a lower level of care, the regulations set forth an elaborate scheme of review before the State Department of Health. See 10 NYCRR §§ 416.9(f), 421.13(f) (1980). These provisions apply even when the attending physician concurs in the determination. The utilization review committee must notify the Department of Health of its adverse finding and "send to the department a written statement setting forth, in specific detail, the changed medical conditions or other circumstances of the individual which support the utilization review agent's decision for transfer, and a copy of the completed patient assessment form (DMS-1) used by the utilization review agent in this review. The department shall review the adverse continued stay finding." §§ 416.9(f)(2)(i), 421.13(f)(2)(i) (emphasis added). See also §§ 421.13(f)(3)(i), 416.9(f)(3)(i). Of course, there is no doubt that the determinations made on this review represent state action because they are performed by state officials. But if the initial determinations were not made according to state-established standards and for the State's purposes, and were in fact "independent" medical decisions as characterized by the Court, it is difficult to understand the State's active role in reviewing the substance of those determinations. 13 Of course, the nursing home operator's power to make transfer decisions for other than medical reasons is severely limited by regulation. He may only discharge or transfer the resident for valid medical reasons, for the welfare of the affected patient or other patients, or for non-payment. 42 CFR §§ 405.1121(k)(4), 442.311(c) (1981); 10 NYCRR § 414.14(4) (1980). 14 The issue presented in this case—the issue that the Court decides presents a live controversy—concerns facility-initiated discharges or transfers. See ante, at 1000. Transfers initiated by the Utilization Review Committee are within the terms of the consent decree entered by the District Court below, and are not before the Court today. These transfers even more clearly show the State's hand in the transfer decision—indeed, it appears that the physicians on the Committees are reimbursed for their services by Medicaid. But there is absolutely no basis upon which to conclude that that decision to transfer a patient to a lower level of care can be made in any meaningful way independently of the state regulatory standards described in text. Of course, we might hypothesize a decision of the resident's personal physician, not premised on the State's view of what constitutes an appropriate level of care for the patient, to remove the patient from the particular facility. In these circumstances, I would agree that the nursing home owner, in simply responding to the personal physician's request, is not a state actor. But it appears to me that the Court's decision sweeps more broadly than that, and clearly reaches transfers based directly upon and arising from the State's procedures and standards.

12

457 U.S. 957 102 S.Ct. 2836 73 L.Ed.2d 508 William P. CLEMENTS, Jr., Governor of the State of Texas, et al., Appellantsv.John L. FASHING et al. No. 80-1290. Argued Jan. 12, 1982. Decided June 25, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1133, 103 S.Ct. 20. Syllabus Article III, § 19, of the Texas Constitution provides that "[n]o judge of any court, Secretary of State, Attorney General, clerk of any court of record, or any person holding a lucrative office under the United States, or this State, or any foreign government shall during the term for which he is elected or appointed, be eligible to the Legislature." As interpreted by the Texas Supreme Court, § 19 requires an officeholder to complete his current term of office—if it overlaps the legislature's term before he may be eligible to serve in the state legislature. Article XVI, § 65, provides that if holders of certain state and county offices whose unexpired term exceeds one year become candidates for any other state or federal office, this shall constitute an automatic resignation of the office then held. Appellees—who challenged these provisions in Federal District Court as violating the First Amendment and the Equal Protection Clause of the Fourteenth Amendment of the Federal Constitution included officeholders subject to § 65, each of whom alleged that he would have announced his candidacy for higher judicial office except that such announcement would constitute an automatic resignation from his current position, and one of whom (Baca), a Justice of the Peace, also alleged that he could not become a candidate for the state legislature because of § 19. The other appellees were voters who alleged that they would vote for the officeholder-appellees were they to become candidates. The District Court held that the challenged provisions denied appellees equal protection, and the Court of Appeals affirmed. Held : The judgment is reversed. 5th Cir., 631 F.2d 731, reversed. Justice REHNQUIST delivered the opinion of the Court with respect to Parts I, II, and V, concluding that: 1. The uncontested allegations in the complaint are sufficient to create an actual case or controversy between the officeholder-appellees and those Texas officials charged with enforcing §§ 19 and 65. Pp. 961-962. 2. Sections 19 and 65 do not violate the First Amendment. The State's interests are sufficient to warrant the de minimis interference with appellees' First Amendment interests in candidacy. In addition, appellees' First Amendment challenge as elected state officeholders contesting restrictions on partisan political activity must fail since §§ 19 and 65 represent a far more limited restriction on political activity than has been upheld with regard tocivil servants. Cf. CSC v. Letter Carriers, 413 U.S. 548, 93 S.Ct. 2880, 37 L.Ed.2d 796; Broadrick v. Oklahoma, 413 U.S. 601, 93 S.Ct. 2908, 37 L.Ed.2d 830; United Public Workers v. Mitchell, 330 U.S. 75, 67 S.Ct. 556, 91 L.Ed. 754. Pp.971-973 Justice REHNQUIST, joined by THE CHIEF JUSTICE, Justice POWELL, and Justice O'CONNOR, concluded in Parts III and IV that neither of the challenged provisions of the Texas Constitution violates the Equal Protection Clause. Pp. 962-971. 1 (a) Candidacy is not a "fundamental right" that itself requires departure from traditional equal protection principles under which state-law classifications need only be drawn in such a manner as to bear some rational relationship to a legitimate state end. Decision in this area of constitutional adjudication is a matter of degree, and involves a consideration of the facts and circumstances behind the law, the interests the State seeks to protect by placing restrictions on candidacy, and the nature of the interests of those who may be burdened by the restrictions. In determining whether the provisions challenged here deserve "scrutiny" more vigorous than that which the traditional principles would require, the nature of the interests affected and the extent of the burden the challenged provisions place on the candidacy of current officeholders must be examined. Pp. 962-966. 2 (b) As applied to Baca, a Justice of the Peace whose term of office is four years whereas a state legislator's term is two years, § 19 simply requires that Baca must wait, at most, two years—one election cycle—before he may run as a candidate for the legislature. In establishing this maximum "waiting period," § 19 places a de minimis burden on the political aspirations of a current officeholder. This sort of insignificant interference with access to the ballot need only rest on a rational predicate in order to survive an equal protection challenge. Section 19 clearly rests on a rational predicate, since it furthers Texas' interests in maintaining the integrity of its Justices of the Peace by ensuring that they will neither abuse their position nor neglect their duties because of aspirations for higher office. Moreover, Texas has a legitimate interest in discouraging its Justices of the Peace from vacating their current terms of office, thereby avoiding the difficulties that accompany interim elections and appointments. Nor is § 19 invalid in that it burdens only those officeholders who desire to run for the legislature. It would be a perversion of the Equal Protection Clause to conclude that Texas must restrict a Justice of the Peace's candidacy for all offices before it can restrict his candidacy for any office. Pp. 966-970. (c) The burdens imposed on candidacy by the automatic-resignation provision of § 65 are even less substantial than those imposed by § 19. Both provisions serve essentially the same state interests. Nor is § 65 invalid on the ground that it applies only to certain elected officials and not to others. Its history shows that the resignation provision was a creature of state electoral reforms, and a regulation is not devoid of a rational predicate simply because it happens to be incomplete. The Equal Protection Clause does not prohibit Texas from restricting one elected officeholder's candidacy for another elected office unless and until it places similar restrictions on other officeholders. Pp. 970-971. 3 James P. Allison, Cooper, Tex., for appellants. 4 Raymond C. Caballero, El Paso, Tex., for appellees. 5 Justice REHNQUIST delivered the opinion of the Court with respect to Parts I, II, and V, and delivered an opinion with respect to Parts III and IV, in which THE CHIEF JUSTICE, Justice POWELL, and Justice O'CONNOR joined. 6 Appellees in this case challenge two provisions of the Texas Constitution that limit a public official's ability to become a candidate for another public office. The primary question in this appeal is whether these provisions violate the Equal Protection Clause of the Fourteenth Amendment. 7 * Article III, § 19, of the Texas Constitution provides: 8 "No judge of any court, Secretary of State, Attorney General, clerk of any court of record, or any person holding a lucrative office under the United States, or this State, or any foreign government shall during the term for which he is elected or appointed, be eligible to the Legislature." 9 Section 19 renders an officeholder ineligible for the Texas Legislature if his current term of office will not expire until after the legislative term to which he aspires begins. Lee v. Daniels, 377 S.W.2d 618, 619 (Tex.1964). Resignation is ineffective to avoid § 19 if the officeholder's current term of office overlaps the term of the legislature to which he seeks election. Ibid. In other words, § 19 requires an officeholder to complete his current term of office before he may be eligible to serve in the legislature. 10 Article XVI, § 65, is commonly referred to as a "resign-to-run" or "automatic resignation" provision. Section 65 covers a wide range of state and county offices.1 It provides in relevant part: 11 "[I]f any of the officers named herein shall announce their candidacy, or shall in fact become a candidate, in any General, Special or Primary Election, for any office of profit or trust under the laws of this State or the United States other than the office then held, at any time when the unexpired term of the office then held shall exceed one (1) year, such announcement or such candidacy shall constitute an automatic resignation of the office then held." Four of the appellees are officeholders subject to the automatic resignation provision of § 65. Fashing is a County Judge, Baca and McGhee are Justices of the Peace, and Ybarra is a Constable. Each officeholder-appellee alleged in the complaint that he is qualified under Texas law to be a candidate for higher judicial office, and that the reason he has not and will not announce his candidacy is that such an announcement will constitute an automatic resignation from his current position. Appellee Baca alleged in addition that he could not become a candidate for the legislature because of § 19. The remaining appellees are 20 voters who allege that they would vote for the officeholder-appellees were they to become candidates. 12 The District Court for the Western District of Texas held that § 19 and § 65 denied appellees equal protection. Fashing v. Moore, 489 F.Supp. 471 (1980). The District Court concluded that § 19 created "classifications that are invidiously discriminatory." Id., at 475. The District Court explained that § 19 draws distinctions between those officials whose terms end concurrently with the beginning of the legislative term and those whose terms overlap the legislative term. The court also found § 19 deficient because "[n]o reciprocal prohibition . . . is placed upon a legislator seeking to run for mayor or judge." Ibid. As to § 65, the District Court determined that the classifications embodied in § 65 "fail[ed] to serve any proper governmental interest" because some state and local officials were covered by § 65 while others were not. The Court of Appeals for the Fifth Circuit affirmed without opinion. Fashing v. Moore, 631 F.2d 731 (1980). We noted probable jurisdiction, 452 U.S. 904, 101 S.Ct. 3028, 69 L.Ed.2d 404 (1981), and now reverse. II 13 Before we may reach the merits of the constitutional issues in this case, we must address appellants' contention that the allegations in the complaint are insufficient to create a "case or controversy" between the officeholder-appellees and those Texas officials charged with enforcing § 19 and § 65. Appellants contend that the dispute in this case is merely hypothetical and therefore not a justiciable controversy within the meaning of Art. III of the United States Constitution. United Public Workers v. Mitchell, 330 U.S. 75, 90-91, 67 S.Ct. 556, 564-565, 91 L.Ed. 754 (1947). 14 We find the uncontested allegations in the complaint sufficient to create an actual case or controversy. The officeholder-appellees have alleged that they have not and will not announce their candidacy for higher judicial office because such action will constitute an automatic resignation of their current offices pursuant to § 65. Unlike the situation in Mitchell, appellees have alleged in a precise manner that, but for the sanctions of the constitutional provision they seek to challenge, they would engage in the very acts that would trigger the enforcement of the provision. Given that § 65 provides for automatic resignation upon an announcement of candidacy, it cannot be said that § 65 presents only a speculative or hypothetical obstacle to appellees' candidacy for higher judicial office. See Regional Rail Reorganization Act Cases, 419 U.S. 102, 143, and n. 29, 95 S.Ct. 335, 358 and n. 29, 42 L.Ed.2d 320 (1974); Turner v. Fouche, 396 U.S. 346, 361-362, n. 23, 90 S.Ct. 532, 540-541, n. 23, 24 L.Ed.2d 567 (1970). 15 Baca's uncontested allegations are sufficient to create a case or controversy with regard to § 19. That provision entirely disables an officeholder from becoming a candidate for the legislature until he completes his present term of office. The gist of Baca's challenge to § 19 is that it renders him ineligible to become a candidate for the legislature because his term as Justice of the Peace overlaps the legislative term. Baca's dispute with appellants over the constitutionality of § 19, therefore, cannot be said to be abstract or hypothetical, since he has sufficiently alleged that § 19 has prevented him from becoming a candidate for the legislature. III 16 The Equal Protection Clause allows the States considerable leeway to enact legislation that may appear to affect similarly situated people differently. Legislatures are ordinarily assumed to have acted constitutionally. Under traditional equal protection principles, distinctions need only be drawn in such a manner as to bear some rational relationship to a legitimate state end. Classifications are set aside only if they are based solely on reasons totally unrelated to the pursuit of the State's goals and only if no grounds can be conceived to justify them. See, e.g., McDonald v. Board of Election Comm'rs, 394 U.S. 802, 808-809, 89 S.Ct. 1404, 1408-1409, 22 L.Ed.2d 739 (1969); McGowan v. Maryland, 366 U.S. 420, 425-426, 81 S.Ct. 1101, 1104-1105, 6 L.Ed.2d 393 (1961). We have departed from traditional equal protection principles only when the challenged statute places burdens upon "suspect classes" of persons or on a constitutional right that is deemed to be "fundamental." San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 17, 93 S.Ct. 1278, 1288, 36 L.Ed.2d 16 (1973). 17 Thus, we must first determine whether the provisions challenged in this case deserve "scrutiny" more vigorous than that which the traditional principles would require. 18 Far from recognizing candidacy as a "fundamental right," we have held that the existence of barriers to a candidate's access to the ballot "does not of itself compel close scrutiny." Bullock v. Carter, 405 U.S. 134, 143, 92 S.Ct. 849, 855, 31 L.Ed.2d 92 (1972). "In approaching candidate restrictions, it is essential to examine in a realistic light the extent and nature of their impact on voters." Ibid. In assessing challenges to state election laws that restrict access to the ballot, this Court has not formulated a "litmus-paper test for separating those restrictions that are valid from those that are invidious under the Equal Protection Clause." Storer v. Brown, 415 U.S. 724, 730, 94 S.Ct. 1274, 1279, 39 L.Ed.2d 714 (1974). Decision in this area of constitutional adjudication is a matter of degree, and involves a consideration of the facts and circumstances behind the law, the interests the State seeks to protect by placing restrictions on candidacy, and the nature of the interests of those who may be burdened by the restrictions. Ibid.; Williams v. Rhodes, 393 U.S. 23, 30, 89 S.Ct. 5, 10, 21 L.Ed.2d 24 (1968). Our ballot access cases, however, do focus on the degree to which the challenged restrictions operate as a mechanism to exclude certain classes of candidates from the electoral process. The inquiry is whether the challenged restriction unfairly or unnecessarily burdens the "availability of political opportunity." Lubin v. Panish, 415 U.S. 709, 716, 94 S.Ct. 1315, 1320, 39 L.Ed.2d 702 (1974). This Court has departed from traditional equal protection analysis in recent years in two essentially separate, although similar, lines of ballot access cases. 19 One line of ballot access cases involves classifications based on wealth.2 In invalidating candidate filing-fee provisions, for example, we have departed from traditional equal protection analysis because such a "system falls with unequal weight on voters, as well as candidates, according to their economic status." Bullock v. Carter, supra, 405 U.S. at 144, 92 S.Ct. at 856. "Whatever may be the political mood at any given time, our tradition has been one of hospitality toward all candidates without regard to their economic status." Lubin v. Panish, supra, 415 U.S. at 717-718, 94 S.Ct. at 1320-1321. Economic status is not a measure of a prospective candidate's qualifications to hold elective office, and a filing fee alone is an inadequate test of whether a candidacy is serious or spurious. Clearly, the challenged provisions in the instant case involve neither filing fees nor restrictions that invidiously burden those of lower economic status. This line of cases, therefore does not support a departure from the traditional equal protection principles. 20 The second line of ballot access cases involves classification schemes that impose burdens on new or small political parties or independent candidates. See, e.g., Illinois State Bd. of Elections Bd. v. Socialist Workers Party, 440 U.S. 173, 99 S.Ct. 983, 59 L.Ed.2d 230 (1979); Storer v. Brown, supra; American Party of Texas v. White, 415 U.S. 767, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974); Jenness v. Fortson, 403 U.S. 431, 91 S.Ct. 1970, 29 L.Ed.2d 554 (1971); Williams v. Rhodes, supra. These cases involve requirements that an independent candidate or minor party demonstrate a certain level of support among the electorate before the minor party or candidate may obtain a place on the ballot. In these cases, the Court has emphasized that the States have important interests in protecting the integrity of their political processes from frivolous or fraudulent candidacies, in ensuring that their election processes are efficient, in avoiding voter confusion caused by an overcrowded ballot, and in avoiding the expense and burden of run-off elections. To this end, the Court has upheld reasonable level-of-support requirements and classifications that turn on the political party's success in prior elections. See Storer v. Brown, supra; American Party of Texas v. White, supra; Jenness v. Fortson, supra. The Court has recognized, however, that such requirements may burden First Amendment interests in ensuring freedom of association, as these requirements classify on the basis of a candidate's association with particular political parties. Consequently, the State may not act to maintain the "status quo" by making it virtually impossible for any but the two major parties to achieve ballot positions for their candidates. SeeWilliams v. Rhodes, supra, 393 U.S. at 25, 89 S.Ct. at 7. 21 The provisions of the Texas Constitution challenged in this case do not contain any classification that imposes special burdens on minority political parties or independent candidates. The burdens placed on those candidates subject to § 19 and § 65 in no way depend upon political affiliation or political viewpoint. 22 It does not automatically follow, of course, that we must apply traditional equal protection principles in examining § 19 and § 65 merely because these restrictions on candidacy do not fall into the two patterns just described. But this fact does counsel against discarding traditional principles without first examining the nature of the interests that are affected and the extent of the burden these provisions place on candidacy. See Bullock v. Carter, supra, 405 U.S. at 143, 92 S.Ct. at 855; Storer v. Brown, supra, 415 U.S. at 730, 94 S.Ct. at 1279. Not all ballot access restrictions require "heightened" equal protection scrutiny. The Court, for example, applied traditional equal protection principles to uphold a classification scheme that denied absentee ballots to inmates in jail awaiting trial. McDonald v. Board of Election Comm'rs, 394 U.S., at 807-811, 89 S.Ct., at 1407-1409. Thus, it is necessary to examine the provisions in question in terms of the extent of the burdens that they place on the candidacy of current holders of public office. IV A. 23 Section 19 applies only to candidacy for the Texas Legislature. Of the appellees, only Baca, a Justice of the Peace, alleged that he would run for the Texas Legislature. Of the plaintiffs in this case, only appellee Baca's candidacy for another public office has in any fashion been restricted by § 19. The issue in this case, therefore, is whether § 19 may be applied to a Justice of the Peace in a manner consistent with the Equal Protection Clause.3 24 Section 19 merely prohibits officeholders from cutting short their current term of office in order to serve in the legislature. In Texas, the term of office for a Justice of the Peace is four years, while legislative elections are held every two years. See Tex.Const., Art. V, § 18; Art. III, §§ 3, 4. Therefore, § 19 simply requires Baca to complete his 4-year term as Justice of the Peace before he may be eligible for the legislature. At most, therefore, Baca must wait two years—one election cycle—before he may run as a candidate for the legislature.4 25 In making an equal protection challenge, it is the claimant's burden to "demonstrate in the first instance a discrimination against [him] of some substance." American Party of Texas v. White, 415 U.S., at 781, 94 S.Ct., at 1306. Classification is the essence of all legislation, and only those classifications which are invidious, arbitrary, or irrational offend the Equal Protection Clause of the Constitution. Williamson v. Lee Optical Co., 348 U.S. 483, 489, 75 S.Ct. 461, 465, 99 L.Ed. 563 (1955). 26 In establishing a maximum "waiting period" of two years for candidacy by a Justice of the Peace for the legislature, § 19 places a de minimis burden on the political aspirations of a current officeholder. Section 19 discriminates neither on the basis of political affiliation nor on any factor not related to a candidate's qualifications to hold political office. Unlike filing fees or the level-of-support requirements, § 19 in no way burdens access to the political process by those who are outside the "mainstream" of political life. In this case, § 19 burdens only a candidate who has successfully been elected to one office, but whose political ambitions lead him to pursue a seat in the Texas Legislature. 27 A "waiting period" is hardly a significant barrier to candidacy. In Storer v. Brown, 415 U.S., at 733-737, 94 S.Ct., at 1280-1281, we upheld a statute that imposed a flat disqualification upon any candidate seeking to run in a party primary if he had been registered or affiliated with another political party within the 12 months preceding his declaration of candidacy. Similarly, we upheld a 7-year durational residency requirement for candidacy in Chimento v. Stark, 414 U.S. 802, 94 S.Ct. 125, 38 L.Ed.2d 39 (1973), summarily aff'g 353 F.Supp. 1211 (NH). We conclude that this sort of insignificant interference with access to the ballot need only rest on a rational predicate in order to survive a challenge under the Equal Protection Clause. See Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S., at 189, 99 S.Ct., at 993 (STEVENS, J., concurring in part and in judgment). 28 Section 19 clearly rests on a rational predicate. That provision furthers Texas' interests in maintaining the integrity of the State's Justices of the Peace.5 By prohibiting candidacy for the legislature until completion of one's term of office, § 19 seeks to ensure that a Justice of the Peace will neither abuse his position nor neglect his duties because of his aspirations for higher office. The demands of a political campaign may tempt a Justice of the Peace to devote less than his full time and energies to the responsibilities of his office. A campaigning Justice of the Peace might be tempted to render decisions and take actions that might serve more to further his political ambitions than the responsibilities of his office. The State's interests are especially important with regard to judicial officers. It is a serious accusation to charge a judicial officer with making a politically motivated decision. By contrast, it is to be expected that a legislator will vote with due regard to the views of his constituents. 29 Texas has a legitimate interest in discouraging its Justices of the Peace from vacating their current terms of office. By requiring Justices of the Peace to complete their current terms of office, the State has eliminated one incentive to vacate one's office prior to the expiration of the term. The State may act to avoid the difficulties that accompany interim elections and appointments. "[T]he Constitution does not require the State to choose ineffectual means to achieve its aims." Storer v. Brown, supra, 415 U.S. at 736, 94 S.Ct. at 1282. Under traditional equal protection principles, a classification is not deficient simply because the State could have selected another means of achieving the desired ends. Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 316, 96 S.Ct. 2562, 2568, 49 L.Ed.2d 520 (1976); Mathews v. Diaz, 426 U.S. 67, 83, 96 S.Ct. 1883, 1893, 48 L.Ed.2d 478 (1976); San Antonio Independent School Dist. v. Rodriguez, 411 U.S., at 51, 93 S.Ct., at 1306. 30 Finally, it is no argument that § 19 is invalid because it burdens only those officeholders who desire to run for the legislature. In Broadrick v. Oklahoma, 413 U.S. 601, 607, n. 5, 93 S.Ct. 2908, 2913, n. 5, 37 L.Ed.2d 830 (1973), we rejected the contention that Oklahoma's restrictions on political activity by public employees violated the Equal Protection Clause: 31 "Appellants also claim that § 818 violates the Equal Protection Clause of the Fourteenth Amendment by singling out classified service employees for restrictions on partisan political expression while leaving unclassified personnel free from such restrictions. The contention is somewhat odd in the context of appellants' principal claim, which is that § 818 reaches too far rather than not far enough. In any event, the legislature must have some leeway in determining which of its employment positions require restrictions on partisan political activities and which may be left unregulated. See McGowan v. Maryland, 366 U.S. 420 [81 S.Ct. 1101, 6 L.Ed.2d 393] (1961). And a State can hardly be faulted for attempting to limit the positions upon which such restrictions are placed." 32 It would indeed be a perversion of the Equal Protection Clause were we to conclude that Texas must restrict a Justice of the Peace's candidacy for all offices before it can restrict a Justice of the Peace's candidacy for any office. 33 The Equal Protection Clause allows the State to regulate "one step at a time, addressing itself to the phase of the problem which seems most acute." Williamson v. Lee Optical Co., 348 U.S., at 489, 75 S.Ct., at 465. The State "need not run the risk of losing an entire remedial scheme simply because it failed, through inadvertence or otherwise, to cover every evil that might conceivably have been attacked." McDonald v. Board of Election Comm'rs, 394 U.S., at 809, 89 S.Ct., at 1408 (citation omitted). B 34 Article XVI, § 65, of the Texas Constitution provides that the holders of certain offices automatically resign their positions if they become candidates for any other elected office, unless the unexpired portion of the current term is one year or less. The burdens that § 65 imposes on candidacy are even less substantial than those imposed by § 19. The two provisions, of course, serve essentially the same state interests. The District Court found § 65 deficient, however, not because of the nature or extent of the provision's restriction on candidacy, but because of the manner in which the offices are classified. According to the District Court, the classification system cannot survive equal protection scrutiny because Texas has failed to explain sufficiently why some elected public officials are subject to § 65 and why others are not. As with the case of § 19, we conclude that § 65 survives a challenge under the Equal Protection Clause unless appellees can show that there is no rational predicate to the classification scheme. 35 The history behind § 65 shows that it may be upheld consistent with the "one step at a time" approach that this Court has undertaken with regard to state regulation not subject to more vigorous scrutiny than that sanctioned by the traditional principles. Section 65 was enacted in 1954 as a transitional provision applying only to the 1954 election. 2 G. Braden, The Constitution of the State of Texas: An Annotated and Comparative Analysis 812 (1977). Section 65 extended the terms of those offices enumerated in the provision from two to four years. The provision also staggered the terms of other offices so that at least some county and local offices would be contested at each election. Ibid. The automatic resignation proviso to § 65 was not added until 1958. In that year, a similar automatic resignation provision was added in Art. XI, § 11, which applies to officeholders in home rule cities who serve terms longer than two years. Section 11 allows home rule cities the option of extending the terms of municipal offices from two to up to four years. 36 Thus, the automatic resignation provision in Texas is a creature of the State's electoral reforms of 1958. That the State did not go further in applying the automatic resignation provision to those officeholders whose terms were not extended by § 11 or § 65, absent an invidious purpose, is not the sort of malfunctioning of the State's lawmaking process forbidden by the Equal Protection Clause. See McDonald v. Board of Election Comm'rs, supra, at 809, 89 S.Ct., at 1408. A regulation is not devoid of a rational predicate simply because it happens to be incomplete. See Williamson v. Lee Optical Co., supra, at 489, 75 S.Ct., at 465. The Equal Protection Clause does not forbid Texas to restrict one elected officeholder's candidacy for another elected office unless and until it places similar restrictions on other officeholders. Broadrick v. Oklahoma, 413 U.S., at 607, n. 5, 93 S.Ct., at 2913 n. 5. Cf. Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 466, 101 S.Ct. 715, 725, 66 L.Ed.2d 659 (1981). The provision's language and its history belie any notion that § 65 serves the invidious purpose of denying access to the political process to identifiable classes of potential candidates. V 37 As an alternative ground to support the judgments of the courts below, appellees contend that § 19 and § 65 violate the First Amendment. Our analysis of appellees' challenge under the Equal Protection Clause disposes of this argument. We have concluded that the burden on appellees' First Amendment interests in candidacy are so insignificant that the classifications of § 19 and § 65 may be upheld consistent with traditional equal protection principles. The State's interests in this regard are sufficient to warrant the de minimis interference with appellees' interests in candidacy.6 38 There is another reason why appellees' First Amendment challenge must fail. Appellees are elected state officeholders who contest restrictions on partisan political activity. Section 19 and § 65 represent a far more limited restriction on political activity than this Court has upheld with regard to civil servants. See CSC v. Letter Carriers, 413 U.S. 548, 93 S.Ct. 2880, 37 L.Ed.2d 796 (1973); Broadrick v. Oklahoma, supra; United Public Workers v. Mitchell, 330 U.S. 75, 67 S.Ct. 556, 91 L.Ed. 754 (1947). These provisions in no way restrict appellees' ability to participate in the political campaigns of third parties. They limit neither political contributions nor expenditures. They do not preclude appellees from holding an office in a political party. Consistent with § 19 and § 65, appellees may distribute campaign literature and may make speeches on behalf of a candidate. 39 In this case, § 19 operates merely to require appellee Baca to await the conclusion of his 4-year term as Justice of the Peace before he may run for the Texas Legislature. By virtue of § 65, appellees in this case will automatically resign their current offices if they announce their candidacy for higher judicial office so long as the unexpired term of their current office exceeds one year. In this sense, § 19 and § 65 are in reality no different than the provisions we upheld in Mitchell, Letter Carriers, and Broadrick, which required dismissal of any civil servant who became a political candidate. See 413 U.S., at 556, 93 S.Ct., at 2886; 413 U.S., at 617, 93 S.Ct., at 2918. 40 Neither the Equal Protection Clause nor the First Amendment authorizes this Court to review in cases such as this the manner in which a State has decided to govern itself. Constitutional limitations arise only if the classification scheme is invidious or if the challenged provision significantly impairs interests protected by the First Amendment. Our view of the wisdom of a state constitutional provision may not color our task of constitutional adjudication. The judgment of the Court of Appeals is 41 Reversed. 42 Justice STEVENS, concurring in part and concurring in the judgment. 43 In cases presenting issues under the Equal Protection Clause, the Court often plunges directly into a discussion of the "level of scrutiny" that will be used to review state action that affects different classes of persons differently. Unfortunately that analysis may do more to obfuscate than to clarify the inquiry. This case suggests that a better starting point may be a careful identification of the character of the federal interest in equality that is implicated by the State's discriminatory classification. In my opinion, the disparate treatment in this case is not inconsistent with any federal interest that is protected by the Equal Protection Clause. With respect to the state action at issue, there is no federal requirement that the different classes be treated as though they were the same. 44 It is first helpful to put to one side the claim that the burdens imposed on certain Texas officeholders are inconsistent with the First Amendment. I am satisfied that the State's interest in having its officeholders faithfully perform the public responsibilities they have voluntarily undertaken is adequate to justify the restrictions placed on their ability to run for other offices. Nor is the First Amendment violated by the fact that the restrictions do not apply equally to all offices; while that Amendment requires a State's treatment of speech to be evenhanded, there is no suggestion here that the State's classification of offices operates to promote a certain viewpoint at the expense of another. The federal constitutional inquiry thus is limited to the question whether the State's classification offends any interest in equality that is protected by the Equal Protection Clause. 45 In considering that question, certain preliminary observations are important. The complaining officeholders do not object to the fact that they are treated differently from members of the general public.1 The only complaint is that certain officeholders are treated differently from other officeholders. Moreover, appellees do not claim that the classes are treated differently because of any characteristic of the persons who happen to occupy the various offices at any particular time or of the persons whom those officeholders serve; there is no suggestion that the attributes of the offices have been defined to conceal an intent to discriminate on the basis of personal characteristics or to provide governmental services of differing quality to different segments of the community. In this case, the disparate treatment of different officeholders is entirely a function of the different offices that they occupy. 46 The question presented then is whether there is any federal interest in requiring a State to define the benefits and burdens of different elective state offices in any particular manner. In my opinion there is not. As far as the Equal Protection Clause is concerned, a State may decide to pay a justice of the peace a higher salary than a Supreme Court justice. It may require game wardens to work longer hours than park rangers. It may require meat inspectors to wear uniforms without requiring building inspectors to do so. In addition, I see no reason why a State may not provide that certain offices will be filled on a part-time basis and that others will be filled by persons who may not seek other office until they have fulfilled their duties in the first. There may be no explanation for these classifications that a federal judge would find to be "rational." But they do not violate the Equal Protection Clause because there is no federal requirement that a State fit the emoluments or the burdens of different elective state offices into any particular pattern.2 The reason, then, that appellees may be treated differently from other officeholders is that they occupy different offices. Cf. Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 189, 99 S.Ct. 983, 993, 59 L.Ed.2d 230 (STEVENS, J., concurring in part and in judgment).3 47 As in so many areas of the law, it is important to consider each case individually. In the situation presented, however, I believe that there is no federal interest in equality that requires the State of Texas to treat the different classes as though they were the same.4 This reasoning brings me to the same conclusion that Justice REHNQUIST has reached. It avoids, however, the danger of confusing two quite different questions.5 Justice REHNQUIST has demonstrated that there is a "rational basis" for imposing the burdens at issue on the offices covered by §§ 19 and 65. He has not, however, adequately explained the reasons, if any, for imposing those burdens on some offices but not others. With respect to the latter inquiry, the plurality is satisfied to note that the State may approach its goals "one step at a time." Ante, at 969, 970. In my judgment, this response is simply another way of stating that there need be no justification at all for treating two classes differently during the interval between the first step and the second step—an interval that, of course, may well last forever. Although such an approach is unobjectionable in a case involving the differences between different public offices, I surely could not subscribe to Justice REHNQUIST's formulation of the standard to be used in evaluating state legislation that treats different classes of persons differently.6 Accordingly, while I join the Court's judgment, I join only Parts I, II, and V of Justice REHNQUIST's opinion. 48 Justice BRENNAN, with whom Justice MARSHALL and Justice BLACKMUN join, and with whom Justice WHITE joins as to Part I, dissenting. 49 In rejecting appellees' equal protection challenge on the basis that the State is proceeding "one step at a time," the plurality today gives new meaning to the term "legal fiction."1 The Court's summary dismissal of appellees' First Amendment claim vastly oversimplifies the delicate accommodations that must be made between the interests of the State as employer and the constitutionally protected rights of state employees. I dissent. 50 * Putting to one side the question of the proper level of equal protection scrutiny to be applied to these restrictions on candidacy for public office,2 I find it clear that no genuine justification exists that might support the classifications embodied in either Art. III, § 19, or Art. XVI, § 65. 51 The State seeks to justify both provisions on the basis of its interest in discouraging abuse of office and neglect of duties by current officeholders campaigning for higher office during their terms. The plurality posits an additional justification not asserted by the State for § 19: That section also discourages certain officeholders "from vacating their current terms of office." Ante, at 968 . But neither the State nor the plurality offers any justification for differential treatment of various classes of officeholders, and the search for such justification makes clear that the classifications embodied in these provisions lack any meaningful relationship to the State's asserted or supposed interests. Article III, § 19, provides: 52 "No judge of any court, Secretary of State, Attorney General, clerk of any court of record, or any person holding a lucrative office under the United States, or this state, or any foreign government shall during the term for which he is elected or appointed, be eligible to the Legislature." 53 And the Texas Election Code provides that persons ineligible to hold an office shall not be permitted to campaign for that office. Tex.Rev.Civ.Stat.Ann., Arts. 1.05, 1.06 (Vernon Supp.1982). Article III, § 19, creates, in effect, two classes of officeholders. Officeholders of state, federal, and even foreign offices seeking Texas legislative office whose terms overlap with the legislative term are barred from campaigning during their terms, and even after they have resigned, see n. 4, infra ; those officeholders seeking any other office and those officeholders whose terms do not overlap the legislative term are free to launch campaigns from their current offices, even while they still hold office. 54 What relationship does the plurality find between the burden placed on the class of all state, federal, and foreign officeholders seeking legislative seats and the asserted state interests? If it faced the question, the plurality would of course have to acknowledge that Texas has no interest in protecting, for example, federal officials—particularly those serving the electorate of another State—from the corrupting influence of a state legislative campaign. The only conceivable state interest in barring these candidacies would be the purely impermissible one of protecting Texas legislative seats against outside competition. But the plurality does not address this question or purport to find any justification for the broad reach of § 19. Instead it defines the equal protection challenge to § 19 as "whether § 19 may be applied to a [Texas] Justice of the Peace," ante, at 966, and acknowledges that § 19 would not necessarily survive constitutional scrutiny with regard to any other officeholder, ante, at 968, n.5. The plurality defines the question in this manner because Baca, the appellee challenging this provision, is a Justice of the Peace. But the State has defined the class of persons restricted by § 19 as all persons "holding a lucrative office under the United States, or [Texas], or any foreign government." And it has always been my understanding that " '[e]qual protection' . . . emphasizes disparity in treatment by a State between classes of individuals," in contrast to " '[d]ue process'," which "emphasizes fairness between the State and the individual dealing with the State, regardless of how other individuals in the same situation may be treated." Ross v. Moffitt, 417 U.S. 600, 609, 94 S.Ct. 2437, 2443, 41 L.Ed.2d 341 (1974). Accordingly, our equal protection cases have always assessed the legislative purpose in light of the class as the legislature has drawn it, rather than on the basis of some judicially drawn subclass for which it is possible to posit some legitimate purpose for discriminatory treatment. See, e.g., Lubin v. Panish, 415 U.S. 709, 717-718, 94 S.Ct. 1315, 1320-1321, 39 L.Ed.2d 702 (1974).3 When the class of persons burdened by § 19, as the State has drawn it, is viewed in light of the asserted purposes of discouraging abuse of office and neglect of duty, it is beyond dispute that the class is substantially overbroad. 55 The plurality cannot, in the same manner that it avoids the overbreadth of the class, avoid the irrationality in the fact that § 19 applies only to candidacy for the Texas Legislature. Officeholders are free to run for President, the United States Senate, governor, mayor, city council, and many other offices. The distracting and corrupting effects of campaigning are obviously present in all campaigns, not only those for the legislature. The plurality responds to this characteristic of the legislative scheme by stating that "[t]he Equal Protection Clause allows the State to regulate 'one step at a time. . . .' " Ante, at 969, quoting Williamson v. Lee Optical Co., 348 U.S. 483, 489, 75 S.Ct. 461, 465, 99 L.Ed. 563 (1955). But the record in this case belies any assertion by the State that it is proceeding "one step at a time." Article III, § 19, has existed in its present form since 1876. There is no legislative history to explain its intended purpose or to suggest that it is part of a larger, more equitable regulatory scheme.4 And in the 106 years that have passed since § 19's adoption, the Texas Legislature has adopted no comparable bar to candidacy for other offices. 56 A state legislature may implement a program step by step, and an underinclusive regulation may be upheld where the record demonstrates that such "one step at a time" regulation is in fact being undertaken. See, e.g., Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 465-466, 101 S.Ct. 715, 724-725, 66 L.Ed.2d 659 (1981); McDonald v. Board of Election Comm'rs, 394 U.S. 802, 809-811, 89 S.Ct. 1404, 1408-1409, 22 L.Ed.2d 739 (1969). I cannot subscribe, however, to the plurality's wholly fictional one-step-at-a-time justification. As Justice STEVENS points out in his concurrence, the plurality's one-step-at-a-time response in this case "is simply another way of stating that there need be no justification at all for treating two classes differently during the interval between the first step and the second step—an interval that, of course, may well last forever." Ante, at 2850. 57 Section 19's haphazard reach and isolated existence strikes me as the very sort of "arbitrary scheme or plan" that we distinguished from an as-yet-uncompleted design in McDonald v. Board of Election Comm'rs, supra, 394 U.S., at 811, 89 S.Ct., at 1409, a case the plurality relies on to support the classification in this case, see ante, at 971. In McDonald the record demonstrated that in providing absentee ballots to certain classes of persons the State was in fact proceeding step by step. The State had demonstrated "a consistent and laudable state policy of adding, over a 50-year period, groups to the absentee coverage as their existence comes to the attention of the legislature." 394 U.S., at 811, 89 S.Ct., at 1409. Article III, § 19, stands in stark contrast to the provision reviewed in McDonald. In this case, it is pure fiction for the plurality to declare that § 19 is one step in a broader and more equitable scheme that due to legislative delay and inadvertence is yet to be completed. 58 Appellants, unlike the plurality, at least attempt to justify the distinction between legislative campaigns and other campaigns. They argue that an officeholder-candidate will not enforce legislative policy if he or she is campaigning for a legislative seat. Brief for Appellants 9. But this attempted justification is unpersuasive. Appellants' argument apparently rests on the tenuous premise that a candidate is likely to choose the strategy of undermining the program of an incumbent opponent in order to advance his own prospects. It is plain that whatever force there is to this premise cannot be limited to a candidate for the legislature; it may as logically be argued that a judge will further his ambition for higher judicial office by failing to follow judicial decisions of a higher court, or that a state legislator with gubernatorial aspirations will use his present position to sabotage the program of the present administration. Even assuming that the State has a particular interest in protecting state legislative policy, and accepting appellants' somewhat dubious premise, it is still apparent to me that this asserted purpose is ill-served by the group of officeholders covered by § 19. Only those officeholders whose terms happen to overlap with the legislative term are prohibited from running for the legislature.5 The District Court noted that this prohibition is most likely to bar the candidacy of mayors and city councilmen—persons who have little if anything to do with carrying out state legislative policy. Fashing v. Moore, 489 F.Supp. 471, 475 (WD Tex.1980). Appointed administrators, District Attorneys, and District Judges to name just a few—whose terms do not overlap with that of the legislature, but who are directly charged with carrying out legislative policy, are left free to campaign for the legislature while remaining in office. See, e.g., Chapa v. Whittle, 536 S.W.2d 681 (Tex.Civ.App.1976). It is thus clear that the prohibition on legislative campaigns in § 19 furthers in no substantial way the State's asserted interest in fidelity to legislative policy. In short, I can discern neither in the appellants' argument nor in the plurality's hypothesis any rational basis for the discriminatory burden placed upon this class of potential candidates. 59 I turn now to Art. XVI, § 65. That section applies only to persons holding any of approximately 16 enumerated offices.6 With respect to persons holding these offices, Art. XVI, § 65, provides: 60 "[I]f any of the officers named herein shall announce their candidacy, or shall in fact become a candidate, in any General, Special or Primary Election, for any office of profit or trust under the laws of this State or the United States other than the office then held, at any time when the unexpired term of the office then held shall exceed one (1) year, such announcement or such candidacy shall constitute an automatic resignation of the office then held . . . ." 61 Other officeholders, performing similar if not identical duties, are not within the reach of this or any similar restriction and are thus free to campaign for one office while holding another. Article XVI, § 65, while lacking § 19's broad sweep into areas completely beyond the purview of the State's concerns, restricts the candidacy only of an unexplained and seemingly inexplicable collection of administrative, executive, and judicial officials. The only distinguishing features of the officeholders collected in § 65 is that in 1954 their terms of office were increased from two to four years, and they all happen to be precinct, county, and district officials as opposed to members of the legislature or statewide elected officials. See 2 G. Braden, The Constitution of the State of Texas: An Annotated and Comparative Analysis 813 (1977). Neither appellants nor the plurality offer any explanation why the State has a greater interest in having the undivided attention of a "Public Weigher" than of a state criminal court judge, or any reason why the State has a greater interest in preventing the abuse of office by an "Inspector of Hides and Animals," than by a justice of the Texas Supreme Court. Yet in each instance § 65 applies to the former office and not to the latter. Again the plurality opines that the State is legislating "one step at a time." But while Art. XVI, § 65, is of more recent vintage than Art. III, § 19, it has been part of the Texas Constitution for 24 years without prompting any corresponding rule applicable to holders of statewide office. Thus § 65, like § 19, cannot in any realistic sense be upheld as one step in an evolving scheme. 62 In short, in my view, neither Art. III, § 19, nor Art. XVI, § 65, can survive even minimal equal protection scrutiny.7 II 63 I also believe that Art. III, § 19, violates the First Amendment. The Court dismisses this contention by stating that this provision is a more limited restriction on political activities of public employees than we have upheld in prior cases. But none of our precedents presented a restriction on campaigning that applied even after an official had resigned from public office or to officials who did not serve in the regulating government. Moreover, the Court does not go on to address what is for me the crucial question: What justification does the State have for this restriction and how does this provision address the State's asserted interests? 64 The Court acknowledges that Art. III, § 19, restrains government employees' pursuit of political office. Such pursuit is clearly protected by the First Amendment and restrictions on it must be justified by the State's interest in ensuring the continued proper performance of current public duties. As the Court notes, similar competing considerations were considered in CSC v. Letter Carriers, 413 U.S. 548, 93 S.Ct. 2880, 37 L.Ed.2d 796 (1973), Broadrick v. Oklahoma, 413 U.S. 601, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973), and United Public Workers v. Mitchell, 330 U.S. 75, 67 S.Ct. 556, 91 L.Ed. 754 (1947). 65 In United Public Workers, the Court upheld § 9(a) of the Hatch Act, 5 U.S.C. § 7324(a)(2), which prohibits certain federal civil service employees from taking "an active part in political management or political campaigns." In Letter Car riers 6¢s the Court reaffirmed United Public Workers, and in Broadrick the Court upheld a similar state provision. In these cases, the Court determined that the restrictions were necessary to foster and protect efficient and effective government by keeping partisan politics out of the civil service. The Court recognized that "the government has an interest in regulating the conduct and 'the speech of its employees that differ[s] significantly from those it possesses in connection with regulation of the speech of the citizenry in general.' " Letter Carriers, supra, 413 U.S. at 564, 93 S.Ct. at 2889, quoting Pickering v. Board of Education, 391 U.S. 563, 568, 88 S.Ct. 1731, 1734, 20 L.Ed.2d 811 (1968). 66 At the same time, this Court has unequivocally rejected the premise that one surrenders the protection of the First Amendment by accepting the responsibilities of public employment. Elrod v. Burns, 427 U.S. 347, 96 S.Ct. 2673, 49 L.Ed.2d 547 (1976); Pickering v. Board of Education, supra. And the Court has clearly recognized that restrictions on candidacy impinge on First Amendment rights. See, e.g., Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 99 S.Ct. 983, 59 L.Ed.2d 230 (1979); Lubin v. Panish, 415 U.S. 709, 94 S.Ct. 1315, 39 L.Ed.2d 702 (1974); American Party of Texas v. White, 415 U.S. 767, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974); Bullock v. Carter, 405 U.S. 134, 142-143, 92 S.Ct. 849, 855-856, 31 L.Ed.2d 92 (1972); Williams v. Rhodes, 393 U.S. 23, 34, 89 S.Ct. 5, 12, 21 L.Ed.2d 24 (1968).8 Our precedents establish the guiding principle for applying the strictures of the First Amendment to restrictions of expressional conduct of state employees: The Court must arrive at an accommodation " 'between the interests of the [employee] . . . and the interest of the [government], as an employer.' " CSC v. Letter Carriers, supra, at 564, 93 S.Ct., at 2890, quoting Pickering v. Board of Education, supra, at 568, 93 S.Ct., at 1734. And in striking the required balance, "[t]he gain to the subordinating interest provided by the means must outweigh the incurred loss of protected rights." Elrod v. Burns, supra, 427 U.S., at 362, 96 S.Ct., at 2684 (plurality opinion). See also United Public Workers v. Mitchell, supra, 330 U.S. at 96, 67 S.Ct. at 567.9 67 In undertaking this balance, I acknowledge, of course, that the State has a vital interest in ensuring that public officeholders perform their duties properly, and that a State requires substantial flexibility to develop both direct and indirect methods of serving that interest. But if the State's interest is not substantially furthered by the challenged restrictions, then the restrictions are an unnecessary intrusion into employee rights. If the restriction is effective, but interferes with protected activity more than is reasonably necessary to further the asserted state interest, then the overintrusive aspects of the restriction lack constitutional justification. In short, to survive scrutiny under the First Amendment, a restriction on political campaigning by government employees must be narrowly tailored and substantially related to furthering the State's asserted interests. 68 It is clear to me that Art. III, § 19, is not narrowly tailored to conform to the State's asserted interests. Nor does it further those interests in a meaningful way. I have discussed briefly the broad sweep and thus the absence of narrow tailoring of § 19 in Part I, supra. Section 19 bars the candidacy of a wide class of state, federal, and foreign officeholders. The offices enumerated in § 19 include the judges of all courts, the Secretary of State, the Attorney General, the clerks of any court of record, and all persons holding any "lucrative" office under the United States, Texas, or any foreign government. Section 19 by its terms would bar, for example, a retired United States District Court Judge, appointed for life, whose District was outside of Texas, from running for the Texas State Legislature. The Texas courts have interpreted "lucrative" broadly enough to include any office that yields profit, gain, revenue, or salary, regardless of the adequacy of the compensation. See Willis v. Potts, 377 S.W.2d 622, 625-627 (Tex.1964). The state courts have also held that offices created by political bodies subordinate to the State, such as cities, are covered by § 19. See, id., at 624-625. 69 Section 19 is not merely a resign-to-run law, or a prohibition on dual officeholding. Rather, the Texas Supreme Court has construed the phrase, "during the term for which he was elected or appointed," to bar candidacy for the legislature even after an official has resigned from his current office. See n. 4, supra. As one commentator has noted, § 19 "has trapped the unwary who believed (not unreasonably) that by resigning their present office they would be eligible to run for the legislature." 1 G. Braden, The Constitution of the State of Texas: An Annotated and Comparative Analysis 135 (1977). 70 In many of its applications § 19 has absolutely no connection to Texas' interest in how Texas public officials perform their current duties. This provision applies to persons holding office under the United States or any foreign government and would thus bar a person holding federal office from resigning from that office and running for the Texas Legislature.10 Even with respect to persons who, like Baca, are currently Texas public officials, § 19 continues to operate after their resignations from current positions have taken effect and their responsibility to the Texas electorate has ceased. A provision directed only at Texas officeholders, that gave those officeholders a choice between resigning and serving out their current terms would serve all of the asserted state interests; yet Texas has inexplicably chosen this far more restrictive alternative.11 71 The same irrationality evident to me when I analyzed § 19 under the Equal Protection Clause convinces me that it is not substantially related to furthering the asserted state interests. Appellants contend that § 19 promotes attention to current duties by officeholders and prevents abuse of their current office in the attempt to further political aspirations. But § 19 prohibits the enumerated officeholders from engaging only in Texas legislative campaigns. It has absolutely no effect on an officeholder who misuses his current office in order to undertake a campaign for any other office. Even if no improper motive underlies the restriction, it is obvious that § 19 is far more likely to discourage officeholders from running for the state legislature than it is to encourage them to serve properly in their current positions. See supra, at 980-983. 72 In sum, the prohibition of § 19 furthers in no substantial way any of the asserted state interests said to support it, and is not narrowly tailored to avoid unnecessary interference with the First Amendment interests of government employees. Accordingly, in my view, this provision is invalid as an unjustified infringement on appellees' First Amendment rights.12 73 Because the Court finds neither an equal protection nor a First Amendment violation in either of these restrictions on candidacy, I respectfully dissent. 1 Section 65 covers District Clerks, County Clerks, County Judges, County Treasurers, Criminal District Attorneys, County Surveyors, Inspectors of Hides and Animals, County Commissioners, Justices of the Peace, Sheriffs, Assessors and Collectors of Taxes, District Attorneys, County Attorneys, Public Weighers, and Constables. Section 65 altered the terms of these offices. See infra, at 970. 2 Bullock v. Carter, 405 U.S. 134, 92 S.Ct. 849, 31 L.Ed.2d 92 (1972); Lubin v. Panish, 415 U.S. 709, 94 S.Ct. 1315, 39 L.Ed.2d 702 (1974). 3 A litigant has standing to challenge the constitutionality of a statute only insofar as it adversely affects his own rights. Ulster County Court v. Allen, 442 U.S. 140, 154-155, 99 S.Ct. 2213, 2223, 60 L.Ed.2d 777 (1979). "Embedded in the traditional rules governing constitutional adjudication is the principle that a person to whom a statute may constitutionally be applied will not be heard to challenge that statute on the ground that it may conceivably be applied unconstitutionally to others, in other situations not before the Court." Broadrick v. Oklahoma, 413 U.S. 601, 610, 93 S.Ct. 2908, 2914, 37 L.Ed.2d 830 (1973). Therefore, Baca may not argue that § 19 may not be applied to restrict a Justice of the Peace's candidacy for the legislature because the State's interests in restricting candidacy by a different class of officeholders are insufficient to survive constitutional scrutiny. See Storer v. Brown, 415 U.S. 724, 737, 94 S.Ct. 1274, 1282, 39 L.Ed.2d 714 (1974). Cf. Cabell v. Chavez-Salido, 454 U.S. 432, 442, 102 S.Ct. 735, 741, 70 L.Ed.2d 677 (1982). 4 In the case of local elected officials whose terms of office typically end in nonelection years, the "waiting period" of § 19 is even shorter. 5 The State's particular interest in maintaining the integrity of the judicial system could support § 19, even if such a restriction could not survive constitutional scrutiny with regard to any other officeholder. See n. 3, supra. 6 Baca may not utilize the "overbreadth" doctrine to challenge § 19. Baca may not challenge the provision's application to him on the grounds that the provision might be unconstitutional as applied to a class of officeholders not before the Court. Broadrick v. Oklahoma, 413 U.S., at 612-616, 93 S.Ct., at 2915-2916. The First Amendment will not suffer if the constitutionality of § 19 is litigated on a case-by-case basis. 1 The fact that appellees hold state office is sufficient to justify a restriction on their ability to run for other office that is not imposed on the public generally. 2 The Federal Constitution does, of course, impose significant constraints on a state government's employment practices. For example, the First Amendment limits the State's power to discharge employees who make controversial speeches. Pickering v. Board of Education, 391 U.S. 563, 88 S.Ct. 1731, 20 L.Ed.2d 811. The Due Process Clause affords procedural safeguards to tenured employees. Board of Regents v. Roth, 408 U.S. 564, 92 S.Ct. 2701, 33 L.Ed.2d 548. The Equal Protection Clause prohibits the State from classifying applicants for employment in an arbitrary manner. Sugarman v. Dougall, 413 U.S. 634, 93 S.Ct. 2842, 37 L.Ed.2d 853. I find no comparable federal interest, however, in this case. 3 In Vance v. Bradley, 440 U.S. 93, 99 S.Ct. 939, 59 L.Ed.2d 171, the Court held that a statutory classification that treated employees of the Foreign Service differently from employees of the Civil Service did not violate the equal protection component of the Due Process Clause of the Fifth Amendment. In my view, such a classification—without more—could not violate equal protection requirements. 4 In defining the interests in equality protected by the Equal Protection Clause, one cannot ignore the State's legitimate interest in structuring its own form of government. The Equal Protection Clause certainly was not intended to require the States to justify every decision concerning the terms and conditions of state employment according to some federal standard. 5 See Westen, The Empty Idea of Equality, 95 Harv.L.Rev. 537 (1982). Professor Westen's article is valuable because it illustrates the distinction between concern with the substantive import of a state restriction and concern with any disparate impact that it may produce. In recognizing that distinction, however, it is important not to lose sight of the fact that the Equal Protection Clause has independent significance in protecting the federal interest in requiring States to govern impartially. 6 The plurality frames the test that should ordinarily be applied in this way: "Classifications are set aside only if they are based solely on reasons totally unrelated to the pursuit of the State's goals and only if no grounds can be conceived to justify them." Ante, at 963. 1 I note that a majority of the Court today rejects the plurality's mode of equal protection analysis. See ante, at 976. (STEVENS, J., concurring in part and in judgment). 2 It is worth noting, however, that the plurality's analysis of the level of scrutiny to be applied to these restrictions gives too little consideration to the impact of our prior cases. Although we have never defined candidacy as a fundamental right, we have clearly recognized that restrictions on candidacy impinge on First Amendment rights of candidates and voters. See, e.g., Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 184, 99 S.Ct. 983, 990, 59 L.Ed.2d 230 (1979); Lubin v. Panish, 415 U.S. 709, 716, 94 S.Ct. 1315, 1320, 39 L.Ed.2d 702 (1974); American Party of Texas v. White, 415 U.S. 767, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974); Bullock v. Carter, 405 U.S. 134, 142-143, 92 S.Ct. 849, 855-856, 31 L.Ed.2d 92 (1972); Williams v. Rhodes, 393 U.S. 23, 31, 89 S.Ct. 5, 10, 21 L.Ed.2d 24 (1968). With this consideration in mind, we have applied strict scrutiny in reviewing most restrictions on ballot access; thus we have required the State to justify any discrimination with respect to candidacy with a showing that the differential treatment is "necessary to further compelling state interests." American Party of Texas v. White, supra, 440 U.S. at 780, 94 S.Ct. at 1305. See also Bullock v. Carter, supra, 405 U.S. at 144, 92 S.Ct. at 856. The plurality dismisses our prior cases as dealing with only two kinds of ballot access restrictions—classifications based on wealth and classifications imposing burdens on new or small political parties or independent candidates. Ante, at 964-965. But strict scrutiny was required in those cases because of their impact on the First Amendment rights of candidates and voters, see Storer v. Brown, 415 U.S. 724, 729, 94 S.Ct. 1274, 1278, 39 L.Ed.2d 714 (1974), not because the class of candidates or voters that was burdened was somehow suspect. Compare Lubin v. Panish, 415 U.S., at 717-718, 94 S.Ct., at 1320-1321, with id., at 719, 94 S.Ct., at 1321 (Douglas, J., concurring) (strict scrutiny demanded because classification based on wealth). The plurality offers no explanation as to why the restrictions at issue here, which completely bar some candidates from running and require other candidates to give up their present employment, are less "substantial" in their impact on candidates and their supporters than, for example, the $700 filing fee at issue in Lubin. In my view, some greater deference may be due the State because these restrictions affect only public employees, see Part II, infra, but this does not suggest that, in subjecting these classifications to equal protection scrutiny, we should completely disregard the vital interests of the candidates and the citizens who they represent in a political campaign. 3 The plurality's sudden focus on the fairness of the restriction to the individual as opposed to the class, is as episodic as it is novel. For in writing for the Court in Weinberger v. Salfi, 422 U.S. 749, 781, 95 S.Ct. 2457, 2474, 45 L.Ed.2d 522 (1975), Justice REHNQUIST refused to hold that an otherwise valid legislative classification should be invalidated on the basis of the characteristics of the individual plaintiff. 4 Indeed, it may be that Art. III, § 19, was intended to do no more than prohibit dual officeholding. If it had been so construed, there would be no equal protection problem for there are blanket prohibitions in Texas against holding two elected offices at the same time. See Art. II, § 1; Art. XVI, § 40. In Lee v. Daniels, 377 S.W.2d 618 (1964), the Texas Supreme Court construed the language in § 19, "during the term for which he was elected or appointed," to mean that even after an otherwise qualified candidate for the legislature had resigned his current position, he could not hold legislative office. The dissent in Lee argued that § 19 was simply a prohibition on dual officeholding and the phrase, "during the term for which he is elected or appointed," simply "negates any basis for the contention that a person" who once held one of the offices covered by the section was still ineligible for the legislature after the completion of his term. Id., at 621 (Steakley, J., dissenting). The Texas Supreme Court was unaided by any legislative history on this provision. We are of course bound by the state court's construction of this state provision, but I point out its ambiguity to highlight the dubious nature of the plurality's hypothesis that Art. III, § 19, marks one step in what will become more complete regulation of a perceived evil. 5 For example, in Lee v. Daniels, supra, a County Commissioner resigned on February 1, 1964, and he sought thereafter to run for the legislature. However, his term did not expire until December 31, 1964; the legislative term commenced in November 1964, and the court therefore held that his name could not be placed on the legislative ballot. In contrast, in Chapa v. Whittle, 536 S.W.2d 681 (Tex.Civ.App.1976), the Director of a Social Culture Intervention Program began campaigning in February 1966. He resigned from his current office in May of that year. Because the Director had no set term, the complainant could not show that the Director's term overlapped the legislative term, beginning in November 1966, and the court therefore allowed the Director to run for the legislature. 6 The assortment of offices restricted by Art. XVI, § 65, are: District Clerks; County Clerks; various County Judges; County Treasurers; Criminal District Attorneys; County Surveyors; Inspectors of Hides and Animals; County Commissioners; Justices of the Peace; Sheriffs; Assessors and Collectors of Taxes; District Attorneys; County Attorneys; Public Weighers; and Constables. 7 Justice STEVENS argues in his concurrence that there is no federal interest in requiring the State to treat different elective state offices in a fair and equitable manner. Ante, at 974. I agree with Justice STEVENS that the State may define many of the "benefits and burdens of different elective state offices" in a dissimilar manner without offering an explanation for the classifications that a federal judge will find to be rational, so long as such classifications do not mask any racial or otherwise impermissible discrimination. Ibid. But where the differential treatment concerns a restriction on the right to seek public office—a right protected by the First Amendment—that Amendment supplies the federal interest in equality that may be lacking where the State is simply determining salary, hours, or working conditions of its own employees. 8 Such restrictions affect not only the expressional and associational rights of candidates, but those of voters as well. Voters generally assert their views on public issues by casting their ballots for the candidate of their choice. "By limiting the choices available to voters, the State impairs the voters' ability to express their political preferences." Illinois Elections Bd. v. Socialist Workers Party, 440 U.S., at 184, 99 S.Ct., at 990. The effect on voters from restrictions on candidacy is illustrated in this case by the fact that 20 of the appellees are voters who allege that they would vote for the officeholder-appellees were they to become candidates. See ante, at 961. 9 "[T]his Court must balance the extent of the guarantees of freedom against a congressional enactment to protect a democratic society against the supposed evil of political partisanship by classified employees of government." 330 U.S., at 96, 67 S.Ct., at 567. 10 The Court, citing Broadrick v. Oklahoma, 413 U.S. 601, 612-616, 93 S.Ct. 2908, 2915-2918, 37 L.Ed.2d 830 (1973), states that Baca may not utilize the "overbreadth" doctrine to "challenge the provision's application to him on the grounds that the provision might be unconstitutional as applied to a class of officeholders not before the Court." Ante, at 972, n.6. But all that Broadrick holds is "that the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute's plainly legitimate sweep." 413 U.S., at 615, 93 S.Ct. at 2917. In my view, the overbreadth of Art. III, § 19, is clearly substantial, particularly when its breadth is viewed in relationship to its relatively tenuous "legitimate sweep." 11 The less intrusive means for accomplishing the State's goals are obvious. A statute requiring persons to take a leave of absence would also preclude an officeholder from misusing his current office during a campaign. See Bolin v. Minnesota, 313 N.W.2d 381, 384 (Minn.1981). Appellants assert an interest in ensuring that defeated candidates do not return to office and administer their old position vindictively or halfheartedly. Brief for Appellants 9. But this would be satisfied by a resign-to-run statute—giving candidates a choice between running for a new office or holding their present position. Appellants suggest that even before an actual announcement of candidacy a potential candidate may begin to abuse his current office. Id., at 13. They thus appear to suggest that a resign-to-run provision is not necessarily adequate because it allows the candidate to stay in his current position until his formal announcement of candidacy. Even if this is a sufficient state concern to justify further intrusion on the interests of potential candidates, it would be fully served by a statute that simply required all potential candidates to resign some period of time before they formally announced their candidacy for a new office. Unlike the plurality, I refuse to assume that the State has an interest in having officeholders who no longer desire to hold their office serve out their terms. See ante, at 968-969. Indeed, appellants have not asserted this interest in this Court or in the courts below. 12 Article XVI, § 65, also affects appellees' right to run for political office; it has a lesser impact on that right for it merely requires that candidates resign before embarking on political campaigns. Moreover, it bears a more substantial relationship to the State's asserted purposes because it bans political campaigns for all offices. That provision does not in my view violate the First Amendment. Because it applies only to an inexplicable group of elected officials, it does, however, violate the Equal Protection Clause.

12

457 U.S. 830 102 S.Ct. 2764 73 L.Ed.2d 418 Sheila RENDELL-BAKER, et al., Petitionersv.Sandra KOHN et al. No. 80-2102. Argued April 19, 1982. Decided June 25, 1982. Syllabus Respondent school is a privately operated school for maladjusted high school students. In recent years, nearly all of the students have been referred to the school by city school committees under a Massachusetts statute or by a state agency. When the students are referred to the school by the city committees, these cities pay for the students' education. The school also receives funds from a number of state and federal agencies. Public funds have recently accounted for at least 90% of the school's operating budget. To be eligible for tuition funding under the state statute, the school must comply with a variety of state regulations, but these regulations impose few specific personnel requirements. Similarly, the school's contracts with the State and the city committees generally do not cover personnel policies. Petitioners, a former vocational counselor and teachers at the school, brought separate actions in Federal District Court under 42 U.S.C. § 1983, claiming that they had been discharged by the school in violation of their First, Fifth, and Fourteenth Amendment rights. The court dismissed the counselor's action but denied a motion to dismiss the teachers' action, reaching conflicting conclusions as to whether the school had acted under color of state law so as to be subject to liability under § 1983. On appeal the cases were consolidated, and the Court of Appeals held that it was error to conclude that the school acted under color of state law, since, although regulated by the State, it was not dominated by the State, especially with respect to decisions involving discharge of personnel. Held : Respondent school did not act under color of state law when it discharged petitioner employees, and hence petitioners have not stated a claim for relief under § 1983. Pp. 837-843. (a) The ultimate issue in determining whether a person is subject to suit under § 1983 is the same question posed in cases arising under the Fourteenth Amendment: Is the alleged infringement of federal rights fairly attributable to the State? Pp. 837-838 (b) The school's receipt of public funds does not make the discharge decisions acts of the State. Cf. Blum v. Yaretsky, 457 U.S. 991, 102 S.Ct. 2777, 73 L.Ed.2d 534. The school is not fundamentally different from many private corporations whose business depends primarily on contracts with the government, and whose acts do not become acts of the government by reason of their significant or even total engagement in performing public contracts. The decision to discharge petitioners was not compelled or even influenced by any state regulation, and the fact that the school performs a public function in educating maladjusted high school students does not make its acts state action. Moreover, since the school's fiscal relationship with the State is not any different from that of many contractors performing services for the government, there is no "symbiotic relationship" between the school and the State. Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45, distinguished. Pp 839-843. 641 F.2d 14 (C.A.1 1981) affirmed. Zachary R. Karol, Boston, Mass., for petitioners. Matthew H. Feinberg, Boston, Mass., for respondents. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to decide whether a private school, whose income is derived primarily from public sources and which is regulated by public authorities, acted under color of state law when it discharged certain employees. 2 * A. 3 Respondent Kohn is the director of the New Perspectives School, a nonprofit institution located on privately owned property in Brookline, Massachusetts. The school was founded as a private institution and is operated by a board of directors, none of whom are public officials or are chosen by public officials. The school specializes in dealing with students who have experienced difficulty completing public high schools; many have drug, alcohol, or behavioral problems, or other special needs. In recent years, nearly all of the students at the school have been referred to it by the Brookline or Boston School Committees, or by the Drug Rehabilitation Division of the Massachusetts Department of Mental Health. The school issues high school diplomas certified by the Brookline School Committee. 4 When students are referred to the school by Brookline or Boston under Chapter 766 of the Massachusetts Acts of 1972, the School Committees in those cities pay for the students' education.1 The school also receives funds from a number of other state and federal agencies. In recent years, public funds have accounted for at least 90%, and in one year 99%, of respondent school's operating budget. There were approximately 50 students at the school in those years and none paid tuition.2 5 To be eligible for tuition funding under Chapter 766, the school must comply with a variety of regulations, many of which are common to all schools. The State has issued detailed regulations concerning matters ranging from recordkeeping to student-teacher ratios. Concerning personnel policies, the Chapter 766 regulations require the school to maintain written job descriptions and written statements describing personnel standards and procedures, but they impose few specific requirements. 6 The school is also regulated by Boston and Brookline as a result of its Chapter 766 funding. By its contract with the Boston School Committee, which refers to the school as a "contractor," the school must agree to carry out the individualized plan developed for each student referred to the school by the Committee. See n. 1, supra. The contract specifies that school employees are not city employees.3 7 The school also has a contract with the State Drug Rehabilitation Division. Like the contract with the Boston School Committee, that agreement refers to the school as a "contractor." It provides for reimbursement for services provided for students referred to the school by the Drug Rehabilitation Division, and includes requirements concerning the services to be provided. Except for general requirements, such as an equal employment opportunity requirement, the agreement does not cover personnel policies. 8 While five of the six petitioners were teachers at the school, petitioner Rendell-Baker was a vocational counselor hired under a grant from the federal Law Enforcement Assistance Administration, whose funds are distributed in Massachusetts through the State Committee on Criminal Justice. As a condition of the grant, the Committee on Criminal Justice must approve the school's initial hiring decisions. The purpose of this requirement is to insure that the school hires vocational counselors who meet the qualifications described in the school's grant proposal to the Committee; the Committee does not interview applicants for counselor positions. B 9 Rendell-Baker was discharged by the school in January 1977, and the five other petitioners were discharged in June 1978. Rendell-Baker's discharge resulted from a dispute over the role of a student-staff council in making hiring decisions. A dispute arose when some students presented a petition to the school's board of directors in December 1976, seeking greater responsibilities for the student-staff council. Director Kohn opposed the proposal, but Rendell-Baker supported it and so advised the board. On December 13, Kohn notified the State Committee on Criminal Justice, which funded Rendell-Baker's position, that she intended to dismiss Rendell-Baker and employ someone else. Kohn notified Rendell-Baker of her dismissal in January 1977. 10 Rendell-Baker then advised the board of directors that she had been discharged without due process because she exercised her First Amendment rights. She demanded reinstatement or a hearing. The school agreed to apply a new policy, calling for appointment of a grievance committee, to consider her claims. Rendell-Baker also complained to the State Committee on Criminal Justice, which asked the school to provide a written explanation for her discharge. After the school complied, the Committee responded that it was satisfied with the explanation, but notified the school that it would not pay any backpay or other damages award Rendell-Baker might obtain from it as a result of her discharge. The Committee told Rendell-Baker that it had no authority to order a hearing, although it would refuse to approve the hiring of another counselor if the school disregarded its agreement to apply its new grievance procedure in her case. At this point Rendell-Baker objected to the composition of the grievance committee, and its proceedings apparently never went forward. Rendell-Baker filed this suit in July 1977 under 42 U.S.C. § 1983, alleging that she had been discharged in violation of her rights under the First, Fifth, and Fourteenth Amendments. 11 In the spring of 1978, students and staff voiced objections to Kohn's policies. The five petitioners other than Rendell-Baker, who were all teachers at the school, wrote a letter to the board of directors urging Kohn's dismissal. When the board affirmed its confidence in Kohn, students from the school picketed the home of the president of the board. The students were threatened with suspension; a local newspaper then ran a story about the controversy at the school. In response to the story, the five petitioners wrote a letter to the editor in which they stated that they thought the prohibition of picketing was unconstitutional. On the day the letter to the editor appeared, the five teachers told the president of the board that they were forming a union. Kohn discharged the teachers the next day. They brought suit against the school and its directors in December 1978. Like Rendell-Baker, they sought relief under § 1983, alleging that their rights under the First, Fifth, and Fourteenth Amendments had been violated. C 12 On April 16, 1980, the District Court for the District of Massachusetts, 488 F.Supp. 764, granted the defendant's motion for summary judgment in the suit brought by Rendell-Baker. A claim may be brought under § 1983 only if the defendant acted "under color" of state law.4 The District Court took as its standard " 'whether there is a sufficiently close nexus between the State and the challenged action of the regulated entity so that the action of the latter may be fairly treated as that of the State itself,' " quoting Jackson v. Metropolitan Edison Co., 419 U.S. 345, 351, 95 S.Ct. 449, 453, 42 L.Ed.2d 477 (1974). Noting that, although the State regulated the school in many ways, it imposed few conditions on the school's personnel policies, the District Court concluded that the nexus between the school and the State was not sufficiently close so that the action of the school in discharging Rendell-Baker could be considered action of the Commonwealth of Massachusetts. 13 Nine days earlier, on April 7, 1980, a different judge of the District Court for the District of Massachusetts had reached a contrary conclusion on the same question in the case brought by the other five petitioners. His opinion stressed the school's dependency on public funding and its regulation by numerous public entities. It also noted that although education was not a uniquely public function, it is primarily a public function, and that Brookline did not maintain a school to serve maladjusted adolescents with drug, alcohol, or emotional problems. The District Court, following the guidelines of Burton v. Wilmington Parking Authority, 365 U.S. 715, 722, 81 S.Ct. 856, 860, 6 L.Ed.2d 45 (1961), concluded that the school performed a "public function," as described in Jackson, supra, at 352, 95 S.Ct., at 454. Accordingly, it held that the defendants acted under color of state law and denied the motion to dismiss. However, on June 13, 1980, noting that there was substantial ground for disagreement on that holding, the District Court certified its order as immediately appealable pursuant to 28 U.S.C. § 1292(b). D 14 The Court of Appeals for the First Circuit consolidated the two actions. It noted that the school's funding, regulation, and function show that it has a close relationship with the State. However, it stressed that the school is managed by a private board and that the State has relatively little involvement in personnel matters. It concluded that the school, although regulated by the State, was not dominated by the State, especially with respect to decisions involving the discharge of personnel. The Court of Appeals then concluded that the District Court which certified the question in the action brought by the five teachers had erred in concluding that the defendants acted under color of state law. 15 The Court of Appeals separately considered Rendell-Baker's claim that she was discharged under color of state law since her position was funded directly by the Committee on Criminal Justice. The court rejected her claim, noting that the Committee had the power to insure that those hired had the qualifications described in the grant proposal, but that it did not have any other control over the school's personnel decisions. It therefore affirmed the District Court's dismissal of her action. 641 F.2d 14 (1981). 16 We granted certiorari, 454 U.S. 891, 102 S.Ct. 385, 70 L.Ed.2d 205 (1981), and we affirm. II A. 17 Petitioners do not claim that their discharges were discriminatory in violation of Title VII of the Civil Rights Act of 1964. Nor do they claim that their discharges were unfair labor practices in violation of the National Labor Relations Act. Rather, they allege that respondents violated 42 U.S.C. § 1983, see n. 4, supra, by discharging them because of their exercise of their First Amendment right of free speech and without the process due them under the Fourteenth Amendment. Although Title VII and the National Labor Relations Act govern action by private parties making personnel decisions, it is fundamental that the First Amendment prohibits governmental infringement on the right of free speech. Similarly, the Fourteenth Amendment, which prohibits the states from denying federal constitutional rights and which guarantees due process, applies to acts of the states, not to acts of private persons or entities. Civil Rights Cases, 109 U.S. 3, 11, 3 S.Ct. 18, 21, 27 L.Ed. 835 (1883); Shelley v. Kraemer, 334 U.S. 1, 13, 68 S.Ct. 836, 842, 92 L.Ed. 1161 (1948).5 And § 1983, which was enacted pursuant to the authority of Congress to enforce the Fourteenth Amendment, prohibits interference with federal rights under color of state law. 18 In United States v. Price, 383 U.S. 787, 794, n. 7, 86 S.Ct. 1152, 1157, n. 7, 16 L.Ed.2d 267 (1966), the Court stated: 19 "In cases under § 1983, 'under color' of law has consistently been treated as the same thing as the 'state action' required under the Fourteenth Amendment." 20 See also, United States v. Classic, 313 U.S. 299, 326, 61 S.Ct. 1031, 1043, 85 L.Ed. 1368 (1941). The ultimate issue in determining whether a person is subject to suit under § 1983 is the same question posed in cases arising under the Fourteenth Amendment: is the alleged infringement of federal rights "fairly attributable to the State?" Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S.Ct. 2744, 2754, 73 L.Ed.2d 482. The core issue presented in this case is not whether petitioners were discharged because of their speech or without adequate procedural protections, but whether the school's action in discharging them can fairly be seen as state action.6 If the action of the respondent school is not state action, our inquiry ends. B 21 In Blum v. Yaretsky, 457 U.S. 991, 102 S.Ct. 2777, 73 L.Ed.2d 534, the Court analyzed the state action requirement of the Fourteenth Amendment. The Court considered whether certain nursing homes were state actors for the purpose of determining whether decisions regarding transfers of patients could be fairly attributed to the State, and hence be subjected to Fourteenth Amendment due process requirements. The challenged transfers primarily involved decisions, made by physicians and nursing home administrators, to move patients from "skilled nursing facilities" to less expensive "health related facilities." 457 U.S., at 1005, 102 S.Ct., at 2786. Like the New Perspectives School, the nursing homes were privately owned and operated. Id., at 1003, 102 S.Ct., at 2785. Relying on Flagg Brothers, Inc. v. Brooks, 436 U.S. 149, 98 S.Ct. 1729, 56 L.Ed.2d 185 (1978); Jackson v. Metropolitan Edison Co., 419 U.S. 345, 95 S.Ct. 449, 42 L.Ed.2d 477 (1974); Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972); and Adickes v. S. H. Kress Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970), the Court held that, "a State normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement, either overt or covert, that the choice must in law be deemed to be that of the State." 457 U.S. at 1004, 102 S.Ct., at 2786. In determining that the transfer decisions were not actions of the State, the Court considered each of the factors alleged by petitioners here to make the discharge decisions of the New Perspectives School fairly attributable to the State. 22 First, the nursing homes, like the school, depended on the State for funds; the State subsidized the operating and capital costs of the nursing homes, and paid the medical expenses of more than 90% of the patients. 457 U.S., at 1011, 102 S.Ct., at 2789. Here the Court of Appeals concluded that the fact that virtually all of the school's income was derived from government funding was the strongest factor to support a claim of state action. 641 F.2d, at 24. But in Blum v. Yaretsky, we held that the similar dependence of the nursing homes did not make the acts of the physicians and nursing home administrators acts of the State, and we conclude that the school's receipt of public funds does not make the discharge decisions acts of the State. 23 The school, like the nursing homes, is not fundamentally different from many private corporations whose business depends primarily on contracts to build roads, bridges, dams, ships, or submarines for the government. Acts of such private contractors do not become acts of the government by reason of their significant or even total engagement in performing public contracts. 24 The school is also analogous to the public defender found not to be a state actor in Polk County v. Dodson, 454 U.S. 312, 102 S.Ct. 445, 70 L.Ed.2d 509 (1981). There we concluded that, although the State paid the public defender, her relationship with her client was "identical to that existing between any other lawyer and client." Id., at 318, 102 S.Ct., at 449. Here the relationship between the school and its teachers and counselors is not changed because the State pays the tuition of the students. 25 A second factor considered in Blum v. Yaretsky was the extensive regulation of the nursing homes by the State. There the State was indirectly involved in the transfer decisions challenged in that case because a primary goal of the State in regulating nursing homes was to keep costs down by transferring patients from intensive treatment centers to less expensive facilities when possible. Both state and federal regulations encouraged the nursing homes to transfer patients to less expensive facilities when appropriate. 457 U.S., at 1007-1008, 1009-1110, 102 S.Ct., at 2787-2789. The nursing homes were extensively regulated in many other ways as well. The Court relied on Jackson, where we held that state regulation, even if "extensive and detailed," 419 U.S., at 350, 95 S.Ct., at 453, did not make a utility's actions state action. 26 Here the decisions to discharge the petitioners were not compelled or even influenced by any state regulation. Indeed, in contrast to the extensive regulation of the school generally, the various regulators showed relatively little interest in the school's personnel matters. The most intrusive personnel regulation promulgated by the various government agencies was the requirement that the Committee on Criminal Justice had the power to approve persons hired as vocational counselors. Such a regulation is not sufficient to make a decision to discharge, made by private management, state action. See n. 6, supra. 27 The third factor asserted to show that the school is a state actor is that it performs a "public function." However, our holdings have made clear that the relevant question is not simply whether a private group is serving a "public function." We have held that the question is whether the function performed has been "traditionally the exclusive prerogative of the State." Jackson, supra, at 353, 95 S.Ct., at 454; quoted in Blum v. Yaretsky, 457 U.S., at 1011, 102 S.Ct., at 2789 (emphasis added). There can be no doubt that the education of maladjusted high school students is a public function, but that is only the beginning of the inquiry. Chapter 766 of the Massachusetts Acts of 1972 demonstrates that the State intends to provide services for such students at public expense. That legislative policy choice in no way makes these services the exclusive province of the State. Indeed, the Court of Appeals noted that until recently the State had not undertaken to provide education for students who could not be served by traditional public schools. 641 F.2d, at 26. That a private entity performs a function which serves the public does not make its acts state action.7 28 Fourth, petitioners argue that there is a "symbiotic relationship" between the school and the State similar to the relationship involved in Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961). Such a claim is rejected in Blum v. Yaretsky, and we reject it here. In Burton, the Court held that the refusal of a restaurant located in a public parking garage to serve Negroes constituted state action. The Court stressed that the restaurant was located on public property and that the rent from the restaurant contributed to the support of the garage. 365 U.S., at 723, 81 S.Ct., at 860. In response to the argument that the restaurant's profits, and hence the State's financial position, would suffer if it did not discriminate, the Court concluded that this showed that the State profited from the restaurant's discriminatory conduct. The Court viewed this as support for the conclusion that the State should be charged with the discriminatory actions. Here the school's fiscal relationship with the State is not different from that of many contractors performing services for the government. No symbiotic relationship such as existed in Burton exists here. C 29 We hold that petitioners have not stated a claim for relief under 42 U.S.C. § 1983; accordingly, the judgment of the Court of Appeals for the First Circuit is 30 Affirmed. 31 Justice WHITE, concurring in the judgments. 32 The issue in Blum v. Yaretsky, 457 U.S. 991, 102 S.Ct. 2777, 73 L.Ed.2d 534, is whether a private nursing home's decision to discharge or transfer a Medicaid patient satisfies the state-action requirement of the Fourteenth Amendment. To satisfy this requirement, respondents must show that the transfer or discharge is made on the basis of some rule of decision for which the State is responsible. Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S.Ct. 2744, 2754, 73 L.Ed.2d 482. It is not enough to show that the State takes certain actions in response to this private decision. The rule of decision implicated in the actions at issue here appears to be nothing more than a medical judgment. This is the clear import of the majority's conclusion that the "decisions ultimately turn on medical judgments made by private parties according to professional standards that are not established by the State," 457 U.S., at 1008, 102 S.Ct., at 2788, with which I agree. 33 Similarly, the allegations of the petitioners in Rendell-Baker v. Kohn, No. 80-2102 [457 U.S. 830, 102 S.Ct. 2764, 73 L.Ed.2d 418], fail to satisfy the state-action requirement. In this case, the question of state action focuses on an employment decision made by a private school that receives most of its funding from public sources and is subject to state regulation in certain respects. For me, the critical factor is the absence of any allegation that the employment decision was itself based upon some rule of conduct or policy put forth by the State. As the majority states, "in contrast to the extensive regulation of the school generally, the various regulators showed relatively little interest in the school's personnel matters." Ante, at 841. The employment decision remains, therefore, a private decision not fairly attributable to the State. 34 Accordingly, I concur in the judgments. 35 Justice MARSHALL, with whom Justice BRENNAN joins, dissenting. 36 Petitioners in these consolidated cases, former teachers and a counselor at the New Perspectives School in Brookline, Mass., were discharged by the school's administrators when they criticized certain school policies. They commenced actions under 42 U.S.C. § 1983, claiming that they had been discharged in violation of the First, Fifth, and Fourteenth Amendments. The Court today holds that their suits must be dismissed because the school did not act "under color" of state law. According to the majority, the decision of the school to discharge petitioners cannot fairly be regarded as a decision of the Commonwealth of Massachusetts. 37 In my view, this holding simply cannot be justified. The State has delegated to the New Perspectives School its statutory duty to educate children with special needs. The school receives almost all of its funds from the State, and is heavily regulated. This nexus between the school and the State is so substantial that the school's action must be considered state action. I therefore dissent. 38 * The critical facts of this case deserve restatement. Chapter 766 of the Massachusetts Acts of 1972, Mass.Gen.Laws Ann., ch. 71B, §§ 1-14 (West 1981), provides that all students with special needs are entitled to a suitable publicly funded education under the supervision of the state and local governments. The school committee of every city, town, or school district in Massachusetts must identify all children who, because of physical or emotional disability, have special educational needs. It must prepare an individualized educational program tailored to meet those needs, and arrange for the implementation of that program. The school committee may offer the programs through existing public schools, or it may contract with private schools to implement the programs. If the school committee decides to place a child in a private school, it must bear all the expenses associated with the placement; parents need not pay the tuition. 39 If a school committee decides to place a child in a private school, it must closely monitor the child's educational progress. Every three months it must determine whether the child can be transferred to a less restrictive environment, such as a public school. 603 Code Mass.Regs. § 28, &Par; 502.4(i), 804.2 (1979). In general, special education programs must be provided in the least restrictive environment possible. Id., at ¶ 322.2. If the parents object to the placement of their child in private school, the child may remain in public school unless he is disruptive or dangerous. Parents may also place their child in a private school of their own choice. If they do so, however, they must pay the tuition. 40 As of 1978, all 50 students enrolled at the New Perspectives School were children with alcohol, drug, behavioral, or other special problems. They had been placed there pursuant to Chapter 766 by the town of Brookline, the city of Boston, or the Drug Rehabilitation Division of the Massachusetts Department of Mental Health. None of the students pays tuition. When they graduate, they receive a diploma certified by the Town of Brookline School Committee. 41 The New Perspectives School is funded almost entirely by governmental agencies. In fiscal year 1975-1976, public funds accounted for 91% of the school's budget. In fiscal year 1976-1977, public funds accounted for 99% of the budget. The school has received money from the town of Brookline, the Massachusetts Department of Mental Health, the Massachusetts Department of Youth Services, the Massachusetts Division of Family and Children's Services, the Massachusetts Office for Children, and the federal Law Enforcement Assistance Administration. See 641 F.2d 14, 17 (C.A.1 1981). 42 In order to remain eligible for placements and funding under Chapter 766, the New Perspectives School must comply with a variety of regulations. The Massachusetts Department of Education has promulgated "Guidelines for Approval of Day Educational Component in Private Schools under Chapter 766." These guidelines cover almost every aspect of a private school's operations, including financial recordkeeping, student discipline, medical examinations for students, parent involvement, health care, subjects of instruction, teacher-student ratio, student records, confidentiality of records, transportation, insurance, nutrition, food preparation, toileting procedures, physical facilities, and classroom equipment. The guidelines also address personnel policies. They set forth minimum standards for staff training, use of volunteers, teacher qualifications, and teacher evaluations. They further require that the school maintain written job descriptions and a written policy on criteria and procedures for hiring and dismissal, and procedures for handling staff complaints. And they require that the school provide vacations and other benefits. 43 The New Perspectives School is subject to additional regulation under contracts with each of the governmental units that refers students. A contract with the Massachusetts Department of Mental Health, Drug Rehabilitation Division, requires the school to provide counseling, educational, and vocational services for drug abusers. Under a contract with the city of Boston, the school must carry out the educational plan devised by the Boston School Committee for each Boston student placed with the school. The school must submit periodic reports to the city and is subject to inspection at any time during normal business hours. Finally, the school is bound by regulations contained in contracts with the Massachusetts Department of Youth Services and the Brookline School Committee. See 641 F.2d, at 18. II 44 The decisions of this Court clearly establish that where there is a symbiotic relationship between the State and a privately owned enterprise, so that the State and a privately owned enterprise are participants in a joint venture, the actions of the private enterprise may be attributable to the State. "Conduct that is formally 'private' may become so entwined with governmental policies or so impregnated with a governmental character" that it can be regarded as governmental action. Evans v. Newton, 382 U.S. 296, 299, 86 S.Ct. 486, 488, 15 L.Ed.2d 373 (1966). See Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961); see also Jackson v. Metropolitan Edison Co., 419 U.S. 345, 351, 95 S.Ct. 449, 453, 42 L.Ed.2d 477 (1974); Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 175, 92 S.Ct. 1965, 1972, 32 L.Ed.2d 627 (1972). The question whether such a relationship exists "can be determined only in the framework of the peculiar facts or circumstances present." Burton, supra, at 726, 81 S.Ct., at 862. Here, an examination of the facts and circumstances leads inexorably to the conclusion that the actions of the New Perspectives School should be attributed to the State; it is difficult to imagine a closer relationship between a government and a private enterprise. 45 The New Perspectives School receives virtually all of its funds from state sources. This financial dependence on the State is an important indicium of governmental involvement. The school's very survival depends on the State. If the State chooses, it may exercise complete control over the school's operations simply by threatening to withdraw financial support if the school takes action that it considers objectionable. 46 The school is heavily regulated and closely supervised by the State. This fact provides further support for the conclusion that its actions should be attributed to the State. The school's freedom of decisionmaking is substantially circumscribed by the Massachusetts Department of Education's guidelines and the various contracts with state agencies. For example, the school is required to develop and comply with written rules for hiring and dismissal of personnel. Almost every decision the school makes is substantially affected in some way by the State's regulations.1 47 The fact that the school is providing a substitute for public education is also an important indicium of state action. The provision of education is one of the most important tasks performed by government: it ranks at the very apex of the function of a State. Ambach v. Norwick, 441 U.S. 68, 77, 99 S.Ct. 1589, 1594, 60 L.Ed.2d 49 (1979).2 Of course, as the majority emphasizes, ante, at 2772, performance of a public function is by itself sufficient to justify treating a private entity as a state actor only where the function has been "traditionally the exclusive prerogative of the State." Jackson, supra, at 353, 95 S.Ct., at 454. See Marsh v. Alabama, 326 U.S. 501, 66 S.Ct. 276, 90 L.Ed. 265 (1946); Smith v. Allwright, 321 U.S. 649, 64 S.Ct. 757, 88 L.Ed. 987 (1944). But the fact that a private entity is performing a vital public function, when coupled with other factors demonstrating a close connection with the State, may justify a finding of state action. Cf. Evans v. Newton, supra. 48 The school's provision of a substitute for public education deserves particular emphasis because of the role of Chapter 766. Under this statute, the State is required to provide a free education to all children, including those with special needs. Clearly, if the State had decided to provide the service itself, its conduct would be measured against constitutional standards. The State should not be permitted to avoid constitutional requirements simply by delegating its statutory duty to a private entity.3 In my view, such a delegation does not convert the performance of the duty from public to private action when the duty is specific and the private institution's decisionmaking authority is significantly curtailed. 49 When an entity is not only heavily regulated and funded by the State, but also provides a service that the State is required to provide, there is a very close nexus with the State. Under these circumstances, it is entirely appropriate to treat the entity as an arm of the State. Cf. Smith v. Allwright, supra; Terry v. Adams, 345 U.S. 461, 469, 73 S.Ct. 809, 813, 97 L.Ed. 1152 (1953) (opinion of Black, J.). Here, since the New Perspectives School exists solely to fulfill the State's obligations under Chapter 766, I think it fully reasonable to conclude that the school is a state actor. 50 Indeed, I would conclude that the actions challenged here were under color of state law, even if I believed that the sole basis for state action was the fact that the school was providing Chapter 766 services. Petitioners claim that they were discharged because they supported student demands for increased responsibilities in school affairs, that is, because they criticized the school's educational policies. If petitioners' allegations are true, then the school has adopted a specific view of the sort of education that should be provided under the statute, and refuses to tolerate departures from that view.4 The State, by refusing to intervene, has effectively endorsed that view of its duties under Chapter 766. In short, because petitioners' criticism was directly addressed to the State's responsibilities under Chapter 766, a finding of state action is justified.5 51 The majority repeatedly compares the school to a private contractor that "depends primarily on contracts to build roads, bridges, dams, ships, or submarines for the government." Ante, at 840-841. The New Perspectives School can be readily distinguished, however. Although shipbuilders and dambuilders, like the school, may be dependent on government funds, they are not so closely supervised by the government. And unlike most private contractors, the school is performing a statutory duty of the State. 52 The majority also focuses on the fact that the actions at issue here are personnel decisions. It would apparently concede that actions directly affecting the students could be treated as under color of state law, since the school is fulfilling the State's obligations to those children under Chapter 766. It suggests, however, that the State has no interest in personnel decisions. As I have suggested, I do not share this narrow view of the school's obligations; the personnel decisions challenged here are related to the provision of Chapter 766 education. In any event, since the school is funded almost entirely by the State, is closely supervised by the State, and exists solely to perform the State's statutory duty to educate children with special needs—since the school is really just an arm of the State its personnel decisions may appropriately be considered state action. III 53 Even though there are myriad indicia of state action in this case, the majority refuses to find that the school acted under color of state law when it discharged petitioners. The decision in this case marks a return to empty formalism in state action doctrine. Because I believe that the state action requirement must be given a more sensitive and flexible interpretation than the majority offers, I dissent. 1 Chapter 766, 1972 Mass.Acts, Mass.Gen.Laws Ann., ch. 71B, § 3 (West Supp.1981), requires school committees to identify students with special needs and to develop suitable educational programs for such students. Massachusetts Gen.Laws Ann., ch. 71B, § 4 (West Supp.1981), provides that school committees may "enter into an agreement with any public or private school, agency, or institution to provide the necessary special education" for these students. A student identified as having special needs and recommended for placement in private school may remain in public school, if his parents object to a placement in a particular private school, unless he is especially disruptive or dangerous. Parents who object to placement in a particular private school may also elect to place their child in a private school of their choice; in such cases, they must pay the tuition. 2 Amicus curiae Massachusetts Association of 766 Approved Private Schools, Inc., of which the New Perspectives School is a member, informs the Court that many of its members have a student population which is more or less evenly divided between students referred and paid for by the State and students referred and paid for by their parents or guardians. Brief as Amicus Curiae 3. 3 The record does not contain details of the school's contract with the Brookline School Committee. 4 Title 42 U.S.C. § 1983 provides: "Every person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured in an action at law, suit in equity, or other proper proceeding for redress." 5 The Fourteenth Amendment provides, in pertinent part: "No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law. . . ." 6 The Court has concluded that the acts of a private party are fairly attributable to the state on certain occasions when the private party acted in concert with state actors. For example, in Adickes v. S. H. Kress & Co., 398 U.S. 144, 155-156, 90 S.Ct. 1598, 1607, 26 L.Ed.2d 142 (1970), the issue was whether a restaurant violated § 1983 by refusing service to a white teacher who was in the company of six Negro students; the town sheriff arrested the white teacher for vagrancy as a result of her request to be served lunch in their company. The Court concluded that the restaurant acted under color of state law because it conspired with the sheriff, a state actor, in depriving the white teacher of federal rights. Similarly, Flagg Brothers, Inc. v. Brooks, 436 U.S. 149, 98 S.Ct. 1729, 56 L.Ed.2d 185 (1978), and Lugar, 457 U.S. 922, 102 S.Ct. 2744, 73 L.Ed.2d 482, illustrate the relevance of whether action was taken in concert with a state actor. The issue in Flagg Brothers was whether a warehouseman could be sued under § 1983 because it sought to execute a lien by selling goods in its possession pursuant to § 7-210 of the New York Uniform Commercial Code. While the sale was authorized by a state statute, and hence appeared to be threatened under color of state law, the Court did not reach that issue. Instead, it concluded that the warehouseman's decision to threaten to sell the goods was not "properly attributable to the State of New York," 436 U.S., at 156, 98 S.Ct., at 1733, since no state actor was involved. Since the respondent in Flagg Brothers claimed that the warehouseman violated her Fourteenth Amendment rights to due process and equal protection, and the Fourteenth Amendment is only offended by action of the state, we held that no claim for relief had been stated. In Lugar, a lessee obtained an ex parte writ of attachment pursuant to a state statute, which was executed by a sheriff. The Court held that § 1983 applied because the involvement of the sheriff distinguished the case from Flagg Brothers. 457 U.S., at 941, 102 S.Ct., at 2756. The lessee thus acted under color of state law and the sheriff's involvement satisfied the state action requirement. The limited role played by the Massachusetts Committee on Criminal Justice in the discharge of Rendell-Baker is not comparable to the role played by the public officials in Adickes and Lugar. The uncontradicted evidence presented by the school showed that the Committee had the power only initially to review the qualifications of a counselor selected by the school to insure that the counselor met the requirements described in the school's grant application. 641 F.2d 14, 28 (1981). The Committee had no power to hire or discharge a counselor who had the qualifications specified in the school's grant application. Moreover, the Committee did not take any part in discharging Rendell-Baker; on the contrary, it attempted to use leverage to aid her. It requested an explanation for her discharge from the school and stated that it would not approve the appointment of a successor unless a grievance committee considered Rendell-Baker's case. As the Court of Appeals correctly concluded, there is no evidence that the Committee had any authority to take even those steps. Ibid. 7 There is no evidence that the State has attempted to avoid its constitutional duties by a sham arrangement which attempts to disguise provision of public services as acts of private parties. Cf. Evans v. Newton, 382 U.S. 296, 86 S.Ct. 486, 15 L.Ed.2d 373 (1966) (private trustees appointed to manage previously public park for white persons only). 1 The majority argues that the fact that the school receives almost all of its funds from the state is not enough, by itself, to justify a finding of state action. It also contends that the fact that the school is closely supervised and heavily regulated is not enough, by itself, to justify such a finding. Ante, at 840-842. I am in general agreement with both propositions. However, when these two factors are present in the same case, and when other indicia of state action are also present, a finding of state action may very well be justified. By analyzing the various indicia of state action separately, without considering their cumulative impact, the majority commits a fundamental error. See also ante, at 842-843. 2 This Court has repeatedly recognized the unique role that education plays in American society. See Plyler v. Doe, 457 U.S. 202, 221, 102 S.Ct. 2382, 2397, 72 L.Ed.2d 786 (1982) (public education is not "merely some governmental 'benefit' indistinguishable from other forms of social welfare legislation"); Wisconsin v. Yoder, 406 U.S. 205, 221, 92 S.Ct. 1526, 1536, 32 L.Ed.2d 15 (1972) (education is necessary to "prepare citizens to participate effectively and intelligently in our open political system"); Abington School District v. Schempp, 374 U.S. 203, 230, 83 S.Ct. 1560, 1575, 10 L.Ed.2d 844 (1963) (BRENNAN, J., concurring) (public education is a "most vital civic institution for the preservation of a democratic system of government"); Meyer v. Nebraska, 262 U.S. 390, 400, 43 S.Ct. 625, 627, 67 L.Ed. 1042 (1923) ("The American people have always regarded education and acquisition of knowledge as matters of supreme importance"). 3 A State may not deliberately delegate a task to a private entity in order to avoid its constitutional obligations. Terry v. Adams, 345 U.S. 461, 73 S.Ct. 809, 97 L.Ed. 1152 (1953). But a State's decision to delegate a duty to a private entity should be carefully examined even when it has acted, not in bad faith, but for reasons of convenience. The doctrinal basis for the state action requirement is that exercises of state authority pose a special threat to constitutional values. A private entity vested with state authority poses that threat just as clearly as a state agency. 4 This Court has previously emphasized the close relationship between teachers' free speech and the educational process. See Givhan v. Western Line Consolidated School District, 439 U.S. 410, 99 S.Ct. 693, 58 L.Ed.2d 619 (1979); Pickering v. Board of Education, 391 U.S. 563, 88 S.Ct. 1731, 20 L.Ed.2d 811 (1968); Keyishian v. Board of Regents, 385 U.S. 589, 87 S.Ct. 675, 17 L.Ed.2d 629 (1967); Shelton v. Tucker, 364 U.S. 479, 81 S.Ct. 247, 5 L.Ed.2d 231 (1960). The Commonwealth of Massachusetts has recently promulgated regulations recognizing that the role of teachers of special needs students is not limited to course instruction. These regulations provide: "[T]he candidate will demonstrate that he or she: 1. responds to the needs of individual students so as to enhance their self-esteem and development 2. establishes constructive relationships with parents and others primarily concerned with the well-being of his or her students 3. works to develop a learning environment which is favorable to openness of inquiry and devoid of ridicule." 603 Code Mass.Regs. § 7, ¶ 7.04(40)(f) (1982). 5 In my view, this connection between the teacher's role and the provision of Chapter 766 education would justify a finding that the State had acted under color of state law, even if the school did not depend solely on Chapter 766 placements. If the school had only one special needs student, and petitioners were discharged for criticizing the school's education of that child, a finding of state action might be justified.

12

458 U.S. 119 102 S.Ct. 3002 73 L.Ed.2d 647 UNION LABOR LIFE INSURANCE COMPANY, Petitioner,v.A. Alexander PIRENO. NEW YORK STATE CHIROPRACTIC ASSOCIATION, Petitioner, v. A. Alexander PIRENO. Nos. 81-389, 81-390. Argued April 27, 1982. Decided June 28, 1982. Syllabus As required by New York law, petitioner Union Labor Life Insurance Co. (ULL) issues health insurance policies covering certain policyholder claims for chiropractic treatments. Some ULL policies limit the company's liability to "reasonable" charges for "necessary" medical care and services. In order to determine whether particular chiropractors' treatments and fees were necessary and reasonable, ULL arranged with petitioner New York State Chiropractic Association (NYSCA), a professional association of chiropractors, to use the advice of its Peer Review Committee, which was established primarily to aid insurers in evaluating claims for chiropractic treatments, and which is composed of 10 practicing New York chiropractors. Respondent is a licensed chiropractor practicing in New York. On a number of occasions ULL referred his treatments of ULL policyholders, and his charges for those treatments, to the Committee for review. The Committee sometimes concluded that respondent's treatments were unnecessary or his charges unreasonable. Respondent brought suit in Federal District Court, alleging that petitioners' peer review practices violated § 1 of the Sherman Act because petitioners had used the Committee as the vehicle for their conspiracy to fix the prices that chiropractors would be permitted to charge for their services. The District Court granted petitioners' motion for summary judgment, dismissing respondent's complaint on the ground that ULL's use of NYSCA's Peer Review Committee was exempted from antitrust scrutiny by § 2(b) of the McCarran-Ferguson Act, which applies to the "business of insurance." The Court of Appeals reversed and remanded the action for further proceedings. Held : ULL's use of NYSCA's Peer Review Committee does not constitute the "business of insurance" within the meaning of § 2(b) of the McCarran-Ferguson Act, and thus is not exempt from antitrust scrutiny. Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261, controlling. Pp. 126-134. (a) There are three criteria relevant in determining whether a particular practice is part of the "business of insurance" exempted from the antitrust laws by § 2(b): first, whether the practice has the effect of transferring or spreading a policyholder's risk; second, whether the practice is an integral part of the policy relationship between the insurer and the insured; and third, whether the practice is limited to entities within the insurance industry. Royal Drug Co., supra. Pp. 126-129. (b) With regard to the first criterion, petitioners' arrangement plays no part in the spreading and underwriting of a policyholder's risk, because it is logically and temporally unconnected to the contract entered by the policyholder and ULL, which was the actual risk-transferring event. As to the second criterion, ULL's use of NYSCA's Peer Review Committee is distinct from ULL's contracts with its policyholders, and constitutes a separate arrangement between the insurer and third parties not engaged in the business of insurance. Nor does the challenged arrangement satisfy this criterion on the asserted ground that it directly involves the "interpretation" and "enforcement" of the insurance contract, because ULL's procedure for deciding whether claims are covered is a matter of indifference to the policyholder, whose only concern is whether his claim is paid, not why it is paid. As respects the third criterion, it may be assumed that the challenged arrangement need not be denied the § 2(b) exemption solely because it involves parties outside the insurance industry—namely, practicing chiropractors serving on the Peer Review Committee. But such arrangements can hardly be said to lie at the center of the legislative concern underlying § 2(b), which was with the protection of intra -industry cooperation in the underwriting of risks. More importantly, such arrangements may prove contrary to the spirit as well as the letter of § 2(b), because they have the potential to restrain competition in noninsurance markets. Pp. 130-134. 2nd Cir., 650 F.2d 387, affirmed. T. Richard Kennedy, New York City, for petitioners. Susan M. Jenkins, Washington, D. C., for respondent. Barry Grossman, Washington, D. C., for U. S. as amicus curiae, by special leave of Court. Justice BRENNAN delivered the opinion of the Court. 1 In these cases we consider an alleged conspiracy to eliminate price competition among chiropractors, by means of a "peer review committee" that advised an insurance company whether particular chiropractors' treatments and fees were "necessary" and "reasonable." The question presented is whether the alleged conspiracy is exempt from federal antitrust laws as part of the "business of insurance" within the meaning of the McCarran-Ferguson Act.1 2 * Petitioners are the New York State Chiropractic Association (NYSCA), a professional association of chiropractors, and the Union Labor Life Insurance Co. (ULL), a Maryland insurer doing business in New York. As required by New York law, ULL's health insurance policies cover certain policyholder claims for chiropractic treatments. But certain ULL policies limit the company's liability to "the reasonable charges" for "necessary medical care and services." App. 19a, 22a (emphasis added). Accordingly, when presented with a policyholder claim for reimbursement for chiropractic treatments, ULL must determine whether the treatments were necessary and whether the charges for them were reasonable. In making some of these determinations, ULL has arranged with NYSCA to use the advise of NYSCA's Peer Review Committee. 3 The Committee was established by NYSCA in 1971, primarily to aid insurers in evaluating claims for chiropractic treatments.2 It is composed of 10 practicing New York chiropractors, who serve on a voluntary basis. At the request of an insurer, the Committee will examine a chiropractor's treatments and charges in a particular case, and will render an opinion on the necessity for the treatments and the reasonableness of the charges made for them. The opinion will be based upon such considerations as the treating chiropractor's experience and specialty degrees; the location of his office; the number of visits and time spent with the patient; the patient's age, occupation, general physical condition, and history of previous treatment; and X-ray findings. 4 Respondent is a chiropractor licensed and practicing in the State of New York. On a number of occasions his treatments of ULL policyholders, and his charges for those treatments, have been referred by ULL to the Committee, which has sometimes concluded that his treatments were unnecessary or his charges unreasonable. Petitioners assert that respondent has treated his patients "in a manner calculated to maximize the number of treatments for a particular condition, and that his fees for these treatments are unusually high." 650 F.2d 387, 389 (CA2 1981). Respondent, for his part, contends that the members of the Committee "practice 'antiquated' techniques that they seek to impose on their more innovative competitors." Ibid. This dispute resulted in the present suit, brought by respondent in the United States District Court for the Southern District of New York. Respondent alleged that the peer review practices of petitioners violated § 1 of the Sherman Act.3 In particular, he claimed that petitioners and others had used the Committee as the vehicle for a conspiracy to fix the prices that chiropractors, including respondent, would be permitted to charge for their services. He concluded that he had been restrained from providing his chiropractic services to the public freely and fully, and that would-be recipients of chiropractic services had been deprived of the benefits of competition. Respondent requested, inter alia, declaratory and injunctive relief against ULL's continued use of NYSCA's Peer Review Committee in evaluating policyholders' claims. 5 After extensive discovery, the District Court granted petitioners' motion for summary judgment dismissing respondent's complaint, concluding that ULL's use of NYSCA's Peer Review Committee was exempted from antitrust scrutiny by the McCarran-Ferguson Act. App. to Pet. for Cert. in No. 81-389, pp. 20a-37a. The court noted that three requirements must be met in order to obtain the McCarran-Ferguson exemption: The challenged practices (1) must constitute the "business of insurance," (2) must be regulated by state law, and (3) must not amount to a "boycott, coercion, or intimidation." Id., at 27a-28a. In the court's view, all three of these requirements were satisfied in the present case. In particular, the court held that petitioners' peer review practices constituted the "business of insurance" because they served "to define the precise extent of ULL's contractual obligations . . . under [its] policies." Id., at 29a-30a. Moreover, the court determined that the peer review practices "involve[d] the spreading of risk, an indispensable element of the 'business of insurance.' " Id., at 30a.4 Respondents' Sherman Act claim was accordingly dismissed with prejudice. 6 The Court of Appeals for the Second Circuit reversed. 650 F.2d 387 (1981). Relying upon this Court's recent opinion in Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261 (1979), the Court of Appeals concluded that the District Court had erred in holding that ULL's use of NYSCA's Peer Review Committee constituted the "business of insurance."5 Accordingly, the Court of Appeals remanded the action for further proceedings. We granted certiorari to resolve a conflict among the Courts of Appeals on the question presented.6 454 U.S. 1052, 102 S.Ct. 595, 70 L.Ed.2d 587 (1981). II 7 The only issue before us is whether petitioners' peer review practices are exempt from antitrust scrutiny as part of the "business of insurance." "It is axiomatic that conduct which is not exempt from the antitrust laws may nevertheless be perfectly legal." Group Life & Health Ins. Co. v. Royal Drug Co., supra, 440 U.S., at 210, n. 5, 99 S.Ct., at 1072, n. 5. Thus in deciding these cases we have no occasion to address the merits of respondent's Sherman Act claims. However, the Sherman Act does express a "longstanding congressional commitment to the policy of free markets and open competition." Community Communications Co. v. Boulder, 455 U.S. 40, 56, 102 S.Ct. 835, 843, 70 L.Ed.2d 810 (1982); see also United States v. Topco Associates, Inc., 405 U.S. 596, 610, 92 S.Ct. 1126, 1134, 31 L.Ed.2d 515 (1972). Accordingly, our precedents consistently hold that exemptions from the antitrust laws must be construed narrowly. FMC v. Seatrain Lines, Inc., 411 U.S. 726, 733, 93 S.Ct. 1773, 1778, 36 L.Ed.2d 620 (1973). This principle applies not only to implicit exemptions, see Group Life & Health Ins. Co. v. Royal Drug Co., supra, 440 U.S., at 231, 99 S.Ct., at 1083, but also to express statutory exemptions, see United States v. McKesson & Robbins, Inc., 351 U.S. 305, 316, 76 S.Ct. 937, 943, 100 L.Ed. 1209 (1956). In Royal Drug, supra, this Court had occasion to reexamine the scope of the express antitrust exemption provided for the "business of insurance" by § 2(b) of the McCarran-Ferguson Act. We hold that decision of the question before us is controlled by Royal Drug. 8 The principal petitioner in Royal Drug was a Texas insurance company, Blue Shield, that offered policies entitling insured persons to purchase prescription drugs for $2 each from any pharmacy participating in a "Pharmacy Agreement" with Blue Shield; policyholders were also allowed to purchase prescription drugs from a nonparticipating pharmacy, but in that event they would have to pay full price for the drugs and would be reimbursed by Blue Shield for only a part of that price. Blue Shield offered Pharmacy Agreements to all licensed pharmacies in Texas, but participating pharmacies were required to sell prescription drugs to Blue Shield's policyholders for $2 each, and were reimbursed only for their cost in acquiring the drugs thus sold. "Thus, only pharmacies that [could] afford to distribute prescription drugs for less than this $2 markup [could] profitably participate in the plan." 440 U.S., at 209, 99 S.Ct., at 1072 (footnote omitted). 9 Respondents in Royal Drug were the owners of nonparticipating pharmacies. They sued Blue Shield and several participating pharmacies under § 1 of the Sherman Act, alleging that the Pharmacy Agreements were the instrument by which Blue Shield had conspired with participating pharmacies to fix the retail prices of prescription drugs. Respondents also alleged that the Agreements encouraged Blue Shield's policyholders to avoid nonparticipating pharmacies, thus constituting an unlawful group boycott. The District Court granted summary judgment to Blue Shield and the other petitioners, holding that the challenged Agreements were exempt under § 2(b) of the McCarran-Ferguson Act. But the Court of Appeals disagreed, holding that the Agreements were not the "business of insurance" within the meaning of that Act, and reversed. 440 U.S., at 210, 99 S.Ct., at 1072. This Court affirmed. Looking to "the structure of the Act and its legislative history," id., at 211, 99 S.Ct., at 1073, the Court discussed three characteristics of the business of insurance that Congress had intended to exempt through § 2(b). 10 First, after noting that one "indispensable characteristic of insurance" is the "spreading and underwriting of a policyholder's risk," id., at 211-212, 99 S.Ct., at 1073,7 the Court observed that parts of the legislative history of the McCarran-Ferguson Act "strongly suggest that Congress understood the business of insurance to be the underwriting and spreading of risk," id., at 220-221, 99 S.Ct., at 1077-1078. The Court then dismissed Blue Shield's contention that its Pharmacy Agreements involved such activities. 11 "The Pharmacy Agreements . . . are merely arrangements for the purchase of goods and services by Blue Shield. By agreeing with pharmacies on the maximum prices it will pay for drugs, Blue Shield effectively reduces the total amount it must pay to its policyholders. The Agreements thus enable Blue Shield to minimize costs and maximize profits. Such cost-savings arrangements may well be sound business practice, and may well inure ultimately to the benefit of policyholders in the form of lower premiums, but they are not the 'business of insurance.' " Id., at 214, 99 S.Ct., at 1074 (footnote omitted). 12 Second, the Court identified "the contract between the insurer and the insured" as "[a]nother commonly understood aspect of the business of insurance." Id., at 215, 99 S.Ct., at 1075. The Court noted that, in enacting the McCarran-Ferguson Act, Congress had been concerned with the " 'relationship between insurer and insured, the type of policy which could be issued, its reliability, interpretation, and enforcement—these were the core of the "business of insurance." ' " Id., at 215-216, 99 S.Ct., at 1075, quoting SEC v. National Securities, Inc., 393 U.S. 453, 460, 89 S.Ct. 564, 568, 21 L.Ed.2d 668 (1969). The Court then rejected Blue Shield's argument that its Pharmacy Agreements were so closely related to the "reliability, interpretation, and enforcement" of its policies as to fall within the intended scope of § 2(b): "This argument . . . proves too much." 440 U.S., at 216, 99 S.Ct., at 1075. 13 "At the most, the petitioners have demonstrated that the Pharmacy Agreements result in cost savings to Blue Shield which may be reflected in lower premiums if the cost savings are passed on to policyholders. But, in that sense, every business decision made by an insurance company has some impact on its reliability, its ratemaking, and its status as a reliable insurer . . . [and thus] could be included in the 'business of insurance.' Such a result would be plainly contrary to the statutory language, which exempts the 'business of insurance' and not the 'business of insurance companies.' " Id., at 216-217, 99 S.Ct., at 1075. 14 Finally, the Court noted that in enacting the McCarran-Ferguson Act, "the primary concern of both representatives of the insurance industry and the Congress was that cooperative ratemaking efforts be exempt from the antitrust laws." Id., at 221, 99 S.Ct., at 1078. This was so because of "the widespread view that it [was] very difficult to underwrite risks in an informed and responsible way without intra-industry cooperation." Ibid. The Court was thus reluctant to extend the § 2(b) exemption to the case before it, "because the Pharmacy Agreements involve parties wholly outside the insurance industry." Id., at 231, 99 S.Ct., at 1083. 15 "There is not the slightest suggestion in the legislative history that Congress in any way contemplated that arrangements such as the Pharmacy Agreements in this case, which involve the mass purchase of goods and services from entities outside the insurance industry, are the 'business of insurance.' " Id., at 224, 99 S.Ct., at 1079 (footnote omitted). 16 In sum, Royal Drug identified three criteria relevant in determining whether a particular practice is part of the "business of insurance" exempted from the antitrust laws by § 2(b): first, whether the practice has the effect of transferring or spreading a policyholder's risk; second, whether the practice is an integral part of the policy relationship between the insurer and the insured; and third, whether the practice is limited to entities within the insurance industry. None of these criteria is necessarily determinative in itself, but examining the arrangement between petitioners NYSCA and ULL with respect to all three criteria, we do not hesitate to conclude that it is not a part of the "business of insurance." Plainly, ULL's use of NYSCA's Peer Review Committee plays no part in the "spreading and underwriting of a policyholder's risk." Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S., at 211, 99 S.Ct., at 1072. Both the "spreading" and the "underwriting" of risk refer in this context to the transfer of risk characteristic of insurance. See n. 7, supra. And as the Court of Appeals below observed: 17 "The risk that an insured will require chiropractic treatment has been transferred from the insured to [ULL] by the very purchase of insurance. Peer review takes place only after the risk has been transferred by means of the policy, and then it functions only to determine whether the risk of the entire loss (the insured's cost of treatment) has been transferred to [ULL]—that is, whether the insured's loss falls within the policy limits." 650 F.2d, at 393. 18 Petitioner ULL argues that the Court of Appeals' analysis is "semantic and unrealistic." Brief for Petitioner ULL 17. Petitioner reasons that "[i]t is inconceivable that Congress would have included risk transfer within the 'business of insurance' but excluded a device that helps 'determine whether the risk . . . has been transferred' and acts as 'an aid in determining the scope of the transfer.' " Ibid. We find no merit in this argument, because the challenged peer review arrangement is logically and temporally unconnected to the transfer of risk accomplished by ULL's insurance policies. The transfer of risk from insured to insurer is effected by means of the contract between the parties the insurance policy—and that transfer is complete at the time that the contract is entered. See 9 G. Couch, Cyclopedia of Insurance Law §§ 39:53, 39:63 (2d ed. 1962). If the policy limits coverage to "necessary" treatments and "reasonable" charges for them, then that limitation is the measure of the risk that has actually been transferred to the insurer: To the extent that the insured pays unreasonable charges for unnecessary treatments, he will not be reimbursed, because the risk of incurring such treatments and charges was never transferred to the insurer, but was instead always retained by the insured. Petitioner's argument contains the unspoken premise that the transfer of risk from an insured to his insurer actually takes place not when the contract between those parties is completed, but rather only when the insured's claim is settled. This premise is contrary to the fundamental principle of insurance that the insurance policy defines the scope of risk assumed by the insurer from the insured. See id., at § 39:3; R. Keeton, Insurance Law § 5.1(a) (1971). 19 Turning to the second Royal Drug criterion, it is clear that ULL's use of NYSCA's Peer Review Committee is not an integral part of the policy relationship between insurer and insured. In the first place, the challenged arrangement between ULL and NYSCA is obviously distinct from ULL's contracts with its policyholders. In this sense the challenged arrangement resembles the Pharmacy Agreements in Royal Drug. There the Court rejected the proposition that the Agreements were " 'between insurer and insured.' " Group Life & Health Ins. Co. v. Royal Drug Co., supra, 440 U.S., at 215, 99 S.Ct., at 1075, quoting SEC v. National Securities, Inc., 393 U.S., at 460, 89 S.Ct., at 568. Rather, it recognized those Agreements as "separate contractual arrangements between Blue Shield and pharmacies engaged in the sale and distribution of goods and services other than insurance." 440 U.S., at 216, 99 S.Ct., at 1075. Similarly, ULL's use of NYSCA's Peer Review Committee is a separate arrangement between the insurer and third parties not engaged in the business of insurance. 20 Petitioner ULL argues that the challenged peer review practices satisfy this criterion because peer review "directly involves the 'interpretation' and 'enforcement' of the insurance contract." Brief for Petitioner ULL 16. But this argument is essentially identical to one made and rejected in Royal Drug. Blue Shield there contended that its Pharmacy Agreements "so closely affect[ed] the 'reliability, interpretation, and enforcement' of the insurance contract . . . as to fall within the exempted area." 440 U.S., at 216, 99 S.Ct., at 1075 (footnote omitted). This Court noted, however: 21 "The benefit promised to Blue Shield policyholders is that their premiums will cover the cost of prescription drugs except for a $2 charge for each prescription. So long as that promise is kept, policyholders are basically unconcerned with arrangements made between Blue Shield and participating pharmacies." Id., at 213-214, 99 S.Ct., at 1074 (footnotes omitted). 22 Similarly, when presented with policyholder claims for reimbursement, ULL must decide whether the claims are covered by its policies. But these decisions are entirely ULL's, and its use of NYSCA's Peer Review Committee as an aid in its decisionmaking process is a matter of indifference to the policyholder, whose only concern is whether his claim is paid, not why it is paid. As in Royal Drug, petitioners have shown, at the most, that the challenged peer review practices result in "cost savings to [ULL] which may be reflected in lower premiums if the cost savings are passed on to policyholders." Id., at 216, 99 S.Ct., at 1075. To grant the practices a § 2(b) exemption on such a showing "would be plainly contrary to the statutory language, which exempts the 'business of insurance' and not the 'business of insurance companies.' " Id., at 217, 99 S.Ct., at 1076. 23 Finally, as respects the third Royal Drug criterion, it is plain that the challenged peer review practices are not limited to entities within the insurance industry. On the contrary, ULL's use of NYSCA's Peer Review Committee inevitably involves third parties wholly outside the insurance industry—namely, practicing chiropractors. Petitioners do not dispute this fact, but instead deprecate its importance. They argue that we should not conclude "that ULL's use of the peer review process is outside the scope of the 'business of insurance' simply because NYSCA is not an insurance company." Brief for Petitioner ULL 25. In petitioners' view: 24 "There is nothing in the McCarran-Ferguson Act that limits the 'business of insurance' to the business of insurance companies. As this Court has stated, '[the Act's] language refers not to the persons or companies who are subject to state regulation, but to laws "regulating the business of insurance." ' National Securities, 393 U.S. at 459 [89 S.Ct. at 568]." Ibid. (emphasis in original of quoted opinion). 25 Asserting that "the [New York] Superintendent of Insurance effectively can regulate the peer review process through his authority over the claims adjustment procedures of ULL," id., at 26, petitioners conclude that the process is part of the "business of insurance" despite the necessary involvement of third parties outside the insurance industry. 26 We may assume that the challenged peer review practices need not be denied the § 2(b) exemption solely because they involve parties outside the insurance industry. But the involvement of such parties, even if not dispositive, constitutes part of the inquiry mandated by the Royal Drug analysis. As the Court noted there, § 2(b) was intended primarily to protect "intra -industry cooperation" in the underwriting of risks. 440 U.S., at 221, 99 S.Ct., at 1078 (emphasis added). Arrangements between insurance companies and parties outside the insurance industry can hardly be said to lie at the center of that legislative concern. More importantly, such arrangements may prove contrary to the spirit as well as the letter of § 2(b), because they have the potential to restrain competition in noninsurance markets. Indeed, the peer review practices challenged in the present cases assertedly realize precisely this potential: Respondent's claim is that the practices restrain competition in a provider market—the market for chiropractic services—rather than in an insurance market. App. 8a. Thus we cannot join petitioners in depreciating the fact that parties outside the insurance industry are intimately involved in the peer review practices at issue in these cases.8 III 27 In sum, we conclude that ULL's use of NYSCA's Peer Review Committee does not constitute the "business of insurance" within the meaning of § 2(b) of the McCarran-Ferguson Act.9 The judgment of the Court of Appeals is accordingly 28 Affirmed. 29 Justice REHNQUIST, with whom THE CHIEF JUSTICE and Justice O'CONNOR join, dissenting. 30 Purporting to rely upon our recent decision in Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 99 S.Ct. 1067, 59 L.Ed.2d 261 (1979), the Court today exposes to antitrust liability an aspect of the business of insurance designed to promote fair and efficient claims settlement. The Court reaches this conclusion by determining that the peer review process does not spread risk, is not an integral part of the insurance relationship, and is not limited to entities within the insurance industry. Because I find the claims adjustment function of the Peer Review Committee to be at the heart of the relationship between insurance companies and their policyholders, I conclude that such committees are clearly within the sphere of insurance activity which the McCarran-Ferguson Act intended to protect from the effect of the antitrust laws.1 This conclusion finds support in the legislative history of the Act and in Royal Drug and its predecessors. 31 For many years statutes such as the Sherman Act were thought not applicable to the business of insurance, this Court having held in Paul v. Virginia, 8 Wall. 168, 183, 19 L.Ed. 357 (1869), that "[i]ssuing a policy of insurance is not a transaction of commerce." When this Court held in United States v. South-Eastern Underwriters Assn., 322 U.S. 533, 64 S.Ct. 1162, 88 L.Ed. 1440 (1944), that the business of insurance was a part of interstate commerce subject to the Sherman Act, Congress responded quickly to reestablish the preeminence of States in regulating such business. Congress' response—the McCarran-Ferguson Act—sought primarily to protect the contractual relationship between the insurer and the insured: 32 "Under the regime of Paul v. Virginia, supra, States had a free hand in regulating the dealings between insurers and their policyholders. Their negotiations, and the contract which resulted, were not considered commerce and were, therefore, left to state regulation. The South-Eastern Underwriters decision threatened the continued supremacy of the States in this area. The McCarran-Ferguson Act was an attempt to turn back the clock, to assure that the activities of insurance companies in dealing with their policyholders would remain subject to state regulation." SEC v. National Securities, Inc., 393 U.S. 453, 459, 89 S.Ct. 564, 568, 21 L.Ed.2d 668 (1969). 33 We recognized this congressional purpose in Royal Drug : 34 " 'The relationship between insurer and insured, the type of policy which could be issued, its reliability, interpretation, and enforcement—these were the core of the "business of insurance." Undoubtedly, other activities of insurance companies relate so closely to their status as reliable insurers that they too must be placed in the same class. But whatever the exact scope of the statutory term, it is clear where the focus was—it was on the relationship between the insurance company and the policyholder.' " Group Life & Health Ins. Co. v. Royal Drug Co., supra, at 215-216, 99 S.Ct., at 1075 (quoting SEC v. National Securities, Inc., supra, at 460, 89 S.Ct., at 568.) 35 Thus, whatever else was said in Royal Drug about the indispensable characteristic of risk-spreading, the Court found the contractual relationship between the insurer and the insured to be the essence of the "business of insurance." 36 Central to this contractual relationship is the process of claims adjustment—the determination of the actual payments to be made to the insured for losses covered by the insurance contract. The key representation of the insurance company and the principal expectation of the policyholder is that prompt payment will be made when the event insured against actually occurs. As one commentator has stated: 37 "Up until the time there is a claim and a payment is made, the only tangible evidence of insurance is a piece of paper. In other words, the real product of insurance is the claims proceeds. Selection of the prospect, qualifying him for coverage that suits his needs, delivery of a policy, collecting premiums for perhaps years, making changes in coverage to meet changing situations, all of these are but preambles to the one purpose for which the insurance was secured, namely to collect dollars if and when an unforeseen event takes place." J. Wickman, Evaluating the Health Insurance Risk 57 (1965).2 38 It is the claims adjustor—in this case petitioners' Peer Review Committee—which determines whether and to what extent an insured's losses will be covered. The Court thus plainly errs when it concludes that the role of petitioners' Peer Review Committee "is not an integral part of the policy relationship between insurer and insured," ante, at 131, and "is a matter of indifference to the policyholder." Ante, at 132. Few insurance matters could be of greater importance to policyholders than whether their claims will be paid, and it is the Peer Review Committee which in effect makes that determination. Being a critical component of the relationship between an insurer and an insured, claims adjustment is part and parcel of the "business of insurance" protected by the McCarran-Ferguson Act.3 39 This conclusion finds support in a source of guidance completely disregarded by the Court—the legislative history of McCarran-Ferguson. The passage of the Act was preceded by the introduction in the Senate Committee of a report and a bill prepared by the National Association of Insurance Commissioners. "The views of the NAIC are particularly significant, because the Act ultimately passed was based in large part on the NAIC bill." Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S., at 221, 99 S.Ct., at 1078 (footnote omitted). Included in that bill were seven specific insurance practices to which the Sherman Act was not to apply, and to which the Court in Royal Drug looked for guidance as to the meaning of the phrase "business of insurance." See id., at 222, 99 S.Ct., at 1078. Among those seven protected practices was the process of claims adjustment: "the said Sherman Act shall not apply . . . to any cooperative or joint service, adjustment, investigation, or inspection agreement relating to insurance." 90 Cong.Rec. A4406 (1944) (emphasis added). Other statements in the legislative history support the conclusion that claims adjustment was to be protected: "[W]e come squarely to the question of whether State regulation is adequate to handle insurance, or whether that business should be subject to the provisions of the antitrust laws. . . . A great number of fire-insurance companies have cooperated in mutual agreement—and of necessity—through the Southeastern Underwriters Association and rating bureaus, adjusting policy rates to risks, classifying insurable property either in co-insurance or in re-insurance, making appraisals of losses, and working out systems of inspection to improve protection against fires. All of this has been done with splendid success. It would be a pity indeed, after all these years, to have the government intervene. The business of insurance involves long contracts. The fidelity of performance of those contracts will not brook intervention." Id., at 6530 (remarks of Rep. Satterfield) (emphasis added). 40 See also id., at 6543 (remarks of Rep. Jennings); id., at 6550-6551 (remarks of Rep. Ploeser). 41 The role of claims adjustment in the insurance relationship and the legislative history of the Act thus unmistakably demonstrate that claims settlement procedures such as petitioners' Peer Review Committee were to be accorded protection from the antitrust laws as the "business of insurance." Few practices followed by insurance companies today present a fairer or more efficient means of claims resolution than professional peer review committees. Insurance claimants seek reimbursement for virtually every form of medical treatment and care, and determining the reasonableness and necessity of such expenses requires the expertise of a practicing physician. Because the entire spectrum of human ailments are involved, the views of one physician are seldom sufficient; specialists from many fields of medicine must be consulted. Few if any insurance companies can afford to staff their claims settlement departments with such a broad range of physicians. The companies thus must either make less than satisfactory claims determinations, or must turn to an outside group of experts such as petitioners' Committee. 42 Although the Court protests that its decision says nothing about petitioners' antitrust liability, there can be little doubt that today's decision will vastly curtail the peer review process. Few professionals or companies will be willing to expose themselves to possible antitrust liability through such activity. The Court thus not only misreads the McCarran-Ferguson Act and our prior precedents, but also eliminates an aspect of the American insurance industry which has long redounded to the benefit of insurance companies and policyholders alike. 1 59 Stat. 33, as amended, 15 U.S.C. §§ 1011—1015. The Act provides in relevant part: "(a) The business of insurance, and every person engaged therein, shall be subject to the laws of the several States which relate to the regulation or taxation of such business. "(b) No Act of Congress shall be construed to invalidate, impair, or supersede any law enacted by any State for the purpose of regulating the business of insurance, . . . unless such Act specifically relates to the business of insurance. . . ." § 2, 15 U.S.C. §§ 1012(a), (b). "(b) Nothing contained in this Act shall render the . . . Sherman Act inapplicable to any agreement to boycott, coerce, or intimidate, or act of boycott, coercion, or intimidation." § 3, 15 U.S.C. § 1013(b). 2 The Committee's advice is also available to patients, governmental agencies, and chiropractors themselves, but insurers are the principal users. 650 F.2d 387, 388 (CA2 1981). 3 15 U.S.C. § 1, which provides in pertinent part that "[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal. . . ." 4 The court then turned to the Act's second requirement, that the challenged practices be "regulated by state law." The court held that that requirement had been met, as well, observing that New York had "enacted a pervasive scheme of regulation and supervision of insurance," had prohibited "the unfair settlement of claims," and had proscribed "the conduct alleged in the complaint" in its state antitrust law, the Donnelly Act, which by its terms applied to insurers. App. to Pet. for Cert. in No. 81-389, pp. 31a-32a. Finally, the court determined that respondent had neither alleged a "boycott" on petitioners' part, nor offered evidentiary support for such a claim. Id., at 33a-35a. The court thus concluded that the Act's third requirement was satisfied in the present case, and that petitioners' actions were consequently "exempt from application of the antitrust laws." Id., at 36a. 5 Since it reached this conclusion, the Court of Appeals did not definitively address the other holdings of the District Court. See n. 4, supra. The court did note, however, that petitioner NYSCA did not itself "appear to be regulated by state law in the manner § 2(b) requires." 650 F.2d, at 390, n. 5. 6 As noted by the Court of Appeals, id., at 395, n. 13, the decision below is contrary to that of the Court of Appeals for the Fourth Circuit "in a factually identical case." See Bartholomew v. Virginia Chiropractors Assn., 612 F.2d 812 (1979). 7 As the Court explained: " 'It is characteristic of insurance that a number of risks are accepted, some of which involve losses, and that such losses are spread over all the risks so as to enable the insurer to accept each risk at a slight fraction of the possible liability upon it.' 1 G. Couch, Cyclopedia of Insurance Law § 1:3 (2d ed. 1959). See also R. Keeton, Insurance Law § 1.2(a) (1971) ('Insurance is an arrangement for transferring and distributing risk'); 1 G. Richards, The Law of Insurance § 2 (W. Freedman 5th ed. 1952)." 440 U.S., at 211, 99 S.Ct., at 1073 (footnote omitted). 8 The premise of the dissent is that NYSCA's Peer Review Committee actually constitutes "the claims adjustor" in these cases. See post, at 137. From this premise the dissent reasons that since "claims adjustment is part and parcel of the 'business of insurance' protected by the McCarran-Ferguson Act," post, at 138, it necessarily follows that the peer review practices at issue in these cases must enjoy the Act's exemption. The fatal flaw in this syllogism is that NYSCA's Peer Review Committee is not the claims adjustor. As the Court of Appeals noted: "Opinions of the committee are not binding unless the parties agree beforehand that they will be." 650 F.2d, at 388. Thus in a case such as the present ones, ULL is perfectly free to disregard the Committee's evaluation. Even if ULL were to act upon the Committee's opinion, the nonbinding nature of the Committee's evaluation means that, at most, peer review is merely ancillary to the claims adjustment process. We see no reason that such ancillary activities must necessarily enjoy the McCarran-Ferguson exemption from the antitrust laws. Unlike activities that occur wholly within the insurance industry—such as the claims adjustment process itself—the ancillary peer review practices at issue in these cases "involve parties wholly outside the insurance industry." See Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S., at 231, 99 S.Ct., at 1083. Thus peer review falls afoul of the third Royal Drug criterion in a way in which pure claims adjustment activities cannot. 9 This conclusion renders it unnecessary for us to address the questions whether the conduct challenged in respondent's complaint was "regulated by state law" or constituted a "boycott, coercion, or intimidation." See n. 5, supra. 1 Since the Court declines to reach the question of whether petitioners' Committee is regulated by state law as required by the McCarran-Ferguson Act, I likewise do not discuss it. I note, however, that the District Court found petitioners' Committee to be so regulated. App. to Pet. for Cert. in No. 81-389, pp. 31a-32a. 2 Other commentators agree with this assessment of the importance of claims settlement: "The adjustment (including payment) of claims represents the final act in the insurance process. The payment of a claim by an insurance company brings the insurance contract 'to life' in a fashion far more vivid than does any other single act in connection with the purchase, issuance, and maintenance of the contract." Butler, Loss Adjustment in Fire Insurance, in Property and Liability Insurance Handbook 219 (J. Long & D. Gregg eds. 1965). "Claim administration is the last link in the process of insurance—a process that begins with actuarial analysis and continues through sales, underwriting, investment, and policy service. . . . [T]he expectation of the policyowner that an insurer is willing to meet its obligations, through claims administration, is an important part in the decision to purchase insurance. Indeed, it is the claim administration function that delivers on the product sold to the policyowner." C. Cissley, Claim Administration: Principles and Practices iii (1980). 3 Apparently unable to discern the difference between a mere method of paying a claim and the more fundamental process of determining whether a claim is covered by the insurance agreement, the Court finds that petitioners' peer review procedure "resembles the Pharmacy Agreements in Royal Drug." Ante, at 131. But the Pharmacy Agreement at issue in Royal Drug was simply a method of reimbursing policyowners for medication expenses. The policyowners could obtain medication from participating pharmacies simply by paying the amount that otherwise would not be covered by the insurance plan. The pharmacies thus constituted nothing more than in-kind dispensers of insurance payments; they played no role whatsoever in the more fundamental process of assessing the validity of a claim and determining the amount to be paid. Peer review committees, which fulfill such a fundamental role, are thus quite unlike the arrangements considered by the Court in Royal Drug.

78

458 U.S. 219 102 S.Ct. 3057 73 L.Ed.2d 721 FORD MOTOR COMPANY, Petitionerv.EQUAL EMPLOYMENT OPPORTUNITY COMMISSION. No. 81-300. Argued April 20, 1982. Decided June 28, 1982. Syllabus Held : An employer charged with discrimination in hiring under Title VII of the Civil Rights Act of 1964 can toll the continuing accrual of backpay liability under § 706(g) of Title VII by unconditionally offering the claimant the job previously denied, and is not required to offer seniority retroactive to the date of the alleged discrimination. Thus, absent special circumstances, the rejection of an employer's unconditional job offer ends the accrual of potential back pay liability. Pp. 225-241. (a) This rule serves Title VII's objective of ending discrimination through voluntary compliance, for it gives the employer a strong incentive to hire the claimant. To require a retroactive seniority offer in addition to the unconditional job offer fails to provide the same incentive, because it makes hiring the claimant more costly than hiring another applicant for the same job. Pp. 228-230. (b) An unemployed or underemployed claimant's statutory obligation to minimize damages under § 706(g) requires him to accept an unconditional job offer, even without retroactive seniority. The rule announced here merely embodies such requirement of minimizing damages, without affecting the claimant's right to compensation. Pp. 230-234. (c) The rule announced here also is consistent with the policy of full compensation when the claimant has had the good fortune to find a more attractive job than that offered by the employer charged with discrimination, because the availability of the better job terminates the ongoing ill effect of the latter's refusal to hire the claimant. Pp. 234-236. (d) Since the rule announced here rests both on the statutory requirement that a claimant minimize damages and on the fact that he no longer incurs additional injury if he has been able to find other work at least as attractive as the charged employer's, the rule in almost all circumstances is consistent with Title VII's object of making injured claimants whole. Pp.236-239 (e) To require a retroactive seniority offer in addition to the unconditional job offer would threaten the interests of innocent incumbent employees by disrupting the established seniority hierarchy, with the attendant risk that an innocent employee would be unfairly laid off or disadvantaged because a Title VII claimant has been granted seniority. Pp. 239-240. 645 F.2d 183 (4th Cir.), reversed and remanded. John R. Wester, Charlotte, N. C., for petitioner. David A. Strauss, Washington, D. C., for respondent, pro hac vice, by special leave of Court. Justice O'CONNOR delivered the opinion of the Court. 1 This case presents the question whether an employer charged with discrimination in hiring can toll the continuing accrual of backpay liability under § 706(g) of Title VII, 42 U.S.C. § 2000e-5(g), simply by unconditionally offering the claimant the job previously denied, or whether the employer also must offer seniority retroactive to the date of the alleged discrimination.1 2 The question has considerable practical significance because of the lengthy delays that too often attend Title VII litigation.2 The extended time it frequently takes to obtain satisfaction in the courts may force a discrimination claimant to suffer through years of underemployment or unemployment before being awarded the job the claimant deserves. Court delays, of course, affect all litigants. But for the victim of job discrimination, delay is especially unfortunate. The claimant cannot afford to stand aside while the wheels of justice grind slowly toward the ultimate resolution of the lawsuit. The claimant needs work that will feed a family and restore self-respect. A job is needed—now. In this case, therefore, we must determine how best to fashion the remedies available under Title VII to fulfill this basic need. 3 * A. 4 In June and July 1971, Judy Gaddis, Rebecca Starr, and Zettie Smith applied at a Ford Motor Co. (Ford) parts warehouse located in Charlotte, N.C., for jobs as "picker-packers," "picking" ordered parts from storage, and "packing" them for shipment. At the time, no woman had ever worked in that capacity at the Ford warehouse. All three women were qualified for the positions: Gaddis and Starr recently had been laid off from equivalent jobs at a nearby General Motors (GM) warehouse, and Smith had comparable prior experience. Smith applied before any of the openings were filled, and Gaddis and Starr applied while at least two positions remained available.3 Ford, however, filled the three vacant positions with men, and Gaddis filed a charge with the federal Equal Employment Opportunity Commission (EEOC), claiming that Ford had discriminated against her because of her sex.4 5 In January 1973, GM recalled Gaddis and Starr to their former positions at its warehouse. The following July, while they were still working at GM, a single vacancy opened up at Ford. Ford offered the job to Gaddis, without seniority retroactive to her 1971 application. Ford's offer, however, did not require Gaddis to abandon or compromise her Title VII claim against Ford. Gaddis did not accept the job, in part because she did not want to be the only woman working at the warehouse, and in part because she did not want to lose the seniority she had earned at GM. Ford then made the same unconditional offer to Starr, who declined for the same reasons. Gaddis and Starr continued to work at the GM warehouse, but in 1974 the warehouse was closed and they were laid off. They then unsuccessfully sought new employment until September 1975, when they entered a Government training program for the unemployed. 6 Smith applied again for work at Ford in 1973, but was never hired. She worked elsewhere, though at lower wages than she would have earned at Ford, during much of the time between 1971 and the District Court's decision in 1977. 7 In contrast to Gaddis', Starr's, and Smith's difficulties, at least two of the three men hired by Ford in 1971 were still working at the warehouse at the time of the trial in 1977. B 8 In July 1975, the EEOC sued Ford in the United States District Court for the Western District of North Carolina, alleging that Ford had violated Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp.IV), by refusing to hire women at the Charlotte warehouse. The Commission sought injunctive relief and backpay for the victims.5 9 After trial, the District Court found that Ford had discriminated against the three women on the basis of their sex and awarded them backpay in an amount equal to "the difference between the amount they would have earned had they been hired in August 1971, and the amounts actually earned or reasonably earnable by them" between that date and the date of the court's order. App. to Pet. for Cert. A-170. The District Court rejected Ford's contention that Gaddis and Starr were not entitled to backpay accruing after the dates on which they declined Ford's offer of employment. Id., at A-170 to A-171. 10 The United States Court of Appeals for the Fourth Circuit affirmed the District Court's finding of unlawful discrimination, as well as the court's award to Gaddis and Starr of backpay that had accrued after July 1973, when the women rejected Ford's unconditional job offer. 645 F.2d 183 (1981). The court suggested that, had Ford promised retroactive seniority with its job offer, the offer would have cut off Ford's backpay liability. The court concluded, however, that without the promise of retroactive seniority, Ford's 1973 offer was "incomplete and unacceptable." Id., at 193.6 11 Ford then petitioned this Court for a writ of certiorari, contending, inter alia, that its unconditional job offer to Gaddis and Starr should have cut off the further accrual of backpay liability.7 We granted the writ. 454 U.S. 1030, 102 S.Ct. 565, 70 L.Ed.2d 473 (1981). II 12 Section 706(g) of the Civil Rights Act of 1964, 78 Stat. 261, as amended, 42 U.S.C. § 2000e-5(g), governs the award of backpay in Title VII cases. In pertinent part, § 706(g) provides: 13 "If the court finds that the respondent has intentionally engaged in or is intentionally engaging in an unlawful employment practice charged in the complaint, the court may enjoin the respondent from engaging in such unlawful employment practice, and order such affirmative action as may be appropriate, which may include, but is not limited to, . . . hiring of employees, with or without back pay, . . . or any other equitable relief as the court deems appropriate. . . . Interim earnings or amounts earnable with reasonable diligence by the person or persons discriminated against shall operate to reduce the back pay otherwise allowable" (emphasis added).8 14 Under § 706(g), then, "backpay is not an automatic or mandatory remedy; . . . it is one which the courts 'may' invoke" in the exercise of their sound "discretion [which] is equitable in nature." Albemarle Paper Co. v. Moody, 422 U.S. 405, 415, 416, 95 S.Ct. 2362, 2370, 2371, 45 L.Ed.2d 280 (1975). Nonetheless, while "the power to award backpay is a discretionary power," id., at 447, 95 S.Ct., at 2389 (BLACKMUN, J., concurring in judgment), a "court must exercise this power 'in light of the large objectives of the Act,' " and, in doing so, must be guided by "meaningful standards" enforced by "thorough appellate review." Id., at 416, 95 S.Ct., at 2371 (opinion of the Court) (citations omitted). Moreover, as we emphasized in Albemarle Paper, in Title VII cases 15 "such discretionary choices are not left to a court's 'inclination, but to its judgment; and its judgment is to be guided by sound legal principles.' United States v. Burr, 25 F.Cas. 30, 35 (No. 14,692d) (CC Va. 1807) (Marshall, C. J.). . . . 16 "It is true that '[e]quity eschews mechanical rules . . . [and] depends on flexibility.' Holmberg v. Armbrecht, 327 U.S. 392, 396 [66 S.Ct. 582, 584, 90 L.Ed. 743] (1946). But when Congress invokes the Chancellor's conscience to further transcendent legislative purposes, what is required is the principled application of standards consistent with those purposes and not 'equity [which] varies like the Chancellor's foot.' Important national goals would be frustrated by a regime of discretion that 'produce[d] different results for breaches of duty in situations that cannot be differentiated in policy.' Moragne v. States Marine Lines, 398 U.S. 375, 405 [90 S.Ct. 1772, 1790, 26 L.Ed.2d 339] (1970)." Id., at 416-417, 95 S.Ct., at 2371 (footnote omitted). 17 In this case, Ford and the EEOC offer competing standards to govern backpay liability. Ford argues that if an employer unconditionally offers a claimant the job for which he previously applied, the claimant's rejection of that offer should toll the continuing accrual of backpay liability.9 The EEOC, on the other hand, defends the lower court's rule,10 contending that backpay liability should be tolled only by the rejection of an offer that includes seniority retroactive to the date on which the alleged discrimination occurred. Our task is to determine which of these standards better coincides with the "large objectives" of Title VII. III 18 The "primary objective" of Title VII is to bring employment discrimination to an end, Albemarle Paper, 422 U.S., at 417, 95 S.Ct., at 2371, by " 'achiev[ing] equality of employment opportunities and remov[ing] barriers that have operated in the past to favor an identifiable group . . . over other employees.' " Ibid. (quoting Griggs v. Duke Power Co., 401 U.S. 424, 429-430, 91 S.Ct. 849, 852-853, 28 L.Ed.2d 158 (1971)). See also McDonnell Douglas Corp. v. Green, 411 U.S. 792, 800, 93 S.Ct. 1817, 1823, 36 L.Ed.2d 668 (1973). "[T]he preferred means for achieving" this goal is through "[c]ooperation and voluntary compliance." Alexander v. Gardner-Denver Co., 415 U.S. 36, 44, 94 S.Ct. 1011, 1017, 39 L.Ed.2d 147 (1974). 19 To accomplish this objective, the legal rules fashioned to implement Title VII should be designed, consistent with other Title VII policies, to encourage Title VII defendants promptly to make curative, unconditional job offers to Title VII claimants, thereby bringing defendants into "voluntary compliance" and ending discrimination far more quickly than could litigation proceeding at its often ponderous pace. Delays in litigation unfortunately are now commonplace, forcing the victims of discrimination to suffer years of underemployment or unemployment before they can obtain a court order awarding them the jobs unlawfully denied them. In a better world, perhaps, lawsuits brought under Title VII would speed to judgment so quickly that the effects of legal rules on the behavior of the parties during the pendency of litigation would not be as important a consideration. We do not now live in such a world, however, as this case illustrates. 20 The rule tolling the further accrual of backpay liability if the defendant offers the claimant the job originally sought well serves the objective of ending discrimination through voluntary compliance, for it gives an employer a strong incentive to hire the Title VII claimant. While the claimant may be no more attractive than the other job applicants, a job offer to the claimant will free the employer of the threat of liability for further backpay damages. Since paying backpay damages is like paying an extra worker who never came to work, Ford's proposed rule gives the Title VII claimant a decided edge over other competitors for the job he seeks. 21 The rule adopted by the court below, on the other hand, fails to provide the same incentive, because it makes hiring the Title VII claimant more costly than hiring one of the other applicants for the same job. To give the claimant retroactive seniority before an adjudication of liability, the employer must be willing to pay the additional costs of the fringe benefits that come with the seniority that newly hired workers usually do not receive. More important, the employer must also be prepared to cope with the deterioration in morale, labor unrest, and reduced productivity that may be engendered by inserting the claimant into the seniority ladder over the heads of the incumbents who have earned their places through their work on the job. In many cases, moreover, disruption of the existing seniority system will violate a collective-bargaining agreement, with all that such a violation entails for the employer's labor relations.11 Under the rule adopted by the court below, the employer must be willing to accept all these additional costs if he hopes to toll his backpay liability by offering the job to the claimant. As a result, the employer will be less, rather than more, likely to hire the claimant. 22 In sum, the Court of Appeals' rule provides no incentive to employers to hire Title VII claimants. The rule advocated by Ford, by contrast, powerfully motivates employers to put Title VII claimants to work, thus ending ongoing discrimination as promptly as possible.12 IV 23 Title VII's primary goal, of course, is to end discrimination; the victims of job discrimination want jobs, not lawsuits.13 But when unlawful discrimination does occur, Title VII's secondary, fallback purpose is to compensate the victims for their injuries. To this end, § 706(g) aims " 'to make the victims of unlawful discrimination whole' " by restoring them, " 'so far as possible . . . to a position where they would have been were it not for the unlawful discrimination.' " Albemarle Paper, 422 U.S., at 421, 95 S.Ct., at 2373 (quoting 118 Cong.Rec. 7168 (1972) (remarks of Sen. Williams)). We now turn to consider whether the rule urged by Ford not only better serves the goal of ending discrimination, but also properly compensates injured Title VII claimants. 24 If Gaddis and Starr had rejected an unconditional offer from Ford before they were recalled to their jobs at GM, tolling Ford's backpay liability from the time of Ford's offer plainly would be consistent with providing Gaddis and Starr full compensation for their injuries. An unemployed or underemployed claimant, like all other Title VII claimants, is subject to the statutory duty to minimize damages set out in § 706(g).14 This duty, rooted in an ancient principle of law,15 requires the claimant to use reasonable diligence in finding other suitable employment. Although the unemployed or underemployed claimant need not go into another line of work, accept a demotion, or take a demeaning position,16 he forfeits his right to backpay if he refuses a job substantially equivalent to the one he was denied.17 Consequently, an employer charged with unlawful discrimination often can toll the accrual of backpay liability by unconditionally offering the claimant the job he sought, and thereby providing him with an opportunity to minimize damages.18 25 An employer's unconditional offer of the job originally sought to an unemployed or underemployed claimant, moreover, need not be supplemented by an offer of retroactive seniority to be effective, lest a defendant's offer be irrationally disfavored relative to other employers' offers of substantially similar jobs. The claimant, after all, plainly would be required to minimize his damages by accepting another employer's offer even though it failed to grant the benefits of seniority not yet earned.19 Of course, if the claimant fulfills the requirement that he minimize damages by accepting the defendant's unconditional offer, he remains entitled to full compensation if he wins his case.20 A court may grant him backpay accrued prior to the effective date of the offer,21 retroactive seniority,22 and compensation for any losses suffered as a result of his lesser seniority before the court's judgment.23 26 In short, the unemployed or underemployed claimant's statutory obligation to minimize damages requires him to accept an unconditional offer of the job originally sought, even without retroactive seniority. Acceptance of the offer preserves, rather than jeopardizes, the claimant's right to be made whole; in the case of an unemployed or underemployed claimant, Ford's suggested rule merely embodies the existing requirement of § 706(g) that the claimant minimize damages, without affecting his right to compensation. B 27 Ford's proposed rule also is consistent with the policy of full compensation when the claimant has had the good fortune to find a more attractive job than the defendant's, because the availability of the better job terminates the ongoing ill effects of the defendant's refusal to hire the claimant. For example, if Gaddis and Starr considered their jobs at GM to be so far superior to the jobs originally offered by Ford that, even if Ford had hired them at the outset, they would have left Ford's employ to take the new work, continuing to hold Ford responsible for backpay after Gaddis and Starr lost their GM jobs would be to require, in effect, that Ford insure them against the risks of unemployment in a new and independent undertaking. Such a rule would not merely restore Gaddis and Starr to the " 'position where they would have been were it not for the unlawful discrimination,' " Albemarle Paper Co. v. Moody, 422 U.S., at 421, 95 S.Ct., at 2373 (citation omitted); it would catapult them into a better position than they would have enjoyed in the absence of discrimination. 28 Likewise, even if Gaddis and Starr considered their GM jobs only somewhat better or even substantially equivalent to the positions they would have held at Ford had Ford hired them initially,24 their rejection of Ford's unconditional offer could be taken to mean that they believed that the lingering ill effects of Ford's prior refusal to hire them had been extinguished by later developments. If, for example, they thought that the Ford and GM jobs were identical in every respect, offering identical pay, identical conditions of employment, and identical risks of layoff, Gaddis and Starr would have been utterly indifferent as to which job they had—Ford's or GM's. Assuming that they could work at only one job at a time, the ongoing economic ill effects caused by Ford's prior refusal to hire them would have ceased when they found the identical jobs at GM, and they would have had no reason to accept Ford's offers. As in the case of a claimant who lands a better job, therefore, requiring a defendant to provide what amounts to a form of unemployment insurance to claimants, after they have found identical jobs and refused the defendant's unconditional job offer, would be, absent special circumstances, to grant them something more than compensation for their injuries. 29 In both of these situations, the claimant has the power to accept the defendant's offer and abandon the superior or substantially equivalent replacement job. As in the case of an unemployed or underemployed claimant, under the rule advocated by Ford acceptance of the defendant's unconditional offer would preserve fully the ultimately victorious claimant's right to full redress for the effects of discrimination.25 The claimant who chooses not to follow this path does so, then, not because it provides inadequate compensation, but because the value of the replacement job outweighs the value of the defendant's job supplemented by the prospect of full court-ordered compensation. In other words, the victim of discrimination who finds a better or substantially equivalent job no longer suffers ongoing injury stemming from the unlawful discrimination. C 30 Thus, the rule advocated by Ford rests comfortably both on the statutory requirement that a Title VII claimant must minimize damages and on the fact that a claimant is no longer incurring additional injury if he has been able to find other suitable work that, all things considered, is at least as attractive as the defendant's. For this reason, in almost all circumstances the rule is fully consistent with Title VII's object of making injured claimants whole. 31 The sole question that can be raised regarding whether the rule adequately compensates claimants arises in that narrow category of cases in which the claimant believes his replacement job to be superior to the defendant's job without seniority, but inferior to the defendant's job with the benefits of seniority. In the present case, for example, it is possible that Gaddis and Starr considered their GM jobs more attractive than the jobs offered by Ford, but less satisfactory than the positions they would have held at Ford if Ford had hired them initially. If so, they were confronted with two options. They could have accepted Ford's unconditional offer, preserving their right to full compensation if they prevailed on their Title VII claims, but forfeiting their favorable positions at GM. Alternatively, they could have kept their jobs at GM, retaining the possibility of continued employment there, but, under the operation of the rule advocated here by Ford, losing the right to claim further backpay from Ford after the date of Ford's offer. The court below concluded that under these circumstances Ford's rule would present Gaddis and Starr with an "intolerable choice," 645 F.2d, at 192, depriving them of the opportunity to receive full compensation. 32 We agree that Gaddis and Starr had to choose between two alternatives. We do not agree, however, that their opportunity to choose deprived them of compensation. After all, they had the option of accepting Ford's unconditional offer and retaining the right to seek full compensation at trial, which would comport fully with Title VII's goal of making discrimination victims whole. Under the rule advocated by Ford, if Gaddis and Starr chose the option of remaining at their GM jobs rather than accept Ford's offer, it was because they thought that the GM jobs, plus their claims to backpay accrued prior to Ford's offer, were more valuable to them than the jobs they originally sought from Ford, plus the right to seek full compensation from the court.26 It is hard to see how Gaddis and Starr could have been deprived of adequate compensation because they chose to venture upon a path that seemed to them more attractive than the Ford job plus the right to seek full compensation in court. 33 If the choice presented to Gaddis and Starr was difficult, it was only because it required them to assess their likelihood of prevailing at trial. But surely it cannot be contended for this reason alone that they were deprived of their right to adequate compensation. It is a fact of life that litigation is risky and that a plaintiff with a claim to compensation for his losses must consider the possibility that the claim might be lost at trial, either wrongly, because of litigation error, or rightly, because the defendant was innocent. Ford's rule merely requires the Title VII claimant to decide whether to take the job offered by the defendant, retaining his rights to an award by the court of backpay accrued prior to the effective date of the offer, and any court-ordered retroactive seniority plus compensation for any losses suffered as a result of his lesser seniority before the court's judgment, or, instead, whether to accept a more attractive job from another employer and the limitation of the claim for backpay to the damages that have already accrued. The rule urged by the EEOC and adopted by the court below, by contrast, would have the perverse result of requiring the employer in effect to insure the claimant against the risk that the employer might win at trial. 34 Therefore, we conclude that, when a claimant rejects the offer of the job he originally sought, as supplemented by a right to full court-ordered compensation, his choice can be taken as establishing that he considers the ongoing injury he has suffered at the hands of the defendant to have been ended by the availability of better opportunities elsewhere. For this reason, we find that, absent special circumstances,27 the simple rule that the ongoing accrual of backpay liability is tolled when a Title VII claimant rejects the job he originally sought comports with Title VII's policy of making discrimination victims whole. V 35 Although Title VII remedies depend primarily upon the objectives discussed above, the statute also permits us to consider the rights of "innocent third parties." City of Los Angeles Department of Water & Power v. Manhart, 435 U.S. 702, 723, 98 S.Ct. 1370, 1383, 55 L.Ed.2d 657 (1978). See also Teamsters v. United States, 431 U.S. 324, 371-376, 97 S.Ct. 1843, 1872-1875, 52 L.Ed.2d 396 (1977). The lower court's rule places a particularly onerous burden on the innocent employees of an employer charged with discrimination. Under the court's rule, an employer may cap backpay liability only by forcing his incumbent employees to yield seniority to a person who has not proved, and may never prove, unlawful discrimination. As we have acknowledged on numerous occasions, seniority plays a central role in allocating benefits and burdens among employees.28 In light of the " 'overriding importance' these rights, American Tobacco Co. v. Patterson, 456 U.S. 63, 76, 102 S.Ct. 1534, 1541, 71 L.Ed.2d 748 (1982) (quoting Humphrey v. Moore, 375 U.S. 335, 346, 84 S.Ct. 363, 370, 11 L.Ed.2d 370 (1964)), we should be wary of any rule that encourages job offers that compel innocent workers to sacrifice their seniority to a person who has only claimed, but not yet proved, unlawful discrimination. 36 The sacrifice demanded by the lower court's rule, moreover, leaves the displaced workers without any remedy against claimants who fail to establish their claims. If, for example, layoffs occur while the Title VII suit is pending, an employer may have to furlough an innocent worker indefinitely while retaining a claimant who was given retroactive seniority. If the claimant subsequently fails to prove unlawful discrimination, the worker unfairly relegated to the unemployment lines has no redress for the wrong done him. We do not believe that " 'the large objectives' " of Title VII, Albemarle Paper Co. v. Moody, 422 U.S., at 416, 95 S.Ct., at 2371, (citation omitted), require innocent employees to carry such a heavy burden.29 VI 37 In conclusion, we find that the rule adopted by the court below disserves Title VII's primary goal of getting the victims of employment discrimination into the jobs they deserve as quickly as possible. The rule, moreover, threatens the interests of other, innocent employees by disrupting the established seniority hierarchy, with the attendant risk that an innocent employee will be unfairly laid off or disadvantaged because a Title VII claimant unfairly has been granted seniority. 38 On the other hand, the rule that a Title VII claimant's rejection of a defendant's job offer normally ends the defendant's ongoing responsibility for backpay suffers neither of these disadvantages, while nevertheless adequately satisfying Title VII's compensation goals. Most important, it also serves as a potent force on behalf of Title VII's objective of bringing discrimination to an end more quickly than is often possible through litigation. For these reasons we hold that, absent special circumstances, the rejection of an employer's unconditional job offer ends the accrual of potential backpay liability. We reverse the judgment of the Court of Appeals and remand for proceedings consistent with this opinion. 39 So ordered. 40 Justice BLACKMUN, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 41 After finding that petitioner Ford Motor Company had discriminated unlawfully against Judy Gaddis and Rebecca Starr because of their sex, the Court of Appeals affirmed the District Court's backpay award to the two women "as a proper exercise of discretion founded on not clearly erroneous factual determinations." 645 F.2d 183, 201 (CA4 1981). The Court today reverses this unremarkable holding with a wide-ranging advisory ruling stretching far beyond the confines of this case. The Court's rule provides employers who have engaged in unlawful hiring practices with a unilateral device to cut off their backpay liability to the victims of their past discrimination. 42 To justify its new rule, the Court mischaracterizes the holding of the Court of Appeals, undertakes an intricate economic analysis of hypothetical situations not presented here, and invokes the rights of " 'innocent third parties,' " ante, at 3069, who are not before the Court. By so doing, the Court not only supplants traditional district court discretion to mold equitable relief, but also ensures that Judy Gaddis and Rebecca Starr—the only Title VII claimants whose rights are at issue in this lawsuit will not be made whole for injury they indisputably have suffered. I find the Court's ruling both unnecessary and unfair. I dissent. 43 * A. 44 The Court frames the question presented as "whether an employer charged with discrimination in hiring can toll the continuing accrual of backpay liability . . . simply by unconditionally offering the [Title VII] claimant the job previously denied, or whether the employer also must offer seniority retroactive to the date of the alleged discrimination." Ante, at 220. In my view, the Court simply and completely misstates the issue. The question before us is not which of two inflexible standards should govern accrual of backpay liability in all Title VII cases, but whether the District Court's award of backpay relief to Gaddis and Starr in this case constituted an abuse of discretion. 45 The Court makes frequent and puzzling reference to the "onerous burden[s]" and "sacrifice demanded by the lower court's rule." Ante, at 239-240. See also ante, at 227 ("the lower court's rule"); ante, at 229 ("[t]he rule adopted by the court below"); ibid. ("the Court of Appeals' rule"); ante, at 230, n. 12 ("the rule applied by the court below"); ante, at 238 ("[t]he rule . . . adopted by the court below"); ante, at 241 ("the rule adopted by the court below"). In fact, the Court of Appeals adopted no inflexible "rule" at all. Rather, it simply applied the well-settled and flexible principles of appellate review of Title VII remedies prescribed in Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975), and Franks v. Bowman Transportation Co., 424 U.S. 747, 96 S.Ct. 1251, 47 L.Ed.2d 444 (1976). 46 In Albemarle, this Court directed that, in most Title VII matters, "the standard of [appellate] review will be the familiar one of whether the District Court was 'clearly erroneous' in its factual findings and whether it 'abused' its traditional discretion to locate 'a just result' in light of the circumstances peculiar to the case," 422 U.S., at 424, 95 S.Ct., at 2375 (citation omitted). With regard to Title VII backpay relief, however, the Court specified that " 'the [district] court has not merely the power but the duty to render a decree which will so far as possible eliminate the discriminatory effects of the past as well as bar like discrimination in the future.' " Id., at 418, 95 S.Ct., at 2372, quoting Louisiana v. United States, 380 U.S. 145, 154, 85 S.Ct. 817, 822, 13 L.Ed.2d 709 (1965). To achieve this purpose, "Congress took care to arm the courts with full equitable powers. For it is the historic purpose of equity to 'secur[e] complete justice.' " 422 U.S., at 418, 95 S.Ct., at 2372 (citation omitted).1 47 The Court in Albemarle and Franks made clear that, in Title VII cases, the equitable discretion of district courts should be guided by a heavy presumption in favor of full backpay awards. "Rather than limiting the power of district courts to do equity, the presumption insures that complete equity normally will be accomplished." Franks v. Bowman Transportation Co., 424 U.S., at 786, 96 S.Ct., at 1275 (POWELL, J., concurring in part and dissenting in part). By exercising their discretion to award full backpay relief, district courts further two broad purposes underlying Title VII. First, "the reasonably certain prospect of a backpay award . . . 'provide[s] the spur or catalyst which causes employers . . . to self-examine and to self-evaluate their employment practices and to endeavor to eliminate, so far as possible, the last vestiges' " of discrimination. Albemarle Paper Co. v. Moody, 422 U.S., at 417-418, 95 S.Ct., at 2371-2372 (citation omitted). Second, backpay awards "make persons whole for injuries suffered on account of unlawful employment discrimination." Id., at 418, 95 S.Ct., at 2372. 48 Thus, the goal of appellate review is to ensure that the district courts have exercised their remedial discretion in the way that "allow[s] the most complete achievement of the objectives of Title VII that is attainable under the facts and circumstances of the specific case." Franks v. Bowman Transportation Co., 424 U.S., at 770-771, 96 S.Ct., at 1266-1267. "The courts of appeals must maintain a consistent and principled application of the backpay provision, consonant with [Title VII's] twin statutory objectives, while at the same time recognizing that the trial court will often have the keener appreciation of those facts and circumstances peculiar to particular cases." Albemarle Paper Co. v. Moody, 422 U.S., at 421-422, 95 S.Ct., at 2373. B 49 In this case, the trial court's findings of fact were uncontroverted. In July 1971, Judy Gaddis and Rebecca Starr sought jobs at petitioner Ford's automotive parts warehouse in Charlotte, N. C. "Because of their experience, each was qualified to work at Ford as a 'picker-packer.' " App. to Pet. for Cert. A-159 (District Court's Findings of Fact). Ford's stated hiring practice was to fill job vacancies at the warehouse by "taking the earliest filed applications first," and selecting employees by interviewing qualified candidates. Id., at A-157. At the time Gaddis and Starr applied, however, Ford had never hired any woman to work at the warehouse.2 Id., at A-167—A-168. When Gaddis and Starr received their application forms, "a receptionist at Ford . . . told them in substance that Ford did not hire women to work in the warehouse." Id., at A-159. 50 Despite Gaddis' persistent requests for job interviews, petitioner interviewed neither woman immediately, supposedly because no job vacancy existed. Id., at A-160—A-161. The unit supervisor testified: "Ms. Gaddis called me on several occasions and asked if I was hiring, and I said no, . . . I just have too much work to do to sit down and interview people if I'm not hiring." App. 31. Shortly thereafter, however, in August 1971, Ford hired male applicants to fill four job openings. App. to Pet. for Cert. A-159—A-160. "At least two of the men . . . were offered their jobs after Gaddis and Starr applied." Id., at A-160 (emphasis in original). 51 Gaddis filed a sex discrimination charge with respondent EEOC in September 1971. Id., at A-154. In January 1973, Gaddis and Starr were recalled to jobs at a nearby General Motors warehouse. In July 1973, petitioner made a vague job offer first to Gaddis, then to Starr.3 The District Courtfound as a fact that "[t]he offer to the two women was made after Ford learned that a charge of sex discrimination had been filed with the Commission (and was prompted by a desire to bring some women into the warehouse in response to the charge)." Id., at A-162—A-163.4 52 Gaddis, and then Starr, turned down petitioner's job offer. The District Court found that the offer was "refused by both women since they were at that time back at work in the General Motors warehouse, having been recalled to work in January, 1973. Neither woman wished to lose accrued seniority at General Motors and neither wanted to be the only woman employed in the Ford warehouse." Id., at A-163. 53 Based on its factual findings, the District Court concluded as a matter of law that "Ford discriminated against . . . Gaddis and Starr on the basis of their sex by failing to employ them in its warehouse in the positions filled in August, 1971." Id., at A-167. In rulings not contested here, the District Court also found that 10 other women had established prima facie cases of unlawful sex discrimination by Ford. Id., at A-168. 54 To determine the backpay remedy to which Gaddis and Starr were entitled, the District Court attached no legal significance to the women's decision to decline beginning employment at Ford nearly two years after they unlawfully had been denied those same jobs and six months after they had begun accumulating seniority elsewhere.5 In the ruling which the Court today implicitly deems an abuse of discretion, the District Court held that "[b]ack pay due to Gaddis and Starr shall not be affected by their refusal to accept the single position offered them in July, 1973, inasmuch as neither would have been confronted by that decision and its implications had both been hired in August, 1971." Id., at A-170 A-171. 55 Applying the standard of review specified in Franks, supra, and Albemarle, supra, the Court of Appeals, 645 F.2d, at 200, affirmed "the district court's decision as a proper exercise of discretion founded on not clearly erroneous factual determinations." Id., at 201. In particular, the Court of Appeals found no abuse of discretion in the District Court's failure to terminate the backpay awards in July 1973.6 56 The Court of Appeals rested its narrow ruling on two key facts: that "Gaddis and Starr could accept [Ford's] offer only by forfeiting the seniority they had accumulated at General Motors and without a compensating offer of seniority at Ford to alleviate the effects of the discrimination against them in 1971." Id., at 192. (Emphasis added.) The court expressed no view as to whether Ford's backpay liability would have been tolled if Gaddis and Starr could have accepted Ford's job offer without forfeiting seniority accumulated elsewhere. Nor did the Court of Appeals decide whether the women would have been obliged to accept Ford's offer had it encompassed some compensating offer of seniority, short of full retroactive seniority. 57 Contrary to this Court's suggestion today, the Court of Appeals announced no general rule that an employer's "backpay liability should be tolled only by the rejection of an offer that includes seniority retroactive to the date on which the alleged discrimination occurred." Ante, at 228 (emphasis added). The Court of Appeals merely refused to announce a broad new rule, urged by Ford, requiring victims of Title VII discrimination to "accept job offers which include a loss of seniority in order to preserve their back pay rights." 645 F.2d, at 192. Such an inflexible approach, the court decided, would frustrate Title VII's central purposes by permitting employers to present discriminatees with an "intolerable choice."7 Ibid. II 58 The Court today accepts Ford's invitation, wisely declined by the Court of Appeals, and adopts its broad new rule governing awards of backpay relief in Title VII cases: henceforth, "absent special circumstances, the rejection of an employer's unconditional job offer ends the accrual of potential backpay liability."8 Ante, at 241. This ruling is disturbing in four respects. 59 First: The Court's new rule is flatly inconsistent with Albemarle's unambiguous directive "that, given a finding of unlawful discrimination, backpay should be denied only for reasons which, if applied generally, would not frustrate the central statutory purposes of eradicating discrimination throughout the economy and making persons whole for injuries suffered through past discrimination." 422 U.S., at 421, 95 S.Ct., at 2373. Applied generally, the Court's rule interferes with both objectives. 60 The Court's approach authorizes employers to make "cheap offers" to the victims of their past discrimination. Employers may now terminate their backpay liability unilaterally by extending to their discrimination victims offers they cannot reasonably accept. Once an employer has refused to hire a job applicant, and that applicant has mitigated damages by obtaining and accumulating seniority in another job, the employer may offer the applicant the same job that she was denied unlawfully several years earlier. In this very case, for example, Ford offered Gaddis and Starr jobs only after they had obtained employment elsewhere and only because they had filed charges with the EEOC. If, as here, the applicant declines the offer to preserve existing job security, the employer has successfully cut off all future backpay liability to that applicant. By insulating a discriminating employer from proper liability for his discriminatory acts, the Court's rule reduces his "incentive to shun practices of dubious legality," id., at 417, 95 S.Ct., at 2371, and hinders the eradication of discrimination. 61 The Court's rule also violates Title VII's second objective making victims of discrimination whole. Again, the rule's anomalies are well illustrated by the facts of this case. Had petitioner not discriminated against Gaddis and Starr, both would have begun to work at Ford in August 1971. By July 1973, both would have accumulated nearly two years of seniority. Because of Ford's discrimination, however, each experienced long periods of unemployment and temporary employment before obtaining jobs elsewhere.9 The District Court therefore determined that only full backpay awards, mitigated by wages earned or reasonably earnable elsewhere, would make Gaddis and Starr whole. 62 This Court now truncates those awards simply because Gaddis and Starr refused to accept Ford's offers of beginning employment in 1973. Yet even if Gaddis and Starr had accepted those offers, they would not have been made whole. Deprived of two years of seniority, Gaddis and Starr would have enjoyed lesser health, life, and unemployment insurance benefits, lower wages, less eligibility for promotion and transfer, and greater vulnerability to layoffs than persons hired after they were unlawfully refused employment. See Tr. of Oral Arg. 30; Brief for Respondent 17. Even if Gaddis and Starr had continued to litigate the question of their retroactive seniority after accepting Ford's offer, they still would have spent many years at Ford "subordinate to persons who, but for the illegal discrimination, would have been[,] in respect to entitlement to [competitive seniority] benefits[,] [their] inferiors." Franks v. Bowman Transportation Co., 424 U.S., at 768, 96 S.Ct., at 1266. 63 The Court claims that its new rule "powerfully motivates employers to put Title VII claimants to work, thus ending ongoing discrimination as promptly as possible." Ante, at 230. In fact, the discrimination is not ended, because a discrimination victim who accepts a "cheap offer" will be obliged to work at a seniority disadvantage, and therefore will suffer ongoing effects from the employer's discriminatory act. The Court also alleges that its rule promotes "cooperation and voluntary compliance" with Title VII by giving both employers and claimants incentives to make and accept "unconditional" job offers. Ante, at 228-229. If the Court's rule furthers this end, however, it does so only by weakening the bargaining position of a claimant vis-a-vis the employer. Discrimination victims will be forced to accept otherwise unacceptable offers, because they will know that rejection of those offers truncates their backpay recovery. A rule that shields discriminating employers from liability for their past discrimination and coerces bona fide Title VII claimants to accept incomplete job offers is fundamentally incompatible with the purposes of Title VII. 64 Second: The Court's rule unjustifiably limits a district court's discretion to make individual discrimination victims whole through awards of backpay. The Court suggests that, "absent special circumstances," a district court abuses its discretion per se if it fails to terminate an employer's backpay liability at the point where that employer has extended an unconditional job offer to a discrimination claimant. Yet "[i]n Albemarle Paper the Court read Title VII as creating a presumption in favor of backpay." Franks v. Bowman Transportation Co., 424 U.S., at 786, 96 S.Ct., at 1275 (POWELL, J., concurring in part and dissenting in part) (emphasis added).10 Franks supplied "emphatic confirmation that federal courts are empowered to fashion such relief as the particular circumstances of a case may require to effect restitution, making whole insofar as possible the victims of . . . discrimination in hiring." Id., at 764, 96 S.Ct., at 1264 (opinion of the Court) (emphasis added). 65 The Court recognizes that its new rule interferes with district court discretion to make complete backpay awards in individual cases. Thus, the Court expressly preserves the principle of appellate deference to the "sound discretion" of the trial court in "exceptional circumstances." Ante, at 238-239, n.27. Yet, curiously, the Court offers no explanation why the facts of this very case fail to satisfy its own "exceptional circumstances" test.11 Given the Court's concession that district courts must retain their discretion to make bona fide Title VII claimants whole in some cases, I see no advantage in prescribing a blanket rule that displaces that discretion in other cases where complete relief is equally justified. 66 Third: I am disturbed by the Court's efforts to justify its rule by relying on situations not presented by this case. For example, the Court partially rests its rule on an "unemployed or underemployed claimant's statutory obligation to minimize damages" by accepting an unconditional job offer without seniority. Ante, at 234. Because Gaddis and Starr were fully employed when Ford finally offered them jobs, however, neither the District Court nor the Court of Appeals exempted unemployed or underemployed victims of discrimination from accepting offers like Ford's.12 Similarly, the Court analyzes the hypothetical case of a Title VII claimant who "has had the good fortune to find a more attractive job than the defendant's." Ibid. But, as the Court later recognizes, there is no assurance that the present case fits this category either. After speculating at length about how Gaddis and Starr may have valued the relative worth of their Ford and General Motors jobs, see ante, at 234-236, the Court finally acknowledges that on this paper record, "[w]e cannot infer" how much Gaddis and Starr "valued their GM jobs . . . solely from their rejection of Ford's offer." Ante, at 235, n. 24. 67 Equally unconvincing is the Court's repeated invocation of, and preoccupation with, "the rights of 'innocent third parties,' " ante, at 239, and the "disruption of the existing seniority system[s]," ante, at 229, that would result from adoption of the Court of Appeals' "rule." The Court nowhere demonstrates how petitioner's labor relations would have suffered had it extended offers of retroactive seniority to Gaddis and Starr. The details of Ford's collective-bargaining agreement were not litigated in either the District Court or the Court of Appeals. See Tr. of Oral Arg. 30-31. Thus, those courts never passed on petitioner's obligation to offer retroactive seniority to Gaddis and Starr if such an offer would have disrupted its labor relations or existing seniority systems.13 Nor did the Court of Appeals decide, as a general matter, whether or not offers of retroactive seniority to discrimination claimants adversely affect the rights of incumbent employees.14 The Court cannot justify reversal in the case at hand by vague reference to classes of claimants and third parties who are not before the Court. To the extent that it seeks to do so, its intricate argument is both irrelevant and advisory. 68 Fourth and finally: I am struck by the contrast between the Court's concern for parties who are not here and its studied indifference to the real-life concerns of the parties whose interests are directly affected. When the Court finally confronts the choice that actually faced Gaddis and Starr, ante, at 236-239, it blithely suggests that "[a]fter all, they had the option of accepting Ford's unconditional offer and retaining the right to seek full compensation at trial" in the form of retroactive seniority. Ante, at 237. Yet the Court earlier acknowledges that "[d]elays in litigation unfortunately are now commonplace, forcing the victims of discrimination to suffer years of underemployment or unemployment before they can obtain a court order awarding them the jobs unlawfully denied them." Ante, at 228. 69 "If the choice presented to Gaddis and Starr was difficult," the Court continues, "it was only because it required them to assess their likelihood of prevailing at trial." Ante, at 238. Without consulting the record, the Court then states: 70 "Gaddis and Starr presumably rejected Ford's offer because they thought their jobs at GM were worth more to them than full compensation (Ford's offer plus a court award) discounted by the risks of litigation. . . . Had they known they were going to win [their lawsuit], of course, they would have rejected the Ford job only if they valued the GM jobs more than they valued the combination of Ford's job plus the value of court-ordered compensation un discounted by the risks of litigation." Ante, at 237, n. 26 (emphasis in original). 71 This is a comfortable rationale stated from the sidelines. Unfortunately, the abstract and technical concerns that govern the Court's calculations bear little resemblance to those that actually motivated Judy Gaddis and Rebecca Starr. When asked on cross-examination why she had turned down Ford's 1973 offer, Gaddis testified: "I had seniority [at General Motors] and I knew that I wasn't in danger of any layoff, where if I had accepted the job at Ford I might have worked a week or two weeks and been laid off because I would have been low seniority." App. 47 (emphasis added). Similarly, Starr testified on cross-examination: "I had seniority at General Motors. I had about fifteen people working under me. I could go to work at Ford and work a week and I knew that they could lay me off." 4 Tr. 365-366 (emphasis added). 72 To a person living in the real world, the value of job security today far outstrips the value of full court-ordered compensation many years in the future. The Court's elaborate speculation about the concerns that "presumably" motivated Gaddis and Starr nowhere recognizes what a Ford job without seniority actually meant to Gaddis and Starr—a job from which they could be laid off at any moment. Unlike the Court, Gaddis and Starr recognized that if they traded their jobs with seniority for jobs without seniority, they could quickly become unemployed again, long before they had the chance to vindicate their rights at trial. 73 To people like Gaddis and Starr, the knowledge that they might someday establish their Title VII claims on the merits provides little solace for their immediate and pressing personal needs. Starr's trial testimony reveals just how much job security meant to her: 74 "It was just a couple of days after I had [started working] there [at a temporary job] and this is, I was just wanting that job so bad because you can't, a woman, when you've got three children, I needed the money, and I was wanting the job so bad. I worked so hard. I'll never forget one day when [the unit supervisor] came to me. I'll never forget that, and he said, I had just been there a few days, I'll have to let you go. . . . It broke my heart because I knew I had worked so hard." Id., at 356.15 75 I agree with the Court that "the victims of job discrimination want jobs, not lawsuits." Ante, at 230. See also, ante, at 221 ("The claimant needs work that will feed a family and restore self-respect. A job is needed—now"). When Ford made its 1973 offers to Gaddis and Starr, however, they had jobs, in which they had accumulated seniority despite Ford's discrimination. I therefore cannot accept the Court's conclusion that these women should have traded those jobs for uncertain employment in which back seniority could be won only by lawsuit. Nor can I justify penalizing Gaddis and Starr because they "discounted" the ultimate likelihood of obtaining court-ordered retroactive seniority at a different rate than the Court does today. 76 After hearing all the witnesses and appraising all the evidence, the District Court exercised its equitable discretion to shape complete backpay relief for Gaddis and Starr. In light of all the circumstances, the District Court refused to penalize Gaddis and Starr for declining Ford's 1973 job offer. Applying the correct standard of review over Title VII remedies, the Court of Appeals concluded that the District Court had exercised its remedial discretion properly. Sitting at this remove, I cannot say that Gaddis and Starr acted unreasonably. I would affirm the judgment of the Court of Appeals and thereby, for these two victims of discrimination, fulfill, and not defeat, the promise of Title VII. 1 The dissent asserts that by so "fram[ing] the question presented" we have "simply and completely misstate[d] the issue." Post, at 242. Apparently, neither party agrees with the dissent. The petitioner summarizes the question presented as "whether back pay due an employment discrimination claimant continues to accrue after the claimant has rejected an unconditional job offer that does not include retroactive seniority or back pay." Brief for Petitioner i. The respondent sums up the question presented as "[w]hether an employer who unlawfully refused to hire job applicants because they were women can terminate its liability for back pay by subsequently offering the applicants positions without seniority at a time when they had obtained, and accumulated seniority in, other jobs." Brief for Respondent i. To buttress the assertion that the Court has addressed a question not presented, the dissent claims that we have "misrea[d]" the Court of Appeals' decision, "transform[ing] a narrow Court of Appeals ruling into a broad one, just so [we could] reverse and install a broad new rule of [our] own choosing," post, at 249, n. 8, rather than attempt, as best we are able, to decide the particular case actually before us. Because we believe we have correctly and fairly framed the question, we decline the opportunity to address further this ad hominem argument. 2 The discriminatory refusals to hire involved in this case occurred 11 years ago. 3 When this case came to trial, Ford claimed that Gaddis and Starr applied after men had already been hired and that Smith had not applied at all. The District Court found to the contrary, however, and the Court of Appeals upheld the findings. 4 After Gaddis had filed her complaint, she and Starr continued to seek work at the Ford warehouse. In November 1972, Ford hired them and four other workers for six weeks to fill temporary jobs at the warehouse. 5 Although the EEOC suit involved additional issues and claimants, we are concerned here with only the part of the suit that involved Gaddis, Starr, and Smith. 6 Senior District Judge Walter E. Hoffman, sitting by designation, dissented from this portion of the Court of Appeals' decision. 7 In its petition Ford raised two other issues. First, Ford read the opinion of the Court of Appeals as suggesting that to toll backpay liability an employer must include with his job offer, not just retroactive seniority, but also an offer of already-accrued backpay. The Court of Appeals' opinion did not expressly so hold, however, and before this Court the EEOC concedes that under Title VII such an offer of a lump-sum payment of backpay is not required to toll the continuing accrual of backpay liability. This issue thus is no longer contested by the parties. The second issue is the only one involving Smith. Ford disputed the District Court's finding that Ford discriminated against the three women, claiming that the court reached its conclusion because it erroneously allocated the burden of proof. We are persuaded, however, that the District Court's findings were consistent with Texas Dept. of Community Affairs v. Burdine, 450 U.S. 248, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). In McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), we set forth the basic allocation of burdens and order of presentation of proof in a Title VII case alleging discriminatory treatment. See also Furnco Construction Corp. v. Waters, 438 U.S. 567, 98 S.Ct. 2943, 57 L.Ed.2d 957 (1978), and Board of Trustees v. Sweeney, 439 U.S. 24, 99 S.Ct. 295, 58 L.Ed.2d 216 (1978). Despite these decisions, some confusion continued to exist. In Burdine we reiterated that after a plaintiff has proved a prima facie case of discrimination, "the burden shifts to the defendant 'to articulate some legitimate, nondiscriminatory reason for the employee's rejection.' " 450 U.S., at 253, 101 S.Ct., at 1093 (citation omitted). The "ultimate burden of persuading the trier of fact that the defendant intentionally discriminated against the plaintiff remains at all times with the plaintiff." Ibid. (citation omitted). It was then made clear that: "The defendant need not persuade the Court that it was actually motivated by the proffered reasons. . . . It is sufficient if the defendant's evidence raises a genuine issue of fact as to whether it discriminated against the plaintiff." Id., at 254-255, 101 S.Ct., at 1094 (footnote omitted). As neither the District Court nor the Court of Appeals cited Burdine (apparently because it had only recently been decided), we restate the foregoing principles. We conclude, however, on the basis of the specific findings of fact by the District Court, undisturbed by the Court of Appeals, that the plaintiffs in this case carried their burden of persuasion. As Burdine commands, the District Court did not place the burden of persuasion on Ford. Instead, it began its discussion of liability by straightforwardly declaring that the EEOC "ha[d] established that Ford discriminated against Smith, Gaddis and Starr on the basis of their sex." App. to Pet. for Cert. A-167. The court supported this conclusion by pointing, not only to the EEOC's proof of a prima facie case, but also to "its showing that Ford had never hired women into the warehouse until November, 1972, and that [its] procedures for hiring were vague and were based on highly subjective criteria." Id., at A-167 to A-168. The court, moreover, entered findings of fact discrediting each of Ford's proffered justifications for refusing to hire the women. Id., at A-155 to A-161. This progression of factual findings and legal conclusions indicates that the District Court found by a preponderance of the evidence that Ford's justifications were "unworthy of credence," 450 U.S., at 256, 101 S.Ct., at 1095, and that the company had discriminated on the basis of sex. These findings are fully consistent with Burdine. As Ford points out, the Court of Appeals opinion contains some statements that are arguably inconsistent with Burdine. That court corrected any misimpression generated by these statements, however, with a discussion directly focusing on the burden-of-proof issue. 645 F.2d 183, 189, n.5 (CA4 1981). In light of this discussion, and because it is clear that the trier of fact properly allocated the burden of proof, we find no merit in Ford's burden-of-proof argument. 8 Section 706(g) was "expressly modeled," Albemarle Paper Co. v. Moody, 422 U.S. 405, 419, and n.11, 95 S.Ct. 2362, 2372, and n.11, 45 L.Ed.2d 280 (1975), on the analogous remedial provision of the National Labor Relations Act (NLRA), § 10(c), 49 Stat. 454, as amended, 29 U.S.C. § 160(c). Section 10(c) provides that, if an unfair labor practice has been, or is being, committed, the National Labor Relations Board (NLRB) is empowered to "take such affirmative action, including reinstatement of employees with or without back pay, as will effectuate the policies" of the Act. The principles developed under the NLRA generally guide, but do not bind, courts in tailoring remedies under Title VII. See, e.g., Teamsters v. United States, 431 U.S. 324, 366-367, 97 S.Ct. 1843, 1870, 52 L.Ed.2d 396 (1977); Franks v. Bowman Transportation Co., 424 U.S. 747, 768-770, 96 S.Ct. 1251, 1266, 47 L.Ed.2d 444 (1976); Albemarle Paper Co., supra, 422 U.S., at 419, and n.11, 95 S.Ct., at 2372, and n.11. Therefore, throughout this opinion we refer to cases decided under the NLRA as well as under Title VII. 9 It should be clear that the contested backpay in this suit stems from the period following Ford's offer, and during which Gaddis and Starr were unemployed, i.e., after the GM warehouse closed. Our decision today does not affect their right to claim backpay for the period before they rejected Ford's offers. 10 For reasons of its own, the dissenting opinion reads the decision below narrowly and takes us to task for discerning the outlines of a "general rule" post, at 248 (emphasis deleted), in the opinion of the Court of Appeals. In this regard, we note that already at least one District Court evidently not only has read the opinion below as prescribing a general rule, but in addition has interpreted that rule more broadly than we do. See Saunders v. Hercules, Inc., 510 F.Supp. 1137, 1142 (WD Va.1981) ("in view of the recent Fourth Circuit Court of Appeals decision in Equal Employment Opportunity Commission v. Ford Motor Company, 645 F.2d 183 (4th Cir. 1981) . . . [i]t is clear . . . that a person who has been discriminated against does not have to accept an offer of reemployment where back pay has not been offered"). 11 See American Tobacco Co. v. Patterson, 456 U.S. 63, 76, 102 S.Ct. 1534, 1541, 71 L.Ed.2d 748 (1982) ("Seniority provisions are of 'overriding importance' in collective bargaining, . . . and they 'are universally included in these contracts' ") (quoting Humphrey v. Moore, 375 U.S. 335, 346, 84 S.Ct. 363, 370, 11 L.Ed.2d 370 (1964), and Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 79, 97 S.Ct. 2264, 2274, 53 L.Ed.2d 113 (1977)). 12 In his dissent, Justice BLACKMUN suggests that it is we who speak from the "comfor[t]" of the "sidelines," post, at 256, somewhere outside "the real world," ibid., of sex discrimination. For all the dissent's rhetoric, however, nowhere does the dissent seriously challenge our conclusion that the rule we adopt will powerfully motivate employers to offer Title VII claimants the jobs they have been denied. But Rebecca Starr's trial testimony eloquently explains what claimants need: "I was just wanting that job so bad because you can't, a woman, when you've got three children, I needed the money, and I was wanting the job so bad." 4 Tr. 356. Thus, it is the rule applied by the court below which manifests a "studied indifference to the real-life concerns," post, at 255, of the victims of sex discrimination. 13 See 118 Cong.Rec. 7569 (1972) (remarks of Rep. Dent during debate on 1972 amendments to Title VII) ("Most people just want to work. That is all. They want an opportunity to work. We are trying to see that all of us, no matter of what race, sex, or religious or ethnic background, will have equal opportunity in employment"). 14 The provision expressly states that "[i]nterim earnings or amounts earnable with reasonable diligence by the person or persons discriminated against shall operate to reduce the back pay otherwise allowable." 42 U.S.C. § 2000e-5(g). Claimants often take other lesser or dissimilar work during the pendency of their claims, even though doing so is not mandated by the statutory requirement that a claimant minimize damages or forfeit his right to compensation. See, e.g., Merriweather v. Hercules, Inc., 631 F.2d 1161 (CA5 1980) (voluntary minimization of damages in dissimilar work); Thornton v. East Texas Motor Freight, 497 F.2d 416, 422 (CA6 1974) (voluntary minimization of damages by moonlighting). 15 See generally, e.g., C. McCormick, Law of Damages 127-158 (1935). McCormick summarizes "the general rule" as follows: "Where one person has committed a tort, breach of contract, or other legal wrong against another, it is incumbent upon the latter to use such means as are reasonable under the circumstances to avoid or minimize the damages. The person wronged cannot recover for any item of damage which could thus have been avoided." Id., at 127. In connection with the remedial provisions of the NLRA, we said: "Making the workers whole for losses suffered on account of an unfair labor practice is part of the vindication of the public policy which the Board enforces. Since only actual losses should be made good, it seems fair that deductions should be made not only for actual earnings by the worker but also for losses which he willfully incurred." Phelps Dodge Corp. v. NLRB, 313 U.S. 177, 197-198, 61 S.Ct. 845, 853-854, 85 L.Ed. 1271 (1941). 16 See, e.g., NLRB v. Madison Courier, Inc., 153 U.S.App.D.C. 232, 245-246, 472 F.2d 1307, 1320-1321 (1972) (employee need not "seek employment which is not consonant with his particular skills, background, and experience" or "which involves conditions that are substantially more onerous than his previous position"); Wonder Markets, Inc., 236 N.L.R.B. 787, 787 (1978) (offer of reinstatement ineffective when discharged employee offered a different job, though former position still existed), enf'd, 598 F.2d 666, 676 (CA1 1979), supplemental decision, 249 N.L.R.B. 294 (1980); Good Foods Manufacturing & Processing Corp., 195 N.L.R.B. 418, 419 (1972) (offer of reinstatement ineffective because job offered had different conditions of employment and benefits), supplemental decision, 200 N.L.R.B. 623 (1972), enf'd, 492 F.2d 1302 (CA7 1974); Harvey Carlton, 143 N.L.R.B. 295, 304 (1963) (offer of reinstatement ineffective because employees would return on probation). Some lower courts have indicated, however, that after an extended period of time searching for work without success, a claimant must consider taking a lower-paying position. See, e.g., NLRB v. Madison Courier, Inc., supra, 153 U.S.App.D.C., at 245-246, 472 F.2d, at 1320-1321; NLRB v. Southern Silk Mills, Inc., 242 F.2d 697, 700 (CA6), cert. denied, 355 U.S. 821, 78 S.Ct. 28, 2 L.Ed.2d 37 (1957). If the claimant decides to go into a dissimilar line of work, or to accept a demotion, his earnings must be deducted from any eventual backpay award. See § 706(g); Merriweather v. Hercules, Inc., supra, at 1168; Taylor v. Philips Industries, Inc., 593 F.2d 783, 787 (CA7 1979) (per curiam). 17 NLRB v. Arduini Mfg. Corp., 394 F.2d 420 (CA1 1968). 18 The claimant's obligation to minimize damages in order to retain his right to compensation does not require him to settle his claim against the employer, in whole or in part. Thus, an applicant or discharged employee is not required to accept a job offered by the employer on the condition that his claims against the employer be compromised. See, e.g., NLRB v. St. Marys Sewer Pipe Co., 146 F.2d 995, 996 (CA3 1945). 19 For the same reasons, a defendant's job offer is effective to force minimization of damages by an unemployed or underemployed claimant even without a supplemental offer of backpay, since the claimant would be required to accept another employer's offer of a substantially similar job without a large front-end, lump-sum bonus. See, e.g., NLRB v. Midwest Hanger Co., 550 F.2d 1101, 1103 (CA8) ("It is clear that had the Company's offer of reinstatement been conditioned solely on its refusal to give back pay, as the Company strenuously argues, then the offer of reinstatement would not have been invalidated"), cert. denied, 434 U.S. 830, 98 S.Ct. 112, 54 L.Ed.2d 90 (1977); Reliance Clay Products Co., 105 N.L.R.B. 135, 137 (1953) ("The Board has consistently held that a discriminatorily discharged employee may not refuse" an unconditioned offer of reinstatement even though unaccompanied by backpay; refusal of such an offer tolls the employer's liability for backpay). 20 In tailoring a Title VII remedy a court " 'has not merely the power but the duty to render a decree which will so far as possible eliminate the discriminatory effects of the past as well as bar like discrimination in the future.' " Albemarle Paper Co. v. Moody, 422 U.S., at 418, 95 S.Ct., at 2372 (quoting Louisiana v. United States, 380 U.S. 145, 154, 85 S.Ct. 817, 822, 13 L.Ed.2d 709 (1965)). 21 See, e.g., NLRB v. Huntington Hospital, Inc., 550 F.2d 921, 924 (CA4 1977). 22 See, e.g., Zipes v. Trans World Airlines, Inc., 455 U.S. 385, 102 S.Ct. 1127, 71 L.Ed.2d 234 (1982); Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977); Franks v. Bowman Transportation Co., 424 U.S. 747, 96 S.Ct. 1251, 47 L.Ed.2d 444 (1976). Decisions construing the remedial provision of the NLRA, § 10(c), 29 U.S.C. § 160(c), are in accord. See, e.g., In re Nevada Consolidated Copper Corp., 26 N.L.R.B. 1182, 1235 (1940) (persons unlawfully refused jobs must be offered jobs with "any seniority or other rights and privileges they would have acquired, had the respondent not unlawfully discriminated against them") (quoted in Franks v. Bowman Transportation Co., supra, at 770, 96 S.Ct., at 1267), enf. denied, 122 F.2d 587 (CA10 1941), rev'd, 316 U.S. 105, 62 S.Ct. 960, 86 L.Ed. 1305 (1942). 23 Both Ford and the EEOC agree on this point. See Brief for Respondent 19; Reply Brief for Petitioner 9. 24 It is possible that they did so value the GM jobs, since they applied at Ford only after being laid off at GM, and since after being recalled to the GM jobs they rejected Ford's offer. Therefore, contrary to the dissent's erroneous suggestion, post, at 253, the possibility that Gaddis and Starr considered their GM jobs superior to the positions they would have had at Ford had Ford hired them at the outset is not merely a "hypothetical case." We cannot infer that they so valued their GM jobs, however, solely from their rejection of Ford's offer. 25 See discussion supra, at 232-234. 26 Employees value a job for many reasons besides the rate of pay, including, for example, the presence of other workers of the employee's own sex, the availability of recreational facilities at the worksite, staggered work hours, better health benefits, longer vacations, and so forth. What makes one job better than another varies from one employee to another. Gaddis and Starr presumably rejected Ford's offer because they thought their jobs at GM were worth more to them than full compensation (Ford's offer plus a court award) discounted by the risks of litigation. In essence, the position adopted by the court below and advocated here by the EEOC turns on the fact that we cannot be sure that, had Gaddis and Starr known they were going to win their lawsuit, they still would have rejected Ford's offer. Had they known they were going to win, of course, they would have rejected the Ford job only if they valued the GM jobs more than they valued the combination of Ford's job plus the value of court-ordered compensation un discounted by the risks of litigation. To agree with the EEOC is, in effect, to contend that a claimant is not made whole for purposes of Title VII unless he decided to stay at a replacement job that was worth to him more than the sum of (1) the defendant's job, (2) the right to seek full court-ordered compensation, and, in addition, (3) a sum analogous to insurance against the risk of loss at trial. We discern, however, no reason for concluding that Title VII requires the defendant to insure the claimant against the possibility that the defendant might prevail in the lawsuit. 27 If, for example, the claimant has been forced to move a great distance to find a replacement job, a rejection of the employer's offer might reflect the costs of relocation more than a judgment that the replacement job was superior, all things considered, to the defendant's job. In exceptional circumstances, the trial court, in the exercise of its sound discretion, could give weight to such factors when deciding whether backpay damages accrued after the rejection of an employer's offer should be awarded to the claimant. The dissent attempts to characterize "the loss of their accumulated seniority at [a] replacement jo[b]" as such a cost of relocation. Post, at 252-253, n. 11. By so doing, the dissent simply confuses the costs of changing from one job to another whatever the respective advantages and disadvantages of the two jobs might be—with the differences between the two jobs. 28 Seniority may govern, " 'not only promotion and layoff, but also transfer, demotion, rest days, shift assignments, prerogative in scheduling vacation, order of layoff, possibilities of lateral transfer to avoid layoff, 'bumping' possibilities in the face of layoff, order of recall, training opportunities, working conditions, length of layoff endured without reducing seniority, length of layoff recall rights will withstand, overtime opportunities, parking privileges, and [even] a preferred place in the punch-out line.' " Franks v. Bowman Transportation Co., 424 U.S., at 766-767, 96 S.Ct., at 1265 (quoting Stacy, Title VII Seniority Remedies in a Time of Economic Downturn, 28 Vand.L.Rev. 487, 490 (1975)). 29 In addition to the rights of innocent employees, the rule urged by the EEOC and adopted by the court below burdens innocent employers. An innocent employer—or one who believes himself innocent—has the right to challenge in court claims he considers weak or baseless. The approach endorsed by the lower court undermines this right by requiring the employer, if he wishes to offer some relief to the claimant and toll the mounting backpay bill, to surrender his defense to the charge that the claimant is entitled to retroactive seniority. If the employer offers the claimant retroactive seniority as well as a job, and then prevails at trial, he will have no recourse against the claimant for the costs of the retroactive seniority that the claimant erroneously received. The rule urged by Ford permits the parties to stem the ongoing effects of the alleged discrimination without compelling either claimant or employer to compromise his claims or surrender his defenses. Cf. Moro Motors Ltd., 216 N.L.R.B. 192, 193 (1975) ("were [an employer] required to offer to an employee, allegedly discharged for discriminatory reasons, reinstatement with accrued back pay, the [employer's] right to litigate the issue of whether the discharge was unlawful would for all practical purposes be nullified") (emphasis in original); National Screen Products Co., 147 N.L.R.B. 746, 747-748 (1964). 1 In passing the Equal Employment Opportunity Act of 1972, 86 Stat. 103, Congress specifically rejected several legislative efforts to limit the judicial power to award backpay. See Albemarle Paper Co. v. Moody, 422 U.S. 405, 420, 95 S.Ct. 2362, 2373, 45 L.Ed.2d 280 (1975). The Section-by-Section Analysis accompanying the Conference Committee Report reaffirmed the "make whole" purpose of § 706(g), Title VII's backpay provision: "The provisions of this subsection are intended to give the courts wide discretion exercising their equitable powers to fashion the most complete relief possible. In dealing with the present section 706(g) the courts have stressed that the scope of relief under that section of the Act is intended to make the victims of unlawful discrimination whole, and that the attainment of this objective . . . requires that persons aggrieved by the consequences and effects of the unlawful employment practice be, so far as possible, restored to a position where they would have been were it not for the unlawful discrimination." 118 Cong.Rec. 7168 (1972), quoted in Albemarle Paper Co. v. Moody, 422 U.S., at 421, 95 S.Ct., at 2373. 2 The District Court found, for example, that the job application of Zettie Smith, who sought employment at Ford about a month before Gaddis and Starr, and who was the first woman to apply for a warehouse job there, "was never seriously considered because she is a woman." App. to Pet. for Cert. A-157—A-158. 3 At trial, Gaddis was asked: "Q. Did [the clerk to the warehouse manager] say that the job was being offered to you, or did he discuss simply with you, in the form of an interview, the possibility of hiring you into some job? "A. It was so vague that I couldn't pinpoint anything down. They never did say what type of work it would be, whether it would be [parts] picking or whether it would be in sheet metal or whether it would be putting up stock or whether it would be on a day shift or night shift, whether it was a permanent or temporary job. At the time, I had a good seniority with General Motors and I had a secure job, and so on those grounds, I refused it." App. 43. Similarly, Starr testified on cross-examination: "I remember [the clerk to the warehouse manager] wasn't specific on the job about what it would be. I did have, at General Motors I had fifteen, I don't know if it was fourteen or fifteen people under me. I had seniority, and I also, this is the truth about [it,] I was scared. Whenever I had worked at Ford before, I had been badgered and I don't know, I was just, I wanted to look into the job. Yet, I had a fear to go back. I didn't know what I would be facing." Id., at 54. 4 The trial testimony of Ford's warehouse operations manager illuminates petitioner's motives: "Q. Whose decision was it to call Ms. Gaddis and Ms. Starr? "A. It was my decision. "Q. Why? "A. Well, mainly because we had a suit, EEOC suit filed against us, and we wanted to give one of them an opportunity to go to work for us, and we only had one, maybe two openings at that time. * * * * * "Q. Mr. Ely, you indicated in your testimony that you offered a job to one of the women, either Ms. Gaddis or Ms. Starr, in July, 1973. Is that correct? "A. Yes, that's correct. "Q. You also stated that you offered such job because of the EEOC charge which had been filed against Ford Motor Company. Is that correct? "A. That's correct." Id., at 17-18. 5 The District Court applied two equitable principles to shape relief in this case. It first concluded that an award of all backpay accruing after August 1971 would make Gaddis and Starr whole. The District Court therefore reconstructed a probable employment history at Ford for each woman, calculating what each would have received but for petitioner's unlawful discrimination. Second, the court obliged Gaddis and Starr to take all reasonable steps to mitigate damages. Accordingly, it subtracted from the backpay awards any amounts Gaddis and Starr actually earned or reasonably could have earned after August 1971. App. to Pet. for Cert. A-170. 6 "[T]he district court reached an eminently reasonable result. It did not permit Ford to cut off the back pay period by making Gaddis and Starr an incomplete and unacceptable offer, and it denied Gaddis and Starr a double recovery by deducting their General Motors wages from their back pay awards." 645 F.2d 183, at 193. 7 "[I]f Gaddis and Starr rejected Ford's offer and stayed at General Motors, they would forego their rights to further back pay benefits. On the other hand, if they accepted the job offered by Ford, which they had not held for the previous two years because of Ford's discriminatory hiring policy, they would lose their seniority rights at General Motors." Id., at 192. 8 The Court's explanation for its misreading of the Court of Appeals' decision is that the United States District Court for the Western District of Virginia has interpreted that decision as stating a somewhat different proposition. See ante, at 227, n. 10. But if one District Court in the Fourth Circuit has misconstrued the Fourth Circuit's opinion, surely that is a matter properly to be corrected by the United States Court of Appeals for the Fourth Circuit. This Court is not entitled to transform a narrow Court of Appeals ruling into a broad one, just so that it may reverse and install a broad new rule of its own choosing. 9 Gaddis, for example, sought employment in South Carolina "at various parts places, independent part places, car dealers, such as Chrysler-Plymouth, the Ford place which was Lewis Ford at that time, all the car dealers, . . . some of the hosiery mills, . . . [and] Radiator Specialty Company," 3 Tr. 362, before obtaining her job at General Motors. 10 The Court cites language from Albemarle suggesting that a district court's discretion is not limitless. See ante, at 226-227. But the Court conspicuously omits Albemarle's clear statement that if Congress intended to limit the equitable discretion of district courts in any way, it did so only by leaving " 'little room for the exercise of discretion not to order reimbursement.' " See Albemarle Paper Co. v. Moody, 422 U.S., at 417, 95 S.Ct., at 2371, quoting Mitchell v. DeMario Jewelry, Inc., 361 U.S. 288, 296, 80 S.Ct. 332, 337, 4 L.Ed.2d 323 (1960) (emphasis added). 11 The Court suggests, for example, that if a hypothetical Title VII "claimant has been forced to move a great distance to find a replacement job, a rejection of the employer's offer might reflect the costs of relocation more than a judgment that the replacement job was superior, all things considered, to the defendant's job." Ante, at 238, n. 27. For Gaddis and Starr, however, the loss of their accumulated seniority at their replacement jobs certainly reflected "costs of relocation" at least as substantial as high moving expenses. I expect that federal courts will find no meaningful distinction between a worker's refusal to accept a job offer because he believes that acceptance would force him to incur costs, and a similar refusal based on the worker's judgment that changing jobs would prove costly. In either case, for purposes of awarding Title VII relief, the reasonableness of the worker's refusal should be left to the trial court's discretion. 12 The purpose of § 706(g)'s "mitigation of damages" requirement is to encourage claimants to work while their Title VII claims are being adjudicated. The Court cannot deny that Gaddis and Starr fully mitigated damages by seeking and obtaining other employment while litigating their claims against Ford. 13 The Court of Appeals did not foreclose the possibility that Ford could have terminated its backpay liability to Gaddis and Starr by offering them employment plus an award of provisional seniority, defeasible in the event that they lost their continuing lawsuit for backpay. Nor did the Court of Appeals deny that offering a job without seniority might terminate Ford's backpay liability, should any provision of Ford's collective-bargaining agreement preclude it from making offers of retroactive seniority. Had petitioner pointed to such a collective-bargaining agreement provision, or proved that its incumbent employees actually had objected to offers of retroactive seniority to Title VII claimants, the Court of Appeals would have considered those factors in determining whether the District Court abused its discretion in shaping Gaddis' and Starr's relief. 14 In any event, the Court's claim that offers of retroactive seniority would injure the rights of incumbent employees is vastly overstated. If an employer sued by a Title VII claimant could toll the accrual of backpay liability by making a unilateral offer that included some form of retroactive seniority, he still would have every incentive to make such an offer as soon as possible after the discriminatory act. The amount of retroactive seniority offered would necessarily be small, and the seniority rights of relatively few incumbent employees would be affected. Under the Court's approach, in contrast, employers will no longer have any incentive to offer retroactive seniority. Any awards of retroactive seniority to bona fide Title VII claimants will thus be court-ordered, and will be entered only after "the lengthy delays that too often attend Title VII litigation." Ante, at 221. By delaying awards of retroactive seniority until final judgment in a significant number of cases, the Court's approach ensures that the seniority rights of comparatively greater numbers of incumbent employees will be affected adversely. 15 Without embarrassment, the Court cites Rebecca Starr's testimony to support its argument that the Court of Appeals' "rule," and not its own new rule, is indifferent to the real-life concerns of victims of sex discrimination. See ante, at 230, n. 12. Under the Court of Appeals' "rule," however, Rebecca Starr was awarded full backpay as compensation for Ford's sex discrimination. Under this Court's rule, a large portion of Starr's compensation will simply be cut off. By claiming that the Court of Appeals was somehow more indifferent to Starr's real-life concerns, the Court only confirms how far removed from the real world it is.

12

458 U.S. 50 102 S.Ct. 2858 73 L.Ed.2d 598 NORTHERN PIPELINE CONSTRUCTION CO., Appellant,v.MARATHON PIPE LINE COMPANY and United States. UNITED STATES, Appellant, v. MARATHON PIPE LINE CO. et al. Nos. 81-150, 81-546. Argued April 27, 1982. Decided June 28, 1982. Judgment Stayed Oct. 4, 1982. See 459 U.S. 813, 103 S.Ct. 199, 200. Syllabus The Bankruptcy Act of 1978 (Act) established a United States bankruptcy court in each judicial district as an adjunct to the district court for such district. The bankruptcy court judges are appointed for 14-year terms, subject to removal by the judicial council of the circuit in which they serve on grounds of incompetence, misconduct, neglect of duty, or disability. Their salaries are set by statute and are subject to adjustment. The Act grants the bankruptcy courts jurisdiction over "all civil proceedings arising under title 11 [bankruptcy] [of the United States Code] or arising in or related to cases under title 11." See 28 U.S.C. § 1471(b) (1976 ed., Supp.IV). After it had filed a petition for reorganization in a Bankruptcy Court, appellant Northern Pipeline Construction Co. (Northern) filed in that court a suit against appellee Marathon Pipe Line Co. (Marathon) seeking damages for an alleged breach of contract and warranty, as well as for misrepresentation, coercion, and duress. Marathon sought dismissal of the suit on the ground that the Act unconstitutionally conferred Art. III judicial power upon judges who lacked life tenure and protection against salary diminution. The Bankruptcy Court denied the motion to dismiss, but on appeal the District Court granted the motion. Held: The judgment is affirmed. D.C., 12 B.R. 946, affirmed. Justice BRENNAN, joined by Justice MARSHALL, Justice BLACKMUN, and Justice STEVENS, concluded that: 1 1. Section 1471's broad grant of jurisdiction to bankruptcy judges violates Art. III. Pp. 57-87. 2 (a) The judicial power of the United States must be exercised by judges who have the attributes of life tenure and protection against salary diminution specified by Art. III. These attributes were incorporated into the Constitution to ensure the independence of the Judiciary from the control of the Executive and Legislative Branches. There is no doubt that bankruptcy judges created by the Act are not Art. III judges. Pp. 57-62. 3 (b) Article III bars Congress from establishing under its Art. I powers legislative courts to exercise jurisdiction over all matters arising under the bankruptcy laws. The establishment of such courts does not fall within any of the historically recognized situations—non-Art. III courts of the Territories or of the District of Columbia, courts-martial, and resolution of "public rights" issues—in which the principle of independent adjudication commanded by Art. III does not apply. The bankruptcy courts do not lie exclusively outside the States, like the courts of the Territories or of the District of Columbia, or bear any resemblance to courts-martial, nor can the substantive legal rights at issue in the present action—the right to recover contract damages to augment Northern's estate—be deemed "public rights." There is no persuasive reason in logic, history, or the Constitution, why bankruptcy courts lie beyond the reach of Art. III. Pp. 63-76. 4 (c) Section 1471 impermissibly removed most, if not all, of the essential attributes of the judicial power from the Art. III district court and vested those attributes in a non-Art. III adjunct. Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 and United States v. Raddatz, 447 U.S. 667, 100 S.Ct. 2406, 65 L.Ed.2d 424 distinguished. Congress does not have the same power to create adjuncts to adjudicate constitutionally recognized rights and state-created rights as it does to adjudicate rights that it creates. The grant of jurisdiction to bankruptcy courts cannot be sustained as an exercise of Congress' power to create adjuncts to Art. III courts. Pp. 76-87. 5 2. The above holding that the broad grant of jurisdiction in § 1471 is unconstitutional shall not apply retroactively but only prospectively. Such grant of jurisdiction presents an unprecedented question of interpretation of Art. III, and retroactive application would not further the operation of the holding but would visit substantial injustice and hardship upon those litigants who relied upon the Act's vesting of jurisdiction in the bankruptcy courts. Pp.87-88 6 Justice REHNQUIST, joined by Justice O'CONNOR, concluded that where appellee Marathon Pipe Line Co. has simply been named defendant in appellant Northern Pipeline Construction Co.'s suit on a contract claim arising under state law, the constitutionality of the Bankruptcy Court's exercise of jurisdiction over that kind of suit is all that need be decided in this case; that resolution of any objections Marathon might make to the exercise of authority conferred on bankruptcy courts by the Bankruptcy Act of 1978, on the ground that the suit must be decided by an Art. III court, should await the exercise of such authority; that so much of that Act as enables a Bankruptcy Court to entertain and decide Northern's suit over Marathon's objection violates Art. III; and that the Court's judgment should not be applied retroactively. Pp. 89-92. 7 Sol. Gen. Rex E. Lee, Washington, D. C., for United States. 8 John L. Devney, St. Paul, Minn., for Northern Pipeline Const. Co. 9 Melvin I. Orenstein, Minneapolis, Minn., for Marathon Pipe Line Co. 10 Justice BRENNAN announced the judgment of the Court and delivered an opinion in which Justice MARSHALL, Justice BLACKMUN, and Justice STEVENS joined. 11 The question presented is whether the assignment by Congress to bankruptcy judges of the jurisdiction granted in 28 U.S.C. § 1471 (1976 ed., Supp.IV) by § 241(a) of the Bankruptcy Act of 1978 violates Art. III of the Constitution. 12 * A. 13 In 1978, after almost 10 years of study and investigation, Congress enacted a comprehensive revision of the bankruptcy laws. The Bankruptcy Act of 1978 (Act)1 made significant changes in both the substantive and procedural law of bankruptcy. It is the changes in the latter that are at issue in this case. 14 Before the Act, federal district courts served as bankruptcy courts and employed a "referee" system. Bankruptcy proceedings were generally conducted before referees,2 except in those instances in which the district court elected to withdraw a case from a referee. See Bkrtcy. Rule 102. The referee's final order was appealable to the district court. Bkrtcy. Rule 801. The bankruptcy courts were vested with "summary jurisdiction"—that is, with jurisdiction over controversies involving property in the actual or constructive possession of the court. And, with consent, the bankruptcy court also had jurisdiction over some "plenary" matters—such as disputes involving property in the possession of a third person. 15 The Act eliminates the referee system and establishes "in each judicial district, as an adjunct to the district court for such district, a bankruptcy court which shall be a court of record known as the United States Bankruptcy Court for the district." 28 U.S.C. § 151(a) (1976 ed., Supp.IV). The judges of these courts are appointed to office for 14-year terms by the President, with the advice and consent of the Senate. §§ 152, 153(a) (1976 ed., Supp.IV). They are subject to removal by the "judicial council of the circuit" on account of "incompetency, misconduct, neglect of duty or physical or mental disability." § 153(b) (1976 ed., Supp.IV). In addition, the salaries of the bankruptcy judges are set by statute and are subject to adjustment under the Federal Salary Act, 2 U.S.C. §§ 351-361 (1976 ed. and Supp.IV). 28 U.S.C. § 154 (1976 ed., Supp.IV). 16 The jurisdiction of the bankruptcy courts created by the Act is much broader than that exercised under the former referee system. Eliminating the distinction between "summary" and "plenary" jurisdiction, the Act grants the new courts jurisdiction over all "civil proceedings arising under title 11 [the Bankruptcy title] or arising in or related to cases under title 11." 28 U.S.C. § 1471(b) (1976 ed., Supp.IV) (emphasis added).3 This jurisdictional grant empowers bankruptcy courts to entertain a wide variety of cases involving claims that may affect the property of the estate once a petition has been filed under Title 11. Included within the bankruptcy courts' jurisdiction are suits to recover accounts, controversies involving exempt property, actions to avoid transfers and payments as preferences or fraudulent conveyances, and causes of action owned by the debtor at the time of the petition for bankruptcy. The bankruptcy courts can hear claims based on state law as well as those based on federal law. See 1 W. Collier, Bankruptcy ¶ 3.01, pp. 3-47 to 3-48 (15th ed. 1982).4 17 The judges of the bankruptcy courts are vested with all of the "powers of a court of equity, law, and admiralty," except that they "may not enjoin another court or punish a criminal contempt not committed in the presence of the judge of the court or warranting a punishment of imprisonment." 28 U.S.C. § 1481 (1976 ed., Supp.IV). In addition to this broad grant of power, Congress has allowed bankruptcy judges the power to hold jury trials, § 1480; to issue declaratory judgments, § 2201; to issue writs of habeas corpus under certain circumstances, § 2256; to issue all writs necessary in aid of the bankruptcy court's expanded jurisdiction, § 451 (1976 ed. and Supp.IV); see 28 U.S.C. § 1651; and to issue any order, process or judgment that is necessary or appropriate to carry out the provisions of Title 11, 11 U.S.C. § 105(a) (1976 ed., Supp.IV). 18 The Act also establishes a special procedure for appeals from orders of bankruptcy courts. The circuit council is empowered to direct the chief judge of the circuit to designate panels of three bankruptcy judges to hear appeals. 28 U.S.C. § 160 (1976 ed., Supp.IV). These panels have jurisdiction of all appeals from final judgments, orders, and decrees of bankruptcy courts, and, with leave of the panel, of interlocutory appeals. § 1482. If no such appeals panel is designated, the district court is empowered to exercise appellate jurisdiction. § 1334. The court of appeals is given jurisdiction over appeals from the appellate panels or from the district court. § 1293. If the parties agree, a direct appeal to the court of appeals may be taken from a final judgment of a bankruptcy court. § 1293(b).5 19 The Act provides for a transition period before the new provisions take full effect in April 1984. §§ 401-411, 92 Stat. 2682-2688. During the transition period, previously existing bankruptcy courts continue in existence. § 404(a), 92 Stat. 2683. Incumbent bankruptcy referees, who served 6-year terms for compensation subject to adjustment by Congress, are to serve as bankruptcy judges until March 31, 1984, or until their successors take office. § 404(b), 92 Stat. 2683.6 During this period they are empowered to exercise essentially all of the jurisdiction and powers discussed above. See §§ 404, 405, 92 Stat. 2683-2685. See generally 1 Collier, supra, &Par; 7.04-7.05, pp. 7-23 to 7-65. The procedure for taking appeals is similar to that provided after the transition period. See § 405(c)(1), 92 Stat. 2685.7 B 20 This case arises out of proceedings initiated in the United States Bankruptcy Court for the District of Minnesota after appellant Northern Pipeline Construction Co. (Northern) filed a petition for reorganization in January 1980. In March 1980 Northern, pursuant to the Act, filed in that court a suit against appellee Marathon Pipe Line Co. (Marathon). Appellant sought damages for alleged breaches of contract and warranty, as well as for alleged misrepresentation, coercion, and duress. Marathon sought dismissal of the suit, on the ground that the Act unconstitutionally conferred Art. III judicial power upon judges who lacked life tenure and protection against salary diminution. The United States intervened to defend the validity of the statute. 21 The Bankruptcy Judge denied the motion to dismiss. 6 B.R. 928 (1980). But on appeal the District Court entered an order granting the motion, on the ground that "the delegation of authority in 28 U.S.C. § 1471 to the Bankruptcy Judges to try cases which are otherwise relegated under the Constitution to Article III judges" was unconstitutional. Both the United States and Northern filed notices of appeal in this Court.8 We noted probable jurisdiction. 454 U.S. 1029, 102 S.Ct. 564, 70 L.Ed.2d 472 (1981).9 II A. 22 Basic to the constitutional structure established by the Framers was their recognition that "[t]he accumulation of all powers, legislative, executive, and judiciary, in the same hands, whether of one, a few, or many, and whether hereditary, self-appointed, or elective, may justly be pronounced the very definition of tyranny." The Federalist No. 47, p. 300 (H. Lodge ed. 1888) (J. Madison). To ensure against such tyranny, the Framers provided that the Federal Government would consist of three distinct Branches, each to exercise one of the governmental powers recognized by the Framers as inherently distinct. "The Framers regarded the checks and balances that they had built into the tripartite Federal Government as a self-executing safeguard against the encroachment or aggrandizement of one branch at the expense of the other." Buckley v. Valeo, 424 U.S. 1, 122, 96 S.Ct. 612, 683, 46 L.Ed.2d 659 (1976) (per curiam ). 23 The Federal Judiciary was therefore designed by the Framers to stand independent of the Executive and Legislature—to maintain the checks and balances of the constitutional structure, and also to guarantee that the process of adjudication itself remained impartial. Hamilton explained the importance of an independent Judiciary: 24 "Periodical appointments, however regulated, or by whomsoever made, would, in some way or other, be fatal to [the courts'] necessary independence. If the power of making them was committed either to the Executive or legislature, there would be danger of an improper complaisance to the branch which possessed it; if to both, there would be an unwillingness to hazard the displeasure of either; if to the people, or to persons chosen by them for the special purpose, there would be too great a disposition to consult popularity, to justify a reliance that nothing would be consulted but the Constitution and the laws." The Federalist No. 78, p. 489 (H. Lodge ed. 1888). 25 The Court has only recently reaffirmed the significance of this feature of the Framers' design: "A Judiciary free from control by the Executive and Legislature is essential if there is a right to have claims decided by judges who are free from potential domination by other branches of government." United States v. Will, 449 U.S. 200, 217-218, 101 S.Ct. 471, 481-482, 66 L.Ed.2d 392 (1980). 26 As an inseparable element of the constitutional system of checks and balances, and as a guarantee of judicial impartiality, Art. III both defines the power and protects the independence of the Judicial Branch. It provides that "The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish." Art. III, § 1. The inexorable command of this provision is clear and definites The judicial power of the United States must be exercised by courts having the attributes prescribed in Art. III. Those attributes are also clearly set forth: 27 "The Judges, both of the supreme and inferior Courts, shall hold their Offices during good Behaviour, and shall, at stated Times, receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office." Art. III, § 1. 28 The "good Behaviour" Clause guarantees that Art. III judges shall enjoy life tenure, subject only to removal by impeachment. United States ex rel. Toth v. Quarles, 350 U.S. 11, 16, 76 S.Ct. 1, 4, 100 L.Ed. 8 (1955). The Compensation Clause guarantees Art. III judges a fixed and irreducible compensation for their services. United States v. Will, supra, 449 U.S., at 218-221, 101 S.Ct., at 482-483. Both of these provisions were incorporated into the Constitution to ensure the independence of the Judiciary from the control of the Executive and Legislative Branches of government.10 As we have only recently emphasized, "[t]he Compensation Clause has its roots in the longstanding Anglo-American tradition of an independent Judiciary," 449 U.S., at 217, 101 S.Ct. at 482, while the principle of life tenure can be traced back at least as far as the Act of Settlement in 1701, id., at 218, 101 S.Ct., at 482. To be sure, both principles were eroded during the late colonial period, but that departure did not escape notice and indignant rejection by the Revolutionary generation. Indeed, the guarantees eventually included in Art. III were clearly foreshadowed in the Declaration of Independence, "which, among the injuries and usurpations recited against the King of Great Britain, declared that he had 'made judges dependent on his will alone, for the tenure of their offices, and the amount and payment of their salaries.' " O'Donoghue v. United States, 289 U.S. 516, 531, 53 S.Ct. 740, 743, 77 L.Ed. 1356 (1933). The Framers thus recognized: 29 "Next to permanency in office, nothing can contribute more to the independence of the judges than a fixed provision for their support. . . . In the general course of human nature, a power over a man's subsistence amounts to a power over his will." The Federalist No. 79, p. 491 (H. Lodge ed. 1888) (A. Hamilton) (emphasis in original).11 30 In sum, our Constitution unambiguously enunciates a fundamental principle—that the "judicial Power of the United States" must be reposed in an independent Judiciary. It commands that the independence of the Judiciary be jealously guarded, and it provides clear institutional protections for that independence. B 31 It is undisputed that the bankruptcy judges whose offices were created by the Bankruptcy Act of 1978 do not enjoy the protections constitutionally afforded to Art. III judges. The bankruptcy judges do not serve for life subject to their continued "good Behaviour." Rather, they are appointed for 14-year terms, and can be removed by the judicial council of the circuit in which they serve on grounds of "incompetency, misconduct, neglect of duty, or physical or mental disability." Second, the salaries of the bankruptcy judges are not immune from diminution by Congress. See supra, at 53. In short, there is no doubt that the bankruptcy judges created by the Act are not Art. III judges. 32 That Congress chose to vest such broad jurisdiction in non-Art. III bankruptcy courts, after giving substantial consideration to the constitutionality of the Act, is of course reason to respect the congressional conclusion. See Fullilove v. Klutznick, 448 U.S. 448, 472-473, 100 S.Ct. 2758, 2771-2772, 65 L.Ed.2d 902 (1980) (opinion of BURGER, C. J.); Palmore v. United States, 411 U.S. 389, 409, 93 S.Ct. 1670, 1682, 36 L.Ed.2d 342 (1973). See also National Ins. Co. v. Tidewater Co., 337 U.S. 582, 655, 69 S.Ct. 1173, 1199, 93 L.Ed. 1556 (1949) (Frankfurter, J., dissenting).12 But at the same time, "[d]eciding whether a matter has in any measure been committed by the Constitution to another branch of government, or whether the action of that branch exceeds whatever authority has been committed, is itself a delicate exercise in constitutional interpretation, and is a responsibility of this Court as ultimate interpreter of the Constitution." Baker v. Carr, 369 U.S. 186, 211, 82 S.Ct. 691, 706, 7 L.Ed.2d 663 (1962). 33 With these principles in mind, we turn to the question presented for decision: whether the Bankruptcy Act of 1978 violates the command of Art. III that the judicial power of the United States must be vested in courts whose judges enjoy the protections and safeguards specified in that Article. 34 Appellants suggest two grounds for upholding the Act's conferral of broad adjudicative powers upon judges unprotected by Art. III. First, it is urged that "pursuant to its enumerated Article I powers, Congress may establish legislative courts that have jurisdiction to decide cases to which the Article III judicial power of the United States extends." Brief for United States 9. Referring to our precedents upholding the validity of "legislative courts," appellants suggest that "the plenary grants of power in Article I permit Congress to establish non-Article III tribunals in 'specialized areas having particularized needs and warranting distinctive treatment,' " such as the area of bankruptcy law. Ibid., quoting Palmore v. United States, supra, at 389, 408, 93 S.Ct., at 1681. Second, appellants contend that even if the Constitution does require that this bankruptcy-related action be adjudicated in an Art. III court, the Act in fact satisfies that requirement. "Bankruptcy jurisdiction was vested in the district court" of the judicial district in which the bankruptcy court is located, "and the exercise of that jurisdiction by the adjunct bankruptcy court was made subject to appeal as of right to an Article III court." Brief for United States 12. Analogizing the role of the bankruptcy court to that of a special master, appellants urge us to conclude that this "adjunct" system established by Congress satisfies the requirements of Art. III. We consider these arguments in turn. III 35 Congress did not constitute the bankruptcy courts as legislative courts.13 Appellants contend, however, that the bankruptcy courts could have been so constituted, and that as a result the "adjunct" system in fact chosen by Congress does not impermissibly encroach upon the judicial power. In advancing this argument, appellants rely upon cases in which we have identified certain matters that "congress may or may not bring within the cognizance of [Art. III courts], as it may deem proper." Murray's Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 284, 15 L.Ed. 372 (1856).14 But when properly understood, these precedents represent no broad departure from the constitutional command that the judicial power of the United States must be vested in Art. IIIcourts.15 Rather, they reduce to three narrow situations not subject to that command, each recognizing a circumstance in which the grant of power to the Legislative and Executive Branches was historically and constitutionally so exceptional that the congressional assertion of a power to create legislative courts was consistent with, rather than threatening to, the constitutional mandate of separation of powers. These precedents simply acknowledge that the literal command of Art. III, assigning the judicial power of the United States to courts insulated from Legislative or Executive interference, must be interpreted in light of the historical context in which the Constitution was written, and of the structural imperatives of the Constitution as a whole. 36 Appellants first rely upon a series of cases in which this Court has upheld the creation by Congress of non-Art. III "territorial courts." This exception from the general prescription of Art. III dates from the earliest days of the Republic, when it was perceived that the Framers intended that as to certain geographical areas, in which no State operated as sovereign, Congress was to exercise the general powers of government. For example, in American Ins. Co. v. Canter, 1 Pet. 511, 7 L.Ed. 242 (1828), the Court observed that Art. IV bestowed upon Congress alone a complete power of government over territories not within the States that constituted the United States. The Court then acknowledged Congress' authority to create courts for those territories that were not in conformity with Art. III. Such courts were 37 "created in virtue of the general right of sovereignty which exists in the government, or in virtue of that clause which enables Congress to make all needful rules and regulations, respecting the territory belonging to the United States. The jurisdiction with which they are invested . . . is conferred by Congress, in the execution of those general powers which that body possesses over the territories of the United States. Although admiralty jurisdiction can be exercised in the states in those Courts, only, which are established in pursuance of the third article of the Constitution; the same limitation does not extend to the territories. In legislating for them, Congress exercises the combined powers of the general, and of a state government." 1 Pet., at 546. 38 The Court followed the same reasoning when it reviewed Congress' creation of non-Art. III courts in the District of Columbia. It noted that there was in the District 39 "no division of powers between the general and state governments. Congress has the entire control over the district for every purpose of government; and it is reasonable to suppose, that in organizing a judicial department here, all judicial power necessary for the purposes of government would be vested in the courts of justice." Kendall v. United States, 12 Pet. 524, 619, 9 L.Ed. 1181 (1838).16 40 Appellants next advert to a second class of cases—those in which this Court has sustained the exercise by Congress and the Executive of the power to establish and administer courts-martial. The situation in these cases strongly resembles the situation with respect to territorial courts: It too involves a constitutional grant of power that has been historically understood as giving the political Branches of Government extraordinary control over the precise subject matter at issue. Article I, § 8, cls. 13, 14, confer upon Congress the power "[t]o provide and maintain a Navy," and "[t]o make Rules for the Government and Regulation of the land and naval Forces." The Fifth Amendment, which requires a presentment or indictment of a grand jury before a person may be held to answer for a capital or otherwise infamous crime, contains an express exception for "cases arising in the land or naval forces." And Art. II, § 2, cl. 1, provides that "The President shall be Commander in Chief of the Army and Navy of the United States, and of the Militia of the several States, when called into the actual Service of the United States." Noting these constitutional directives, the Court in Dynes v. Hoover, 20 How. 65, 15 L.Ed. 838 (1857), explained: 41 "These provisions show that Congress has the power to provide for the trial and punishment of military and naval offences in the manner then and now practiced by civilized nations; and that the power to do so is given without any connection between it and the 3d article of the Constitution defining the judicial power of the United States; indeed, that the two powers are entirely independent of each other." Id., at 79.17 Finally, appellants rely on a third group of cases, in which this Court has upheld the constitutionality of legislative courts and administrative agencies created by Congress to adjudicate cases involving "public rights."18 The "public rights" doctrine was first set forth in Murray's Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 15 L.Ed. 372 (1856): 42 "[W]e do not consider congress can either withdraw from judicial cognizance any matter which, from its nature, is the subject of a suit at the common law, or in equity, or admiralty; nor, on the other hand, can it bring under the judicial power a matter which, from its nature, is not a subject for judicial determination. At the same time there are matters, involving public rights, which may be presented in such form that the judicial power is capable of acting on them, and which are susceptible of judicial determination, but which congress may or may not bring within the cognizance of the courts of the United States, as it may deem proper." Id., at 284 (emphasis added). 43 This doctrine may be explained in part by reference to the traditional principle of sovereign immunity, which recognizes that the Government may attach conditions to its consent to be sued. See id., at 283-285; see also Ex parte Bakelite Corp., 279 U.S. 438, 452, 49 S.Ct. 411, 413, 73 L.Ed. 789 (1929). But the public-rights doctrine also draws upon the principle of separation of powers, and a historical understanding that certain prerogatives were reserved to the political Branches of Government. The doctrine extends only to matters arising "between the Government and persons subject to its authority in connection with the performance of the constitutional functions of the executive or legislative departments," Crowell v. Benson, 285 U.S. 22, 50, 52 S.Ct. 285, 292, 76 L.Ed. 598 (1932), and only to matters that historically could have been determined exclusively by those departments, see Ex parte Bakelite Corp., supra, 279 U.S., at 458, 49 S.Ct., at 416. The understanding of these cases is that the Framers expected that Congress would be free to commit such matters completely to nonjudicial executive determination, and that as a result there can be no constitutional objection to Congress' employing the less drastic expedient of committing their determination to a legislative court or an administrative agency. Crowell v. Benson, supra, 285 U.S., at 50, 52 S.Ct., at 292.19 44 The public-rights doctrine is grounded in a historically recognized distinction between matters that could be conclusively determined by the Executive and Legislative Branches and matters that are "inherently . . . judicial." Ex parte Bakelite Corp., supra, 279 U.S., at 458, 49 S.Ct., at 416. See Murray's Lessee v. Hoboken Land & Improvement Co., 18 How., at 280-282. For example, the Court in Murray's Lessee looked to the law of England and the States at the time the Constitution was adopted, in order to determine whether the issue presented was customarily cognizable in the courts. Ibid. Concluding that the matter had not traditionally been one for judicial determination, the Court perceived no bar to Congress' establishment of summary procedures, outside of Art. III courts, to collect a debt due to the Government from one of its customs agents.20 On the same premise, the Court in Ex parte Bakelite Corp., supra, held that the Court of Customs Appeals had been properly constituted by Congress as a legislative court: 45 "The full province of the court under the act creating it is that of determining matters arising between the Government and others in the executive administration and application of the customs laws. . . . The appeals include nothing which inherently or necessarily requires judicial determination, but only matters the determination of which may be, and at times has been, committed exclusively to executive officers." 279 U.S., at 458, 49 S.Ct., at 416 (emphasis added).21 46 The distinction between public rights and private rights has not been definitively explained in our precedents.22 Nor is it necessary to do so in the present cases, for it suffices to observe that a matter of public rights must at a minimum arise "between the government and others." Ex parte Bakelite Corp., supra, at 451, 49 S.Ct., at 413.23 In contrast, "the liability ofone individual to another under the law as defined," Crowell v. Benson, supra, at 51, 52 S.Ct., at 292, is a matter of private rights. Our precedents clearly establish that only controversies in the former category may be removed from Art. III courts and delegated to legislative courts or administrative agencies for their determination. See Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S. 442, 450, n. 7, 97 S.Ct. 1261, 1266, n. 7, 51 L.Ed.2d 464 (1977); Crowell v. Benson, supra, 285 U.S., at 50-51, 52 S.Ct., at 292. See also Katz, Federal Legislative Courts, 43 Harv.L.Rev. 894, 917-918 (1930).24 Private-rights disputes, on the other hand, lie at the core of the historically recognized judicial power. 47 In sum, this Court has identified three situations in which Art. III does not bar the creation of legislative courts. In each of these situations, the Court has recognized certain exceptional powers bestowed upon Congress by the Constitution or by historical consensus. Only in the face of such an exceptional grant of power has the Court declined to hold the authority of Congress subject to the general prescriptions of Art. III.25 48 We discern no such exceptional grant of power applicable in the cases before us. The courts created by the Bankruptcy Act of 1978 do not lie exclusively outside the States of the Federal Union, like those in the District of Columbia and the Territories. Nor do the bankruptcy courts bear any resemblance to courts-martial, which are founded upon the Constitution's grant of plenary authority over the Nation's military forces to the Legislative and Executive Branches. Finally, the substantive legal rights at issue in the present action cannot be deemed "public rights." Appellants argue that a discharge in bankruptcy is indeed a "public right," similar to such congressionally created benefits as "radio station licenses, pilot licenses, or certificates for common carriers" granted by administrative agencies. See Brief for United States 34. But the restructuring of debtor-creditor relations, which is at the core of the federal bankruptcy power, must be distinguished from the adjudication of state-created private rights, such as the right to recover contract damages that is at issue in this case. The former may well be a "public right," but the latter obviously is not. Appellant Northern's right to recover contract damages to augment its estate is "one of private right, that is, of the liability of one individual to another under the law as defined." Crowell v. Benson, 285 U.S., at 51, 52 S.Ct., at 292.26 49 Recognizing that the present cases may not fall within the scope of any of our prior cases permitting the establishment of legislative courts, appellants argue that we should recognize an additional situation beyond the command of Art. III, sufficiently broad to sustain the Act. Appellants contend that Congress' constitutional authority to establish "uniform Laws on the subject of Bankruptcies throughout the United States," Art. I, § 8, cl. 4, carries with it an inherent power to establish legislative courts capable of adjudicating "bankruptcy-related controversies." Brief for United States 14. In support of this argument, appellants rely primarily upon a quotation from the opinion in Palmore v. United States, 411 U.S. 389, 93 S.Ct. 1670, 36 L.Ed.2d 342 (1973), in which we stated that 50 "both Congress and this Court have recognized that . . . the requirements of Art. III, which are applicable where laws of national applicability and affairs of national concern are at stake, must in proper circumstances give way to accommodate plenary grants of power to Congress to legislate with respect to specialized areas having particularized needs and warranting distinctive treatment." Id., 407-408, 93 S.Ct., at 1681. 51 Appellants cite this language to support their proposition that a bankruptcy court created by Congress under its Art. I powers is constitutional, because the law of bankruptcy is a "specialized area," and Congress has found a "particularized need" that warrants "distinctive treatment." Brief for United States 20-33. 52 Appellants' contention, in essence, is that pursuant to any of its Art. I powers, Congress may create courts free of Art. III's requirements whenever it finds that course expedient. This contention has been rejected in previous cases. See, e.g., Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S., at 450, n. 7, 97 S.Ct., at 1266, n. 7; United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955). Although the cases relied upon by appellants demonstrate that independent courts are not required for all federal adjudications, those cases also make it clear that where Art. III does apply, all of the legislative powers specified in Art. I and elsewhere are subject to it. See, e.g., Ex parte Bakelite Corp., 279 U.S., at 449, 49 S.Ct., at 412; United States ex rel. Toth v. Quarles, supra; American Ins. Co. v. Canter, 1 Pet., at 546; Murray's Lessee, 18 How., at 284. Cf. Crowell v. Benson, supra, 285 U.S., at 51, 52 S.Ct., at 292. 53 The flaw in appellants' analysis is that it provides no limiting principle. It thus threatens to supplant completely our system of adjudication in independent Art. III tribunals and replace it with a system of "specialized" legislative courts. True, appellants argue that under their analysis Congress could create legislative courts pursuant only to some "specific" Art. I power, and "only when there is a particularized need for distinctive treatment." Brief for United States 22-23. They therefore assert that their analysis would not permit Congress to replace the independent Art. III Judiciary through a "wholesale assignment of federal judicial business to legislative courts." Ibid. But these "limitations" are wholly illusory. For example, Art. I, § 8, empowers Congress to enact laws, inter alia, regulating interstate commerce and punishing certain crimes. Art. I, § 8, cls. 3, 6. On appellants' reasoning Congress could provide for the adjudication of these and "related" matters by judges and courts within Congress' exclusive control.27 The potential for encroachment upon powers reserved to the Judicial Branch through the device of "specialized" legislative courts is dramatically evidenced in the jurisdiction granted to the courts created by the Act before us. The broad range of questions that can be brought into a bankruptcy court because they are "related to cases under title 11," 28 U.S.C. § 1471(b) (1976 ed., Supp.IV), see supra, at 54, is the clearest proof that even when Congress acts through a "specialized" court, and pursuant to only one of its many Art. I powers, appellants' analysis fails to provide any real protection against the erosion of Art. III jurisdiction by the unilateral action of the political Branches. In short, to accept appellants' reasoning, would require that we replace the principles delineated in our precedents, rooted in history and the Constitution, with a rule of broad legislative discretion that could effectively eviscerate the constitutional guarantee of an independent Judicial Branch of the Federal Government.28 54 Appellants' reliance upon Palmore for such broad legislative discretion is misplaced. In the context of the issue decided in that case, the language quoted from the Palmore opinion, supra, at 72, offers no substantial support for appellants' argument. Palmore was concerned with the courts of the District of Columbia, a unique federal enclave over which "Congress has . . . entire control . . . for every purpose of government." Kendall v. United States, 12 Pet., at 619. The "plenary authority" under the District of Columbia Clause, Art. I, § 8, cl. 17, was the subject of the quoted passage and the powers granted under that Clause are obviously different in kind from the other broad powers conferred on Congress: Congress' power over the District of Columbia encompasses the full authority of government, and thus, necessarily, the Executive and Judicial powers as well as the Legislative. This is a power that is clearly possessed by Congress only in limited geographic areas. Palmore itself makes this limitation clear. The quoted passage distinguishes the congressional powers at issue in Palmore from those in which the Art. III command of an independent Judiciary must be honored: where "laws of national applicability and affairs of national concern are at stake." 411 U.S., at 408, 93 S.Ct., at 1681. Laws respecting bankruptcy, like most laws enacted pursuant to the national powers cataloged in Art. I, § 8, are clearly laws of national applicability and affairs of national concern. Thus our reference in Palmore to "specialized areas having particularized needs" referred only to geographic areas, such as the District of Columbia or territories outside the States of the Federal Union. In light of the clear commands of Art. III, nothing held or said in Palmore can be taken to mean that in every area in which Congress may legislate, it may also create non-Art. III courts with Art. III powers. 55 In sum, Art. III bars Congress from establishing legislative courts to exercise jurisdiction over all matters related to those arising under the bankruptcy laws. The establishment of such courts does not fall within any of the historically recognized situations in which the general principle of independent adjudication commanded by Art. III does not apply. Nor can we discern any persuasive reason, in logic, history, or the Constitution, why the bankruptcy courts here established lie beyond the reach of Art. III. IV 56 Appellants advance a second argument for upholding the constitutionality of the Act: that "viewed within the entire judicial framework set up by Congress," the bankruptcy court is merely an "adjunct" to the district court, and that the delegation of certain adjudicative functions to the bankruptcy court is accordingly consistent with the principle that the judicial power of the United States must be vested in Art. III courts. See Brief for United States 11-13, 37-45. As support for their argument, appellants rely principally upon Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932), and United States v. Raddatz, 447 U.S. 667, 100 S.Ct. 2406, 65 L.Ed.2d 424 (1980), cases in which we approved the use of administrative agencies and magistrates as adjuncts to Art. III courts. Brief for United States 40-42. The question to which we turn, therefore, is whether the Act has retained "the essential attributes of the judicial power," Crowell v. Benson, supra, 285 U.S., at 51, 52 S.Ct., at 292, in Art. III tribunals.29 57 The essential premise underlying appellants' argument is that even where the Constitution denies Congress the power to establish legislative courts, Congress possesses the authority to assign certain factfinding functions to adjunct tribunals. It is, of course, true that while the power to adjudicate "private rights" must be vested in an Art. III court, see Part III, supra, 58 "this Court has accepted factfinding by an administrative agency, . . . as an adjunct to the Art. III court, analogizing the agency to a jury or a special master and permitting it in admiralty cases to perform the function of the special master. Crowell v. Benson, 285 U.S. 22, 5165 [52 S.Ct. 285, 292-298, 76 L.Ed. 598] (1932)." Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S., at 450, n. 7, 97 S.Ct., at 1266, n. 7. 59 The use of administrative agencies as adjuncts was first upheld in Crowell v. Benson, supra. The congressional scheme challenged in Crowell empowered an administrative agency, the United States Employees' Compensation Commission, to make initial factual determinations pursuant to a federal statute requiring employers to compensate their employees for work-related injuries occurring upon the navigable waters of the United States. The Court began its analysis by noting that the federal statute administered by the Compensation Commission provided for compensation of injured employees "irrespective of fault," and that the statute also prescribed a fixed and mandatory schedule of compensation. Id., 285 U.S., at 38, 52 S.Ct., at 287. The agency was thus left with the limited role of determining "questions of fact as to the circumstances, nature, extent and consequences of the injuries sustained by the employee for which compensation is to be made." Id., at 54, 52 S.Ct., at 293. The agency did not possess the power to enforce any of its compensation orders: On the contrary, every compensation order was appealable to the appropriate federal district court, which had the sole power to enforce it or set it aside, depending upon whether the court determined it to be "in accordance with law" and supported by evidence in the record. Id., at 44-45, 48, 52 S.Ct., at 289-290, 291. The Court found that in view of these limitations upon the Compensation Commission's functions and powers, its determinations were "closely analogous to findings of the amount of damages that are made, according to familiar practice, by commissioners or assessors." Id., at 54, 52 S.Ct., at 293. Observing that "there is no requirement that, in order to maintain the essential attributes of the judicial power, all determinations of fact in constitutional courts shall be made by judges," id., at 51, 52 S.Ct., at 292, the Court held that Art. III imposed no bar to the scheme enacted by Congress, id., at 54, 52 S.Ct., at 293. 60 Crowell involved the adjudication of congressionally created rights. But this Court has sustained the use of adjunct factfinders even in the adjudication of constitutional rights— so long as those adjuncts were subject to sufficient control by an Art. III district court. In United States v. Raddatz, supra, the Court upheld the 1978 Federal Magistrates Act, which permitted district court judges to refer certain pretrial motions, including suppression motions based on alleged violations of constitutional rights, to a magistrate for initial determination. The Court observed that the magistrate's proposed findings and recommendations were subject to de novo review by the district court, which was free to rehear the evidence or to call for additional evidence. Id., 447 U.S., at 676-677, 681-683, 100 S.Ct., at 2412-2413, 2415-2416. Moreover, it was noted that the magistrate considered motions only upon reference from the district court, and that the magistrates were appointed, and subject to removal, by the district court. Id., at 685, 100 S.Ct., at 2417 (BLACKMUN, J., concurring).30 In short, the ultimate decisionmaking authority respecting all pretrial motions clearly remained with the district court. Id., at 682, 100 S.Ct., at 2415. Under these circumstances, the Court held that the Act did not violate the constraints of Art. III. Id., at 683-684, 100 S.Ct., at 2416.31 Together these cases establish two principles that aid us in determining the extent to which Congress may constitutionally vest traditionally judicial functions in non-Art. III officers. First, it is clear that when Congress creates a substantive federal right, it possesses substantial discretion to prescribe the manner in which that right may be adjudicated—including the assignment to an adjunct of some functions historically performed by judges.32 Thus Crowell recognized that Art. III does not require "all determinations of fact [to] be made by judges," 285 U.S., at 51, 52 S.Ct., at 292; with respect to congressionally created rights, some factual determinations may be made by a specialized factfinding tribunal designed by Congress, without constitutional bar, id., at 54, 52 S.Ct., at 293. Second, the functions of the adjunct must be limited in such a way that "the essential attributes" of judicial power are retained in the Art. III court. Thus in upholding the adjunct scheme challenged in Crowell, the Court emphasized that "the reservation of full authority to the court to deal with matters of law provides for the appropriate exercise of the judicial function in this class of cases." Ibid. And in refusing to invalidate the Magistrates Act at issue in Raddatz, the Court stressed that under the congressional scheme " '[t]he authority and the responsibility—to make an informed, final determination . . . remains with the judge,' " 447 U.S., at 682, 100 S.Ct., at 2415, quoting Mathews v. Weber, 423 U.S. 261, 271, 96 S.Ct. 549, 554, 46 L.Ed.2d 483 (1976); the statute's delegation of power was therefore permissible, since "the ultimate decision is made by the district court," 447 U.S., at 683, 100 S.Ct., at 2416. 61 These two principles assist us in evaluating the "adjunct" scheme presented in these cases. Appellants assume that Congress' power to create "adjuncts" to consider all cases related to those arising under Title 11 is as great as it was in the circumstances of Crowell. But while Crowell certainly endorsed the proposition that Congress possesses broad discretion to assign factfinding functions to an adjunct created to aid in the adjudication of congressionally created statutory rights, Crowell does not support the further proposition necessary to appellants' argument—that Congress possesses the same degree of discretion in assigning traditionally judicial power to adjuncts engaged in the adjudication of rights not created by Congress. Indeed, the validity of this proposition was expressly denied in Crowell, when the Court rejected "the untenable assumption that the constitutional courts may be deprived in all cases of the determination of facts upon evidence even though a constitutional right may be involved," 285 U.S., at 60-61, 52 S.Ct., at 296 (emphasis added),33 and stated that 62 "the essential independence of the exercise of the judicial power of the United States in the enforcement of constitutional rights requires that the Federal court should determine . . . an issue [of agency jurisdiction] upon its own record and the facts elicited before it." Id., at 64, 52 S.Ct., at 297 (emphasis added).34 63 Appellants' proposition was also implicitly rejected in Raddatz. Congress' assignment of adjunct functions under the Federal Magistrates Act was substantially narrower than under the statute challenged in Crowell. Yet the Court's scrutiny of the adjunct scheme in Raddatz —which played a role in the adjudication of constitutional rights—was far stricter than it had been in Crowell. Critical to the Court's decision to uphold the Magistrates Act was the fact that the ultimate decision was made by the district court. 447 U.S., at 683, 100 S.Ct., at 2416. 64 Although Crowell and Raddatz do not explicitly distinguish between rights created by Congress and other rights, such a distinction underlies in part Crowell's and Raddatz' recognition of a critical difference between rights created by federal statute and rights recognized by the Constitution. Moreover, such a distinction seems to us to be necessary in light of the delicate accommodations required by the principle of separation of powers reflected in Art. III. The constitutional system of checks and balances is designed to guard against "encroachment or aggrandizement" by Congress at the expense of the other branches of government. Buckley v. Valeo, 424 U.S., at 122, 96 S.Ct., at 683. But when Congress creates a statutory right, it clearly has the discretion, in defining that right, to create presumptions, or assign burdens of proof, or prescribe remedies; it may also provide that persons seeking to vindicate that right must do so before particularized tribunals created to perform the specialized adjudicative tasks related to that right.35 Such provisions do, in a sense, affect the exercise of judicial power, but they are also incidental to Congress' power to define the right that it has created. No comparable justification exists, however, when the right being adjudicated is not of congressional creation. In such a situation, substantial inroads into functions that have traditionally been performed by the Judiciary cannot be characterized merely as incidental extensions of Congress' power to define rights that it has created. Rather, such inroads suggest unwarranted encroachments upon the judicial power of the United States, which our Constitution reserves for Art. III courts. 65 We hold that the Bankruptcy Act of 1978 carries the possibility of such an unwarranted encroachment. Many of the rights subject to adjudication by the Act's bankruptcy courts, like the rights implicated in Raddatz, are not of Congress' creation. Indeed, the cases before us, which center upon appellant Northern's claim for damages for breach of contract and misrepresentation, involve a right created by state law, a right independent of and antecedent to the reorganization petition that conferred jurisdiction upon the Bankruptcy Court.36 Accordingly, Congress' authority to control the manner in which that right is adjudicated, through assignment of historically judicial functions to a non-Art. III "adjunct," plainly must be deemed at a minimum. Yet it is equally plain that Congress has vested the "adjunct" bankruptcy judges with powers over Northern's state-created right that far exceed the powers that it has vested in administrative agencies that adjudicate only rights of Congress' own creation. 66 Unlike the administrative scheme that we reviewed in Crowell, the Act vests all "essential attributes" of the judicial power of the United States in the "adjunct" bankruptcy court. First, the agency in Crowell made only specialized, narrowly confined factual determinations regarding a particularized area of law. In contrast, the subject-matter jurisdiction of the bankruptcy courts encompasses not only traditional matters of bankruptcy, but also "all civil proceedings arising under title 11 or arising in or related to cases under title 11." 28 U.S.C. § 1471(c) (1976 ed., Supp.IV) (emphasis added). Second, while the agency in Crowell engaged in statutorily channeled factfinding functions, the bankruptcy courts exercise "all of the jurisdiction" conferred by the Act on the district courts, § 1471(c) (emphasis added). Third, the agency in Crowell possessed only a limited power to issue compensation orders pursuant to specialized procedures, and its orders could be enforced only by order of the district court. By contrast, the bankruptcy courts exercise all ordinary powers of district courts, including the power to preside over jury trials, 28 U.S.C. § 1480 (1976 ed., Supp.IV), the power to issue declaratory judgments, § 2201, the power to issue writs of habeas corpus, § 2256, and the power to issue any order, process, or judgment appropriate for the enforcement of the provisions of Title 11, 11 U.S.C. § 105(a) (1976 ed., Supp.IV).37 Fourth, while orders issued by the agency in Crowell were to be set aside if "not supported by the evidence," the judgments of the bankruptcy courts are apparently subject to review only under the more deferential "clearly erroneous" standard. See n. 5, supra. Finally, the agency in Crowell was required by law to seek enforcement of its compensation orders in the district court. In contrast, the bankruptcy courts issue final judgments, which are binding and enforceable even in the absence of an appeal.38 In short, the "adjunct" bankruptcy courts created by the Act exercise jurisdiction behind the facade of a grant to the district courts, and are exercising powers far greater than those lodged in the adjuncts approved in either Crowell or Raddatz.39 We conclude that 28 U.S.C. § 1471 (1976 ed., Supp.IV), as added by § 241(a) of the Bankruptcy Act of 1978, has impermissibly removed most, if not all, of "the essential attributes of the judicial power" from the Art. III district court, and has vested those attributes in a non-Art. III adjunct. Such a grant of jurisdiction cannot be sustained as an exercise of Congress' power to create adjuncts to Art. III courts. V 67 Having concluded that the broad grant of jurisdiction to the bankruptcy courts contained in 28 U.S.C. § 1471 (1976 ed., Supp.IV) is unconstitutional, we must now determine whether our holding should be applied retroactively to the effective date of the Act.40 Our decision in Chevron Oil Co. v. Huson, 404 U.S. 97, 92 S.Ct. 349, 30 L.Ed.2d 296 (1971), sets forth the three considerations recognized by our precedents as properly bearing upon the issue of retroactivity. They are, first, whether the holding in question "decid[ed] an issue of first impression whose resolution was not clearly foreshadowed" by earlier cases, id., at 106, 92 S.Ct., at 355; second, "whether retrospective operation will further or retard [the] operation" of the holding in question, id., at 107, 92 S.Ct., at 355; and third, whether retroactive application "could produce substantial inequitable results" in individual cases, ibid. In the present cases, all of these considerations militate against the retroactive application of our holding today. It is plain that Congress' broad grant of judicial power to non-Art. III bankruptcy judges presents an unprecedented question of interpretation of Art. III. It is equally plain that retroactive application would not further the operation of our holding, and would surely visit substantial injustice and hardship upon those litigants who relied upon the Act's vesting of jurisdiction in the bankruptcy courts. We hold, therefore, that our decision today shall apply only prospectively.41 68 The judgment of the District Court is affirmed. However, we stay our judgment until October 4, 1982. This limited stay will afford Congress an opportunity to reconstitute the bankruptcy courts or to adopt other valid means of adjudication, without impairing the interim administration of the bankruptcy laws. See Buckley v. Valeo, 424 U.S., at 143, 96 S.Ct., at 693; cf. Georgia v. United States, 411 U.S. 526, 541, 93 S.Ct. 1702, 1711, 36 L.Ed.2d 472 (1973); Fortson v. Morris, 385 U.S. 231, 235, 87 S.Ct. 446, 449, 17 L.Ed.2d 330 (1966); Maryland Committee for Fair Representation v. Tawes, 377 U.S. 656, 675-676, 84 S.Ct. 1429, 1439-1440, 12 L.Ed.2d 595 (1964). 69 It is so ordered. 70 Justice REHNQUIST, with whom Justice O'CONNOR joins, concurring in the judgment. 71 Were I to agree with the plurality that the question presented by these cases is "whether the assignment by Congress to bankruptcy judges of the jurisdiction granted in 28 U.S.C. § 1471 (1976 ed., Supp.IV) by § 241(a) of the Bankruptcy Act of 1978 violates Art. III of the Constitution," ante, at 52, I would with considerable reluctance embark on the duty of deciding this broad question. But appellee Marathon Pipe Line Co. has not been subjected to the full range of authority granted bankruptcy courts by § 1471. It was named as a defendant in a suit brought by appellant Northern Pipeline Construction Co. in a United States Bankruptcy Court. The suit sought damages for, inter alia, breaches of contract and warranty. Marathon moved to dismiss the action on the grounds that the Bankruptcy Act of 1978, which authorized the suit, violated Art. III of the Constitution insofar as it established bankruptcy judges whose tenure and salary protection do not conform to the requirements of Art. III. 72 With the cases in this posture, Marathon has simply been named defendant in a lawsuit about a contract, a lawsuit initiated by appellant Northern after having previously filed a petition for reorganization under the Bankruptcy Act. Marathon may object to proceeding further with this lawsuit on the grounds that if it is to be resolved by an agency of the United States, it may be resolved only by an agency which exercises "[t]he judicial power of the United States" described by Art. III of the Constitution. But resolution of any objections it may make on this ground to the exercise of a different authority conferred on bankruptcy courts by the 1978 Act, see ante, at 54-55, should await the exercise of such authority. 73 "This Court, as is the case with all federal courts, 'has no jurisdiction to pronounce any statute, either of a State or of the United States, void, because irreconcilable with the Constitution, except as it is called upon to adjudge the legal rights of litigants in actual controversies. In the exercise of that jurisdiction, it is bound by two rules, to which it has rigidly adhered, one, never to anticipate a question of constitutional law in advance of the necessity of deciding it; the other never to formulate a rule of constitutional law broader than is required by the precise facts to which it is to be applied.' Liverpool, New York & Philadelphia S.S. Co. v. Commissioners of Emigration, 113 U.S. 33, 39 [5 S.Ct. 352, 355, 28 L.Ed. 899]." United States v. Raines, 362 U.S. 17, 21, 80 S.Ct. 519, 522, 4 L.Ed.2d 524 (1960). 74 Particularly in an area of constitutional law such as that of "Art. III Courts," with its frequently arcane distinctions and confusing precedents, rigorous adherence to the principle that this Court should decide no more of a constitutional question than is absolutely necessary accords with both our decided cases and with sound judicial policy. 75 From the record before us, the lawsuit in which Marathon was named defendant seeks damages for breach of contract, misrepresentation, and other counts which are the stuff of the traditional actions at common law tried by the courts at Westminster in 1789. There is apparently no federal rule of decision provided for any of the issues in the lawsuit; the claims of Northern arise entirely under state law. No method of adjudication is hinted, other than the traditional common-law mode of judge and jury. The lawsuit is before the Bankruptcy Court only because the plaintiff has previously filed a petition for reorganization in that court. 76 The cases dealing with the authority of Congress to create courts other than by use of its power under Art. III do not admit of easy synthesis. In the interval of nearly 150 years between American Insurance Co. v. Canter, 1 Pet. 511, 7 L.Ed. 242 (1828), and Palmore v. United States, 411 U.S. 389, 93 S.Ct. 1670, 36 L.Ed.2d 342 (1973), the Court addressed the question infrequently. I need not decide whether these cases in fact support a general proposition and three tidy exceptions, as the plurality believes, or whether instead they are but landmarks on a judicial "darkling plain" where ignorant armies have clashed by night, as JUSTICE WHITE apparently believes them to be. None of the cases has gone so far as to sanction the type of adjudication to which Marathon will be subjected against its will under the provisions of the 1978 Act. To whatever extent different powers granted under that Act might be sustained under the "public rights" doctrine of Murray's Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 15 L.Ed. 372 (1856), and succeeding cases, I am satisfied that the adjudication of Northern's lawsuit cannot be so sustained. 77 I am likewise of the opinion that the extent of review by Art. III courts provided on appeal from a decision of the bankruptcy court in a case such as Northern's does not save the grant of authority to the latter under the rule espoused in Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932). All matters of fact and law in whatever domains of the law to which the parties' dispute may lead are to be resolved by the bankruptcy court in the first instance, with only traditional appellate review by Art. III courts apparently contemplated. Acting in this manner the bankruptcy court is not an "adjunct" of either the district court or the court of appeals. 78 I would, therefore, hold so much of the Bankruptcy Act of 1978 as enables a Bankruptcy Court to entertain and decide Northern's lawsuit over Marathon's objection to be violative of Art. III of the United States Constitution. Because I agree with the plurality that this grant of authority is not readily severable from the remaining grant of authority to bankruptcy courts under § 1471, see ante, at 87-88, n. 40, I concur in the judgment. I also agree with the discussion in Part V of the plurality opinion respecting retroactivity and the staying of the judgment of this Court. 79 Chief Justice BURGER, dissenting. 80 I join Justice WHITE's dissenting opinion, but I write separately to emphasize that, notwithstanding the plurality opinion, the Court does not hold today that Congress' broad grant of jurisdiction to the new bankruptcy courts is generally inconsistent with Art. III of the Constitution. Rather, the Court's holding is limited to the proposition stated by Justice REHNQUIST in his concurrence in the judgment—that a "traditional" state common-law action, not made subject to a federal rule of decision, and related only peripherally to an adjudication of bankruptcy under federal law, must, absent the consent of the litigants, be heard by an "Art. III court" if it is to be heard by any court or agency of the United States. This limited holding, of course, does not suggest that there is something inherently unconstitutional about the new bankruptcy courts; nor does it preclude such courts from adjudicating all but a relatively narrow category of claims "arising under" or "arising in or related to cases under" the Bankruptcy Act. 81 It will not be necessary for Congress, in order to meet the requirements of the Court's holding, to undertake a radical restructuring of the present system of bankruptcy adjudication. The problems arising from today's judgment can be resolved simply by providing that ancillary common-law actions, such as the one involved in these cases, be routed to the United States district court of which the bankruptcy court is an adjunct. 82 Justice WHITE, with whom THE CHIEF JUSTICE and Justice POWELL join, dissenting. 83 Article III, § 1, of the Constitution is straightforward and uncomplicated on its face: "The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish. The Judges, both of the supreme and inferior Courts, shall hold their Offices during good Behaviour, and shall at stated Times, receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office." 84 Any reader could easily take this provision to mean that although Congress was free to establish such lower courts as it saw fit, any court that it did establish would be an "inferior" court exercising "judicial Power of the United States" and so must be manned by judges possessing both life tenure and a guaranteed minimal income. This would be an eminently sensible reading and one that, as the plurality shows, is well founded in both the documentary sources and the political doctrine of separation of powers that stands behind much of our constitutional structure. Ante, at 57-60. 85 If this simple reading were correct and we were free to disregard 150 years of history, these would be easy cases and the plurality opinion could end with its observation that "[i]t is undisputed that the bankruptcy judges whose offices were created by the Bankruptcy Act of 1978 do not enjoy the protections constitutionally afforded to Art. III judges." Ante, at 60. The fact that the plurality must go on to deal with what has been characterized as one of the most confusing and controversial areas of constitutional law1 itself indicates the gross oversimplification implicit in the plurality's claim that "our Constitution unambiguously enunciates a fundamental principle—that the 'judicial Power of the United States' must be reposed in an independent Judiciary [and] provides clear institutional protections for that independence." Ibid. While this is fine rhetoric, analytically it serves only to put a distracting and superficial gloss on a difficult question. 86 That question is what limits Art. III places on Congress' ability to create adjudicative institutions designed to carry out federal policy established pursuant to the substantive authority given Congress elsewhere in the Constitution. Whether fortunate or unfortunate, at this point in the history of constitutional law that question can no longer be answered by looking only to the constitutional text. This Court's cases construing that text must also be considered. In its attempt to pigeonhole these cases, the plurality does violence to their meaning and creates an artificial structure that itself lacks coherence. 87 * There are, I believe, two separate grounds for today's decision. First, non-Art. III judges, regardless of whether they are labeled "adjuncts" to Art. III courts or "Art. I judges," may consider only controversies arising out of federal law. Because the immediate controversy in these cases—Northern Pipeline's claim against Marathon—arises out of state law, it may only be adjudicated, within the federal system, by an Art. III court.2 Second, regardless of the source of law that governs the controversy, Congress is prohibited by Art. III from establishing Art. I courts, with three narrow exceptions. Adjudication of bankruptcy proceedings does not fall within any of these exceptions. I shall deal with the first of these contentions in this section. 88 The plurality concedes that Congress may provide for initial adjudications by Art. I courts or administrative judges of all rights and duties arising under otherwise valid federal laws. Ante, at 80. There is no apparent reason why this principle should not extend to matters arising in federal bankruptcy proceedings. The plurality attempts to escape the reach of prior decisions by contending that the bankrupt's claim against Marathon arose under state law. Non-Article III judges, in its view, cannot be vested with authority to adjudicate such issues. It then proceeds to strike down 28 U.S.C. § 1471 (1976 ed., Supp.IV) on this ground. For several reasons, the Court's judgment is unsupportable. 89 First, clearly this ground alone cannot support the Court's invalidation of § 1471 on its face. The plurality concedes that in adjudications and discharges in bankruptcy, "the restructuring of debtor-creditor relations, which is at the core of the federal bankruptcy power," ante, at 71, and "the manner in which the rights of debtors and creditors are adjusted," ante, at 84, n. 36, are matters of federal law. Under the plurality's own interpretation of the cases, therefore, these matters could be heard and decided by Art. I judges. But because the bankruptcy judge is also given authority to hear a case like that of appellant Northern against Marathon, which the Court says is founded on state law, the Court holds that the section must be stricken down on its face. This is a grossly unwarranted emasculation of the scheme Congress has adopted. Even if the Court is correct that such a state-law claim cannot be heard by a bankruptcy judge, there is no basis for doing more than declaring the section unconstitutional as applied to the claim against Marathon, leaving the section otherwise intact. In that event, cases such as these would have to be heard by Art. III judges or by state courts—unless the defendant consents to suit before the bankruptcy judge—just as they were before the 1978 Act was adopted. But this would remove from the jurisdiction of the bankruptcy judge only a tiny fraction of the cases he is now empowered to adjudicate and would not otherwise limit his jurisdiction.3 90 Second, the distinction between claims based on state law and those based on federal law disregards the real character of bankruptcy proceedings. The routine in ordinary bankruptcy cases now, as it was before 1978, is to stay actions against the bankrupt, collect the bankrupt's assets, require creditors to file claims or be forever barred, allow or disallow claims that are filed, adjudicate preferences and fraudulent transfers, and make pro rata distributions to creditors, who will be barred by the discharge from taking further actions against the bankrupt. The crucial point to be made is that in the ordinary bankruptcy proceeding the great bulk of creditor claims are claims that have accrued under state law prior to bankruptcy—claims for goods sold, wages, rent, utilities, and the like. "[T]he word debt as used by the Act is not confined to its technical common law meaning but . . . extends to liabilities arising out of breach of contract . . . to torts . . . and to taxes owing to the United States or state or local governments." 1 W. Collier, Bankruptcy ¶ 1.14, p. 88 (14th ed. 1976). Every such claim must be filed and its validity is subject to adjudication by the bankruptcy court. The existence and validity of such claims recurringly depend on state law. Hence, the bankruptcy judge is constantly enmeshed in state-law issues. 91 The new aspect of the Bankruptcy Act of 1978, in this regard, therefore, is not the extension of federal jurisdiction to state law claims, but its extension to particular kinds of state-law claims, such as contract cases against third parties or disputes involving property in the possession of a third person.4 Prior to 1978, a claim of a bankrupt against a third party, such as the claim against Marathon in this case, was not within the jurisdiction of the bankruptcy judge. The old limits were based, of course, on the restrictions implicit within the concept of in rem jurisdiction; the new extension is based on the concept of in personam jurisdiction. "The bankruptcy court is given in personam jurisdiction as well as in rem jurisdiction to handle everything that arises in a bankruptcy case." H.R.Rep.No. 95-595, p. 445 (1977), U.S.Code Cong. & Admin.News 1978, p. 6400. The difference between the new and old Acts, therefore, is not to be found in a distinction between state-law and federal-law matters; rather, it is in a distinction between in rem and in personam jurisdiction. The majority at no place explains why this distinction should have constitutional implications. 92 Third, all that can be left of the majority's argument in this regard is that state-law claims adjudicated within the federal system must be heard in the first instance by Art. III judges. I shall argue below that any such attempt to distinguish Art. I from Art. III courts by the character of the controversies they may adjudicate fundamentally misunderstands the historical and constitutional significance of Art. I courts. Initially, however, the majority's proposal seems to turn the separation-of-powers doctrine, upon which the majority relies, on its head: Since state-law claims would ordinarily not be heard by Art. III judges—i.e., they would be heard by state judges—one would think that there is little danger of a diminution of, or intrusion upon, the power of Art. III courts, when such claims are assigned to a non-Art. III court. The plurality misses this obvious point because it concentrates on explaining how it is that federally created rights can ever be adjudicated in Art. I courts a far more difficult problem under the separation-of-powers doctrine. The plurality fumbles when it assumes that the rationale it develops to deal with the latter problem must also govern the former problem. In fact, the two are simply unrelated and the majority never really explains the separation-of-powers problem that would be created by assigning state law questions to legislative courts or to adjuncts of Art. III courts. 93 One need not contemplate the intricacies of the separation-of-powers doctrine, however, to realize that the majority's position on adjudication of state-law claims is based on an abstract theory that has little to do with the reality of bankruptcy proceedings. Even prior to the present Act, bankruptcy cases were generally referred to bankruptcy judges, previously called referees. Bkrtcy.Rule 102(a). Title 11 U.S.C. § 66 described the jurisdiction of the referees. Their powers included the authority to 94 "consider all petitions referred to them and make the adjudications or dismiss the petitions . . . grant, deny or revoke discharges, determine the dischargeability of debts, and render judgments thereon [and] perform such of the duties as are by this title conferred on courts of bankruptcy, including those incidental to ancillary jurisdiction, and as shall be prescribed by rules or orders of the courts of bankruptcy of their respective districts, except as herein otherwise provided." The bankruptcy judge possessed "complete jurisdiction of the proceedings." 1 W. Collier, Bankruptcy ¶ 1.09, p. 65 (14th ed. 1976). The referee would initially hear and decide practically all matters arising in the proceedings, including the allowance and disallowance of the claims of creditors.5 If a claim was disallowed by the bankruptcy judge and the decision was not reversed on appeal, the creditor was forever barred from further action against the bankrupt. As pointed out above, all of these matters could and usually did involve state-law issues. Initial adjudication of state-law issues by non-Art. III judges is, then, hardly a new aspect of the 1978 Act. 95 Furthermore, I take it that the Court does not condemn as inconsistent with Art. III the assignment of these functions—i.e., those within the summary jurisdiction of the old bankruptcy courts to a non-Art. III judge, since, as the plurality says, they lie at the core of the federal bankruptcy power. Ante, at 71. They also happen to be functions that have been performed by referees or bankruptcy judges for a very long time and without constitutional objection. Indeed, we approved the authority of the referee to allow or disallow claims in Katchen v. Landy, 382 U.S. 323, 86 S.Ct. 467, 15 L.Ed.2d 391 (1966). There, the referee held that a creditor had received a preference and that his claim could therefore not be allowed. We agreed that the referee had the authority not only to adjudicate the existence of the preference, but also to order that the preference be disgorged. We also recognized that the referee could adjudicate counterclaims against a creditor who files his claim against the estate. The 1973 Bankruptcy Rules make similar provision. See Rule 306(c), Rule 701, and Advisory Committee Note to Rule 701, 11 U.S.C., p. 1340. Hence, if Marathon had filed a claim against the bankrupt in this case, the trustee could have filed and the bankruptcy judge could have adjudicated a counterclaim seeking the relief that is involved in these cases. 96 Of course, all such adjudications by a bankruptcy judge or referee were subject to review in the district court, on the record. See 11 U.S.C. § 67(c). Bankruptcy Rule 810, transmitted to Congress by this Court, provided that the district court "shall accept the referee's findings of fact unless they are clearly-erroneous." As the plurality recognizes, ante, at 55, the 1978 Act provides for appellate review in Art. III courts and presumably under the same "clearly erroneous standard." In other words, under both the old and new Acts, initial determinations of state-law questions were to be made by non-Art. III judges, subject to review by Art. III judges. Why the differences in the provisions for appeal in the two Acts are of unconstitutional dimension remains entirely unclear. 97 In theory and fact, therefore, I can find no basis for that part of the majority's argument that rests on the state-law character of the claim involved here. Even if, prior to 1978, the referee could not generally participate in cases aimed at collecting the assets of a bankrupt estate, he nevertheless repeatedly adjudicated issues controlled by state law. There is very little reason to strike down § 1471 on its face on the ground that it extends, in a comparatively minimal way, the referees' authority to deal with state-law questions. To do so is to lose all sense of proportion. II 98 The plurality unpersuasively attempts to bolster its case for facial invalidity by asserting that the bankruptcy courts are now "exercising powers far greater than those lodged in the adjuncts approved in either Crowell or Raddatz." Ante, at 86. In support of this proposition it makes five arguments in addition to the "state-law" issue. Preliminarily, I see no basis for according standing to Marathon to raise any of these additional points. The state-law objection applies to the Marathon case. Only that objection should now be adjudicated.6 99 I also believe that the major premise of the plurality's argument is wholly unsupported: There is no explanation of why Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932), and United States v. Raddatz, 447 U.S. 667, 100 S.Ct. 2406, 65 L.Ed.2d 424 (1980), define the outer limits of constitutional authority. Much more relevant to today's decision are, first, the practice in bankruptcy prior to 1978, which neither the majority nor any authoritative case has questioned, and, second, the practice of today's administrative agencies. Considered from this perspective, all of the plurality's arguments are unsupportable abstractions, divorced from the realities of modern practice. 100 The first three arguments offered by the plurality, ante, at 85, focus on the narrowly defined task and authority of the agency considered in Crowell: The agency made only "specialized, narrowly confined factual determinations" and could issue only a narrow class of orders. Regardless of whether this was true of the Compensation Board at issue in Crowell, it certainly was not true of the old bankruptcy courts, nor does it even vaguely resemble current administrative practice. As I have already said, general references to bankruptcy judges, which was the usual practice prior to 1978, permitted bankruptcy judges to perform almost all of the functions of a bankruptcy court. Referees or bankruptcy judges not only exercised summary jurisdiction but could also conduct adversary proceedings to 101 "(1) recover money or property . . . . (2) determine the validity, priority, or extent of a lien or other interest in property, (3) sell property free of a lien or other interest for which the holder can be compelled to make a money satisfaction, (4) object to or revoke a discharge, (5) obtain an injunction, (6) obtain relief from a stay . . . (7) determine the dischargeability of a debt." Bkrtcy.Rule 701. 102 Although there were some exceptions to the referees' authority, which have been removed by the 1978 Act, the additions to the jurisdiction of the bankruptcy judges were of marginal significance when examined in the light of the overall functions of those judges before and after 1978. In my view, those changes are not sufficient to work a qualitative change in the character of the bankruptcy judge. 103 The plurality's fourth argument fails to point to any difference between the new and old Bankruptcy Acts. While the administrative orders in Crowell may have been set aside by a court if "not supported by the evidence," under both the new and old Acts at issue here, orders of the bankruptcy judge are reviewed under the "clearly-erroneous standard." See Bkrtcy.Rule 810. Indeed, judicial review of the orders of bankruptcy judges is more stringent than that of many modern administrative agencies. Generally courts are not free to set aside the findings of administrative agencies, if supported by substantial evidence. But more importantly, courts are also admonished to give substantial deference to the agency's interpretation of the statute it is enforcing. No such deference is required with respect to decisions on the law made by bankruptcy judges. 104 Finally, the plurality suggests that, unlike the agency considered in Crowell, the orders of a post-1978 bankruptcy judge are final and binding even though not appealed. Ante, at 85-86. To attribute any constitutional significance to this, unless the plurality intends to throw into question a large body of administrative law, is strange. More directly, this simply does not represent any change in bankruptcy practice. It was hornbook law prior to 1978 that the authorized judgments and orders of referees, including turnover orders, were final and binding and res judicata unless appealed and overturned: 105 "The practice before the referee should not differ from that before the judge of the court of bankruptcy and, apart from direct review within the limitation of § 39(c), the orders of the referee are entitled to the same presumption of validity, conclusiveness and recognition in the court of bankruptcy or other courts." 1 W. Collier, Bankruptcy ¶ 1.09, pp. 65-66 (14th ed. 1976). 106 Even if there are specific powers now vested in bankruptcy judges that should be performed by Art. III judges, the great bulk of their functions are unexceptionable and should be left intact. Whatever is invalid should be declared to be such; the rest of the 1978 Act should be left alone. I can account for the majority's inexplicably heavy hand in this case only by assuming that the Court has once again lost its conceptual bearings when confronted with the difficult problem of the nature and role of Art. I courts. To that question I now turn. III A. 107 The plurality contends that the precedents upholding Art. I courts can be reduced to three categories. First, there are territorial courts, which need not satisfy Art. III constraints because "the Framers intended that as to certain geographical areas . . . Congress was to exercise the general powers of government."7 Ante, at 64. Second, there are courts martial, which are exempt from Art. III limits because of a constitutional grant of power that has been "historically understood as giving the political Branches of Government extraordinary control over the precise subject matter at issue." Ante, at 66. Finally, there are those legislative courts and administrative agencies that adjudicate cases involving public rights—controversies between the Government and private parties—which are not covered by Art. III because the controversy could have been resolved by the executive alone without judicial review. See ante, at 68. Despite the plurality's attempt to cabin the domain of Art. I courts, it is quite unrealistic to consider these to be only three "narrow," ante, at 64, limitations on or exceptions to the reach of Art. III. In fact, the plurality itself breaks the mold in its discussion of "adjuncts" in Part IV, when it announces that "when Congress creates a substantive federal right, it possesses substantial discretion to prescribe the manner in which that right may be adjudicated." Ante, at 80. Ajudications of federal rights may, according to the plurality, be committed to administrative agencies, as long as provision is made for judicial review. 108 The first principle introduced by the plurality is geographical: Art. I courts presumably are not permitted within the States.8 The problem, of course, is that both of the other exceptions recognize that Art. I courts can indeed operate within the States. The second category relies upon a new principle: Art. I courts are permissible in areas in which the Constitution grants Congress "extraordinary control over the precise subject matter." Ante, at 66. Preliminarily, I do not know how we are to distinguish those areas in which Congress' control is "extraordinary" from those in which it is not. Congress' power over the Armed Forces is established in Art. I, § 8, cls. 13, 14. There is nothing in those Clauses that creates congressional authority different in kind from the authority granted to legislate with respect to bankruptcy. But more importantly, in its third category, and in its treatment of "adjuncts," the plurality itself recognizes that Congress can create Art. I courts in virtually all the areas in which Congress is authorized to act, regardless of the quality of the constitutional grant of authority. At the same time, territorial courts or the courts of the District of Columbia, which are Art. I courts, adjudicate private, just as much as public or federal, rights. 109 Instead of telling us what it is Art. I courts can and cannot do, the plurality presents us with a list of Art. I courts. When we try to distinguish those courts from their Art. III counterparts, we find—apart from the obvious lack of Art. III judges—a series of nondistinctions. By the plurality's own admission, Art. I courts can operate throughout the country, they can adjudicate both private and public rights, and they can adjudicate matters arising from congressional actions in those areas in which congressional control is "extraordinary." I cannot distinguish this last category from the general "arising under" jurisdiction of Art. III courts. 110 The plurality opinion has the appearance of limiting Art. I courts only because it fails to add together the sum of its parts. Rather than limiting each other, the principles relied upon complement each other; together they cover virtually the whole domain of possible areas of adjudication. Without a unifying principle, the plurality's argument reduces to the proposition that because bankruptcy courts are not sufficiently like any of these three exceptions, they may not be either Art. I courts or adjuncts to Art. III courts. But we need to know why bankruptcy courts cannot qualify as Art. I courts in their own right. B 111 The plurality opinion is not the first unsuccessful attempt to articulate a principled ground by which to distinguish Article I from Article III courts. The concept of a legislative, or Article I, court was introduced by an opinion authored by Chief Justice Marshall. Not only did he create the concept, but at the same time he started the theoretical controversy that has ever since surrounded the concept: 112 "The Judges of the Superior Courts of Florida hold their offices for four years. These Courts, then, are not constitutional Courts, in which the judicial power conferred by the Constitution on the general government, can be deposited. They are incapable of receiving it. They are legislative Courts, created in virtue of the general right of sovereignty which exists in the government, or in virtue of that clause which enables Congress to make all needful rules and regulations, respecting the territory belonging to the United States. The jurisdiction with which they are invested, is not a part of that judicial power which is defined in the 3d article of the Constitution, but is conferred by Congress, in the execution of those general powers which that body possesses over the territories of the United States." American Insurance Co. v. Canter, 1 Pet. 511, 546, 7 L.Ed. 242 (1828). 113 The proposition was simple enough: Constitutional courts exercise the judicial power described in Art. III of the Constitution; legislative courts do not and cannot. 114 There were only two problems with this proposition. First, Canter itself involved a case in admiralty jurisdiction, which is specifically included within the "judicial power of the United States" delineated in Art. III. How, then, could the territorial court not be exercising Art. III judicial power? Second, and no less troubling, if the territorial courts could not exercise Art. III power, how could their decisions be subject to appellate review in Art. III courts, including this one, that can exercise only Art. III "judicial" power? Yet from early on this Court has exercised such appellate jurisdiction. Benner v. Porter, 9 How. 235, 243, 13 L.Ed. 119 (1850); Clinton v. Englebrecht, 13 Wall. 434, 20 L.Ed. 659 (1872); Reynolds v. United States, 98 U.S. 145, 154, 25 L.Ed. 244 (1879); United States v. Coe, 155 U.S. 76, 86, 15 S.Ct. 16, 19, 39 L.Ed. 76 (1894); Balzac v. Porto Rico, 258 U.S. 298, 312-313, 42 S.Ct. 343, 348-49, 66 L.Ed. 627 (1922). The attempt to understand the seemingly unexplainable was bound to generate "confusion and controversy." This analytic framework, however—the search for a principled distinction—has continued to burden the Court. 115 The first major elaboration on the Canter principle was in Murray's Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 15 L.Ed. 372 (1856). The plaintiff in that case argued that a proceeding against a customs collector for the collection of moneys claimed to be due to the United States was an exercise of "judicial power" and therefore had to be carried out by Art. III judges. The Court accepted this premise: "It must be admitted that, if the auditing of this account, and the ascertainment of its balance, and the issuing of this process, was an exercise of the judicial power of the United States, the proceeding was void; for the officers who performed these acts could exercise no part of that judicial power." Id., at 275. Having accepted this premise, the Court went on to delineate those matters which could be determined only by an Art. III court, i.e., those matters that fall within the nondelegable "judicial power" of the United States. The Court's response to this was twofold. First, it suggested that there are certain matters which are inherently "judicial": "[W]e do not consider congress can either withdraw from judicial cognizance any matter which, from its nature, is the subject of a suit at the common law, or in equity, or admiralty." Id., at 284. Second, it suggested that there is another class of issues that, depending upon the form in which Congress structures the decisionmaking process, may or may not fall within "the cognizance of the courts of the United States." Ibid. This latter category consisted of the so-called "public rights." Apparently, the idea was that Congress was free to structure the adjudication of "public rights" without regard to Art. III. 116 Having accepted the plaintiff's premise, it is hard to see how the Court could have taken too seriously its first contention. The Court presented no examples of such issues that are judicial "by nature" and simply failed to acknowledge that Art. I courts already sanctioned by the Court—e.g., territorial courts—were deciding such issues all the time. The second point, however, contains implicitly a critical insight; one that if openly acknowledged would have undermined the entire structure. That insight follows from the Court's earlier recognition that the term "judicial act" is broad enough to encompass all administrative action involving inquiry into facts and the application of law to those facts. Id., at 280. If administrative action can be characterized as "judicial" in nature, then obviously the Court's subsequent attempt to distinguish administrative from judicial action on the basis of the manner in which Congress structures the decision cannot succeed. There need be no Art. III court involvement in any adjudication of a "public right," which the majority now interprets as any civil matter arising between the Federal Government and a citizen. In that area, whether an issue is to be decided by an Art. III court depends, finally, on congressional intent. 117 Although Murray's Lessee implicitly undermined Chief Justice Marshall's suggestion that there is a difference in kind between the work of Art. I and that of Art. III courts, it did not contend that the Court must always defer to congressional desire in this regard. The Court considered the plaintiff's contention that removal of the issue from an Art. III court must be justified by "necessity." Although not entirely clear, the Court seems to have accepted this proposition: "[I]t seems to us that the just inference from the entire law is, that there was such a necessity for the warrant." Id., at 285.9 118 The Court in Murray's Lessee was precisely right: Whether an issue can be decided by a non-Art. III court does not depend upon the judicial or nonjudicial character of the issue, but on the will of Congress and the reasons Congress offers for not using an Art. III court. This insight, however, was completely disavowed in the next major case to consider the distinction between Art. I and Art. III courts, Ex parte Bakelite Corp., 279 U.S. 438, 49 S.Ct. 411, 73 L.Ed. 789 (1929), in which the Court concluded that the Court of Customs Appeals was a legislative court. The Court there directly embraced the principle also articulated in Murray's Lessee that Art. I courts may not consider any matter "which inherently or necessarily requires judicial determination," but only such matters as are "susceptible of legislative or executive determination." 279 U.S., at 453, 49 S.Ct., at 414. It then went on effectively to bury the critical insight of Murray's Lessee, labeling as "fallacious" any argument that "assumes that whether a court is of one class or the other depends on the intention of Congress, whereas the true test lies in the power under which the court was created and in the jurisdiction conferred." 279 U.S., at 459, 49 S.Ct., at 416.10 119 The distinction between public and private rights as the principle delineating the proper domains of legislative and constitutional courts respectively received its death blow, I had believed, in Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932). In that case, the Court approved an administrative scheme for the determination, in the first instance, of maritime employee compensation claims. Although acknowledging the framework set out in Murray's Lessee and Ex parte Bakelite Corp., the Court specifically distinguished the case before it: "The present case does not fall within the categories just described but is one of private right, that is, of the liability of one individual to another under the law as defined."11 285 U.S., at 51, 52 S.Ct., at 292. Nevertheless, the Court approved of the use of an Art. I adjudication mechanism on the new theory that "there is no requirement that, in order to maintain the essential attributes of the judicial power, all determinations of fact in constitutional courts shall be made by judges." Ibid. Article I courts could deal not only with public rights, but also, to an extent, with private rights. The Court now established a distinction between questions of fact and law: "[T]he reservation of full authority to the court to deal with matters of law provides for the appropriate exercise of the judicial function in this class of cases."12 Id., at 54, 52 S.Ct., at 293. 120 Whatever sense Crowell may have seemed to give to this subject was exceedingly short-lived. One year later, the Court returned to this subject, abandoning both the public/private and the fact/law distinction and replacing both with a simple literalism. In O'Donoghue v. United States, 289 U.S. 516, 53 S.Ct. 740, 77 L.Ed. 1356 (1933), considering the courts of the District of Columbia, and in Williams v. United States, 289 U.S. 553, 53 S.Ct. 751, 77 L.Ed. 1372 (1933), considering the Court of Claims, the Court adopted the principle that if a federal court exercises jurisdiction over cases of the type listed in Art. III, § 2, as falling within the "judicial power of the United States," then that court must be an Art. III court: 121 "The provision of this section of the article is that the 'judicial power shall extend' to the cases enumerated, and it logically follows that where jurisdiction over these cases is conferred upon the courts of the District, the judicial power, since they are capable of receiving it, is ipso facto, vested in such courts as inferior courts of the United States." O'Donoghue, supra, 289 U.S., at 545, 53 S.Ct., at 748.13 122 In order to apply this same principle and yet hold the Court of Claims to be a legislative court, the Court found it necessary in Williams, supra, to conclude that the phrase "Controversies to which the United States shall be a party" in Art. III must be read as if it said "controversies to which the United States shall be a party plaintiff or petitioner."14 123 By the time of the Williams decision, this area of the law was mystifying to say the least. What followed helped very little, if at all. In the next two major cases the Court could not agree internally on a majority position. In National Insurance Co. v. Tidewater Co., 337 U.S. 582, 69 S.Ct. 1173, 93 L.Ed. 1556 (1949), the Court upheld a statute giving federal district courts jurisdiction over suits between citizens of the District of Columbia and citizens of a State. A majority of the Court, however, rejected the plurality position that Congress had the authority to assign Art. I powers to Art. III courts, at least outside of the District of Columbia. Only Chief Justice Vinson in dissent reflected on the other side of this problem: whether Art. I courts could be assigned Art. III powers. He entirely disagreed with the conceptual basis for Williams and O'Donoghue, noting that to the extent that Art. I courts consider non-Art. III matters, appellate review by an Art. III court would be precluded. Or conversely, since appellate review is exercised by this Court over Art. I courts, Art. I courts must "exercise federal question jurisdiction." 337 U.S., at 643, 69 S.Ct., at 1208. Having gone this far, the Chief Justice was confronted with the obvious question of whether in fact "the distinction between constitutional and legislative courts is meaningless." Id., at 644, 69 S.Ct., at 1208-09. Although suggesting that outside of the Territories or the District of Columbia there may be some limits on assignment to Art. I courts of matters that fall within Art. III jurisdiction—apart from federal-question jurisdiction—for the most part the Chief Justice ended up relying on the good will of Congress: "[W]e cannot impute to Congress an intent now or in the future to transfer jurisdiction from constitutional to legislative courts for the purpose of emasculating the former." Ibid. 124 Another chapter in this somewhat dense history of a constitutional quandary was provided by Justice Harlan's plurality opinion in Glidden Co. v. Zdanok, 370 U.S. 530, 82 S.Ct. 1459, 8 L.Ed.2d 671 (1962), in which the Court, despite Bakelite and Williams —and relying on an Act of Congress enacted since those decisions—held the Court of Claims and the Court of Customs and Patent Appeals to be Art. III courts. Justice Harlan continued the process of intellectual repudiation begun by Chief Justice Vinson in Tidewater. First, it was clear to him that Chief Justice Marshall could not have meant what he said in Canter on the inability of Art. I courts to consider issues within the jurisdiction of Art. III courts: "Far from being 'incapable of receiving' federal-question jurisdiction, the territorial courts have long exercised a jurisdiction commensurate in this regard with that of the regular federal courts and have been subjected to the appellate jurisdiction of this Court precisely because they do so." 370 U.S., at 545, n. 13, 82 S.Ct., at 1470, n. 13. Second, exceptions to the requirements of Art. III, he thought, have not been founded on any principled distinction between Art. I issues and Art. III issues; rather, a "confluence of practical considerations," id., at 547, 82 S.Ct., at 1471, accounts for this Court's sanctioning of Art. I courts: 125 "The touchstone of decision in all these cases has been the need to exercise the jurisdiction then and there and for a transitory period. Whether constitutional limitations on the exercise of judicial power have been held inapplicable has depended on the particular local setting, the practical necessities, and the possible alternatives." Id., at 547-548, 82 S.Ct., at 1471-72. 126 Finally, recognizing that there is frequently no way to distinguish Art. I and Art. III courts on the basis of the work they do, Justice Harlan suggested that the only way to tell them apart is to examine the "establishing legislation" to see if it complies with the requirements of Art. III. This, however, comes dangerously close to saying that Art. III courts are those with Art. III judges; Art. I courts are those without such judges. One hundred and fifty years of constitutional history, in other words, had led to a simple tautology. IV 127 The complicated and contradictory history of the issue before us leads me to conclude that Chief Justice Vinson and Justice Harlan reached the correct conclusion: There is no difference in principle between the work that Congress may assign to an Art. I court and that which the Constitution assigns to Art. III courts. Unless we want to overrule a large number of our precedents upholding a variety of Art. I courts—not to speak of those Art. I courts that go by the contemporary name of "administrative agencies"—this conclusion is inevitable. It is too late to go back that far; too late to return to the simplicity of the principle pronounced in Art. III and defended so vigorously and persuasively by Hamilton in The Federalist Nos. 78-82. 128 To say that the Court has failed to articulate a principle by which we can test the constitutionality of a putative Art. I court, or that there is no such abstract principle, is not to say that this Court must always defer to the legislative decision to create Art. I, rather than Art. III, courts. Article III is not to be read out of the Constitution; rather, it should be read as expressing one value that must be balanced against competing constitutional values and legislative responsibilities. This Court retains the final word on how that balance is to be struck. 129 Despite the principled, although largely mistaken, rhetoric expanded by the Court in this area over the years, such a balancing approach stands behind many of the decisions holding Art. I courts. Justice Harlan suggested as much in Glidden, although he needlessly limited his consideration to the "temporary" courts that Congress has had to set up on a variety of occasions. In each of these instances, this Court has implicitly concluded that the legislative interest in creating an adjudicative institution of temporary duration outweighed the values furthered by a strict adherence to Art. III. Besides the territorial courts approved in American Insurance Co. v. Canter, 1 Pet. 511, 7 L.Ed. 242 (1828), these courts have included the Court of Private Land Claims, United States v. Coe, 155 U.S. 76, 15 S.Ct. 16, 39 L.Ed. 76 (1894), the Choctaw and Chickasaw Citizenship Court, Stephens v. Cherokee Nation, 174 U.S. 445, 19 S.Ct. 722, 43 L.Ed. 1041 (1899), and consular courts established in foreign countries, In re Ross, 140 U.S. 453, 11 S.Ct. 897, 35 L.Ed. 581 (1891). This same sort of "practical" judgment was voiced, even if not relied upon, in Crowell with respect to the Employees' Compensation Claims Commission, which was not meant to be of limited duration: "[W]e are unable to find any constitutional obstacle to the action of the Congress in availing itself of a method shown by experience to be essential in order to apply its standards to the thousands of cases involved." 285 U.S., at 54, 52 S.Ct., at 293. And even in Murray's Lessee, there was a discussion of the "necessity" of Congress' adopting an approach that avoided adjudication in an Art. III court. 18 How., at 285. 130 This was precisely the approach taken to this problem in Palmore v. United States, 411 U.S. 389, 93 S.Ct. 1670, 36 L.Ed.2d 342 (1973), which, contrary to the suggestion of the plurality, did not rest on any theory of territorial or geographical control. Ante, at 75-76. Rather, it rested on an evaluation of the strength of the legislative interest in pursuing in this manner one of its constitutionally assigned responsibilities—a responsibility not different in kind from numerous other legislative responsibilities. Thus, Palmore referred to the wide variety of Art. I courts, not just territorial courts. It is in this light that the critical statement of the case must be understood: "[T]he requirements of Art. . III, which are applicable where laws of national applicability and affairs of national concern are at stake, must in proper circumstances give way to accommodate plenary grants of power to Congress to legislate with respect to specialized areas having particularized needs and warranting distinctive treatment." 411 U.S., at 407-408, 93 S.Ct., at 1681-1682. 131 I do not suggest that the Court should simply look to the strength of the legislative interest and ask itself if that interest is more compelling than the values furthered by Art. III. The inquiry should, rather, focus equally on those Art. III values and ask whether and to what extent the legislative scheme accommodates them or, conversely, substantially undermines them. The burden on Art. III values should then be measured against the values Congress hopes to serve through the use of Art. I courts. 132 To be more concrete: Crowell, supra, suggests that the presence of appellate review by an Art. III court will go a long way toward insuring a proper separation of powers. Appellate review of the decisions of legislative courts, like appellate review of state-court decisions, provides a firm check on the ability of the political institutions of government to ignore or transgress constitutional limits on their own authority. Obviously, therefore, a scheme of Art. I courts that provides for appellate review by Art. III courts should be substantially less controversial than a legislative attempt entirely to avoid judicial review in a constitutional court. 133 Similarly, as long as the proposed Art. I courts are designed to deal with issues likely to be of little interest to the political branches, there is less reason to fear that such courts represent a dangerous accumulation of power in one of the political branches of government. Chief Justice Vinson suggested as much when he stated that the Court should guard against any congressional attempt "to transfer jurisdiction . . . for the purpose of emasculating" constitutional courts. National Insurance Co. v. Tidewater Co., 337 U.S., at 644, 69 S.Ct., at 1208-1209. V 134 I believe that the new bankruptcy courts established by the Bankruptcy Act of 1978, 28 U.S.C. § 1471 (1976 ed., Supp.IV), satisfy this standard. 135 First, ample provision is made for appellate review by Art. III courts. Appeals may in some circumstances be brought directly to the district courts. 28 U.S.C. § 1334 (1976 ed., Supp.IV). Decisions of the district courts are further appealable to the court of appeals. § 1293. In other circumstances, appeals go first to a panel of bankruptcy judges, § 1482, and then to the court of appeals. § 1293. In still other circumstances—when the parties agree—appeals may go directly to the court of appeals. In sum, there is in every instance a right of appeal to at least one Art. III court. Had Congress decided to assign all bankruptcy matters to the state courts, a power it clearly possesses, no greater review in an Art. III court would exist. Although I do not suggest that this analogy means that Congress may establish an Art. I court wherever it could have chosen to rely upon the state courts, it does suggest that the critical function of judicial review is being met in a manner that the Constitution suggests is sufficient. 136 Second, no one seriously argues that the Bankruptcy Act of 1978 represents an attempt by the political branches of government to aggrandize themselves at the expense of the third branch or an attempt to undermine the authority of constitutional courts in general. Indeed, the congressional perception of a lack of judicial interest in bankruptcy matters was one of the factors that led to the establishment of the bankruptcy courts: Congress feared that this lack of interest would lead to a failure by federal district courts to deal with bankruptcy matters in an expeditious manner. H.R.Rep.No. 95-595, p. 14 (1977). Bankruptcy matters are, for the most part, private adjudications of little political significance. Although some bankruptcies may indeed present politically controversial circumstances or issues, Congress has far more direct ways to involve itself in such matters than through some sort of subtle, or not so subtle, influence on bankruptcy judges. Furthermore, were such circumstances to arise, the Due Process Clause might very well require that the matter be considered by an Art. III judge: Bankruptcy proceedings remain, after all, subject to all of the strictures of that constitutional provision.15 137 Finally, I have no doubt that the ends that Congress sought to accomplish by creating a system of non-Art. III bankruptcy courts were at least as compelling as the ends found to be satisfactory in Palmore v. United States, 411 U.S. 389, 83 S.Ct. 1670, 36 L.Ed.2d 342 (1973), or the ends that have traditionally justified the creation of legislative courts. The stresses placed upon the old bankruptcy system by the tremendous increase in bankruptcy cases were well documented and were clearly a matter to which Congress could respond.16 I do not believe it is possible to challenge Congress' further determination that it was necessary to create a specialized court to deal with bankruptcy matters. This was the nearly uniform conclusion of all those that testified before Congress on the question of reform of the bankruptcy system, as well as the conclusion of the Commission on Bankruptcy Laws established by Congress in 1970 to explore possible improvements in the system.17 138 The real question is not whether Congress was justified in establishing a specialized bankruptcy court, but rather whether it was justified in failing to create a specialized, Art. III bankruptcy court. My own view is that the very fact of extreme specialization may be enough, and certainly has been enough in the past,18 to justify the creation of a legislative court. Congress may legitimately consider the effect on the federal judiciary of the addition of several hundred specialized judges: We are, on the whole, a body of generalists.19 The addition of several hundred specialists may substantially change, whether for good or bad, the character of the federal bench. Moreover, Congress may have desired to maintain some flexibility in its possible future responses to the general problem of bankruptcy. There is no question that the existence of several hundred bankruptcy judges with life tenure would have severely limited Congress' future options. Furthermore, the number of bankruptcies may fluctuate, producing a substantially reduced need for bankruptcy judges. Congress may have thought that, in that event, a bankruptcy specialist should not as a general matter serve as a judge in the countless nonspecialized cases that come before the federal district courts. It would then face the prospect of large numbers of idle federal judges. Finally, Congress may have believed that the change from bankruptcy referees to Art. I judges was far less dramatic, and so less disruptive of the existing bankruptcy and constitutional court systems, than would be a change to Art. III judges. 139 For all of these reasons, I would defer to the congressional judgment. Accordingly, I dissent. 1 Pub.L. 95-598, 92 Stat. 2549, U.S.Code Cong. & Admin.News 1978, p. 5787. The Act became effective October 1, 1979. 2 Bankruptcy referees were redesignated as "judges" in 1973. Bkrtcy.Rule 901(7). For purposes of clarity, however, we refer to all judges under the old Act as "referees." 3 Although the Act initially vests this jurisdiction in district courts, 28 U.S.C. § 1471(a) (1976 ed., Supp.IV), it subsequently provides that "[t]he bankruptcy court for the district in which a case under title 11 is commenced shall exercise all of the jurisdiction conferred by this section on the district courts," § 1471(c) (1976 ed., Supp.IV) (emphasis added). Thus the ultimate repository of the Act's broad jurisdictional grant is the bankruptcy courts. See 1 W. Collier, Bankruptcy ¶ 3.01, pp. 3-37, 3-44 to 3-49 (15th ed. 1982). 4 With respect to both personal jurisdiction and venue, the scope of the Act is also expansive. Although the Act does not in terms indicate the extent to which bankruptcy judges may exercise personal jurisdiction, it has been construed to allow the constitutional maximum. See, e.g., In re Whippany Paper Board Co., 15 B.R. 312, 314-315 (Bkrtcy. NJ 1981). With two exceptions not relevant here, the venue of "a proceeding arising in or related to a case under title 11 [is] in the bankruptcy court in which such case is pending." 28 U.S.C. § 1473(a) (1976 ed., Supp.IV). Furthermore, the Act permits parties to remove many kinds of actions to the bankruptcy court. Parties "may remove any claim or cause of action in a civil action, other than a proceeding before the United States Tax Court or a civil action by a Government unit to enforce such governmental unit's police or regulatory power." § 1478(a) (1976 ed., Supp.IV). The bankruptcy court may, however, remand such actions "on any equitable ground"; the decision to remand or retain an action is unreviewable. § 1478(b). 5 Although no particular standard of review is specified in the Act, the parties in the present cases seem to agree that the appropriate one is the clearly-erroneous standard, employed in old Bankruptcy Rule 810 for review of findings of fact made by a referee. See Brief for United States 41; Tr. of Oral Arg. 27. See also In re Rivers, 19 B.R. 438 (Bkrtcy. ED Tenn.1982); 1 Collier, supra n. 3, ¶ 3.03, p. 3-315. 6 Under the old Bankruptcy Act, referees could be removed by the district court for "incompetency, misconduct, or neglect of duty," 11 U.S.C. § 62(b) (repealed); the same grounds for removal apply during the transition period, see § 404(d), 92 Stat. 2684. 7 It appears, however, that during the transition period an appeal of a bankruptcy judge's decision may be taken to the district court even if an appellate panel of bankruptcy judges has been established. 8 After Northern docketed an appeal in this Court, the District Court supplemented its order with an opinion. 12 B.R. 946, 947 (1981). 9 Two other Bankruptcy Courts have considered the constitutionality of § 1471: The Bankruptcy Court for the District of Puerto Rico determined it to be constitutional, In re Segarra, 14 B.R. 870 (1981), while the Bankruptcy Court for the Eastern District of Tennessee reached the opposite conclusion, In re Rivers, supra. 10 These provisions serve other institutional values as well. The independence from political forces that they guarantee helps to promote public confidence in judicial determinations. See The Federalist No. 78 (A. Hamilton). The security that they provide to members of the Judicial Branch helps to attract well-qualified persons to the federal bench. Ibid. The guarantee of life tenure insulates the individual judge from improper influences not only by other branches but by colleagues as well, and thus promotes judicial individualism. See Kaufman, Chilling Judicial Independence, 88 Yale L.J. 681, 713 (1979). See generally Note, Article III Limits on Article I Courts: The Constitutionality of the Bankruptcy Court and the 1979 Magistrate Act, 80 Colum.L.Rev. 560, 583-585 (1980). 11 Further evidence of the Framers' concern for assuring the independence of the Judicial Branch may be found in the fact that the Constitutional Convention soundly defeated a proposal to allow the removal of judges by the Executive and Legislative Branches. See 2 M. Farrand, Records of the Federal Convention of 1787, pp. 428-429 (1911); P. Bator, P. Mishkin, D. Shapiro, & H. Wechsler, Hart and Wechsler's The Federal Courts and the Federal System 7 (2d ed. 1973). Mr. Wilson, of Pennsylvania, commented that "[t]he Judges would be in a bad situation if made to depend on every gust of faction which might prevail in the two branches of our Govt." 2 Farrand, supra, at 429. 12 It should be noted, however, that the House of Representatives expressed substantial doubts respecting the constitutionality of the provisions eventually included in the Act. The House Judiciary Committee and its Subcommittee on Civil and Constitutional Rights gave lengthy consideration to the constitutional issues surrounding the conferral of broad powers upon the new bankruptcy courts. The Committee, the Subcommittee, and the House as a whole initially concluded that Art. III courts were constitutionally required for bankruptcy adjudications. See H.R. 8200, 95th Cong., 1st Sess. (1977); Hearings on H.R. 31 and H.R. 32 before the Subcommittee on Civil and Constitutional Rights of the House Committee on the Judiciary, 94th Cong., 2d Sess., 2081-2084 (1976); id., at 2682-2706; H.R.Rep.No. 95-595, p. 39 (1977), U.S.Code Cong. & Admin.News 1978, p. 5787, 6000 ("Article III is the constitutional norm, and the limited circumstances in which the courts have permitted departure from the requirements of Article III are not present in the bankruptcy context"); id., at 21-38; Subcommittee on Civil and Constitutional Rights of the House Committee on the Judiciary, Constitutional Bankruptcy Courts, 95th Cong., 1st Sess., 33 (Comm.Print No. 3, 1977) (concluding that the proposed bankruptcy courts should be established "under Article III, with all of the protection that the Framers intended for an independent judiciary"); Subcommittee on Civil and Constitutional Rights of the House Committee on the Judiciary, Report on Hearings on the Court Administrative Structure for Bankruptcy Cases, 95th Cong., 2d Sess., 5 (Comm.Print No. 13, 1978) (same); see generally Klee, Legislative History of the New Bankruptcy Law, 28 De Paul L.Rev. 941, 945-949, 951 (1979). The Senate bankruptcy bill did not provide for life tenure or a guaranteed salary, instead adopting the concept of a bankruptcy court with similarly broad powers but as an "adjunct" to an Art. III court. S. 2266, 95th Cong., 2d Sess. (1978). The bill that was finally enacted, denying bankruptcy judges the tenure and compensation protections of Art. III, was the result of a series of last-minute conferences and compromises between the managers of both Houses. See Klee, supra, at 952-956. 13 The Act designates the bankruptcy court in each district as an "adjunct" to the district court. 28 U.S.C. § 151(a) (1976 ed., Supp.IV). Neither House of Congress concluded that the bankruptcy courts should be established as independent legislative courts. See n. 12, supra. 14 At one time, this Court suggested a rigid distinction between those subjects that could be considered only in Art. III courts and those that could be considered only in legislative courts. See Williams v. United States, 289 U.S. 553, 53 S.Ct. 751, 77 L.Ed. 1372 (1933). But this suggested dichotomy has not withstood analysis. See C. Wright, Law of the Federal Courts 33-35 (3d ed. 1976). Our more recent cases clearly recognize that legislative courts may be granted jurisdiction over some cases and controversies to which the Art. III judicial power might also be extended. E.g., Palmore v. United States, 411 U.S. 389, 93 S.Ct. 1670, 36 L.Ed.2d 342 (1973). See Glidden Co. v. Zdanok, 370 U.S. 530, 549-551, 82 S.Ct. 1459, 1472-1473, 8 L.Ed.2d 671 (1962) (opinion of Harlan, J.). 15 Justice WHITE's dissent finds particular significance in the fact that Congress could have assigned all bankruptcy matters to the state courts. Post, at 116. But, of course, virtually all matters that might be heard in Art. III courts could also be left by Congress to state courts. This fact is simply irrelevant to the question before us. Congress has no control over state-court judges; accordingly the principle of separation of powers is not threatened by leaving the adjudication of federal disputes to such judges. See Krattenmaker, Article III and Judicial Independence: Why the New Bankruptcy Courts are Unconstitutional, 70 Geo.L.J. 297, 304-305 (1981). The Framers chose to leave to Congress the precise role to be played by the lower federal courts in the administration of justice. See Hart and Wechsler's The Federal Courts and the Federal System, supra, n. 11, at 11. But the Framers did not leave it to Congress to define the character of those courts—they were to be independent of the political branches and presided over by judges with guaranteed salary and life tenure. 16 We recently reaffirmed the principle, expressed in these early cases, that Art. I, § 8, cl. 17, provides that Congress shall have power "[t]o exercise exclusive Legislation in all Cases whatsoever, over" the District of Columbia. Palmore v. United States, 411 U.S., at 397, 93 S.Ct., at 1676. See also Wallace v. Adams, 204 U.S. 415, 423, 27 S.Ct. 363, 365, 51 L.Ed. 547 (1907) (recognizing Congress' authority to establish legislative courts to determine questions of tribal membership relevant to property claims within Indian territory); In re Ross, 140 U.S. 453, 11 S.Ct. 897, 35 L.Ed. 581 (1891) (same, respecting consular courts established by concession from foreign countries). See generally 1 J. Moore, J. Lucas, H. Fink, D. Weckstein, & J. Wicker, Moore's Federal Practice 46-49, 53-54 (1982). But see Reid v. Covert, 354 U.S. 1, 77 S.Ct. 1222, 1 L.Ed.2d 1148 (1957). 17 See also Burns v. Wilson, 346 U.S. 137, 139-140, 73 S.Ct. 1045, 1047, 97 L.Ed. 1508 (1953). But this Court has been alert to ensure that Congress does not exceed the constitutional bounds and bring within the jurisdiction of the military courts matters beyond that jurisdiction, and properly within the realm of "judicial power." See, e.g., Reid v. Covert, supra; United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955). 18 Congress' power to create legislative courts to adjudicate public rights carries with it the lesser power to create administrative agencies for the same purpose, and to provide for review of those agency decisions in Art. III courts. See, e.g., Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S. 442, 450, 97 S.Ct. 1261, 1266, 51 L.Ed.2d 464 (1977). 19 See Oceanic Steam Navigation Co. v. Stranahan, 214 U.S. 320, 339, 29 S.Ct. 671, 676, 53 L.Ed. 1013 (1909); Katz, Federal Legislative Courts, 43 Harv.L.Rev. 894, 915 (1930). 20 Doubtless it could be argued that the need for independent judicial determination is greatest in cases arising between the Government and an individual. But the rationale for the public-rights line of cases lies not in political theory, but rather in Congress' and this Court's understanding of what power was reserved to the Judiciary by the Constitution as a matter of historical fact. 21 See also Williams v. United States, 289 U.S. 553, 53 S.Ct. 751, 77 L.Ed. 1372 (1933) (holding that Court of Claims was a legislative court and that salary of a judge of that court could therefore be reduced by Congress). 22 Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932), attempted to catalog some of the matters that fall within the public-rights doctrine: "Familiar illustrations of administrative agencies created for the determination of such matters are found in connection with the exercise of the congressional power as to interstate and foreign commerce, taxation, immigration, the public lands, public health, the facilities of the post office, pensions and payments to veterans." Id., at 51, 52 S.Ct., at 292 (footnote omitted). 23 Congress cannot "withdraw from [Art. III] judicial cognizance any matter which, from its nature, is the subject of a suit at the common law, or in equity, or admiralty." Murray's Lessee v. Hoboken Land & Improvement Co., 18 How. 272, 284 (1856) (emphasis added). It is thus clear that the presence of the United States as a proper party to the proceeding is a necessary but not sufficient means of distinguishing "private rights" from "public rights." And it is also clear that even with respect to matters that arguably fall within the scope of the "public rights" doctrine, the presumption is in favor of Art. III courts. See Glidden Co. v. Zdanok, 370 U.S., at 548-549, and n. 21, 82 S.Ct., at 1471-1472, and n. 21 (opinion of Harlan, J.). See also Currie, The Federal Courts and the American Law Institute, Part 1, 36 U.Chi.L.Rev. 1, 13-14, n. 67 (1968). Moreover, when Congress assigns these matters to administrative agencies, or to legislative courts, it has generally provided, and we have suggested that it may be required to provide, for Art. III judicial review. See Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S., at 455, n. 13, 97 S.Ct., at 1269, n. 13. 24 Of course, the public-rights doctrine does not extend to any criminal matters, although the Government is a proper party. See, e.g., United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955). 25 The "unifying principle" that Justice WHITE's dissent finds lacking in all of these cases, see post, at 105, is to be found in the exceptional constitutional grants of power to Congress with respect to certain matters. Although the dissent is correct that these grants are not explicit in the language of the Constitution, they are nonetheless firmly established in our historical understanding of the constitutional structure. When these three exceptional grants are properly constrained, they do not threaten the Framers' vision of an independent Federal Judiciary. What clearly remains subject to Art. III are all private adjudications in federal courts within the States—matters from their nature subject to "a suit at common law or in equity or admiralty"—and all criminal matters, with the narrow exception of military crimes. There is no doubt that when the Framers assigned the "judicial Power" to an independent Art. III Branch, these matters lay at what they perceived to be the protected core of that power. Although the dissent recognizes that the Framers had something important in mind when they assigned the judicial power of the United States to Art. III courts, it concludes that our cases and subsequent practice have eroded this conception. Unable to find a satisfactory theme in our precedents for analyzing these cases, the dissent rejects all of them, as well as the historical understanding upon which they were based, in favor of an ad hoc balancing approach in which Congress can essentially determine for itself whether Art. III courts are required. See post, at 105-116. But even the dissent recognizes that the notion that Congress rather than the Constitution should determine whether there is a need for independent federal courts cannot be what the Framers had in mind. See post, at 113. 26 This claim may be adjudicated in federal court on the basis of its relationship to the petition for reorganization. See Williams v. Austrian, 331 U.S. 642, 67 S.Ct. 1443, 91 L.Ed. 1718 (1947); Schumacher v. Beeler, 293 U.S. 367, 55 S.Ct. 230, 79 L.Ed. 433 (1934). See also National Ins. Co. v. Tidewater Co., 337 U.S. 582, 611-613, 69 S.Ct. 1173, 1187-1188, 93 L.Ed. 1556 (1949) (Rutledge, J., concurring); Textile Workers v. Lincoln Mills, 353 U.S. 448, 472, 77 S.Ct. 912, 929, 1 L.Ed.2d 972 (1957) (Frankfurter, J., dissenting). Cf. Osborn v. Bank of the United States, 9 Wheat. 738, 6 L.Ed. 204 (1824). But this relationship does not transform the state-created right into a matter between the Government and the petitioner for reorganization. Even in the absence of the federal scheme, the plaintiff would be able to proceed against the defendant on the state-law contractual claims. 27 Nor can appellants' analysis logically be limited to Congress' Art. I powers. For example, appellants' reasoning relies in part upon analogy to our approval of territorial courts in American Ins. Co. v. Canter, 1 Pet. 511, 7 L.Ed. 242 (1828), and of the use of an administrative agency in Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932). Brief for United States 15; Brief for Northern Pipeline Construction Co. 10. In those cases the Court recognized the right of Congress to create territorial courts pursuant to the authority granted under Art. IV, § 3, cl. 2, and to create administrative tribunals to adjudicate rights in admiralty pursuant to the federal authority in Art. III, § 2, over admiralty jurisdiction. See American Ins. Co. v. Canter, supra, at 546; Crowell v. Benson, supra, 285 U.S., at 39, 52 S.Ct., at 287. This reliance underscores the fact that appellants offer no principled means of distinguishing between Congress' Art. I powers and any of Congress' other powers including, for example, those conferred by the various amendments to the Constitution, e.g., U.S.Const., Amdts. 13-16, 19, 23, 24, 26. 28 Justice WHITE's suggested "limitations" on Congress' power to create Art. I courts are even more transparent. Justice WHITE's dissent suggests that Art. III "should be read as expressing one value that must be balanced against competing constitutional values and legislative responsibilities," and that the Court retains the final word on how the balance is to be struck. Post, at 113-114. The dissent would find the Art. III "value" accommodated where appellate review by Art. III courts is provided and where the Art. I courts are "designed to deal with issues likely to be of little interest to the political branches." Post, at 115. But the dissent's view that appellate review is sufficient to satisfy either the command or the purpose of Art. III is incorrect. See n. 39, infra. And the suggestion that we should consider whether the Art. I courts are designed to deal with issues likely to be of interest to the political Branches would undermine the validity of the adjudications performed by most of the administrative agencies, on which validity the dissent so heavily relies. In applying its ad hoc balancing approach to the facts of this case, the dissent rests on the justification that these courts differ from standard Art. III courts because of their "extreme specialization." As noted above, "extreme specialization" is hardly an accurate description of bankruptcy courts designed to adjudicate the entire range of federal and state controversies. See infra, at 84-85. Moreover, the special nature of bankruptcy adjudications is in no sense incompatible with performance of such functions in a tribunal afforded the protection of Art. III. As one witness pointed out to Congress: "Relevant to that question of need, it seems worth noting that Article III itself permits much flexibility; so long as tenure during good behavior is granted, much room exists as regards other conditions. Thus it would certainly be possible to create a special bankruptcy court under Article III and there is no reason why the judges of that court would have to be paid the same salary as district judges or any other existing judges. It would also be permissible to provide that when a judge of that court retired pursuant to statute, a vacancy for a new appointment would not automatically be created. And it would be entirely valid to specify that the judges of that court could not be assigned to sit, even temporarily, on the general district courts or courts of appeals." Hearings on H.R. 31 and H.R. 32 before the Subcommittee on Civil and Constitutional Rights of the House Committee on the Judiciary, 94th Cong., 2d Sess., 2697 (1976) (letter of Paul Mishkin). 29 Justice WHITE's dissent fails to distinguish between Congress' power to create adjuncts to Art. III courts, and Congress' power to create Art. I courts in limited circumstances. See post, at 103-104. Congress' power to create adjuncts and assign them limited adjudicatory functions is in no sense an "exception" to Art. III. Rather, such an assignment is consistent with Art. III, so long as "the essential attributes of the judicial power" are retained in the Art. III court, Crowell v. Benson, 285 U.S., at 51, 52 S.Ct., at 292, and so long as Congress' adjustment of the traditional manner of adjudication can be sufficiently linked to its legislative power to define substantive rights, see infra, at 83-84. Cf. Atlas Roofing Co. v. Occupational Safety and Health Review Comm'n, 430 U.S., at 450, n. 7, 97 S.Ct., at 1266, n. 7. 30 Thus in Raddatz there was no serious threat that the exercise of the judicial power would be subject to incursion by other branches. "[T]he only conceivable danger of a 'threat' to the 'independence' of the magistrate comes from within, rather than without the judicial department." 447 U.S., at 685, 100 S.Ct., at 2417 (BLACKMUN, J., concurring). 31 Appellants and Justice WHITE's dissent also rely on the broad powers exercised by the bankruptcy referees immediately before the Bankruptcy Act of 1978. See post, at 98-103. But those particular adjunct functions, which represent the culmination of years of gradual expansion of the power and authority of the bankruptcy referee, see 1 Collier, supra n. 3, ¶ 1.02, have never been explicitly endorsed by this Court. In Katchen v. Landy, 382 U.S. 323, 86 S.Ct. 467, 15 L.Ed.2d 391 (1966), on which the dissent relies, there was no discussion of the Art. III issue. Moreover, when Katchen was decided the 1973 Bankruptcy Rules had not yet been adopted, and the district judge, after hearing the report of magistrate, was free to "modify it or . . . reject it in whole or in part or . . . receive further evidence or . . . recommit it with instructions." General Order in Bankruptcy No. 47, 305 U.S. 702 (1939). We note, moreover, that the 1978 Act made at least three significant changes from the bankruptcy practice that immediately preceded it. First, of course, the jurisdiction of the bankruptcy courts was "substantially expanded" by the Act. H.R.Rep.No. 95-595, p. 13 (1977). Before the Act the referee had no jurisdiction, except with consent, over controversies beyond those involving property in the actual or constructive possession of the court. 11 U.S.C. § 46(b) (repealed). See MacDonald v. Plymouth Trust Co., 286 U.S. 263, 266, 52 S.Ct. 505, 506, 76 L.Ed. 1093 (1932). It cannot be doubted that the new bankruptcy judges, unlike the referees, have jurisdiction far beyond that which can be even arguably characterized as merely incidental to the discharge in bankruptcy or a plan for reorganization. Second, the bankruptcy judges have broader powers than those exercised by the referees. See infra at 84-86; H.R.Rep. No. 95-595, supra, at 12, and nn. 63-68. Finally, and perhaps most significantly, the relationship between the district court and the bankruptcy court was changed under the 1978 Act. Before the Act, bankruptcy referees were "subordinate adjuncts of the district courts." Id., at 7, U.S.Code Cong. & Admin.News 1978, p. 5968. In contrast, the new bankruptcy courts are "independent of the United States district courts." Ibid.; 1 Collier, supra n. 3, ¶ 1.03, p. 1-9. Before the Act, bankruptcy referees were appointed and removable only by the district court. 11 U.S.C. § 62 (repealed). And the district court retained control over the reference by his power to withdraw the case from the referee. Bkrptcy.Rule 102. Thus even at the trial stage, the parties had access to an independent judicial officer. Although Congress could still lower the salary of referees, they were not dependent on the political Branches of Government for their appointment. To paraphrase Justice BLACKMUN's observation in Raddatz, supra, the primary "danger of a 'threat' to the 'independence' of the [adjunct came] from within, rather than without, the judicial department." 447 U.S., at 685, 100 S.Ct., at 2417 (concurring opinion). 32 Contrary to Justice WHITE's suggestion, we do not concede that "Congress may provide for initial adjudications by Art. I courts or administrative judges of all rights and duties arising under otherwise valid federal laws." See post, at 94. Rather we simply reaffirm the holding of Crowell—that Congress may assign to non-Art. III bodies some adjudicatory functions. Crowell itself spoke of "specialized" functions. These cases do not require us to specify further any limitations that may exist with respect to Congress' power to create adjuncts to assist in the adjudication of federal statutory rights. 33 The Court in Crowell found that the requirement of de novo review as to certain facts was not "simply the question of due process in relation to notice and hearing," but was "rather a question of the appropriate maintenance of the Federal judicial power." 285 U.S., at 56, 52 S.Ct., at 294. The dissent agreed that some factual findings cannot be made by adjuncts, on the ground that "under certain circumstances, the constitutional requirement of due process is a requirement of [Art. III] judicial process." Id., at 87, 52 S.Ct., at 306 (Brandeis, J., dissenting). 34 Crowell § precise holding, with respect to the review of "jurisdictional" and "constitutional" facts that arise within ordinary administrative proceedings, has been undermined by later cases. See St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 53, 56 S.Ct. 720, 726, 80 L.Ed. 1033 (1936). See generally 4 K. Davis, Administrative Law Treatise §§ 29.08, 29.09 (1st ed. 1958). But the general principle of Crowell—distinguishing between congressionally created rights and constitutionally recognized rights—remains valid, as evidenced by the Court's recent approval of Ng Fung Ho v. White, 259 U.S. 276, 42 S.Ct. 492, 66 L.Ed. 938 (1922), on which Crowell relied. See Agosto v. INS, 436 U.S. 748, 753, 98 S.Ct. 2081, 2085, 56 L.Ed.2d 677 (1978) (de novo judicial determination required for claims of American citizenship in deportation proceedings). See also United States v. Raddatz, 447 U.S., at 682-684, 100 S.Ct., at 2415-2416; id., at 707-712, 100 S.Ct., at 2426-2431 (MARSHALL, J., dissenting). 35 Drawing the line between permissible extensions of legislative power and impermissible incursions into judicial power is a delicate undertaking, for the powers of the Judicial and Legislative Branches are often overlapping. As Justice Frankfurter noted in a similar context: "To be sure the content of the three authorities of government is not to be derived from an abstract analysis. The areas are partly interacting, not wholly disjointed." Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 610, 72 S.Ct. 863, 897, 96 L.Ed. 1153 (1952) (concurring opinion). The interaction between the Legislative and Judicial Branches is at its height where courts are adjudicating rights wholly of Congress' creation. Thus where Congress creates a substantive right, pursuant to one of its broad powers to make laws, Congress may have something to say about the proper manner of adjudicating that right. 36 Of course, bankruptcy adjudications themselves, as well as the manner in which the rights of debtors and creditors are adjusted, are matters of federal law. Appellant Northern's state-law contract claim is now in federal court because of its relationship to Northern's reorganization petition. See n. 26, supra. But Congress has not purported to prescribe a rule of decision for the resolution of Northern's contractual claims. 37 The limitations that the judges "may not enjoin another court or punish a criminal contempt not committed in the presence of the judge of the court or warranting a punishment of imprisonment," 28 U.S.C. § 1481 (1976 ed., Supp.IV), are also denied to Art. III judges under certain circumstances. See 18 U.S.C. §§ 401, 402, 3691; 28 U.S.C. § 2283. 38 Although the entry of an enforcement order is in some respects merely formal, it has long been recognized that " '[t]he award of execution . . . is a part, and an essential part of every judgment passed by a court exercising judicial power. It is no judgment in the legal sense of the term, without it.' " ICC v. Brimson, 154 U.S. 447, 484, 14 S.Ct. 1125, 1136, 38 L.Ed. 1047 (1894), quoting Chief Justice Taney's memorandum in Gordon v. United States, 117 U.S. 697, 702 (1864). 39 Appellants suggest that Crowell and Raddatz stand for the proposition that Art. III is satisfied so long as some degree of appellate review is provided. But that suggestion is directly contrary to the text of our Constitution: "The Judges, both of the supreme and inferior Courts, shall hold their Offices during good Behaviour, and shall . . . receive [undiminished] Compensation." Art. III, § 1 (emphasis added). Our precedents make it clear that the constitutional requirements for the exercise of the judicial power must be met at all stages of adjudication, and not only on appeal, where the court is restricted to considerations of law, as well as the nature of the case as it has been shaped at the trial level. The Court responded to a similar suggestion in Crowell by stating that to accept such a regime, "would be to sap the judicial power as it exists under the Federal Constitution, and to establish a government of bureaucratic character alien to our system, wherever fundamental rights depend, as not infrequently they do depend, upon the facts, and finality as to facts becomes in effect finality in law." 285 U.S., at 57, 52 S.Ct., at 295. Cf. Ward v. Village of Monroeville, 409 U.S. 57, 61-62, 93 S.Ct. 80, 83-84, 34 L.Ed.2d 267 (1972); Osborn v. Bank of the United States, 9 Wheat. 738, 883, 6 L.Ed. 204 (1824). Justice WHITE's dissent views the function of the Third Branch as interpreting the Constitution in order to keep the other two Branches in check, and would accordingly find the purpose, if not the language, of Art. III satisfied where there is an appeal to an Art. III court. See post, at 115. But in the Framers' view, Art. III courts would do a great deal more than, in an abstract way, announce guidelines for the other two Branches. While "expounding" the Constitution was surely one vital function of the Art. III courts in the Framers' view, the tasks of those courts, for which independence was an important safeguard, included the mundane as well as the glamorous, matters of common law and statute as well as constitutional law, issues of fact as well as issues of law. As Hamilton noted, "it is not with a view to infractions of the Constitution only, that the independence of the judges may be an essential safeguard against the effects of occasional ill humors in the society." The Federalist No. 78, p. 488 (H. Lodge ed. 1888). In order to promote the independence and improve the quality of federal judicial decisionmaking in all of these areas, the Framers created a system of independent federal courts. See The Federalist Nos. 78-82. 40 It is clear that, at the least, the new bankruptcy judges cannot constitutionally be vested with jurisdiction to decide this state-law contract claim against Marathon. As part of a comprehensive restructuring of the bankruptcy laws, Congress has vested jurisdiction over this and all matters related to cases under Title 11 in a single non-Art. III court, and has done so pursuant to a single statutory grant of jurisdiction. In these circumstances we cannot conclude that if Congress were aware that the grant of jurisdiction could not constitutionally encompass this and similar claims, it would simply remove the jurisdiction of the bankruptcy court over these matters, leaving the jurisdictional provision and adjudicatory structure intact with respect to other types of claims, and thus subject to Art. III constitutional challenge on a claim-by-claim basis. Indeed, we note that one of the express purposes of the Act was to ensure adjudication of all claims in a single forum and to avoid the delay and expense of jurisdictional disputes. See H.R.Rep.No. 95-595, pp. 43-48 (1977); S.Rep.No. 95-989, p. 17 (1978). Nor can we assume, as THE CHIEF JUSTICE suggests, post, at 92, that Congress' choice would be to have these cases "routed to the United States district court of which the bankruptcy court is an adjunct." We think that it is for Congress to determine the proper manner of restructuring the Bankruptcy Act of 1978 to conform to the requirements of Art. III in the way that will best effectuate the legislative purpose. 41 See also Buckley v. Valeo, 424 U.S., 1, 142, 96 S.Ct. 612, 692, 46 L.Ed.2d 659 (1976); Chicot County Drainage Dist. v. Baxter State Bank, 308 U.S. 371, 376-377, 60 S.Ct. 317, 319-320, 84 L.Ed. 329 (1940); Insurance Corp. of Ireland v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 702, n. 9, 102 S.Ct. 2099, 2104, n.9, 72 L.Ed.2d 492 (1982). 1 Glidden Co. v. Zdanok, 370 U.S. 530, 534, 82 S.Ct. 1459, 1464, 8 L.Ed.2d 671 (1962) (plurality opinion of Harlan, J.). 2 Because this is the sole ground relied upon by the Justices concurring in the judgment, this is the effective basis for today's decision. 3 The plurality attempts to justify its sweeping invalidation of § 1471, because of its inclusion of state-law claims, by suggesting that this statutory provision is nonseverable. Ante, at 87-88, n. 40. The Justices concurring in the judgment specifically adopt this argument as the reason for their decision to join the judgment of the Court. The basis for the conclusion of nonseverability, however, is nothing more than a presumption: "Congress has vested jurisdiction over this and all matters related to cases under Title 11 in a single non-Art. III court, and has done so pursuant to a single statutory grant of jurisdiction. In these circumstances, we cannot conclude that if Congress were aware that the grant of jurisdiction could not constitutionally encompass this and similar claims, it would simply remove the jurisdiction of the bankruptcy court over these matters." Ante, at 87, n. 40. Although it is possible, as a historical matter, to find cases of this Court supporting this presumption, see, e.g., Williams v. Standard Oil Co., 278 U.S. 235, 242, 49 S.Ct. 115, 117, 73 L.Ed. 287 (1929), I had not thought this to be the contemporary approach to the problem of severability, particularly when dealing with federal statutes. I would follow the approach taken by the Court in Buckley v. Valeo, 424 U.S. 1, 108, 96 S.Ct. 612, 677, 46 L.Ed.2d 659 (1976): " 'Unless it is evident that the Legislature would not have enacted those provisions which are within its power, independently of that which is not, the invalid part may be dropped if what is left is fully operative as a law.' " Quoting Champlin Refining Co. v. Corporation Comm'n, 286 U.S. 210, 234, 52 S.Ct. 559, 565, 76 L.Ed. 1062 (1932). This presumption seems particularly strong when Congress has already "enacted those provisions which are within its power, independently of that which is not"—i.e., in the old Bankruptcy Act. 4 Even this is not entirely new. Under the old Act, in certain circumstances, the referee could actually adjudicate and order the payment of a claim of the bankrupt estate against another. In Katchen v. Landy, 382 U.S. 323, 86 S.Ct. 467, 15 L.Ed.2d 391 (1966), for example, we recognized that when a creditor files a claim, the referee is empowered to hear and decide a counterclaim against that creditor arising out of the same transaction. A similar situation could arise in adjudicating setoffs under former § 68 of the Bankruptcy Act. 5 "The judicial act of allowance or disallowance is one, of course, that is performed by the referee where the proceedings have been generally referred." 3 W. Collier, Bankruptcy ¶ 57.14, p. 229, n.3 (14th ed. 1977). 6 On this point I am in agreement with the Justices concurring in the judgment. 7 The majority does not explain why the constitutional grant of power over the Territories to Congress is sufficient to overcome the strictures of Art. III, but presumably not sufficient to overcome the strictures of the Presentment Clause or other executive limits on congressional authority. 8 Had the plurality cited only the territorial courts, the principle relied on perhaps could have been the fact that power over the Territories is provided Congress in Art. IV. However, Congress' power over the District of Columbia is an Art. I power. As such, it does not seem to have any greater status than any of the other powers enumerated in Art. I, § 8. 9 By stating that "of this necessity congress alone is the judge," 18 How., at 285, the Court added some serious ambiguity to the standard it applied. Because this statement ends the Court's analysis of the merits of the claim, it does not seem to mean that the Court will simply defer to congressional judgment. Rather, it appears to mean that the Court will review the legislative record to determine whether there appeared to Congress to be compelling reasons for not establishing an Art. III court. 10 The Court did not, however, entirely follow this principle, for it stated elsewhere that "there is propriety in mentioning the fact that Congress always has treated [the Court of Claims as an Art. I court]." 279 U.S., at 454, 49 S.Ct., at 414. 11 The plurality is clearly wrong in citing Crowell in support of the proposition that matters involving private, as opposed to public, rights may not be considered in a non-Art. III court. Ante, at 70. 12 Crowell also suggests that certain facts—constitutional or jurisdictional—must also be subject to de novo review in an Art. III court. I agree with the plurality that this aspect of Crowell has been "undermined by later cases," ante, at 82, n.34. As a matter of historical interest, however, I would contend that Crowell's holding with respect to these "facts" turned more on the questions of law that were inseparably tied to them, than on some notion of the inadequacy of a non-Art. III factfinder. 13 O'Donoghue does not apply this principle wholly consistently: It still recognizes a territorial court exception to Art. III's requirements. It now bases this exception, however, not on any theoretical difference in principle, but simply on the "transitory character of the territorial governments." 289 U.S., at 536, 53 S.Ct., at 745. 14 See P. Bator, P. Mishkin, D. Shapiro, & H. Wechsler, Hart and Wechsler's The Federal Courts and The Federal System 399 (2d ed. 1973) (reviewing the problems of the Williams case and characterizing it as an "intellectual disaster"). 15 See Crowell v. Benson, 285 U.S. 22, 87, 52 S.Ct. 285, 306, 76 L.Ed. 598 (1932) (Brandeis, J., dissenting) ("If there be any controversy to which the judicial power extends that may not be subjected to the conclusive determination of administrative bodies or federal legislative courts, it is not because of any prohibition against the diminution of the jurisdiction of the federal district courts as such, but because, under the circumstances, the constitutional requirement of due process is a requirement of judicial process"). 16 "During the past 30 years, the number of bankruptcy cases filed annually has increased steadily from 10,000 to over 254,000." H.R.Rep.No. 95-595, p. 21 (1977); U.S.Code Cong. & Admin.News 1978, p. 5982. 17 See H.R.Doc.No. 93-137, pt. 1, pp. 85-96 (1973). 18 Consider, for example, the Court of Customs Appeals involved in Ex parte Bakelite Corp., 279 U.S. 438, 49 S.Ct. 411, 73 L.Ed. 789 (1929), or the variety of specialized administrative agencies that engage in some form of adjudication. 19 In 1977, there were approximately 190 full-time and 30 part-time bankruptcy judges throughout the country. H.R.Rep.No. 95-595, at 9.

89

458 U.S. 141 102 S.Ct. 3014 73 L.Ed.2d 664 FIDELITY FEDERAL SAVINGS AND LOAN ASSOCIATION, et al., Appellantsv.Reginald D. de la CUESTA et al. No. 81-750. Argued April 28, 1982. Decided June 28, 1982. Syllabus Section 5(a) of the Home Owners' Loan Act of 1933 (HOLA) empowers the Federal Home Loan Bank Board (Board), under such regulations as it may prescribe, to provide for the organization, operation, and regulation of federal savings and loan associations. Pursuant to this authorization, the Board issued a regulation providing that a federal savings and loan association "continues to have the power to include . . . in its loan instrument" a "due-on-sale" clause, i.e., a provision that permits the association to declare the entire balance of the loan immediately due and payable if the property securing the loan is sold or otherwise transferred without the association's prior written consent. A preamble to the regulation stated that the due-on-sale practices of federal savings and loan associations shall be governed "exclusively by Federal law" and that the association "shall not be bound by or subject to any conflicting State law which imposes different . . . due-on-sale requirements." Appellees each purchased California real property from one who had borrowed money from appellant Fidelity Federal Savings and Loan Association (Fidelity). The borrowers had given Fidelity deeds of trust on the property; each deed contained a due-on-sale clause. Fidelity, not having received prior notice of the purchases, proceeded to enforce the due-on-sale clauses to accelerate payment of the loans, and when they were not paid, instituted nonjudicial foreclosure proceedings. Each appellee then filed suit against Fidelity in California Superior Court, asserting that Fidelity's exercise of the due-on-sale clauses violated the principles announced inWellenkamp v. Bank of America, 21 Cal.3d 943, 148 Cal.Rptr. 379, 582 P.2d 970, which limited a lender's right to exercise such a clause to cases where the lender can demonstrate that the transfer of the property has impaired its security. The Superior Court consolidated the actions and granted Fidelity's motion for summary judgment on the ground that the Federal Government had totally occupied the regulation of federal savings and loan associations. The California Court of Appeal reversed, holding that Wellenkamp was controlling and that federal law had not expressly or impliedly pre-empted state due-on-sale law. Held : The Board's due-on-sale regulation pre-empts conflicting state limitations on the due-on-sale practices of federal savings and loan associations, and thus bars application of the Wellenkamp rule to such associations. Pp. 152-170. (a) The general principles governing pre-emption of state law that conflicts with federal law are not inapplicable here simply because real property is a matter of special concern to the States. And federal regulations have no less pre-emptive effect than federal statutes. Where Congress has empowered an administrator to promulgate regulations, regulations intended to pre-empt state law have that effect unless the administrator exceeded his statutory authority or acted arbitrarily. Pp. 152-154. (b) The language of the Board's regulation and especially the preamble thereto clearly show the Board's intent to pre-empt the Wellenkamp doctrine. The conflict between that doctrine and the regulation does not evaporate because the regulation simply permits, but does not compel, federal savings and loan associations to include a due-on-sale clause in their contracts and to enforce that clause when the security property is transferred. While compliance with both the regulation and the Wellenkamp rule may not be a physical impossibility, that rule forbids a federal savings and loan association to enforce a due-on-sale clause at its option and deprives the association of the flexibility given it by the Board. The rule therefore creates an obstacle to the accomplishment of the regulation's purpose. Pp. 154-159. (c) The Board acted within its statutory authority in issuing the pre-emptive due-on-sale regulation. Both the language and legislative history of the HOLA indicate that the Board was authorized to regulate the lending practices of federal savings and loan associations. Congress delegated power to the Board expressly for the purpose of creating and regulating these associations so as to ensure that they would remain financially sound and able to supply financing for home construction and purchase. Consistent with that purpose, the Board reasonably exercised its authority in promulgating the due-on-sale regulation. Pp. 159-170. 121 Cal.App.3d 328, 175 Cal.Rptr. 467, reversed. Ernest Leff, Beverly Hills, Cal., for appellants. Stephen M. Shapiro, Washington, D. C., for FHLBB and FHLMC as amicus curiae, by special leave of Court. Robert E. Boehmer, Riverside, Cal., for appellees. Justice BLACKMUN delivered the opinion of the Court. 1 At issue in this case is the pre-emptive effect of a regulation, issued by the Federal Home Loan Bank Board (Board), permitting federal savings and loan associations to use "due-on-sale" clauses in their mortgage contracts. Appellees dispute both the Board's intent and its statutory authority to displace restrictions imposed by the California Supreme Court on the exercise of these clauses. 2 * A. 3 The Board, an independent federal regulatory agency, was formed in 1932 and thereafter was vested with plenary authority to administer the Home Owners' Loan Act of 1933 (HOLA), 48 Stat. 128, as amended, 12 U.S.C. § 1461 et seq. (1976 ed. and Supp. IV).1 Section 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp. IV), empowers the Board, "under such rules and regulations as it may prescribe, to provide for the organization, incorporation, examination, operation, and regulation of associations to be known as 'Federal Savings and Loan Associations.' " Pursuant to this authorization, the Board has promulgated regulations governing "the powers and operations of every Federal savings and loan association from its cradle to its corporate grave." People v. Coast Federal Sav. & Loan Assn., 98 F.Supp. 311, 316 (S.D.Cal.1951). 4 In 1976, the Board became concerned about the increasing controversy as to the authority of a federal savings and loan association to exercise a "due-on-sale" clause—a contractual provision that permits the lender to declare the entire balance of a loan immediately due and payable if the property securing the loan is sold or otherwise transferred.2 Specifically, the Board felt that restrictions on a savings and loan's ability to accelerate a loan upon transfer of the security would have a number of adverse effects: (1) that "the financial security and stability of Federal associations would be endangered if . . . the security property is transferred to a person whose ability to repay the loan and properly maintain the property is inadequate"; (2) that "elimination of the due on sale clause will cause a substantial reduction of the cash flow and net income of Federal associations, and that to offset such losses it is likely that the associations will be forced to charge higher interest rates and loan charges on home loans generally"; and (3) that "elimination of the due on sale clause will restrict and impair the ability of Federal associations to sell their home loans in the secondary mortgage market, by making such loans unsalable or causing them to be sold at reduced prices, thereby reducing the flow of new funds for residential loans, which otherwise would be available." 41 Fed.Reg. 6283, 6285 (1976). The Board concluded that "elimination of the due-on-sale clause will benefit only a limited number of home sellers, but generally will cause economic hardship to the majority of home buyers and potential home buyers." Ibid. 5 Accordingly, the Board issued a regulation in 1976 governing due-on-sale clauses. The regulation, now 12 CFR § 545.8-3(f) (1982),3 provides in relevant part: 6 "[A federal savings and loan] association continues to have the power to include, as a matter of contract between it and the borrower, a provision in its loan instrument whereby the association may, at its option, declare immediately due and payable sums secured by the association's security instrument if all or any part of the real property securing the loan is sold or transferred by the borrower without the association's prior written consent. Except as [otherwise] provided in . . . this section . . ., exercise by the association of such option (hereafter called a due-on-sale clause) shall be exclusively governed by the terms of the loan contract, and all rights and remedies of the association and borrower shall be fixed and governed by that contract." 7 In the preamble accompanying final publication of the due-on-sale regulation, the Board explained its intent that the due-on-sale practices of federal savings and loans be governed "exclusively by Federal law." 41 Fed.Reg. 18286, 18287 (1976). The Board emphasized that "[f]ederal associations shall not be bound by or subject to any conflicting State law which imposes different . . . due-on-sale requirements." Ibid.4 B 8 Appellant Fidelity Federal Savings and Loan Association (Fidelity) is a private mutual savings and loan association chartered by the Board pursuant to § 5(a) of the HOLA. Fidelity's principal place of business is in Glendale, Cal. Appellees, de la Cuesta, Moore, and Whitcombe, each made a purchase of California real property from one who had borrowed money from Fidelity. As security for the loan, the borrower had given Fidelity a deed of trust on the property. Each deed of trust contained a due-on-sale clause. Two of the deeds also included a provision, identified as ¶ 15, which stated that the deed "shall be governed by the law of the jurisdiction in which the Property is located." App. 51, 86.5 9 Fidelity was not notified prior to each appellee's purchase of property; when it did learn of the transfer, it gave notice of its intent to enforce the due-on-sale clause. Fidelity expressed a willingness to consent to the transfer, however, if the appellee agreed to increase the interest rate on the loan secured by the property to the then-prevailing market rate. Each appellee refused to accept this condition; Fidelity then exercised its option to accelerate the loan. When the loan was not paid, Fidelity instituted a nonjudicial foreclosure proceeding. 10 In response, each appellee filed suit in the Superior Court of California for Orange County. Each asserted that, under the principles announced by the California Supreme Court in Wellenkamp v. Bank of America, 21 Cal.3d 943, 148 Cal.Rptr. 379, 582 P.2d 970 (1978), Fidelity's exercise of the due-on-sale clause violated California's prohibition of unreasonable restraints on alienation, Cal.Civ.Code Ann. § 711 (West 1982), "unless the lender can demonstrate that enforcement is reasonably necessary to protect against impairment to its security or the risk of default." 21 Cal.3d, at 953, 148 Cal.Rptr., at 386, 582 P.2d, at 977. Each complaint sought (1) a judicial declaration that the due-on-sale clause was not enforceable unless Fidelity first showed that the transfer had harmed its security interest, (2) an injunction against any foreclosure procedures based on the clause, and (3) compensatory and punitive damages. App. 5, 49, 84.6 11 The Superior Court consolidated the three actions and granted appellants' motion for summary judgment. The court explained that "the federal government has totally occupied the subject of regulation of Federal Savings and Loans," and held, therefore, that the decision in Wellenkamp "cannot be extended to [federal] savings and loans." App. to Juris. Statement 29a. 12 The Court of Appeal for the Fourth Appellate District, however, reversed that judgment. In an opinion that adopted substantial portions of a parallel ruling by the Court of Appeal for the First Appellate District, it concluded that the California Supreme Court's opinion in Wellenkamp was controlling. 121 Cal.App.3d 328, 331, 175 Cal.Rptr. 467, 468 (1981), quoting Panko v. Pan American Federal Sav. & Loan Assn., 119 Cal.App.3d 916, 174 Cal.Rptr. 240 (1981), cert. pending, No. 81-922. The court found that Congress had neither expressed an intent to pre-empt state due-on-sale law nor fully occupied the field of federal savings and loan regulation; for example, the court pointed out, federal associations traditionally have been governed by state real property and mortgage law with respect to title, conveyancing, recording, priority of liens, and foreclosure proceedings. 13 The Court of Appeal likewise rejected appellants' contention that the Board's 1976 regulation expressly had pre-empted the Wellenkamp doctrine. Although the court recognized that the preamble accompanying 12 CFR § 545.8-3(f) (1982) manifested the Board's intent that its due-on-sale regulation supersede conflicting state law, it refused to "equate the Board's expression of intent with the requisite congressional intent." 121 Cal.App.3d, at 339, 175 Cal.Rptr., at 474 (emphasis in original).7 14 Finally, the Court of Appeal found no evidence that federal law impliedly had pre-empted state law, reasoning that California's due-on-sale law was not incompatible with federal law. The Wellenkamp doctrine, the court observed, "is a substantive rule of California property and mortgage law," and not a form of "regulation" over federal savings and loans. 121 Cal.App.3d, at 341, 175 Cal.Rptr., at 474. Moreover, the court noted, the Board's regulation "merely authorizes and does not compel savings and loan associations to include a due-on-sale clause in their loan contracts and to exercise their rights thereunder." Ibid., 175 Cal.Rptr., at 475. The Court of Appeal likewise discovered no conflict between the Wellenkamp doctrine and the purposes of the HOLA because both were designed to assist financially distressed homeowners. 15 The court derived "further support," 121 Cal.App.3d, at 342, 175 Cal.Rptr., at 475, for its decision from ¶ 15, which was included in two of the deeds of trust and which provided that the deeds would be "governed by the law of the jurisdiction in which the Property is located." See n. 5, supra. That language, the court ruled, evinced an unmistakable intent that state law should govern the interpretation, validity, and enforcement of the deeds.8 16 The California Supreme court denied appellants' petition for review. App. to Juris. Statement 28a. 17 Because the majority of courts to consider the question have concluded, in contrast to the decision of the Court of Appeal, that the Board's regulations, including § 545.8-3(f), do pre-empt state regulation of federal savings and loans,9 we noted probable jurisdiction. 455 U.S. 917, 102 S.Ct. 1272, 71 L.Ed.2d 458 (1982). II 18 The pre-emption doctrine, which has its roots in the Supremacy Clause, U.S.Const., Art. VI, cl. 2, requires us to examine congressional intent. Pre-emption may be either express or implied, and "is compelled whether Congress' command is explicitly stated in the statute's language or implicitly contained in its structure and purpose." Jones v. Rath Packing Co., 430 U.S. 519, 525, 97 S.Ct. 1305, 1309, 51 L.Ed.2d 604 (1977). Absent explicit pre-emptive language, Congress' intent to supersede state law altogether may be inferred because "[t]he scheme of federal regulation may be so pervasive as to make reasonable the inference that Congress left no room for the States to supplement it," because "the Act of Congress may touch a field in which the federal interest is so dominant that the federal system will be assumed to preclude enforcement of state laws on the same subject," or because "the object sought to be obtained by federal law and the character of obligations imposed by it may reveal the same purpose." Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230, 67 S.Ct. 1146, 1152, 91 L.Ed. 1447 (1947). 19 Even where Congress has not completely displaced state regulation in a specific area, state law is nullified to the extent that it actually conflicts with federal law. Such a conflict arises when "compliance with both federal and state regulations is a physical impossibility." Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-143, 83 S.Ct. 1210, 1217, 10 L.Ed.2d 248 (1963), or when state law "stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress," Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 404, 85 L.Ed. 581 (1941). See also Jones v. Rath Packing Co., 430 U.S., at 526, 97 S.Ct., at 1310; Bethlehem Steel Co. v. New York Labor Relations Bd., 330 U.S. 767, 773, 67 S.Ct. 1026, 1029, 91 L.Ed. 1234 (1947). These principles are not inapplicable here simply because real property law is a matter of special concern to the States: "The relative importance to the State of its own law is not material when there is a conflict with a valid federal law, for the Framers of our Constitution provided that the federal law must prevail." Free v. Bland, 369 U.S. 663, 666, 82 S.Ct. 1089, 1092, 8 L.Ed.2d 180 (1962); see also Ridgway v. Ridgway, 454 U.S. 46, 54-55, 102 S.Ct. 49, 55, 70 L.Ed.2d 39 (1981). 20 Federal regulations have no less pre-emptive effect than federal statutes. Where Congress has directed an administrator to exercise his discretion, his judgments are subject to judicial review only to determine whether he has exceeded his statutory authority or acted arbitrarily. United States v. Shimer, 367 U.S. 374, 381-382, 81 S.Ct. 1554, 1559-1560, 6 L.Ed.2d 908 (1961). When the administrator promulgates regulations intended to pre-empt state law, the court's inquiry is similarly limited: 21 "If [h]is choice represents a reasonable accommodation of conflicting policies that were committed to the agency's care by the statute, we should not disturb it unless it appears from the statute or its legislative history that the accommodation is not one that Congress would have sanctioned." Id., at 383, 81 S.Ct., at 1560. 22 See also Blum v. Bacon, 457 U.S. 132, 145-146, 102 S.Ct. 2355, 2363-2364, 72 L.Ed.2d 728 (1982); Ridgway v. Ridgway, 454 U.S., at 57, 102 S.Ct., at 56 (regulations must not be "unreasonable, unauthorized, or inconsistent with" the underlying statute); Free v. Bland, 369 U.S., at 668, 82 S.Ct., at 1093. 23 A pre-emptive regulation's force does not depend on express congressional authorization to displace state law; moreover, whether the administrator failed to exercise an option to promulgate regulations which did not disturb state law is not dispositive. See United States v. Shimer, 367 U.S., at 381-383, 81 S.Ct., at 1559-1560. Thus, the Court of Appeal's narrow focus on Congress' intent to supersede state law was misdirected. Rather, the questions upon which resolution of this case rests are whether the Board meant to pre-empt California's due-on-sale law, and, if so, whether that action is within the scope of the Board's delegated authority. III 24 As even the Court of Appeal recognized, the Board's intent to pre-empt the Wellenkamp doctrine is unambiguous. The due-on-sale regulation plainly provides that a federal savings and loan "continues to have the power" to include a due-on-sale clause in a loan instrument and to enforce that clause "at its option." 12 CFR § 545.8-3(f) (1982). The California courts, in contrast, have limited a federal association's right to exercise a due-on-sale provision to those cases where the lender can demonstrate that the transfer has impaired its security. 25 The conflict does not evaporate because the Board's regulation simply permits, but does not compel, federal savings and loans to include due-on-sale clauses in their contracts and to enforce those provisions when the security property is transferred. The Board consciously has chosen not to mandate use of due-on-sale clauses "because [it] desires to afford associations the flexibility to accommodate special situations and circumstances." 12 CFR § 556.9(f)(1) (1982),10 Although compliance with both § 545.8-3(f) and the Wellenkamp rule may not be "a physical impossibility," Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S., at 142-143, 83 S.Ct., at 1217, the California courts have forbidden a federal savings and loan to enforce a due-on-sale clause solely "at its option" and have deprived the lender of the "flexibility" given it by the Board. 26 Moreover, the Board recently has "reiterat[ed] its long-standing policy" of authorizing federal savings and loan associations to enforce due-on-sale clauses "subject only to express limitations imposed by the Board." 46 Fed.Reg. 39123, 39124 (1981). The only restrictions specified in the Board's regulation are contained in 12 CFR § 545.8-3(g) (1982).11 That provision, unlike the Wellenkamp doctrine, does not confine a federal association's right to accelerate a loan to cases where the lender's security is impaired. In addition, Wellenkamp explicitly bars a federal savings and loan from exercising a due-on-sale clause to adjust a long-term mortgage's interest rate towards current market rates—a due-on-sale practice the Board has approved and views as critical to "the financial stability of the association." See Schott Advisory Opinion, at 27. 27 By further limiting the availability of an option the Board considers essential to the economic soundness of the thrift industry, the State has created "an obstacle to the accomplishment and execution of the full purposes and objectives" of the due-on-sale regulation. Hines v. Davidowitz, 312 U.S., at 67, 61 S.Ct., at 404. Cf. Franklin Nat. Bank v. New York, 347 U.S. 373, 378, 74 S.Ct. 550, 553, 98 L.Ed. 767 (1954) (finding a "clear conflict" between federal law, which authorized national banks to receive savings deposits but did not specifically permit—much less require—advertising by such banks, and New York law, which forbade them to use the word "savings" in their advertising or business). 28 Contending that the Wellenkamp doctrine is not inconsistent with the due-on-sale regulation, however, appellees point to the regulation's second sentence, which provides in pertinent part: 29 "[E]xercise by the association of such option (hereafter called a due-on-sale clause) shall be exclusively governed by the terms of the loan contract, and all rights and remedies of the association and borrower shall be fixed and governed by that contract." 12 CFR § 545.8-3(f) (1982). 30 Appellees interpret this language as incorporating state contract law—and therefore any state law restricting the exercise of a due-on-sale clause. We note, however, that the incorporation of state law does not signify the inapplicability of federal law, for "a fundamental principle in our system of complex national polity" mandates that "the Constitution, laws, and treaties of the United States are as much a part of the law of every State as its own local laws and Constitution." Hauenstein v. Lynham, 100 U.S. 483, 490, 25 L.Ed. 628 (1880). See also Testa v. Katt, 330 U.S. 386, 390-392, 67 S.Ct. 810, 812-813 (1947).12 Moreover, in our view, the second sentence of § 545.8-3(f) simply makes clear that the regulation does not empower federal savings and loans to accelerate a loan upon transfer of the security property unless the parties to the particular loan instrument, as a matter of contract, have given the lender that right. Similarly, if the parties to a given contract agree somehow to limit the association's right to exercise a due-on-sale provision, the second sentence of § 545.8-3(f) precludes the lender from relying on the first sentence as authorizing more expansive use of the clause. 31 Any ambiguity in § 545.8-3(f)'s language is dispelled by the preamble accompanying and explaining the regulation. The preamble unequivocally expresses the Board's determination to displace state law: 32 "Finally, it was and is the Board's intent to have . . . due-on-sale practices of Federal associations governed exclusively by Federal law. Therefore, . . . exercise of due-on-sale clauses by Federal associations shall be governed and controlled solely by [§ 545.8-3] and the Board's new Statement of Policy. Federal associations shall not be bound by or subject to any conflicting State law which imposes different . . . due-on-sale requirements, nor shall Federal associations attempt to . . . avoid the limitations on the exercise of due-on-sale clauses delineated in [§ 545.8-3(g) ] on the ground that such . . . avoidance of limitations is permissible under State law." 41 Fed.Reg. 18286, 18287 (1976) (emphasis added).13 33 In addition, the Board recently has "confirm[ed]" that the due-on-sale practices of federal savings and loans "shall be governed exclusively by the Board's regulations in pre-emption of and without regard to any limitations imposed by state law on either their inclusion or exercise." 12 CFR § 556.9(f)(2) (1982). Thus, we conclude that the Board's due-on-sale regulation was meant to pre-empt conflicting state limitations on the due-on-sale practices of federal savings and loans, and that the California Supreme Court's decision in Wellenkamp creates such a conflict.14 IV 34 The question remains whether the Board acted within its statutory authority in issuing the pre-emptive due-on-sale regulation. The language and history of the HOLA convince us that Congress delegated to the Board ample authority to regulate the lending practices of federal savings and loans so as to further the Act's purposes, and that § 545.8-3(f) is consistent with those purposes. A. 35 The HOLA, a product of the Great Depression of the 1930's, was intended "to provide emergency relief with respect to home mortgage indebtedness" at a time when as many as half of all home loans in the country were in default. H.R.Conf.Rep.No.210, 73d Cong., 1st Sess., 1 (1933). See 77 Cong.Rec. 2499 (1933) (remarks of Rep. Hancock); id., at 2570 (remarks of Rep. Reilly); Home Owners' Loan Act: Hearings on S. 1317 before a Subcommittee of the Senate Committee on Banking and Currency 9, 73 Cong., 1st Sess., (1933) (Senate Hearings) (statement of Horace Russell, one of the drafters of the bill and General Counsel, Federal Home Loan Bank Board, Atlanta, Ga.). Local institutions that had previously supplied funds to finance homes had ceased doing business or had discontinued such long-term loans, so that more than half the counties in the country, containing almost one-fifth of the total population, were without home-financing institutions. See id., at 7, 19; see also H.R.Rep.No.55, 73d Cong., 1st Sess., 2 (1933); S.Rep.No.91, 73d Cong., 1st Sess., 2 (1933); Home Owners' Loan Act: Hearings on H.R. 4980 before the House Committee on Banking and Currency, 73d Cong., 1st Sess., 16-17 (1933) (House Hearings) (statement of William F. Stevenson, Chairman, Federal Home Loan Bank Board); Comment, 11 Pac.L.J. 1085, 1103 (1980) (by 1933, 1,700 state-chartered savings and loans had failed, causing losses of some $200 million, about one-third the value of savings in these associations). 36 In order to ameliorate these conditions, Congress enacted the HOLA, "a radical and comprehensive response to the inadequacies of the existing state systems." Conference of Federal Sav. & Loan Assns. v. Stein, 604 F.2d 1256, 1257 (CA9 1979), summarily aff'd, 445 U.S. 921, 100 S.Ct. 1304, 63 L.Ed.2d 754 (1980). The Act provided for the creation of a system of federal savings and loan associations, which would be regulated by the Board so as to ensure their vitality as "permanent associations to promote the thrift of the people in a cooperative manner, to finance their homes and the homes of their neighbors." S.Rep.No.91, 73d Cong., 1st Sess., 2 (1933); see also H.R.Rep.No.55, 73d Cong., 1st Sess., 2 (1933); 77 Cong.Rec. 4974 (1933) (remarks of Sen. Bulkley). 37 Thus, in § 5(a) of the Act, Congress gave the Board plenary authority to issue regulations governing federal savings and loans: 38 "In order to provide local mutual thrift institutions in which people may invest their funds and in order to provide for the financing of homes, the Board is authorized, under such rules and regulations as it may prescribe, to provide for the organization, incorporation, examination, operation, and regulation of associations to be known as 'Federal Savings and Loan Associations', or 'Federal mutual savings banks' . . ., and to issue charters therefor, giving primary consideration to the best practices of local mutual thrift and home-financing institutions in the United States." 12 U.S.C. § 1464(a)(1) (1976 ed., Supp.IV) (emphasis added). 39 The broad language of § 5(a) expresses no limits on the Board's authority to regulate the lending practices of federal savings and loans. As one court put it, "[i]t would have been difficult for Congress to give the Bank Board a broader mandate." Glendale Federal Sav. & Loan Assn. v. Fox, 459 F.Supp. 903, 910 (CD Cal.1978), final summary judgment granted, 481 F.Supp. 616 (1979), order reversing and remanding, 663 F.2d 1078 (CA9 1981), cert. pending, No. 81-1192. And Congress' explicit delegation of jurisdiction over the "operation" of these institutions must empower the Board to issue regulations governing mortgage loan instruments, for mortgages are a central part of any savings and loan's "operation." See Schott Advisory Opinion, at 21; House Hearings 16 (Apr. 20, 1933) (statement of William F. Stevenson, Chairman, Federal Home Loan Bank Board) ("We are loaning [savings associations] seven million dollars a week and they are lending it pretty largely on homes of the type contemplated in the Act"); Tr. of Oral Arg. 4 (approximately 78% of savings and loan associations' assets are invested in mortgage loan contracts). 40 Moreover, Congress directed that, in regulating federal savings and loans, the Board consider "the best practices of local mutual thrift and home-financing institutions in the United States," which were at that time all state-chartered. § 5(a) of the HOLA, 12 U.S.C. § 1464(a). By so stating, Congress plainly envisioned that federal savings and loans would be governed by what the Board—not any particular State—deemed to be the "best practices." See also First Federal Sav. & Loan Assn. v. Massachusetts Tax Comm'n, 437 U.S. 255, 258, n. 3, 98 S.Ct. 2333, 2335, n. 3, 57 L.Ed.2d 187 (1978) (observing that the HOLA "protects federal associations from being forced into the state regulatory mold"). Thus, the statutory language suggests that Congress expressly contemplated, and approved, the Board's promulgation of regulations superseding state law. 41 Appellees, however, point to the various sections of the HOLA explicitly pre-empting15 and incorporating16 state law, and contend that the Board has no additional authority to adopt regulations displacing state law. Although Congress made decisions about the applicability of certain aspects of state law to federal savings and loans, these provisions do not imply that Congress intended no further pre-emption of state law. Rather, Congress invested the Board with broad authority to regulate federal savings and loans so as to effect the statute's purposes, and plainly indicated that the Board need not feel bound by existing state law. § 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp.IV). We cannot read this broad delegation of power as confining the Board's authority to pre-empt state law to those areas "specifically described by the Act's other provisions." United States v. Southwestern Cable Co., 392 U.S. 157, 172, 88 S.Ct. 1994, 2002, 20 L.Ed.2d 1001 (1968); see also Phelps Dodge Corp. v. NLRB, 313 U.S. 177, 193-194, 61 S.Ct. 845, 852, 85 L.Ed. 1271 (1941). 42 Furthermore, if federal savings and loans were expected to conform to state law except where explicitly pre-empted in the Act itself, the provisions incorporating specific aspects of state law were needlessly repetitive. We decline to construe the Act so as to render these provisions nugatory, "thereby offending the well-settled rule that all parts of a statute, if possible, are to be given effect." American Textile Mfrs. Institute, Inc. v. Donovan, 452 U.S. 490, 513, 101 S.Ct. 2478, 2492, 69 L.Ed.2d 185 (1981). See also Jarecki v. G. D. Searle & Co., 367 U.S. 303, 307-308, 81 S.Ct. 1579, 1582, 6 L.Ed.2d 859 (1961); cf. Franklin Nat. Bank v. New York, 347 U.S., at 378, 74 S.Ct., at 553 ("We find no indication that Congress intended to make this phase of national banking [i.e., advertising] subject to local restrictions, as it has done by express language in several other instances").17 B 43 Because of the exigencies of the times, the HOLA was enacted hurriedly and its legislative history, concededly, is somewhat sparse.18 But that history does confirm our reading of the statutory language and the Board's plenary authority to regulate the operations of federal savings and loans. Attempting to provide for the "relief of the man who is about to lose his home," Congress set out the general framework and left many of the details to the Board. House Hearings 13 (Apr. 20, 1933) (statement of William F. Stevenson, Chairman, Federal Home Loan Bank Board). Thus, references to the Board's broad discretion to regulate the newly created federal savings and loans appear throughout the legislative history. Nowhere is there a suggestion of any intent somehow to limit the Board's authority. 44 Chairman Stevenson's testimony during the HOLA hearings suggests that the Act contemplated that federal law would govern the terms of the loan instruments used by federal savings and loans. Discussing § 5(c) of the HOLA, as amended, 12 U.S.C. § 1464(c), Representative Hancock noted: "You are departing from uniformity with respect to loan associations throughout the United States when you say that the thrift associations cannot loan on a piece of real estate in excess of $20,000." House Hearings 14 (Apr. 21, 1933). The Chairman replied: "That may be true. We are departing in a good many ways. We have a good many [thrift associations] that are in dire straits because they have loaned on property way up yonder in value, and they have their money tied up in hotels, apartment houses and things of that kind, which puts them in a desperate situation." Ibid. 45 Similarly, in response to concern expressed during the Senate hearings that the Act did not prohibit borrowers from obtaining financing and then renting the property, Chairman Stevenson observed: "That would be a matter of regulation. That could be covered by regulation under the bill." Senate Hearings 14. Asked whether the Board would have authority to promulgate such a regulation, Stevenson replied: 46 "If the Federal Home Loan Bank Board should choose to make that kind of a regulation it could put that in. A great many of these local private institutions would put that kind of a clause in their loans." Ibid. 47 See also House Hearings 5 (Apr. 20, 1933) (statement of Chairman Stevenson) (referring to "the regulations as to the use of the property after the loan is once obtained"); id., at 9 (Apr. 21, 1933) (statement of Mr. Stevenson) ("[I]t is in the discretion of the Board when it will grant [a 3-year] extension [of loan payments]."); id., at 18-19 (colloquy between Stevenson and Rep. Reilly) (noting that the Board has discretion in determining whether to charter a federal association). 48 The subsequent debates confirm that Congress accepted Chairman Stevenson's offer and furnished the Board with broad power to regulate the federal savings and loans. Thus, Representative Luce, ranking minority member of the House Committee on Banking and Currency, observed that the federal savings and loan associations "will be formed in accordance with the best building-and-loan practice, and I feel sure we may rely upon [Chairman Stevenson] and his Board to carry out that promise." 77 Cong.Rec. 2480 (1933). "It is contemplated by the bill before us to put the machinery in the hands of the Home Loan Bank Board," and "[w]e give the board great power to administer the act," Representative Luce continued. Id., at 2480, 2481. See also id., at 2481 ("We leave such things [as limitations on conversion of federal home loan banks to federal savings and loans] to the judgment of the board"); id., at 2501 ("The prudent course is to leave this to the judgment of the board, by imposing a maximum [rate of interest] in the bill—4 percent upon what we borrow, 5 percent upon what we lend—and trust this Board . . . to get lower rates for borrowing or make lower rates for lending as the opportunity may come"); id., at 4987 (colloquy between Sens. Hebert and Bulkley) (observing that the Board has discretion in determining when savings and loans should be chartered in areas with existing local thrift institutions). 49 Thus, the HOLA did not simply incorporate existing local loan practices. Rather, Congress delegated to the Board broad authority to establish and regulate "a uniform system of [savings and loan] institutions where there are not any now," and to "establish them with the force of the government behind them, with a national charter." House Hearings 15 (Apr. 21, 1933) (statement of Chairman Stevenson); id., at 17 (Apr. 20, 1933).19 And the Board has exercised that discretion, regulating comprehensively the operations of these associations, including their lending practices and, specifically, the terms of loan instruments.20 C 50 As we noted above, a savings and loan's mortgage lending practices are a critical aspect of its "operation," over which the Board unquestionably has jurisdiction. Although the Board's power to promulgate regulations exempting federal savings and loans from the requirements of state law may not be boundless, in this case we need not explore the outer limits of the Board's discretion. We have no difficulty concluding that the due-on-sale regulation is within the scope of the Board's authority under the HOLA and consistent with the Act's principal purposes. 51 Congress delegated power to the Board expressly for the purpose of creating and regulating federal savings and loans so as to ensure that they would remain financially sound institutions able to supply financing for home construction and purchase. Thus, in testifying during the House hearings on the HOLA, the Board's Chairman observed: "The new corporations that we propose to set up, we want them set up on a sound basis as they will be of very material assistance in home financing for all time, if properly managed." House Hearings 12 (Apr. 21, 1933). And the relevant House and Senate Reports referred to the federal associations as "permanent" institutions. S.Rep.No.91, 73d Cong., 1st Sess., 2 (1933); H.R.Rep.No.55, 73d Cong., 1st Sess., 2 (1933). 52 The due-on-sale regulation was promulgated with these purposes in mind. The Board has determined that due-on-sale clauses are "a valuable and often an indispensable source of protection for the financial soundness of Federal associations and for their continued ability to fund new home loan commitments." 12 CFR § 556.9(f)(1) (1982). Specifically, the Board has concluded that the due-on-sale clause is "an important part of the mortgage contract" and that its elimination "will have an adverse [e]ffect on the earning power and financial stability of Federal associations, will impair the ability of Federal associations to sell their loans in the secondary markets, will reduce the amount of home-financing funds available to potential home buyers, and generally will cause a rise in home loan interest rates." Schott Advisory Opinion, at 2, 17-18. 53 The Board's analysis proceeds as follows: It observes that the federal associations' practice of borrowing short and lending long—obtaining funds on a short-term basis and investing them in long-term real estate loans, which typically have a 25- to 30-year term—combined with rising interest rates, has increased the cost of funds to these institutions and reduced their income. Exercising due-on-sale clauses enables savings and loans to alleviate this problem by replacing longterm, low-yield loans with loans at the prevailing interest rates and thereby to avoid increasing interest rates across the board. See id., at 21-22. Moreover, the Board has determined that restrictions like the Wellenkamp doctrine lengthen the expected maturity date of a lender's mortgages, thus reducing their marketability in the secondary mortgage market. As a result, the Board fears, "the financial stability of Federal associations in California will be eroded and the flow of home loan funds into California will be reduced." Schott Advisory Opinion at 34.21 54 Admittedly, the wisdom of the Board's policy decision is not uncontroverted.22 But neither is it arbitrary or capricious. As judges, it is neither our function, nor within our expertise, to evaluate the economic soundness of the Board's approach. In promulgating the due-on-sale regulation, the Board reasonably exercised the authority, given it by Congress, so as to ensure the financial stability of "local mutual thrift institutions in which people . . . invest their funds and . . . [which] provide for the financing of homes." § 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp.IV).23 By so doing, the Board intended to pre-empt conflicting state restrictions on due-on-sale practices like the California Supreme Court's Wellenkamp doctrine. 55 Our inquiry ends there. Accordingly, we hold that the Board's due-on-sale regulation bars application of the Wellenkamp rule to federal savings and loan associations.24 The judgment of the Court of Appeal is reversed. 56 It is so ordered. Justice POWELL took no part in the consideration or decision of this case. 57 Justice O'CONNOR, concurring. 58 I join in the Court's opinion but write separately to emphasize that the authority of the Federal Home Loan Bank Board to pre-empt state laws is not limitless.* Although Congress delegated broad power to the Board to ensure that federally chartered savings and loan institutions "would remain financially sound," ante, at 168, it is clear that HOLA does not permit the Board to pre-empt the application of all state and local laws to such institutions. Nothing in the language of § 5(a) of HOLA, which empowers the Board to "provide for the organization, incorporation, examination, operation, and regulation" of federally chartered savings and loans, remotely suggests that Congress intended to permit the Board to displace local laws, such as tax statutes and zoning ordinances, not directly related to savings and loan practices. Accordingly, in my view, nothing in the Court's opinion should be read to the contrary. 59 Justice REHNQUIST, with whom Justice STEVENS joins, dissenting. 60 The Court today concludes that in § 5(a) of the Home Owners' Loan Act of 1933 (HOLA), 12 U.S.C. § 1464(a) (1976 ed., Supp.IV), Congress authorized the Federal Home Loan Bank Board to preempt by administrative fiat California's limitations upon the enforceability of "due-on-sale" clauses in real estate mortgages held by federal savings and loan institutions. The Court reaches this extraordinary result by concluding that due-on-sale clauses relate to a savings and loan's mortgage lending practices which "are a critical aspect of its 'operation' over which the Board unquestionably has jurisdiction." Ante, at 167. Because I conclude that Congress has not authorized the Board to promulgate a regulation such as 12 CFR § 545.8-3(f) (1982), I dissent. 61 Section 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp.IV), unquestionably grants broad authority to the Board to regulate the mortgage lending practices of federal savings and loans. In order to perform this role, the Board may take into account state property and contract law which governs real estate transactions in general and the enforceability and interpretation of mortgage lending instruments in particular. Thus, it would be within the Board's power to determine that it constitutes an unsafe lending practice for a federal savings and loan to conclude a real property mortgage without a fully enforceable due-on-sale clause. It would be within the authority delegated to it by Congress for the Board to conclude that a due-on-sale clause must be included in a mortgage instrument as a means of enabling a federal savings and loan to remove unprofitable loans from its portfolio. 62 Such a regulation would be entirely consistent with the approach taken by Congress in regulating the savings and loan industry. In § 8 of the Federal Home Loan Bank Act of 1932 (FHLBA), 12 U.S.C. § 1428, the precursor to HOLA, Congress has required the Board to examine state law "relating to the conveying or recording of land titles, or to homestead and other rights, or to the enforcement of the rights of holders of mortgages on lands securing loans." (Emphasis added.) Section 8 provides further: 63 "If any such examination shall indicate, in the opinion of the board, that under the laws of any such State . . . there would be inadequate protection to a Federal Home Loan Bank in making or collecting advances under this chapter, the board may withhold or limit the operation of any Federal Home Loan Bank in such State until satisfactory conditions of law . . . shall be established." 12 U.S.C. § 1428 (emphasis added). 64 Thus, there is no indication in the FHLBA that the Board may, by promulgating regulations, pre-empt those state laws that are deemed to be economically unsound. Instead, if the Board concludes that California's limitations upon the enforceability of due-on-sale clauses endangers the soundness of the system established by the HOLA and the FHLBA, then the response contemplated by Congress is for the Board to "withhold or limit the operation" of the system in California. 65 In declaring the due-on-sale clause enforceable as a matter of federal law, however, the Board has departed from the approach contemplated by Congress. Although Congress has authorized the Board to regulate the lending activities of federal savings and loan associations, there is no indication in the HOLA itself, or in its legislative history, that Congress has empowered the Board to determine whether and when federal law shall govern the enforceability of particular provisions contained in mortgages concluded by federal savings and loan associations. If anything, § 8 of the FHLBA indicates that it was Congress' understanding in 1932 that the enforceability of provisions in mortgages is a matter of state law. Contract and real property law are traditionally the domain of state law. Aronson v. Quick Point Pencil Co., 440 U.S. 257, 262, 99 S.Ct. 1096, 1099, 59 L.Ed.2d 296 (1979); Butner v. United States, 440 U.S. 48, 55, 99 S.Ct. 914, 918, 59 L.Ed.2d 136 (1979). In the HOLA, Congress did not intend to create a federal common law of mortgages. See Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 101 S.Ct. 2061, 68 L.Ed.2d 500 (1981).* 66 The Board's attempt to enforce due-on-sale clauses as a matter of federal law cannot be upheld as a regulation of mortgage lending practices of federal savings and loan associations. In § 545.8-3(f), the Board has gone beyond regulating how, when, and in what manner a federal savings and loan may lend mortgage money. Instead, as the Court recognizes, ante, at 146-147, the Board's regulation purports to create a rule of law which will govern the rights and obligations of the parties to the mortgage instrument. This regulation does not simply delineate those provisions a federal savings and loan must or must not include in a mortgage instrument. Section 545.8-3(f) purports to guarantee the enforceability of a contractual provision notwithstanding state law to the contrary. In this case, the Board is not regulating the operation of federal savings and loan associations, but the operation of due-on-sale clauses. Without a congressional authorization more explicit than that relied upon by the Court, I conclude that the Board has entered a domain in which it is not authorized to override state laws. 67 The limitations the California courts have placed upon the enforceability of due-on-sale clauses do not impair the ability of the Board to regulate the manner in which federal savings and loan associations engage in mortgage lending. California has not interfered with the Board's determination that it constitutes an unsafe lending practice for a federal savings and loan to enter a loan agreement without a fully enforceable due-on-sale clause. California's rule regarding due-on-sale clauses is not invalid pursuant to the Supremacy Clause simply because it makes it difficult for lenders to eliminate unprofitable mortgage loans from their portfolios. 68 Although the Board has concluded that the California courts' limitations upon the enforceability of due-on-sale clauses is economically unsound, I cannot agree that Congress has enabled the Board to insulate federal savings and loans from California mortgage law merely by promulgating a regulation that declares these clauses to be enforceable. Discharge of its mission to ensure the soundness of federal savings and loans does not authorize the Federal Home Loan Bank Board to intrude into the domain of state property and contract law that Congress has left to the States. 1 The Board came into being under § 17 of the earlier Federal Home Loan Bank Act, 47 Stat. 736, as amended, 12 U.S.C. § 1437, the statute which created the federal home loan bank system. The three members of the Board are appointed by the President, with the advice and consent of the Senate, for 4-year terms. See note following 12 U.S.C. § 1437. In addition to providing for the establishment of federal savings and loan associations, the HOLA, by its § 3, 48 Stat. 129, repealed § 4(d) of the Federal Home Loan Bank Act, 47 Stat. 727, which had authorized federal home loan banks to make loans directly to homeowners. The HOLA, by its § 4, 48 Stat. 129, instructed the Board to create the Home Owners' Loan Corporation; this agency was to exchange its bonds for mortgages held by financial institutions, including state-chartered savings and loans, and to provide funds to needy homeowners for accrued taxes, maintenance, and repairs. 2 The due-on-sale clause used in many loan instruments is ¶ 17 of the uniform mortgage instrument developed by the Federal Home Loan Mortgage Corporation and the Federal National Mortgage Association. Paragraph 17 appears in two of the deeds of trust at issue in this case and reads: "17. Transfer of the Property; Assumption. If all or any part of the Property or an interest therein is sold or transferred by Borrower without Lender's prior written consent, excluding (a) the creation of a lien or encumbrance subordinate to this Deed of Trust, (b) the creation of a purchase money security interest for household appliances, (c) a transfer by devise, descent or by operation of law upon the death of a joint tenant or (d) the grant of any leasehold interest of three years or less not containing an option to purchase, Lender may, at Lender's option, declare all the sums secured by this Deed of Trust to be immediately due and payable. Lender shall have waived such option to accelerate if, prior to the sale or transfer, Lender and the person to whom the Property is to be sold or transferred reach agreement in writing that the credit of such person is satisfactory to Lender and that the interest payable on the sums secured by this Deed of Trust shall be at such rate as Lender shall request. If Lender has waived the option to accelerate provided in this paragraph 17 and if Borrower's successor in interest has executed a written assumption agreement accepted in writing by Lender, Lender shall release Borrower from all obligations under this Deed of Trust and the Note. "If Lender exercises such option to accelerate, Lender shall mail Borrower notice of acceleration in accordance with paragraph 14 hereof. Such notice shall provide a period of not less than 30 days from the date the notice is mailed within which Borrower may pay the sums declared due. If Borrower fails to pay such sums prior to the expiration of such period, Lender may, without further notice or demand on Borrower, invoke any remedies permitted by paragraph 18 hereof." App. 50-51, 85-86 (emphasis added). 3 The due-on-sale regulation was codified initially in 12 CFR § 545.6-11(f) (1982). See 44 Fed.Reg. 39108, 39149 (1979). 4 Even before adopting the due-on-sale regulation, the Board had interpreted 12 CFR § 545.8-3(a) (1982)—a regulation promulgated in 1948 that requires all loan instruments to "provide for full protection to the Federal association"—as authorizing federal savings and loans to exercise due-on-sale provisions, despite any state law to the contrary, because such clauses help ensure "full protection" to the lender. See the Board's Advisory Opinion, Resolution No. 75-647, in Schott v. Mission Federal Sav. & Loan Assn. (Schott Advisory Opinion), No. Civ-75-366, pp. 13-15 (CD Cal. July 30, 1975), reprinted as Exhibit A to Defendants' Memorandum of Points and Authorities in Opposition to Plaintiffs' Motion for Preliminary Injunction. 5 Paragraph 15 is also part of the uniform mortgage instrument developed by the Federal Home Loan Mortgage Corporation and the Federal National Mortgage Association. See n. 2, supra. The paragraph reads in full: "15. Uniform Deed of Trust; Governing Law; Severability. This form of deed of trust combines uniform covenants for national use and non-uniform covenants with limited variations by jurisdiction to constitute a uniform security instrument covering real property. This Deed of Trust shall be governed by the law of the jurisdiction in which the Property is located. In the event that any provision or clause of this Deed of Trust or the Note conflicts with applicable law, such conflicts shall not affect other provisions of this Deed of Trust or the Note which can be given effect without the conflicting provision, and to this end the provisions of the Deed of Trust and the Note are declared to be severable." App. 51-52, 86-87. 6 Each complaint also included a slander count, alleging that Fidelity had maliciously published false charges that the appellee was in default under the deed of trust. Id., at 9, 54, 89. 7 In addition, the Court of Appeal noted that two of the three deeds of trust at issue were executed prior to the effective date of § 545.8-3(f). Therefore, the court reasoned, the Board's due-on-sale regulation was not applicable to those loan instruments and could not pre-empt state law with respect to those deeds. See 121 Cal.App.3d, at 344, 345, 175 Cal.Rptr., at 476-477. 8 The Court of Appeal refused to ascribe any weight to the absence of ¶ 15 in the third deed of trust at issue here. The court described its earlier discussion of ¶ 15 as "not based so much on an agreement between the parties for the application of state law as on the conclusion that the general use of a provision containing such language by federal savings and loan associations with the approval of the Board persuasively evidences a recognition by the Board and federal savings and loan associations that state law would govern the interpretation, validity and enforcement of security instruments." Id., at 346, 175 Cal.Rptr., at 477. Nor did the court find significant the fact that this deed covered commercial rather than residential property. 9 A number of Federal District Courts have concluded that the Board's due-on-sale regulation pre-empts state law. See, e.g., Price v. Florida Federal Sav. & Loan Assn., 524 F.Supp. 175, 178 (MD Fla.1981) (§ 545.8-3(f) is pre-emptive of any state regulation); First Federal Sav. & Loan Assn. v. Peterson, 516 F.Supp. 732, 740 (ND Fla.1981) (§ 545.8-3(f) pre-empts Florida due-on-sale restrictions similar to those imposed by California); Dantus v. First Federal Sav. & Loan Assn., 502 F.Supp. 658, 661 (Colo.1980) (analogous ruling with respect to Colorado law); Bailey v. First Federal Sav. & Loan Assn., 467 F.Supp. 1139, 1141 (CD Ill.1979) (§ 545.8-3(f) forecloses any state regulation of due-on-sale practices of federal savings and loans), appeal dism'd, 636 F.2d 1221 (CA7 1980); Glendale Federal Sav. & Loan Assn. v. Fox, 459 F.Supp. 903, 907 (CD Cal.1978) (same), final summary judgment granted, 481 F.Supp. 616 (1979), order reversing and remanding, 663 F.2d 1078 (CA9 1981), cert. pending, No. 81-1192. One court appears to have agreed with the California Court of Appeal. See Holiday Acres No. 3 v. Midwest Federal Sav. & Loan Assn., 308 N.W.2d 471 (Minn.1981) (§ 545.8-3(f) does not pre-empt state regulation of due-on-sale clauses). In addition, at least three Federal Courts of Appeals, several District Courts, and one State Supreme Court have ruled that various other Board regulations supersede state law. See, e.g., Conference of Federal Sav. & Loan Assns. v. Stein, 604 F.2d 1256, 1260 (CA9 1979), ("In our judgment the regulatory control of the Bank Board over federal savings and loan associations is so pervasive as to leave no room for state regulatory control") summarily aff'd, 445 U.S. 921, 100 S.Ct. 1304, 63 L.Ed.2d 754 (1980); First Federal Sav. & Loan Assn. v. Greenwald, 591 F.2d 417, 425-426 (CA1 1979) (Board regulation specifying the conditions under which federal savings and loans must pay interest on escrow accounts pre-empts state law imposing greater interest requirements); Kupiec v. Republic Federal Sav. & Loan Assn., 512 F.2d 147, 150-152 (CA7 1975) (Board regulation supersedes any common-law right to inspect savings and loan's membership list); Meyers v. Beverly Hills Federal Sav. & Loan Assn., 499 F.2d 1145, 1147 (CA9 1974) (Board regulation pre-empts the field of prepayments of real estate loans to federal associations); Rettig v. Arlington Heights Federal Sav. & Loan Assn., 405 F.Supp. 819 (ND Ill.1975) (Board regulations and policy statements pre-empt the field of fiduciary duties of federal savings and loan officers); Lyons Sav. & Loan Assn. v. Federal Home Loan Bank Bd., 377 F.Supp. 11 (ND Ill.1974) (Board regulation displaces state law regarding branching of federal savings and loans); People v. Coast Federal Sav. & Loan Assn., 98 F.Supp. 311, 318 (SD Cal.1951) (federal regulation of savings and loans pre-empts the field); Kaski v. First Federal Sav. & Loan Assn., 72 Wis.2d 132, 141-142, 240 N.W.2d 367, 373 (1976) (federal law supersedes state regulation of federal savings and loans' lending practices). But see Derenco, Inc. v. Benjamin Franklin Federal Sav. & Loan Assn., 281 Or. 533, 577 P.2d 477 (Board regulation authorizing federal savings and loans to maintain reserve accounts for tax and insurance payments does not occupy the field of reserve accounts or pre-empt state law requiring payment of interest on such accounts), cert. denied, 439 U.S. 1051, 99 S.Ct. 733, 58 L.Ed.2d 712 (1978). Cf. Gulf Federal Sav. & Loan Assn. v. Federal Home Loan Bank Bd., 651 F.2d 259, 266 (CA5 1981) (Board has authority only over internal management of federal savings and loans, and not over disputed loan agreement provisions), cert. pending, No. 81-1744. 10 As a practical matter, however, few mortgage instruments are written without due-on-sale clauses. The Federal Home Loan Mortgage Corporation and the Federal National Mortgage Association, which purchase the bulk of mortgages sold in the secondary mortgage market, both require, in the mortgages they buy, either a due-on-sale clause or a provision enabling the lender to demand payment of the loan in seven years. The marketability of a mortgage in the secondary market is critical to a savings and loan, for it thereby can sell mortgages to obtain funds to make additional home loans. See Schott Advisory Opinion, at 28-34; Kinzler, Due-on-Sale Clauses: The Economic and Legal Issues, 43 U.Pitt.L.Rev. 441, 452-453 (1982); Comment, 9 Fla. State L.Rev. 645, 646, 650 (1981). 11 Title 12 CFR § 545.8-3(g) (1982), which applies to loans made after July 31, 1976, and secured by a home occupied or to be occupied by the borrower, prohibits the exercise of a due-on-sale clause in the same four circumstances listed in ¶ 17 of the uniform mortgage instrument, see n. 2, supra : when a lien subordinate to the lender's security instrument is created; when a purchase money security interest for household appliances is created; when a transfer occurs by devise, descent, or operation of law on the death of a joint tenant; or when a leasehold interest of not more than three years is granted with no option to purchase. Section 545.8-3(g) also bars the association from imposing a prepayment penalty when a loan is accelerated by means of a due-on-sale clause, and provides that, under specified circumstances, the lender waives its option to exercise a due-on-sale provision. 12 This principle likewise leads us to reject appellees' contention that, with respect to the two deeds of trust containing ¶ 15, see n. 5, supra, appellants did in fact agree to be bound by local law. Paragraph 15 provides that the deed is to be governed by the "law of the jurisdiction" in which the property is located; but the "law of the jurisdiction" includes federal as well as state law. Moreover, like ¶ 17—the due-on-sale clause in the uniform mortgage instrument, see n. 2, supra —¶ 15 typically must be included in any mortgage the Federal Home Loan Mortgage Corporation or the Federal National Mortgage Association purchases in the secondary mortgage market. See n. 10, supra. Paragraph 15 was added to the uniform mortgage instrument not to elevate state law over federal law, but to provide a uniform choice-of-law provision to be used when interstate disputes arose regarding the interpretation of a mortgage. See App. to Brief for Federal Home Loan Bank Board and Federal Home Loan Mortgage Corporation as Amici Curiae 2a (letter from Henry L. Judy, General Counsel, Federal Home Loan Mortgage Corporation); see also S.Rep.No.91-761, p. 25 (1970) (letter from Arthur F. Burns, Chairman of the Board of Governors, Federal Reserve System), U.S.Code Cong. & Admin.News 1970, p. 3488. 13 Citing Chrysler Corp. v. Brown, 441 U.S. 281, 315-316, 99 S.Ct. 1705, 1724, 60 L.Ed.2d 208 (1979), appellees characterize the preamble as an interpretative regulation that does not have the binding force of law and therefore cannot pre-empt state law. But Chrysler Corp. is not on point because we conclude that § 545.8-3(f) itself supersedes contrary state due-on-sale law; we look to the preamble only for the administrative construction of the regulation, to which "deference is . . . clearly in order." Udall v. Tallman, 380 U.S. 1, 16, 85 S.Ct. 792, 801, 13 L.Ed.2d 616 (1965). We need not consider, therefore, the pre-emptive effect of the preamble standing alone. 14 Because we find an actual conflict between federal and state law, we need not decide whether the HOLA or the Board's regulations occupy the field of due-on-sale law or the entire field of federal savings and loan regulation. 15 See § 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp.IV) (exempting federal mutual savings banks formerly organized under state law from "any numerical limitations of State law on the establishment of branch offices and other facilities"); and § 5(h) of the Act, § 1464(h) (pre-empting state taxes on federal savings and loans greater than those imposed on "other similar local mutual or cooperative thrift and home financing institutions"). Cf. § 13 of the Federal Home Loan Bank Act, 12 U.S.C. § 1433 (exempting Federal Home Loan Bank bonds from taxation). 16 See § 5(a) of the HOLA, 12 U.S.C. § 1464(a) (1976 ed., Supp.IV) (providing that any federal mutual savings bank which was formerly a state-chartered institution is subject to state laws pertaining to discrimination in lending based on neighborhood or geographic area, and to requirements imposed under the Consumer Credit Protection Act, 15 U.S.C. § 1601 et seq.); § 5(b)(3) of the Act, § 1464(b)(3) (1976 ed., Supp.IV) (authorizing federal savings and loans to borrow funds from a state mortgage finance agency "to the same extent as" state law permits state-chartered savings and loans to do so); and § 5(c)(4)(A) of the Act, § 1464(c)(4)(A) (1976 ed., Supp.IV) (permitting federal associations to invest in, or lend to, any business development credit corporation incorporated in the State "to the same extent as" state-chartered savings and loans are authorized to do so). 17 Likewise, we find nothing in § 8 of the Federal Home Loan Bank Act of 1932, 12 U.S.C. § 1428, relied on by the dissent, see post, at 173, that suggests any limit on the Board's authority to issue regulations pre-empting state law. That provision, which is not even part of the HOLA, speaks only to the Board's authority to examine state laws governing the operation of federal home loan banks, not federal savings and loans, for the purpose of ensuring "[a]dequate protection to a Federal Home Loan Bank in making or collecting advances under th[at] chapter . . . ." 12 U.S.C. § 1428. It does not purport to constrict the Board's power to regulate the operations of federal savings and loans and does not negate the explicit language and history of the HOLA. 18 On April 13, 1933, President F. D. Roosevelt wrote Congress, asking for "legislation to protect small home owners from foreclosure and to relieve them of a portion of the burden of excessive interest and principal payments incurred during the period of higher values and higher earning power." H.R.Doc.No.19, 73d Cong., 1st Sess., 1 (1933). Hearings were held by the House Committee on Banking and Currency on April 20 and 21, 1933, and by the Senate Committee on Banking and Currency on April 20 and 22. The bill was approved by the House on April 28, see 77 Cong.Rec. 2585, and passed the Senate on June 5, see id., at 4995. The President signed the bill into law on June 9, 1933, see id., at 6198, less than two months after he had first requested the legislation. 19 The postenactment history of the HOLA corroborates the Board's broad authority to regulate the lending practices of federal savings and loans. As part of the Financial Institutions Regulatory and Interest Rate Control Act of 1978, Pub.L. 95-630, 92 Stat. 3641, Congress amended § 5(a) of the HOLA to permit state mutual savings banks to obtain federal charters. During debate in the House, Representative Hanley introduced an amendment providing that those mutual savings banks opting to convert to federally chartered institutions would continue to be subject to state law pertaining to lending discrimination and to regulations imposed under the Consumer Credit Protection Act, 82 Stat. 146, as amended, 15 U.S.C. § 1601 et seq., if the Board determined that state law imposed more stringent requirements than federal law. See 124 Cong.Rec. 33847 (1978). Representative Hanley explained: "In no way, of course, would the use of State law requirements for Federal mutual savings banks be interpreted to erode the Bank Board's long-standing plenary authority over Federal savings and loan associations; Federal law alone would continue to govern these institutions in such areas as branching, anti-discrimination, and lending authority." Id., at 33848. Representative St Germain, chairman of the Subcommittee on Financial Institutions Supervision, Regulation, and Insurance of the House Committee on Banking, Finance, and Urban Affairs and chief sponsor of the bill, agreed: "This restriction applies only to converted mutual savings banks, and Congress in no way intends to interfere with the longstanding, all-inclusive power of the Bank Board over the activities of Federal savings and loan associations, including branching authority." Id., at 33849. The amendment was agreed to. Ibid. Similar views were expressed during the Senate debate on the bill. Senator Brooke observed that "we do not intend to interfere with the Bank Board's plenary authority over Federal savings and loan associations, and in this area, Federal law alone would continue to govern." Id., at 36148. Then, during debate in the House on the Depository Institutions Deregulation and Monetary Control Act of 1980, Pub.L. 96-221, 94 Stat. 132, one Congressman expressed concern that permitting federal savings and loans to make residential real estate loans to the same extent national banking associations were authorized to do so might be interpreted as making "federal savings and loans . . . subject to State requirements." 126 Cong.Rec. 6981 (1980) (remarks of Rep. Patterson). Representative St Germain responded that the Act would expand the federal associations' investment powers "[o]nly if the Federal Home Loan Bank Board permits. Under the Home Owners' Act, the Bank Board has complete authority to determine by regulation the lending practices of Federal associations." Ibid. Although these postenactment events cannot be accorded the weight of contemporary history, they do provide further confirmation of Congress' intent to delegate to the Board broad discretion in regulating the lending practices of federal savings and loans. See NLRB v. Bell Aerospace Co., 416 U.S. 267, 275, 94 S.Ct. 1757, 1762, 40 L.Ed.2d 134 (1974); Red Lion Broadcasting Co. v. FCC, 395 U.S. 367, 380-381, 89 S.Ct. 1794, 1801, 23 L.Ed.2d 371 (1969). 20 The Board's extensive regulations govern, for example, fair credit requirements, the types and amount of loans, collateral required, repayment schedules, initial loan charges, assignment of rents, escrow accounts and interest paid on those accounts, late charges, servicing of loans, and loan payments and prepayments. See 12 CFR §§ 545.6, 545.8 (1982). 21 The Board's Due-on-Sale Task Force estimates that the California Supreme Court's restrictions on the exercise of due-on-sale clauses accounted for 40% of the total losses suffered in 1981 by state-chartered associations in the State—some $200 million. See Federal Home Loan Bank Board, Due-on-Sale Task Force Report 2, 15 (1982). The Task Force projects that imposition of such restrictions nationwide would create, within two years, annual losses of $600 to $800 million for federal savings and loans, and $1 to $1.3 billion for all federal and state associations. See id., at 2, 18, 25. 22 Those subscribing to the opposite view contend that the unrestricted exercise of due-on-sale clauses may preclude the assumption of mortgages at lower interest rates, thus preventing the sale of homes and transferring the burden of an inflationary market from the lender to the homeowner and prospective homeowner. See, e.g., Patton v. First Federal Sav. & Loan Assn., 118 Ariz. 473, 578 P.2d 152 (1978); Wellenkamp v. Bank of America, 21 Cal.3d 943, 148 Cal.Rptr. 379, 582 P.2d 970 (1978); Nichols v. Ann Arbor Federal Sav. & Loan Assn., 73 Mich.App. 163, 250 N.W.2d 804 (1977). A number of courts, however, have agreed with the Board's approach. See, e.g., Williams v. First Federal Sav. & Loan Assn., 651 F.2d 910 (CA4 1981); Tierce v. APS Co., 382 So.2d 485 (Ala.1979); Malouff v. Midland Federal Sav. & Loan Assn., 181 Colo. 294, 509 P.2d 1240 (1973); Martin v. Peoples Mutual Sav. & Loan Assn., 319 N.W.2d 220 (Iowa 1982); Occidental Savings & Loan Assn. v. Venco Partnership, 206 Neb. 469, 293 N.W.2d 843 (1980); Crockett v. First Federal Sav. & Loan Assn., 289 N.C. 620, 224 S.E.2d 580 (1976); Gunther v. White, 489 S.W.2d 529 (Tenn.1973). 23 We therefore reject appellees' contention that the Board's power to regulate federal savings and loans extends only to the associations' internal management and not to any external matters, such as their relationship with borrowers. Although one federal and one state court have drawn this distinction, see Gulf Federal Sav. & Loan Assn. v. Federal Home Loan Bank Bd., 651 F.2d, at 266; Holiday Acres No. 3 v. Midwest Federal Sav. & Loan Assn., 308 N.W.2d, at 478, we find no support in the language of the HOLA or its legislative history for such a restriction on the Board's authority. Moreover, whatever validity the distinction has in theory, it makes little sense here. As the Wisconsin Supreme Court recognized, "[t]he regulation of loan practices directly affects the internal management and operations of federal associations and therefore requires uniform federal control." Kaski v. First Federal Sav. & Loan Assn., 72 Wis.2d, at 142, 240 N.W.2d, at 373. In fact, as discussed in the text, the Board's due-on-sale policy is based on the view that due-on-sale clauses are essential to the financial soundness of federal savings and loans; preservation of the associations' very existence is obviously related to their internal management and is one of the functions delegated to the Board by Congress. 24 Pointing out that two of the deeds of trust were executed prior to the 1976 effective date of § 545.8-3(f), appellees argue that the due-on-sale regulation may not be applied so as to destroy vested rights. Therefore, appellees reason, California law does not conflict with federal law with respect to those two deeds. Appellants respond that § 545.8-3(f) did not interfere with appellees' rights because it merely codified pre-existing law. See n. 4, supra. When the two deeds of trust were executed in 1971 and 1972, California law permitted the unrestricted exercise of due-on-sale clauses upon outright transfer of the security property, as occurred here. The Board's due-on-sale regulation was then issued in 1976, reinforcing Fidelity's right to enforce the due-on-sale provisions. Not until Wellenkamp was decided in 1978 was a lender's right under California law to accelerate a loan in response to an outright transfer limited to cases where the security was impaired. The California Supreme Court's prior cases, which forbade the automatic enforcement of due-on-sale provisions when the borrower further encumbered the property securing the loan, La Sala v. American Savings & Loan Assn., 5 Cal.3d 864, 97 Cal.Rptr. 849, 489 P.2d 1113 (1971), and when the borrower entered into an installment land contract covering all or part of the security property, Tucker v. Lassen Savings & Loan Assn., 12 Cal.3d 629, 116 Cal.Rptr. 633, 526 P.2d 1169 (1974), permitted the unrestricted exercise of due-on-sale clauses in cases of outright transfers of the security. See 5 Cal.3d, at 880, 97 Cal.Rptr., at 859, 489 P.2d, at 1123; 12 Cal.3d, at 637-638, 116 Cal.Rptr., at 638-639, 526 P.2d, at 1174-1175. Because we find the Wellenkamp doctrine pre-empted by a previously promulgated federal regulation and therefore inapplicable to federal savings and loans, appellees are deprived of no vested rights if Fidelity is permitted to enforce the due-on-sale clauses in the two pre-1976 deeds: the savings and loan had the right to accelerate the loans, pursuant to California law, when the deeds were executed, and that power was never diminished by state law. We have no occasion, therefore, to consider whether § 545.8-3(f) may be applied so as to give a savings and loan broader authority to enforce a due-on-sale clause than it had when the deed of trust was executed, or to address appellants' contention that § 545.8-3(f) effected no change in the law. * At one point in today's opinion, the Court states that "we need not decide whether the HOLA or the Board's regulations occupy . . . the entire field of federal savings and loan regulation." Ante, at 159, n. 14. * The Board, however, has argued that federal common law does govern the contractual relationship between federal savings and loan institutions and their mortgagors. See Gulf Federal Sav. & Loan v. Federal Home Loan Bank Bd., 651 F.2d 259, 266 (CA5 1981), cert. pending, No. 81-1744; Brief for Federal Home Loan Bank Board et al. as Amici Curiae 26, n. 21.

910

458 U.S. 1 102 S.Ct. 2977 73 L.Ed.2d 563 John S. TOLL, President, University of Maryland et al., Petitionersv.Juan Carlos MORENO et al. No. 80-2178. Argued March 2, 1982. Decided June 28, 1982. Syllabus Held: The University of Maryland's policy of categorically denying domiciled nonimmigrant aliens who hold G-4 visas (visas issued to nonimmigrant aliens who are officers or employees of certain international organizations and to members of their immediate families) in-state status under which preferential treatment is given to domiciled citizen and immigrant alien students for purposes of tuition and fees, is invalid under the Supremacy Clause. Pp. 10-19. (a) "[S]tate regulation not congressionally sanctioned that discriminates against aliens lawfully admitted to the country is impermissible if it imposes additional burdens not contemplated by Congress." De Canas v. Bica, 424 U.S. 351, 358, n. 6, 96 S.Ct. 933, 938, n. 6, 47 L.Ed.2d 43. Here, in light of Congress' explicit decision in the Immigration and Nationality Act of 1952 not to bar G-4 aliens from acquiring domicile in the United States, the State's decision to deny "in-state" status to G-4 aliens, solely on account of such aliens' immigration status, amounts to an ancillary "burden not contemplated by Congress" in admitting these aliens to the United States. Moreover, by imposing on domiciled G-4 aliens higher tuition and fees than are imposed on other domiciliaries of the State, the University's policy frustrates the federal policies embodied in the special tax exemptions afforded G-4 aliens by various treaties, international agreements, and federal statutes. Pp. 10-17. (b) The Eleventh Amendment did not preclude the District Court from ordering the University to pay refunds to various G-4 alien class members who would have obtained in-state status but for the stay, pending appeal, of that court's original order granting partial summary judgment in favor of the named plaintiffs, where the University, in seeking the stay, represented that if the order was affirmed on appeal, it would make appropriate refunds. Contrary to petitioners' contention, the order was not vacated when this Court vacated the Court of Appeals' judgment affirming the District Court and remanded the case to the District Court for reconsideration. Pp. 17-19. 4 Cir., 645 F.2d 217, affirmed. Robert A. Zarnoch, Annapolis, Md., for petitioners. James R. Bieke, Washington, D. C., for respondents. Justice BRENNAN delivered the opinion of the Court. 1 The state-operated University of Maryland grants preferential treatment for purposes of tuition and fees to students with "in-state" status. Although citizens and immigrant aliens may obtain in-state status upon a showing of domicile within the State, nonimmigrant aliens, even if domiciled, are not eligible for such status. The question in this case is whether the University's in-state policy is invalid under the Supremacy Clause of the Constitution, insofar as the policy categorically denies in-state status to domiciled nonimmigrant aliens who hold G-4 visas. 2 * The factual and procedural background of this case, which has prompted two prior decisions of this Court, requires some elaboration. The focus of the controversy has been a policy adopted by the University in 1973 governing the eligibility of students for in-state status with respect to admission and fees. The policy provides in relevant part: 3 "1. It is the policy of the University of Maryland to 4 grant in-state status for admission, tuition and charge-differential purposes to United States citizens, and to immigrant aliens lawfully admitted for permanent residence in accordance with the laws of the United States, in the following cases: "a. Where a student is financially dependent upon a parent, parents, or spouse domiciled in Maryland for at least six consecutive months prior to the last day available for registration for the forthcoming semester. 5 "b. Where a student is financially independent for at least the preceding twelve months, and provided the student has maintained his domicile in Maryland for at least six consecutive months immediately prior to the last day available for registration for the forthcoming semester." App. to Pet. for Cert. 167a-168a. 6 In 1975, when this action was filed, respondents Juan Carlos Moreno, Juan Pablo Otero, and Clare B. Hogg were students at the University of Maryland. Each resided with, and was financially dependent on, a parent who was a nonimmigrant alien holding a "G-4" visa. Such visas are issued to nonimmigrant aliens who are officers or employees of certain international organizations, and to members of their immediate families. 66 Stat. 168, 8 U.S.C. § 1101(a)(15)(G)(iv).1 Despite respondents' residence in the State, the University denied them in-state status pursuant to its policy of excluding all nonimmigrant aliens. Seeking declaratory and injunctive relief, the three respondents filed a class action against the University of Maryland and its President.2 They contended that the University's policy violated various federal laws, the Due Process and Equal Protection Clauses of the Fourteenth Amendment, and the Supremacy Clause. 7 The District Court granted partial summary judgment in favor of the three named plaintiffs and the class of G-4 visa holders represented by them.3 In the view of the District Court, the University's denial of in-state status to these plaintiffs rested upon an irrebuttable presumption that a G-4 alien cannot establish Maryland domicile. Concluding that the presumption was "not universally true" as a matter of either federal or Maryland law, the District Court held that under Vlandis v. Kline, 412 U.S. 441, 93 S.Ct. 2230, 37 L.Ed.2d 63 (1973), the in-state policy violated the Due Process Clause of the Fourteenth Amendment. Moreno v. University of Maryland, 420 F.Supp. 541, 559 (Md.1976). Accordingly, in an order dated July 13, 1976, the District Court enjoined the President of the University4 from denying respondents the opportunity to establish in-state status solely on the basis of an "irrebuttable presumption of non-domicile." Id., at 565.5 The court stayed its order pending appeal in reliance on the University's representation that it would make appropriate refunds "in the event the Court's Order of July 13, 1976, were finally affirmed on appeal." App. to Pet. for Cert. 100a. The Court of Appeals for the Fourth Circuit affirmed, adopting the reasoning of the District Court. Id., at 102a.6 Affirmance order reported at 556 F.2d 573 (1977). 8 We reviewed the case on writ of certiorari. Elkins v. Moreno, 435 U.S. 647, 98 S.Ct. 1338, 55 L.Ed.2d 614 (1978). We held that "[b]ecause petitioner makes domicile the 'paramount' policy consideration and because respondents' contention is that they can be domiciled in Maryland but are conclusively presumed to be unable to do so, this case is squarely within Vlandis as limited by [Weinberger v.] Salfi, [422 U.S. 749, 95 S.Ct. 2457, 45 L.Ed.2d 522 (1975) ]." Id., at 660, 98 S.Ct., at 1346.7 It was therefore necessary to decide whether the presumption was universally true. With respect to federal law, we concluded that G-4 visa holders could "adopt the United States as their domicile." Id., at 666, 98 S.Ct., at 1349.8 We were thus left with the "potentially dispositive" question whether G-4 aliens are as a matter of state law incapable of becoming domiciliaries of Maryland. We certified this question to the Maryland Court of Appeals.9 The state court answered the certified question in the negative, advising us that "nothing in the general Maryland law of domicile renders G-4 visa holders, or their dependents, incapable of becoming domiciled in this State." Toll v. Moreno, 284 Md. 425, 444, 397 A.2d 1009, 1019 (1979). 9 After our certification, but before the state court's response, the University adopted a "clarifying resolution" concerning its in-state policy.10 By its terms the resolution did not offer a new definition of "in-state" students; rather, it purported to "reaffirm" the existing policy.11 The resolution indicated, however, that the University's policy, "insofar as it denies in-state status to nonimmigrant aliens, serves a number of substantial purposes and interests, whether or not it conforms to the generally or otherwise applicable definition of domicile under the Maryland common law." App. to Pet. for Cert. 173a. The interests assertedly served by the policy were described in the following terms: 10 "(a) limiting the University's expenditures by granting a higher subsidy toward the expenses of providing educational services to that class of persons who, as a class, are more likely to have a close affinity to the State and to contribute more to its economic well-being; 11 "(b) achieving equalization between the affected classes of the expenses of providing educational services; 12 "(c) efficiently administering the University's in-state determination and appeals process; and "(d) preventing disparate treatment among categories of nonimmigrants with respect to admissions, tuition, and charge-differentials." Id., at 173a-174a. 13 Following the Maryland Court of Appeals' decision, the case returned to this Court. But we declined to restore the case to the active docket for full briefing and argument, concluding that the University's clarifying resolution had "fundamentally altered the posture of the case." Toll v. Moreno, 441 U.S. 458, 461, 99 S.Ct. 2044, 2045, 60 L.Ed.2d 354 (1979) (per curiam ). We noted that "if domicile [was] not the 'paramount' policy consideration of the University, this case [was] no longer 'squarely within Vlandis as limited by Salfi,' " and thus raised "new issues of constitutional law which should be addressed in the first instance by the District Court." Id., at 461-462, 99 S.Ct., at 2045-2046, quoting Elkins v. Moreno, supra, at 660, 98 S.Ct., at 1346.12 Accordingly, we vacated the judgment of the Court of Appeals and remanded the case "to the District Court for further consideration in light of our opinion and judgment in Elkins, the opinion and judgment of the Maryland Court of Appeals in Toll, and the Board of Regents' clarifying resolution of June 23, 1978." 441 U.S., at 462, 99 S.Ct., at 2046. 14 On remand, the District Court determined that the clarifying resolution constituted a change in the University's position. Before that resolution, the University's primary concern had in fact been domicile; after the resolution, domicile was no longer "the paramount consideration in the University's policy." 480 F.Supp. 1116, 1124 (Md.1979). Thus, with respect to the period preceding the issuance of the resolution, the District Court reaffirmed its earlier determination that insofar as the policy precluded G-4 aliens (or their dependents) from acquiring in-state status, it denied due process under Vlandis. 480 F.Supp., at 1122-1125. With respect to the period following the promulgation of the resolution, however, the court held that Vlandis did not control: The University had abandoned its position that G-4 aliens could not establish domicile in Maryland. 480 F.Supp., at 1125. Nevertheless, the District Court concluded that the revised in-state policy was constitutionally invalid, basing its conclusion on two alternative grounds. First, the court held that the policy ran afoul of the Equal Protection Clause of the Fourteenth Amendment. According to the court, the challenged portion of the University's policy contained a classification based on alienage, requiring strict scrutiny, an analysis which the policy did not survive, since the policy did not further any compelling interest. 489 F.Supp. 658, 660-667 (Md.1980). Alternatively, the court held that the in-state policy violated the Supremacy Clause by encroaching upon Congress' prerogatives with respect to the regulation of immigration. Id., at 667-668.13 15 The Court of Appeals affirmed for "reasons sufficiently stated" by the District Court. Moreno v. University of Maryland, 645 F.2d 217, 220 (1981) (per curiam ). We granted certiorari. 454 U.S. 815, 102 S.Ct. 91, 70 L.Ed.2d 83 (1981). For the reasons that follow, we hold that the University of Maryland's in-state policy, as applied to G-4 aliens and their dependents, violates the Supremacy Clause of the Constitution,14 and on that ground affirm the judgment of the Court of Appeals. We therefore have no occasion to consider whether the policy violates the Due Process or Equal Protection Clauses. II 16 Our cases have long recognized the preeminent role of the Federal Government with respect to the regulation of aliens within our borders. See, e.g., Mathews v. Diaz, 426 U.S. 67, 96 S.Ct. 1883, 48 L.Ed.2d 478 (1976); Graham v. Richardson, 403 U.S. 365, 377-380, 91 S.Ct. 1848, 1854-1856, 29 L.Ed.2d 534 (1971); Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 418-420, 68 S.Ct. 1138, 1142-1143, 92 L.Ed. 1478 (1948); Hines v. Davidowitz, 312 U.S. 52, 62-68 (1941); Truax v. Raich, 239 U.S. 33, 42, 36 S.Ct. 7, 11, 60 L.Ed. 131 (1915). Federal authority to regulate the status of aliens derives from various sources, including the Federal Government's power "[t]o establish [a] uniform Rule of Naturalization," U.S.Const., Art. I, § 8, cl. 4, its power "[t]o regulate Commerce with foreign Nations", id., cl. 3, and its broad authority over foreign affairs, see United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 318, 57 S.Ct. 216, 220, 81 L.Ed. 255 (1936); Mathews v. Diaz, supra, at 81, n. 17, 96 S.Ct., at 1892, n. 17; Harisiades v. Shaughnessy, 342 U.S. 580, 588-589, 72 S.Ct. 512, 518-519, 96 L.Ed. 586 (1952). 17 Not surprisingly, therefore, our cases have also been at pains to note the substantial limitations upon the authority of the States in making classifications based upon alienage. In Takahashi v. Fish & Game Comm'n, supra, we considered a California statute that precluded aliens who were "ineligible for citizenship under federal law" from obtaining commercial fishing licenses, even though they "met all other state requirements" and were lawful inhabitants of the State. 334 U.S., at 414, 68 S.Ct., at 1140.15 In seeking to defend the statute, the State argued that it had "simply followed the Federal Government's lead" in classifying certain persons as "ineligible for citizenship." Id., at 418, 68 S.Ct., at 1142. We rejected the argument, stressing the delicate nature of the federal-state relationship in regulating aliens: 18 "The Federal Government has broad constitutional powers in determining what aliens shall be admitted to the United States, the period they may remain, regulation of their conduct before naturalization, and the terms and conditions of their naturalization. Under the Constitution the states are granted no such powers; they can neither add to nor take from the conditions lawfully imposed by Congress upon admission, naturalization and residence of aliens in the United States or the several states. State laws which impose discriminatory burdens upon the entrance or residence of aliens lawfully within the United States conflict with this constitutionally derived federal power to regulate immigration, and have accordingly been held invalid." Id., at 419, 68 S.Ct., at 1142 (emphasis added) (citation and footnote omitted).16 19 The decision in Graham v. Richardson, supra, followed directly from Takahashi. In Graham we held that a State may not withhold welfare benefits from resident aliens "merely because of their alienage." 403 U.S., at 378, 91 S.Ct., at 1855. Such discrimination, the Court concluded, would not only violate the Equal Protection Clause, but would also encroach upon federal authority over lawfully admitted aliens. In support of the latter conclusion, the Court noted that Congress had "not seen fit to impose any burden or restriction on aliens who become indigent after their entry into the United States," id., at 377, 91 S.Ct., at 1855, but rather had chosen to afford "lawfully admitted resident aliens . . . the full and equal benefit of all state laws for the security of persons and property," id., at 378, 91 S.Ct., at 1855. The States had thus imposed an "auxiliary burde[n] upon the entrance or residence of aliens" that was never contemplated by Congress. Id., at 379, 91 S.Ct., at 1855. 20 Read together, Takahashi and Graham stand for the broad principle17 that "state regulation not congressionally sanctioned that discriminates against aliens lawfully admitted to the country is impermissible if it imposes additional burdens not contemplated by Congress." De Canas v. Bica, 424 U.S. 351, 358, n. 6, 96 S.Ct. 933, 938, n. 6, 47 L.Ed.2d 43 (1976).18 To be sure, when Congress has done nothing more than permit a class of aliens to enter the country temporarily, the proper application of the principle is likely to be a matter of some dispute. But the instant case does not present such a situation, and there can be little doubt regarding the invalidity of the challenged portion of the University's in-state policy. 21 The Immigration and Nationality Act of 1952, 66 Stat. 163, as amended, 8 U.S.C. § 1101 et seq. (1976 ed. and Supp.IV), represents "a comprehensive and complete code covering all aspects of admission of aliens to this country, whether for business or pleasure, or as immigrants seeking to become permanent residents." Elkins v. Moreno, 435 U.S., at 664, 98 S.Ct., at 1348. The Act recognizes two basic classes of aliens, immigrant and nonimmigrant.19 With respect to the nonimmigrant class, the Act establishes various categories, the G-4 category among them. For many of these nonimmigrant categories, Congress has precluded the covered alien from establishing domicile in the United States. Id., at 665, 98 S.Ct., at 1349.20 But significantly, Congress has allowed G-4 aliens—employees of various international organizations, and their immediate families to enter the country on terms permitting the establishment of domicile in the United States. Id., at 666, 98 S.Ct., at 1349. In light of Congress' explicit decision not to bar G-4 aliens from acquiring domicile, the State's decision to deny "in-state" status to G-4 aliens, solely on account of the G-4 alien's federal immigration status, surely amounts to an ancillary "burden not contemplated by Congress" in admitting these aliens to the United States. We need not rely, however, simply on Congress' decision to permit the G-4 alien to establish domicile in this country; the Federal Government has also taken the additional affirmative step of conferring special tax privileges on G-4 aliens. 22 As a result of an array of treaties, international agreements, and federal statutes, G-4 visaholders employed by the international organizations described in 8 U.S.C. § 1101(a)(15)(G)(iv) are relieved of federal and, in many instances, state and local taxes on the salaries paid by the organizations. For example, the international agreements governing the international banks for which the parents of the named respondents are employed specifically exempt the parents from all taxes on their organizational salaries. See Articles of Agreement of the International Bank for Reconstruction and Development, Art. VII, § 9(b), 60 Stat. 1458, T.I.A.S. No. 1502 (1945) ("No tax shall be levied on or in respect of salaries and emoluments paid by the Bank to executive directors, alternates, officials or employees of the Bank who are not local citizens, local subjects, or other local nationals"); Agreement Establishing the Inter-American Development Bank, Art. XI, § 9(b), [1959] 10 U.S.T. 3029, 3096, T.I.A.S. No. 4397 (1959) ("No tax shall be levied on or in respect of salaries and emoluments paid by the Bank to . . . employees of the Bank who are not local citizens or other local nationals").21 Not only have some of the specific tax exemptions contained in international agreements been incorporated into a federal statute, see 22 U.S.C. § 286h, but also the International Organizations Immunities Act has explicitly afforded a federal tax exemption for those G-4 visaholders employed by international organizations for which no treaty or international agreement has provided a tax exemption for foreign employees.22 § 4(b), 59 Stat. 670, reenacted, 68A Stat. 284, as § 893 of the Internal Revenue Code of 1954, 26 U.S.C. § 893 ("Wages, fees, or salary of any employee [except citizens of the United States and of the Republic of the Philippines] of . . . an international organization . . ., received as compensation for official services to such . . . international organization shall not be included in gross income and shall be exempt from [federal] taxation"). 23 In affording G-4 visaholders such tax exemption, the Federal Government has undoubtedly sought to benefit the employing international organizations by enabling them to pay salaries not encumbered by the full panoply of taxes, thereby lowering the organizations' costs. See 41 Op.Atty.Gen. 170, 172-173 (1954). The tax benefits serve as an inducement for these organizations to locate significant operations in the United States. See, e.g., H.R.Rep.No.1203, 79th Cong., 1st Sess., 2-3 (1945); S.Rep.No.861, 79th Cong., 1st Sess., 2-3 (1945). By imposing on those G-4 aliens who are domiciled in Maryland higher tuition and fees than are imposed on other domiciliaries of the State, the University's policy frustrates these federal policies. Petitioners' very argument in this Court only buttresses this conclusion. One of the grounds on which petitioners have sought to justify the discriminatory burden imposed on the named respondents is that the salaries their parents receive from the international banks for which they work are exempt from Maryland income tax. Indeed, petitioners suggest that the "dollar differential . . . at stake here [is] an amount roughly equivalent to the amount of state income tax an international bank parent is spared by treaty each year." Brief for Petitioners 23 (footnote omitted). But to the extent this is indeed a justification for the University's policy with respect to the named respondents, it is an impermissible one: The State may not recoup indirectly from respondents' parents the taxes that the Federal Government has expressly barred the State from collecting.23 In sum, the Federal Government has not merely admitted G-4 aliens into the country; it has also permitted them to establish domicile and afforded significant tax exemptions on organizational salaries. In such circumstances, we cannot conclude that Congress ever contemplated that a State, in the operation of a university, might impose discriminatory tuition charges and fees solely on account of the federal immigration classification.24 We therefore conclude that insofar as it bars domiciled G-4 aliens (and their dependents) from acquiring in-state status, the University's policy violates the Supremacy Clause.25 III 24 Finally, we must address petitioners' contention that the Eleventh Amendment precluded the District Court from ordering the University to pay refunds to various class members who would have obtained in-state status but for the stay of the District Court's original order of July 13, 1976. As petitioners concede, in seeking a stay of that order the University made the representation to the District Court that in the event the 1976 order was "finally affirmed on appeal," it would make appropriate refunds. This representation was incorporated in the stay orders of both the District Court and Court of Appeals. It is petitioners' contention, however, that the 1976 order was "effectively" vacated when this Court, in Toll v. Moreno, 441 U.S. 458, 99 S.Ct. 2044, 60 L.Ed.2d 354 (1979), vacated the judgment of the Court of Appeals and remanded the case to the District Court for reconsideration. Petitioners therefore conclude that the terms of the University's waiver of sovereign immunity can no longer be satisfied. 25 Petitioners' argument is not persuasive. We do not interpret Toll as having vacated the judgment of the District Court. In Toll the Court recognized that the University had altered its position through the promulgation of the clarifying resolution, raising "new issues of constitutional law which should be addressed in the first instance by the District Court." Id., at 462, 99 S.Ct., at 2046. The Court declined, however, to decide whether the District Court, in issuing its 1976 order, had improperly relied on due process grounds, and whether continuation of the order was justified on equal protection or pre-emption grounds. Thus, while we vacated "the judgment of the Court of Appeals," ibid., we left the judgment of the District Court undisturbed.26 And contrary to petitioners' suggestion, a vacatur of the District Court's judgment was not necessary to give the District Court jurisdiction to reconsider the case. See Goldberg v. United States, 425 U.S. 94, 111-112, 96 S.Ct. 1338, 1348-1349, 47 L.Ed.2d 603 (1976); Campbell v. United States, 365 U.S. 85, 98-99, 81 S.Ct. 421, 428, 5 L.Ed.2d 428 (1961); 28 U.S.C. § 2106 ("The Supreme Court . . . may affirm, modify, vacate, set aside or reverse any judgment . . . and may . . . require such further proceedings to be had as may be just under the circumstances").27 IV 26 For the foregoing reasons, the judgment of the Court of Appeals is 27 Affirmed. 28 Justice BLACKMUN, concurring. 29 I join the Court's opinion. Its action today provides an eloquent and sufficient answer to Justice REHNQUIST's dissent: despite the vehemence with which his opinion is written, Justice REHNQUIST has persuaded only one Justice to his position. But because the dissent attempts to plumb the Court's psyche, see post, at 41-42, n. 12,1 I feel compelled to add comments addressed to Justice REHNQUIST's ruminations on equal protection. In particular, I cannot leave unchallenged his suggestion that the Court's decisions holding resident aliens to be a "suspect class" no longer are good law. 30 Justice REHNQUIST's analysis on this point is based on a simple syllogism. Alienage classifications have been subjected to strict scrutiny, he suggests, because "aliens [are] barred from asserting their interests in the governmental body responsible for imposing burdens upon them." Post, at 40. But "[m]ore recent decisions," he continues, have established that "the political powerlessness of aliens is itself the consequence of distinctions on the basis of alienage that are constitutionally permissible." Ibid. This prompts Justice REHNQUIST to pose what one supposes to be a rhetorical question: "whether political powerlessness is any longer a legitimate reason for treating aliens as a 'suspect class' deserving of 'heightened judicial solicitude.' " Post, at 41. The reader would infer from this analysis that Justice REHNQUIST would uphold state enactments disadvantaging aliens unless those enactments are wholly irrational. 31 With respect, in my view it is Justice REHNQUIST's analysis that is wholly irrational; simply to state his proposition is to demonstrate its logical flaws. Most obviously, his exegesis of the Court's reasons for according aliens "suspect class" status is simplistic to the point of caricature. By labeling aliens a " 'discrete and insular' minority," Graham v. Richardson, 403 U.S. 365, 372, 91 S.Ct. 1848, 1852, 29 L.Ed.2d 534 (1971), the Court did something more than provide an historical description of their political standing. That label also reflected the Court's considered conclusion that for most legislative purposes there simply are no meaningful differences between resident aliens and citizens, see Ambach v. Norwick, 441 U.S. 68, 75, 99 S.Ct. 1589, 1593, 60 L.Ed.2d 49 (1979), so that aliens and citizens are "persons similarly circumstanced" who must "be treated alike." F. S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415, 40 S.Ct. 560, 562, 64 L.Ed. 989 (1920). At the same time, both common experience and the unhappy history reflected in our cases, see Cabell v. Chavez-Salido, 454 U.S. 432, 462-463, 102 S.Ct. 735, 747-748, 70 L.Ed.2d 677 (1982) (dissenting opinion); Ambach v. Norwick, 441 U.S., at 82, 99 S.Ct., at 1597 (dissenting opinion), demonstrate that aliens often have been the victims of irrational discrimination. 32 In combination, these factors—disparate treatment accorded a class of "similarly circumstanced" persons who historically have been disabled by the prejudice of the majority the Court to conclude that alienage classifications "in themselves supply a reason to infer antipathy," Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 272, 99 S.Ct. 2282, 2292, 60 L.Ed.2d 870 (1979), and therefore demand close judicial scrutiny. This understanding, which is at the heart of the Court's modern alienage decisions, was unreservedly reaffirmed this Term in Cabell v. Chavez-Salido, 454 U.S., at 438, 102 S.Ct., at 739 ("citizenship is not a relevant ground for the distribution of economic benefits"). 33 Justice REHNQUIST nevertheless suggests that the Court's original understanding somehow has been undercut by "more recent decisions" recognizing that aliens may be excluded from the governmental process. For this proposition he cites Cabell v. Chavez-Salido, supra; Ambach v. Norwick, supra ; and Foley v. Connelie, 435 U.S. 291, 98 S.Ct. 1067, 55 L.Ed.2d 287 (1978). Again, with all due respect, Justice REHNQUIST is simply wrong. The idea that aliens may be denied political rights is not a recently discovered concept or a newly molded principle that can be said to have eroded the prior understanding. To the contrary, the Court always has recognized that aliens may be denied use of the mechanisms of self-government, and all of the alienage cases have been decided against the backdrop of that principle. Indeed, this aspect of the alienage-equal protection doctrine was explored at length in Sugarman v. Dougall, 413 U.S. 634, 647-649, 93 S.Ct. 2842, 2850-2851, 37 L.Ed.2d 853 (1973), the second of the Court's modern decisions in the area.2 See Cabell v. Chavez-Salido, 454 U.S., at 438-442, 102 S.Ct., at 739-741 (citing Sugarman ); Ambach v. Norwick, 441 U.S., at 74, 99 S.Ct., at 1593 (citing Sugarman ); Foley v. Connelie, 435 U.S., at 294-296, 98 S.Ct., at 1069-1070 (citing Sugarman). Yet in cases contemporary with or postdating Sugarman the Court has experienced no noticeable discomfort in applying strict scrutiny to alienage classifications that did not involve political interests. See In re Griffiths, 413 U.S. 717, 93 S.Ct. 2851, 37 L.Ed.2d 910 (1973); Examining Board v. Flores de Otero, 426 U.S. 572, 96 S.Ct. 2264, 49 L.Ed.2d 65 (1976); Nyquist v. Mauclet, 432 U.S. 1, 97 S.Ct. 2120, 53 L.Ed.2d 63 (1977). 34 It is not surprising, then, that none of the "more recent decisions" relied on by Justice REHNQUIST so much as suggested that the Court's earlier analysis had been undercut. Instead, those cases pointedly have declined to "retrea[t] from the position that restrictions on lawfully resident aliens that primarily affect economic interests are subject to heightened judicial scrutiny." Cabell v. Chavez-Salido, 454 U.S., at 439, 102 S.Ct., at 739. See Ambach v. Norwick, 441 U.S., at 75, 99 S.Ct., at 1593 (that aliens may be denied political rights "is an exception to the general standard applicable to classifications based on alienage"); Foley v. Connelie, 435 U.S., at 296, 98 S.Ct., at 1070. This reflects the Court's proper judgment that the alienage cases are not irreconcilable or inconsistent with one another. For while the Court has recognized, as the Constitution suggests, that alienage may be taken into account when it is relevant—that is, when classifications bearing on political interests are involved—"[t]he distinction between citizens and aliens . . . ordinarily [is] irrelevant to private activity," Ambach v. Norwick, 441 U.S., at 75, 99 S.Ct., at 1593 (emphasis added). And it hardly need be demonstrated that governmental distinctions based on irrelevant characteristics cannot stand. If this dual aspect of alienage doctrine is unique, it is because aliens constitute a unique class.3 35 Finally, even were I to accept Justice REHNQUIST's view that powerlessness is the end-all of alienage-equal protection doctrine, I would find preposterous his further suggestion that, because States do not violate the Constitution when they exclude aliens from participation in the government of the community, the alien's powerlessness therefore is constitutionally irrelevant. From the moment the Court began constructing modern equal protection doctrine in United States v. Carolene Products Co., 304 U.S. 144, 58 S.Ct. 778, 82 L.Ed. 1234 (1938), it never has been suggested that the reason for a discrete class' political powerlessness is significant; instead, the fact of powerlessness is crucial, for in combination with prejudice it is the minority group's inability to assert its political interests that "curtail[s] the operation of those political processes ordinarily to be relied upon to protect minorities." Id., at 152-153, n. 4, 58 S.Ct., at 783-784, n. 4. The very powerlessness of a discrete minority, then, is itself the factor that overcomes the usual presumption that " 'even improvident decisions [affecting minorities] will eventually be rectified by the democratic process.' " Personnel Administrator of Massachusetts v. Feeney, 442 U.S., at 272, 99 S.Ct., at 2292, quoting Vance v. Bradley, 440 U.S. 93, 97, 99 S.Ct. 939, 942, 59 L.Ed.2d 171 (1979). If anything, the fact that aliens constitutionally may be—and generally are—formally and completely barred from participating in the process of self-government makes particularly profound the need for searching judicial review of classifications grounded on alienage. I might add that the Court explicitly has endorsed this seemingly self-evident proposition: in Hampton v. Mow Sun Wong, 426 U.S. 88, 96 S.Ct. 1895, 48 L.Ed.2d 495 (1976), after noting that "[s]ome of [an alien's] disadvantages stem directly from the Constitution itself," the Court declared that "[t]he legitimacy of the delineation of the affected class [of aliens] buttresses the conclusion that it is 'a "discrete and insular" minority' . . . and, of course, is consistent with the premise that the class is one whose members suffer special disabilities." Id., at 102, n. 22, 96 S.Ct., at 1905, n. 22. I find Justice REHNQUIST's attempt to stand this principle on its head perplexing, to say the least. 36 One of the few assertions that can be made with complete confidence about the Court's alienage-equal protection decisions is that no opinion for the Court has ever so much as suggested that Justice REHNQUIST's lone dissent in Sugarman, 413 U.S., at 649, 93 S.Ct., at 2851—which espoused a view similar to the one he hints at today—expressed the proper approach for deciding these cases. Of course, one cannot condemn another for sticking to his guns. Barring a radical change in the Court's reasoning in cases concerning alienage, however, one can expect that today's equal protection writing by Justice REHNQUIST will join his opinion in Sugarman, to use his phrase, as "lifeless words on the pages of these Reports." Post, at 48. 37 Justice O'CONNOR, concurring in part and dissenting in part. 38 I concur in the Court's opinion insofar as it holds that the State may not charge out-of-state tuition to nonimmigrant aliens who, under federal law, are exempt from both state and federal taxes, and who are domiciled in the State. Imposition of out-of-state tuition on such aliens conflicts with federal law exempting them from state taxes, since, after all, the University admits that it seeks to charge the higher tuition in order to recover costs that state income taxes normally would cover. 39 I cannot join the remainder of the Court's opinion, however, for it wholly fails to address the criticisms leveled in Justice REHNQUIST's dissenting opinion. As Justice REHNQUIST makes clear, the class of G-4 aliens is not homogenous: some G-4 aliens are exempt under federal law from state taxes, while other G-4 aliens are not. Moreover, the legislative history of § 4(b) of the International Organizations Immunities Act, later reenacted as § 893 of the Internal Revenue Code of 1954, 26 U.S.C. § 893, from which many G-4 aliens derive their federal tax immunity, demonstrates that Congress did not intend to exempt such aliens from state taxes, choosing instead to leave the matter to the state and local authorities. Thus, I disagree with the Court when it states that the "State may not recoup indirectly from respondents' parents the taxes that the Federal Government has expressly barred the State from collecting," ante, at 16, for in fact Congress has not barred the State from collecting state taxes from many G-4 aliens. Accordingly, I conclude that the Supremacy Clause does not prohibit the University from charging out-of-state tuition to those G-4 aliens who are exempted by federal law from federal taxes only. 40 Justice REHNQUIST, with whom THE CHIEF JUSTICE joins, dissenting. 41 Despite rather broad dicta regarding the conditions under which federal power over immigration will pre-empt state statutes that adversely affect aliens, the Court's holding is narrow. Purporting to rely on a collection of treaties and statutes that concern the tax liability of certain nonimmigrant aliens, it concludes that no room is left for the State of Maryland to charge such aliens nonresident tuition for attending the State's university. The Court's dicta seems to me inconsistent with our prior cases, and its conclusion about the effect of the statutes and treaties is strained at best. In short, the Court reaches a result that I find quite out of step with our normal approach to federal pre-emption of state law. 42 Its holding has the additional vice of foreclosing governmental autonomy in an area plainly within the State's traditional responsibilities—education. And it acts, not on behalf of a disadvantaged minority, but at the behest of a group of individuals who have been accorded a status by the Federal Government superior to that of the average citizen, and in a case where the State has demonstrated, by virtue of its favorable treatment of resident aliens, that its policy is not the result of an invidious or irrational motive. I find the Court's actions unjustified and unnecessary and, accordingly, I dissent. Because I would reverse the judgment of the Court of Appeals, I also address other grounds relied on by the lower courts and argued by respondents in support of their judgments. 43 * Our prior decisions indicate that "when a State's exercise of its police power is challenged under the Supremacy Clause, 'we start with the assumption that the historic police powers of the States were not to be superseded by the Federal Act unless that was the clear and manifest purpose of Congress.' Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230 [67 S.Ct. 1146, 1153, 91 L.Ed. 1447] (1947)." Ray v. Atlantic Richfield Co., 435 U.S. 151, 157, 98 S.Ct. 988, 994, 55 L.Ed.2d 179 (1978). State laws will survive such a challenge unless there is "such actual conflict between the two schemes of regulation that both cannot stand in the same area, [or] evidence of a congressional design to preempt the field." Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 141, 83 S.Ct. 1210, 1217, 10 L.Ed.2d 248 (1963). 44 Unquestionably, federal power over immigration and naturalization is plenary and exclusive. Our decision in De Canas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976), however, unambiguously forecloses any argument that this power, either unexercised or as manifested in the Immigration and Nationality Act, preempts the field of regulations affecting aliens once federal authorities have admitted them into this country. In light of the Court's expansive observations in the instant case, that opinion bears quoting at some length: 45 "[T]he Court has never held that every state enactment which in any way deals with aliens is a regulation of immigration and thus per se pre-empted by [the Federal Government's] constitutional power, whether latent or exercised. For example, Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 415-422 [68 S.Ct. 1138, 1140-1144, 92 L.Ed. 1478] (1948), and Graham v. Richardson, 403 U.S. 365, 372-373 [91 S.Ct. 1848, 1852, 29 L.Ed.2d 534] (1971), cited a line of cases that upheld certain discriminatory state treatment of aliens lawfully within the United States. Although the 'doctrinal foundations' of the cited cases, which generally arose under the Equal Protection Clause 'were undermined in Takahashi,' they remain authority that, standing alone, the fact that aliens are the subject of a state statute does not render it a regulation of immigration, which is essentially a determination of who should or should not be admitted into the country, and the conditions under which a legal entrant may remain." Id., at 355, 96 S.Ct., at 936 (citations omitted; emphasis added). 46 In De Canas the Court also held that Congress' enactment of the Immigration and Nationality Act (INA) was insufficient to oust "harmonious state regulation touching on aliens in general." Id., at 358, 96 S.Ct., at 938. 47 Thus, neither Congress' unexercised constitutional power over immigration and naturalization, nor its exercise of that power in passing the INA, precludes the States from enforcing laws and regulations that prove burdensome to aliens. Under our precedents, therefore, state law is invalid only if there is "such actual conflict between the two schemes of regulation that both cannot stand in the same area," Florida Lime & Avocado Growers, Inc. v. Paul, supra, at 141, 83 S.Ct., at 1217,1 or if Congress has in some other way unambiguously declared its intention to foreclose the state law in question, see Ray v. Atlantic Richfield Co., supra, at 157-158, 98 S.Ct., at 994. In the absence of a conflict, "we are not to conclude that Congress legislated the ouster of [a state law] in the absence of an unambiguous congressional mandate to that effect." Florida Lime & Avocado Growers, supra, 373 U.S., at 146-147, 83 S.Ct., at 1219. 48 Notwithstanding these settled principles, the Court suggests in dicta that any state law which discriminates against lawfully admitted aliens is void, presumably without regard to the strength of the State's justification, if Congress did not contemplate such a law. Ante, at 12-13. This standard seems to me clearly to reverse the presumption that normally prevails when state laws are challenged under the Supremacy Clause. The Court relies on language in three cases to support this proposition. On closer inspection, none of the three offers the precedential support for which the Court obviously grasps. 49 The first case, Takahashi v. Fish & Game Comm'n, 334 U.S. 410, 68 S.Ct. 1138, 92 L.Ed. 1478 (1948), involved a California statute that prohibited the issuance of commercial fishing licenses to aliens who were ineligible for citizenship. The language emphasized by the Court explains that "[s]tate laws which impose discriminatory burdens upon the entrance or residence of aliens lawfully within the United States conflict with this constitutionally derived federal power to regulate immigration, and have accordingly been held invalid." Id., at 419, 68 S.Ct., at 1142 (footnote omitted). In the Takahashi opinion, this statement is immediately followed by three citations, which the Court omits. These citations explain, and qualify, the otherwise broad language quoted by the Court. In the first of these cases, Chy Lung v. Freeman, 92 U.S. 275, 23 L.Ed. 550 (1876), the Court considered a California law that, with certain extremely limited exceptions, prohibited any alien who was, or would likely become, "a public charge," from entering the State through any of its ports. The Court held that the statute was pre-empted by federal law: "The passage of laws which concern the admission of citizens and subjects of foreign nations to our shores belongs to Congress, and not to the States." Id., at 280 (emphasis added). 50 The second case cited in Takahashi, Truax v. Raich, 239 U.S. 33, 36 S.Ct. 7, 60 L.Ed. 131 (1915), concerned an Arizona statute limiting virtually all employment opportunities in the State to citizens. Although Truax involved an asserted repugnancy to the Equal Protection Clause, the Court also suggested that the challenged statute was in conflict with federal law. It is important to note that the Court interpreted the statute as "deny[ing] to lawful inhabitants, because of their race or nationality, the ordinary means of earning a livelihood." Id., at 41, 36 S.Ct., at 10. The Court subsequently stated: "The assertion of an authority to deny to aliens the opportunity of earning a livelihood when lawfully admitted to the State would be tantamount to the assertion of the right to deny them entrance and abode, for in ordinary cases they cannot live where they cannot work." Id., at 42, 36 S.Ct., at 11 (emphasis added). 51 The final case relied on in Takahashi is Hines v. Davidowitz, 312 U.S. 52, 61 S.Ct. 399, 85 L.Ed. 581 (1941). The Pennsylvania statute at issue there required adult aliens to register with the State and to carry an identification card, which they were required to present on demand to state agents. The Court held that the statute was pre-empted by the federal Alien Registration Act of 1940, finding that "[t]he basic subject of the state and federal laws [was] identical," id., at 61, 61 S.Ct., at 401, and that the state law embodied requirements that Congress had studiously avoided in passing the federal Act, id., at 70-74, 61 S.Ct., at 405-407. 52 Thus, in each of these cases, the Court found either a clear encroachment on exclusive federal power to admit aliens into the country or a clear conflict with a specific congressional purpose. It was with these cases in mind that the Court in Takahashi condemned "[s]tate laws which impose discriminatory burdens upon the entrance or residence of aliens lawfully within the United States." 334 U.S., at 419, 68 S.Ct., at 1142. It is most unlikely, therefore, that the Court intended with one stroke of the pen to reverse the normal presumption applicable in cases challenging state enactments under the Supremacy Clause, and to declare such laws invalid without regard to the existence of a conflict with federal statutes or a usurpation of federal power over immigration. 53 The Court also relies on Graham v. Richardson, 403 U.S. 365, 91 S.Ct. 1848, 29 L.Ed.2d 534 (1971), which struck down as a denial of equal protection a California law that withheld welfare benefits from lawfully resident aliens. As an alternative ground, the Court also declared the law invalid as an encroachment on federal power. On the basis of specific federal statutes barring the admission of aliens likely to become public charges, and providing for the deportation of aliens who become public charges because of factors that existed prior to entry, the Court inferred a congressional purpose not "to impose any burden or restriction on aliens who become indigent after their entry into the United States." Id., at 377, 91 S.Ct., at 1855. The Court also concluded, relying on Truax, supra, that the law denied indigent aliens the "necessities of life," and therefore "equate[d] with the assertion of a right, inconsistent with federal policy, to deny entrance and abode." The holding in Graham, therefore, offers no support for a presumption that all state laws burdening aliens conflict with amorphous federal power over immigration. 54 Finally, the Court quotes from dictum appearing in a footnote in De Canas v. Bica, 424 U.S., at 358, n. 6, 96 S.Ct., at 938, n. 6, that " 'state regulation not congressionally sanctioned that discriminates against aliens lawfully admitted to the country is impermissible if it imposes additional burdens not contemplated by Congress.' " Ante, at 12-13. The principal support for this assertion was the passage previously quoted by the Court from Takahashi. As I have already indicated, that passage in context means a good deal less than it does out of context. Most important, however, De Canas itself suggests that the quoted footnote is not a fair description of the law. Although the statute at issue only affected illegal aliens, the principles recognized in the Court's opinion were not so limited. Thus, the Court emphasized that "the fact that aliens are the subject of a state statute does not render it a regulation of immigration," 424 U.S., at 355, 96 S.Ct., at 936, that Takahashi, Graham, and Hines found preemption only after examining specific congressional enactments, 424 U.S., at 355, 96 S.Ct., at 936, that it was necessary to look for some "specific indication . . . that Congress intended to preclude even harmonious state regulation touching on aliens in general," id., at 358, 96 S.Ct., at 937, and that pre-emption should be found only when it is possible to say " 'either that the nature of the regulated subject matter permits no other conclusion or that the Congress has unmistakably so ordained,' " id., at 356, 96 S.Ct., at 937 (quoting Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S., at 142, 83 S.Ct., at 1217). 55 In sum, the fact that a state statute can be said to discriminate against aliens does not, standing alone, demonstrate that the statute is pre-empted, absent some form of congressional sanction. The statute in De Canas discriminated against aliens, yet the Court found no strong evidence that Congress intended to pre-empt it. Obviously, the fact that the aliens were in this country illegally was an important factor in ascertaining Congress' intent. But, just as clearly, the fact that disadvantaged aliens are lawfully in the country does not authorize the Court to dispense with the particularized inquiry into congressional intent that pre-emption analysis traditionally has demanded.2 Discriminatory legislation may well be invalid under the federal civil rights laws as a denial of equal treatment, but under our precedents such a conclusion is possible only after an examination of the classification drawn by the State and its justification for doing so. Under the Court's summary of pre-emption principles applicable to laws discriminating against aliens, these factors would be irrelevant.3 I cannot agree that such a summary accurately reflects the law. 56 The Court concedes that the proper application of its preemption principle "is likely to be a matter of some dispute," ante, at 13, and then proceeds to resolve the case by finding a conflict between Maryland's tuition policy and a collection of treaties and statutes that address the tax liability of certain nonimmigrant aliens. Although I find this conclusion quite unconvincing, it is gratifying to learn that in practice perhaps the Court's new principle still demands proof of a conflict with federal law, just as traditional pre-emption cases instruct. Because the Court's judgment relies on the asserted presence of such a conflict, its statements suggesting that such a particularized inquiry is unnecessary must be regarded as dicta, though unwise dicta at that. With this said, I turn to the Court's discovery of a conflict with federal law. II 57 The Court relies on two features of federal law. First, it notes that Congress has permitted nonimmigrant aliens holding G-4 visas to establish domicile in the United States. Ante, at 14. It then reasons that denying these aliens in-state tuition conflicts with Congress' decision. The Court offers no evidence that Congress' intent in permitting respondents to establish "domicile in the United States" has any bearing at all on the tuition available to them at state universities. Federal law does not require the States to make residence or domicile the determinant of their tuition policies, and as the Court recognizes, Maryland has chosen not to do so in the case of nonimmigrant aliens. Moreover, unlike the state laws scrutinized in Truax and Graham, Maryland's policy does not deprive respondents of a livelihood or the means of subsistence such that it could fairly be characterized as denying respondents "entrance and abode," 239 U.S., at 42, 36 S.Ct., at 11. The Court's reference to "domicile in the United States," therefore, is little more than a restatement of its more general principle that any laws burdensome to aliens who have been lawfully admitted are presumptively pre-empted absent congressional intent to "sanction" them. As I have already suggested, this turns pre-emption analysis on its head. 58 The second feature of federal law on which the Court relies consists of certain statutes and treaties that affect the tax liability of G-4 visaholders. The Court considers these statutes and treaties as an amorphous whole and concludes that the University's policy "frustrates" the policies embodied in them. "The State may not recoup indirectly from respondents' parents the taxes that the Federal Government has expressly barred the State from collecting." Ante, at 16. There are two serious flaws in this argument. First, the Federal Government has not barred the States from collecting taxes from many, if not most, G-4 visaholders. Second, as to those G-4 nonimmigrants who are immune from state income taxes by treaty, Maryland's tuition policy cannot fairly be said to conflict with those treaties in a manner requiring its pre-emption. 59 The individual respondents in this case represent a class of G-4 visaholders or their dependents who are or may become students at the University of Maryland. The Court, contrary to the teaching of our cases,4 reasons as though the class members were a homogenous group. They are not, and the Court's ignorance of relevant differences leads it into error. The named class representatives are dependents of employees of either the Inter-American Development Bank or the International Bank for Reconstruction and Development (the World Bank). As the Court notes, the salaries paid employees of these organizations are exempt by international agreement from taxation by any country other than their own. Ante, at 15. As the Court also notes, the exemption contained in the agreement establishing the World Bank has by statute been given the force of federal law in the United States. 22 U.S.C. § 286h; see ante, at 15-16. 60 Most G-4 visaholders, however, derive whatever tax immunity they enjoy in this country from § 4(b) of the International Organizations Immunities Act (IOIA or Act), later reenacted as § 893 of the Internal Revenue Code of 1954, 26 U.S.C. § 893. That statute exempts the salaries paid to alien employees of international organizations from federal income tax. The principal purpose of the Act as a whole, which is now divided among many Titles of the United States Code, was to extend governmental privileges and immunities to international organizations and their officers and employees located in this country. H.R.Rep. No. 1203, 79th Cong., 1st Sess., 4 (1945), U.S.Code Cong.Serv. 1945, p. 946. As noted, § 4 amended the Internal Revenue Code to exempt the salaries of such officers and employees from federal income tax. As the relevant Committee Reports demonstrate, the exemption was strictly limited to salaries; income derived from commercial activities, investments, and other similar sources was not to enjoy an exemption, and all federal taxes other than those applicable to income remained fully effective. Ibid.; S.Rep. No. 861, 79th Cong., 1st Sess., 4-5 (1945). 61 Section 6 of the bill, as originally introduced in the House, provided an exemption from state and local taxes as well.5 The Senate Committee deleted the exemption, reasoning that "this matter should be properly dealt with by the State and local authorities." S.Rep. No. 861, supra, at 5. The House eventually agreed to the amendment, and the bill as enacted contains no exemption from state or local taxes.6 Floor debates confirm what the Committee amendment implied: although the Act provides an exemption from the federal income tax, it was not intended to foreclose the States from taxing employees of international organizations.7 Accordingly, employees of international organizations whose tax immunity derives solely from the IOIA can claim no federal immunity from state taxes. According to petitioners, approximately three-quarters of the international organizations whose employees hold G-4 visas fall into that category. Brief for Petitioners 29, n. 22. Therefore, even if one were to accept the Court's reasoning that immunity from state taxes implies a right to in-state college tuition, many, if not most of the class members cannot benefit from the argument.8 62 The Court's reasoning is flawed, however, and cannot help even those class members whose parents' tax immunity is based on a treaty or international agreement.9 The State's tuition policy is void under the Supremacy Clause only "to the extent that it actually conflicts with a valid federal statute," Ray v. Atlantic Richfield Co., 435 U.S., at 158, 98 S.Ct., at 994, or, of course, a valid treaty. As the Court stated in Ray, ibid.: 63 "A conflict will be found 'where compliance with both federal and state regulations is a physical impossibility . . .,' Florida Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-143 [83 S.Ct. 1210, 1217, 10 L.Ed.2d 248] (1963), or where the state 'law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.' Hines v. Davidowitz, 312 U.S. 52, 67 [61 S.Ct. 399, 404, 85 L.Ed. 581] (1941); Jones v. Rath Packing Co. [430 U.S.], at 526, 540-541. Accord, De Canas v. Bica, 424 U.S. 351, 363 [96 S.Ct. 933, 940, 47 L.Ed.2d 43] (1976)." 64 There is, of course, no physical impossibility in the coexistence of the two policies. The treaties and agreements insure that signatory nations will not tax the salaries of foreign nationals employed by the designated organizations. The State of Maryland does not tax these salaries. It merely charges tuition for enrollment in its University that is higher than the tuition charged to American citizens and other foreign nationals who have been admitted to this country as immigrants. 65 The remaining question is whether Maryland's tuition policy "stands as an obstacle to the accomplishment and execution of the full purposes and objectives" of the treaties and agreements. Hines v. Davidowitz, 312 U.S., at 67, 61 S.Ct., at 404. In answering this question, it is well to bear in mind certain guideposts that the Court appears to have forgotten: "It is, of course, true that even treaties with foreign nations will be carefully construed so as not to derogate from the authority and jurisdiction of the States of this nation unless clearly necessary to effectuate the national policy." United States v. Pink, 315 U.S. 203, 230, 62 S.Ct. 552, 565, 86 L.Ed. 796 (1942). "Even the language of a treaty wherever reasonably possible will be construed so as not to override state laws or to impair rights arising under them." Guaranty Trust Co. v. United States, 304 U.S. 126, 143, 58 S.Ct. 785, 794, 82 L.Ed. 1224 (1938). In this case, the Court has gone out of its way to raise the banner of federal supremacy over the State's University, without support in the language of the treaties and without examining the intent of the negotiating parties. 66 It is one thing to exempt employees of an international organization from tax liability on their salaries, which otherwise would be incurred by the employees simply by doing what they came to this country to do—working for international organizations such as the World Bank. It is another matter to restrict the State's ability to recover its costs in providing educational services, which respondents were certainly not required to use. Cf. Hamilton v. Regents of the University of California, 293 U.S. 245, 262, 55 S.Ct. 197, 204, 79 L.Ed. 343 (1934). Although a college education over the years has become accessible to increasing numbers of Americans, it can hardly be characterized as an unavoidable feature of life in this country. Thus, although the negotiating parties undoubtedly intended to lower the costs of international organizations by exempting employees from income taxes, it does not at all follow that they further intended to require the States to subsidize the cost of services which those employees or their families might choose to use.10 67 Indeed, the United States, which unlike the State of Maryland negotiated the agreements in question, clearly does not understand them to require that education for G-4 visaholders be subsidized to the same extent as education for citizens or resident aliens. For example, the Federal Guaranteed Student Loan Program, which provides significant aid to students attending qualifying colleges and graduate schools, is available to American citizens and permanent resident aliens, but not to nonimmigrant aliens such as respondents. See 34 CFR § 682.201(a)(2) (1981). If this reflects the federal policy embodied in the treaties on which the Court relies, I fail to see how Maryland's tuition policy "frustrates" it. III 68 The lower courts' principal basis for invalidating Maryland's tuition policy was not the Supremacy Clause, but the Equal Protection Clause. Those courts interpreted the State's policy as a classification based on alienage, and therefore subjected it to "strict scrutiny" on the authority of Graham v. Richardson, 403 U.S. 365, 91 S.Ct. 1848, 29 L.Ed.2d 534 (1971), and later cases. In light of several recent decisions, however, it is clear that not every alienage classification is subject to strict scrutiny. In my view, the classification relied upon by the State in this case cannot fairly be called "suspect," and therefore I would ask only whether it rests upon a rational basis. Because I believe it does, I cannot agree with the lower courts that it denies the equal protection of the laws. 69 The Equal Protection Clause of the Fourteenth Amendment has been interpreted by this Court as embodying the principle that "all persons similarly circumstanced shall be treated alike." F. S. Royster Guano Co. v. Virginia, 253 U.S. 412, 415, 40 S.Ct. 560, 562, 64 L.Ed. 989 (1920). By the same token, however, "[t]he Constitution does not require things which are different in fact or opinion to be treated in law as though they were the same." Tigner v. Texas, 310 U.S. 141, 147, 60 S.Ct. 879, 882, 84 L.Ed. 1124 (1940). 70 All laws classify, and, unremarkably, the characteristics that distinguish the classes so created have been judged relevant by the legislators responsible for the enactment. The Equal Protection Clause, however, reflects the judgment of its Framers that some distinguishing characteristics may seldom, if ever, be the basis for difference in treatment by the legislature. The key question in all equal protection cases, of course, is whether the distinguishing characteristics on which the State relies are constitutional. 71 In the vast majority of cases our judicial function permits us to ask only whether the judgment of relevance made by the State is rational. See McGowan v. Maryland, 366 U.S. 420, 425-426, 81 S.Ct. 1101, 1104-1105, 6 L.Ed.2d 393 (1961).11 In a very few other cases, we have required that the State pass a more demanding test because of the judgment that the classification drawn by the State is virtually never permissible from a constitutional perspective. Such classifications are deemed "suspect" and strictly scrutinized. Until 1971, only race and national origin had been so classified by the Court. See Brown v. Board of Education, 347 U.S. 483, 494, 74 S.Ct. 686, 691, 98 L.Ed. 873 (1954); Strauder v. West Virginia, 100 U.S. 303, 25 L.Ed. 664 (1880); Oyama v. California, 332 U.S. 633, 68 S.Ct. 269, 92 L.Ed. 249 (1948). 72 In Graham v. Richardson, supra, the Court added alienage to this select list. Apart from the abbreviated conclusion that "[a]liens as a class are a prime example of a 'discrete and insular' minority," id., at 372, 91 S.Ct., at 1852, the Court did not elaborate on the justification for "heightened judicial solicitude," ibid. Subsequently, the Court observed that aliens, unlike other members of the community, were subject to the particular disadvantage of being unable to vote, and thus were barred from participating formally in the process of self-government. Hampton v. Mow Sun Wong, 426 U.S. 88, 102, 96 S.Ct. 1895, 1905, 48 L.Ed.2d 495 (1976). One could infer that rigorous judicial scrutiny normally was necessary because aliens were barred from asserting their interests in the governmental body responsible for imposing burdens upon them. 73 More recent decisions have established, however, that the political powerlessness of aliens is itself the consequence of distinctions on the basis of alienage that are constitutionally permissible. 74 "[I]t is clear that a State may deny aliens the right to vote, or to run for elective office, for these lie at the heart of our political institutions. See [Sugarman v. Dougall, 413 U.S. 634, 647-649, 93 S.Ct. 2842, 2850-2851, 37 L.Ed.2d 853 (1973) ]. Similar considerations support a legislative determination to exclude aliens from jury service. See Perkins v. Smith, 370 F.Supp. 134 (Md.1974), aff'd, 426 U.S. 913 [96 S.Ct. 2616, 49 L.Ed.2d 368] (1976). Likewise, we have recognized that citizenship may be a relevant qualification for fulfilling those 'important nonelective executive, legislative, and judicial positions,' held by 'officers who participate directly in the formulation, execution, or review of broad public policy.' Dougall, supra, at 647 [93 S.Ct., at 2850]." Foley v. Connelie, 435 U.S. 291, 296, 98 S.Ct. 1067, 1070, 55 L.Ed.2d 287 (1978). As the Court explained earlier this Term: 75 "The exclusion of aliens from basic governmental processes is not a deficiency in the democratic system but a necessary consequence of the community's process of political self-definition. Self-government, whether direct or through representatives, begins by defining the scope of the community of the governed and thus of the governors as well: Aliens are by definition those outside of this community. Judicial incursions into this area may interfere with those aspects of democratic self-government that are most essential to it." Cabell v. Chavez-Salido, 454 U.S. 432, 439-440, 102 S.Ct. 735, 740, 70 L.Ed.2d 677 (1982). 76 If the exclusion of aliens from the political processes is legitimate, as it clearly is, there is reason to doubt whether political powerlessness is any longer a legitimate reason for treating aliens as a "suspect class" deserving of "heightened judicial solicitude." Indeed, in Foley v. Connelie, supra, Ambach v. Norwick, 441 U.S. 68, 99 S.Ct. 1589, 60 L.Ed.2d 49 (1979), and Cabell v. Chavez-Salido, supra, the Court plainly eschewed the application of strict scrutiny to the States' exclusion of aliens from particular public offices.12 In my view, these decisions merely reflect the judgment that alienage, or the other side of the coin, citizenship, is for certain important state purposes a constitutionally relevant characteristic and therefore cannot always be considered invidious in the same manner as race or national origin.13 IV 77 The State's policy in this case is to provide in-state tuition to residents of the State who are citizens and immigrant aliens lawfully admitted for permanent residence. In-state tuition is not available to certain students, however, regardless of whether they have established residence within the State. Within this class are citizens who are financially dependent either on parents or on a spouse who is not domiciled in the State, as well as citizens who are members of the Armed Forces and have been assigned by the military to attend the University.14 Also within the class are nonimmigrant aliens, who have not been admitted to this country for permanent residence. 78 In each case in which the Court has tested state alienage classifications under the Equal Protection Clause, the question has been the extent to which the States could permissibly distinguish between citizens and permanent resident aliens. See Graham v. Richardson, 403 U.S. 365, 91 S.Ct. 1848, 29 L.Ed.2d 534 (1971); Sugarman v. Dougall, 413 U.S. 634, 93 S.Ct. 2842, 37 L.Ed.2d 853 (1973); In re Griffiths, 413 U.S. 717, 93 S.Ct. 2851, 37 L.Ed.2d 910 (1973); Examining Board v. Flores de Otero, 426 U.S. 572, 96 S.Ct. 2264, 49 L.Ed.2d 65 (1976); Nyquist v. Mauclet, 432 U.S. 1, 97 S.Ct. 2120, 53 L.Ed.2d 63 (1977). We recently summarized these decisions as implying that "there would be few—if any—areas in which a State could legitimately distinguish between its citizens and lawfully resident aliens." Cabell v. Chavez-Salido, 454 U.S., at 438, 102 S.Ct., at 739 (emphasis added). In this case, however, the question is whether the State can distinguish between two groups, each of which consists of citizens and aliens. For two reasons, the State's classification should not be deemed "suspect" and subjected to strict scrutiny. 79 First, unlike immigrant aliens, nonimmigrants such as G-4 visaholders are significantly different from citizens in certain important respects. Our previous decisions have emphasized that immigrant aliens have been lawfully admitted to this country for permanent residence and share many of the normal burdens of citizenship, such as the duty to pay taxes and to serve in the Armed Forces. Nyquist v. Mauclet, supra, 432 U.S., at 12, 97 S.Ct., at 2126; Hampton v. Mow Sun Wong, 426 U.S., at 107, n. 30, 96 S.Ct., at 1907 n. 30; Sugarman v. Dougall, supra, 413 U.S., at 645, 93 S.Ct., at 2849, Graham v. Richardson, supra, 403 U.S., at 376, 91 S.Ct., at 1854. Implicit in these cases is the judgment that because permanent resident aliens are in so many respects situated similarly to citizens, distinctions between them are to be carefully scrutinized.15 Although there is legitimate doubt whether these decisions have survived Foley, Ambach, and Cabell intact, their judgment about the need for strict scrutiny simply does not apply to state policies that distinguish between permanent resident aliens and nonimmigrants. 80 As noted earlier, nonimmigrant aliens holding G-4 visas, unlike resident aliens, are exempt from Maryland's income tax, by operation of either international agreement or a combination of federal and state law.16 The University is substantially supported by general state revenues appropriated by the legislature, and of this sum nearly half is generated by the state income tax. See Brief for Petitioners 29, n. 23. Consequently, for the purpose of assessing tuition to the State's University, G-4 nonimmigrant aliens are not situated similarly either to most citizens or to permanent resident aliens. They are distinguished by a trait that is obviously quite relevant from the State's perspective, and legitimately so. Other nonimmigrant aliens are subject to state income taxes, but, as respondents concede, Brief for Respondents 12, 14, 23, they are admitted to this country only temporarily and for limited purposes. These aliens are also not situated similarly to resident citizens or to permanent resident aliens because most are admitted on the condition that they cannot establish domicile in the United States. See Elkins v. Moreno, 435 U.S. 647, 665, 98 S.Ct. 1338, 1349, 55 L.Ed.2d 614 (1978). As a group, then, nonimmigrant aliens are sufficiently different from citizens in relevant respects that distinctions between them and citizens or immigrant aliens should not call for heightened scrutiny. 81 Second, the State's tuition policy, as it applies to G-4 visaholders, simply cannot be broadly characterized as a classification that discriminates on the basis of alienage. It is more accurately described as a policy that classifies on the basis of financial contribution toward the costs of operating the University. In one class are citizens and permanent resident aliens, all of whom have lived in the State and have contributed to state revenues through the payment of income taxes. To these students the State offers its in-state tuition, which covers only a portion of the cost of educating each student. The remainder is subsidized through state revenues, to which the students themselves have contributed by paying the full spectrum of state taxes. 82 In the other class is an equally mixed group of citizens and aliens. Some of these citizens do not reside in the State and therefore do not pay state taxes. Others do reside in the State, but are financially dependent on parents or a spouse who is domiciled elsewhere and therefore do not help finance the operation of the University through income taxes. Nonimmigrant aliens holding G-4 visas also reside in the State but, like citizens in this class, do not pay state income taxes.17 To all members of this class the State charges a higher, so-called "out-of-state" tuition, although one that still does not fully cover the cost of education. Just as it may seem unfair for a State to deny to a resident alien the right to participate in public benefits to which he has contributed through taxes, it might seem equally unfair to allow G-4 visaholders to participate, on a par with taxpaying resident citizens and permanent resident aliens, in public benefits to which they have not contributed. Whether or not such a judgment is correct, a policy justified in such terms cannot fairly be called the product of xenophobic prejudice. Given the State's decision to treat immigrant aliens on a par with citizens, its decision to require a higher tuition of G-4 nonimmigrant aliens cannot be characterized as a classification on the basis of alienage.18 83 Consequently, for either of these reasons, the "strict scrutiny" authorized by Graham v. Richardson, 403 U.S. 365, 91 S.Ct. 1848, 29 L.Ed.2d 534 (1971), even if it is still applicable to discrimination against permanent resident aliens, has no proper application to the State's policy in this case. The only question, therefore, is whether "the State's classification rationally furthers the purpose identified by the State." Massachusetts Board of Retirement v. Murgia, 427 U.S. 307, 314, 96 S.Ct. 2562, 2567, 49 L.Ed.2d 520 (1976). The State has articulated several purposes for its policy of denying in-state tuition to nonimmigrant aliens. One purpose is roughly to equalize the cost of higher education borne by those students who do and those who do not financially contribute to the University through income tax payments. The purpose surely is a legitimate one, and I should think it evident that the State's classification rationally furthers that purpose.19 V 84 On June 23, 1978, approximately two months after our decision in Elkins v. Moreno, 435 U.S. 647, 98 S.Ct. 1338, 55 L.Ed.2d 614 (1978), the University's Board of Regents adopted a "clarifying" resolution establishing beyond doubt that the State's policy excluding G-4 visaholders from eligibility for in-state tuition was not based on their lack of domicile. For this reason, we remanded the case to the District Court for further proceedings, having concluded that this case was no longer controlled by Vlandis v. Kline, 412 U.S. 441, 93 S.Ct. 2230, 37 L.Ed.2d 63 (1973), as limited by Weinberger v. Salfi, 422 U.S. 749, 771, 95 S.Ct. 2457, 2469, 45 L.Ed.2d 522 (1975). Toll v. Moreno, 441 U.S. 458, 461-462, 99 S.Ct. 2044, 2045-2046, 60 L.Ed.2d 354 (1979). On remand, the District Court concluded that although the clarifying resolution adopted on June 23, 1978, eliminated the "conclusive presumption" that respondents could not establish domicile, the existence of such a presumption before that date denied respondents due process under the teaching of Vlandis v. Kline, supra. 85 There is legitimate doubt whether at this late date anything remains of Vlandis v. Kline but its lifeless words on the pages of these Reports. Such doubts, however, need not be resolved in this case. The University has made clear that domicile is not the principal consideration underlying its tuition policy as applied to respondents, and in my view that policy is rationally related to other legitimate purposes proffered by the State. The classification challenged by respondents did not change on June 23, 1978. If the classification is valid today, as I believe it is, then it was valid before the State issued its "clarifying" resolution. A statute's consistency with the Due Process Clause or the Equal Protection Clause should not depend on which purpose state officials choose to emphasize at a particular time, as long as one of the State's purposes is rationally served by the statute. See McGowan v. Maryland, 366 U.S., at 426, 81 S.Ct., at 1105 ("A statutory discrimination will not be set aside if any state of facts reasonably may be conceived to justify it"). 86 For the foregoing reasons, I would reverse the judgment of the Court of Appeals. 1 The international organizations covered by the provision are those that are entitled to the privileges, exemptions, and immunities conferred under the International Organizations Immunities Act, 59 Stat. 669, 22 U.S.C. § 288 et seq. At the time suit was brought, the named plaintiffs in this case were dependents of employees of either the Inter-American Development Bank or the International Bank for Reconstruction and Development (World Bank). 2 A fourth individual, Rene Otero, Jr., a respondent in this Court, was made a named plaintiff in 1980 when a supplemental complaint was filed. 3 The court certified a class of G-4 visaholders or their dependents who, "residing in Maryland, . . . are current students at the University of Maryland, or . . . chose not to apply to the University of Maryland because of the challenged policies but would now be interested in attending if given an opportunity to establish 'in-state' status, or . . . are currently students in senior high schools in Maryland." Moreno v. University of Maryland, 420 F.Supp. 541, 563 (Md.1976). 4 Citing Monroe v. Pape, 365 U.S. 167, 81 S.Ct. 473, 5 L.Ed.2d 492 (1961), the District Court dismissed the claim against the University itself. 420 F.Supp., at 548-550. The plaintiffs did not appeal that dismissal. 5 The District Court did not order the University to grant the named plaintiffs in-state status. Rather, it merely barred the University from denying them and the members of the class "the opportunity to demonstrate that they or any of them are entitled to 'in-state' status for purposes of tuition and charge differential determinations." Id., at 565. 6 The Court of Appeals stayed its mandate "on the same terms as the district court originally granted its stay." App. to Pet. for Cert. 103a-104a. 7 Salfi limited Vlandis "to those situations in which a State 'purport[s] to be concerned with [domicile, but] at the same time den[ies] to one seeking to meet its test of [domicile] the opportunity to show factors clearly bearing on that issue.' " Elkins v. Moreno, 435 U.S., at 660, 98 S.Ct., at 1346, quoting Weinberger v. Salfi, 422 U.S. 749, 771, 95 S.Ct. 2457, 2469, 45 L.Ed.2d 522 (1975). 8 We noted that as to some categories of nonimmigrant aliens, Congress had "expressly conditioned admission . . . on an intent not to abandon a foreign residence or, by implication, on an intent not to seek domicile in the United States." 435 U.S., at 665, 98 S.Ct., at 1349. See, e.g., 8 U.S.C. §§ 1101(a)(15)(B), (F), (H). With respect to G-4 nonimmigrant aliens, however, we concluded that Congress had deliberately declined to "impose restrictions on intent," thereby permitting them to "adopt the United States as their domicile." 435 U.S., at 666, 98 S.Ct., at 1349. 9 The certified question was phrased as follows: "Are persons residing in Maryland who hold or are named in a visa under 8 U.S.C. § 1101(a)(15)(G)(iv) (1976 ed.), or who are financially dependent upon a person holding or named in such a visa, incapable as a matter of state law of becoming domiciliaries of Maryland?" Id., at 668-669, 98 S.Ct., at 1351 (footnote omitted). 10 It was entitled "A Resolution Clarifying the Purposes, Meaning, and Application of the Policy of the University of Maryland for Determination of In-State Status for Admission, Tuition, and Charge-Differential Purposes, Insofar as It Denies In-State Status to Nonimmigrant Aliens." App. to Pet. for Cert. 172a. 11 "Reaffirmation of In-State Policy. Regardless of whether or not the policy approved by the Board of Regents on September 21, 1973, conforms with the generally or otherwise applicable definition of domicile under the Maryland common law, the Board of Regents reaffirms that policy . . . ." Id., at 174a. 12 We further noted: "Our decision in Elkins rests on the premise that 'the University apparently has no interest in continuing to deny in-state status to G-4 aliens as a class if they can become Maryland domiciliaries since it has indicated both here and in the District Court that it would redraft its policy "to accommodate" G-4 aliens were the Maryland courts to hold that G-4 aliens can' acquire such domicile. 435 U.S., at 661, 98 S.Ct., at 1347. After the clarifying resolution, this premise no longer appears to be true." 441 U.S., at 461, 99 S.Ct., at 2046. 13 The District Court's pre-emption holding rested in part on its equal protection analysis; according to the court, "the standard utilized to uphold a state regulation dealing with benefits to be accorded to aliens is essentially the strict scrutiny analysis" of equal protection. 489 F.Supp., at 668. 14 "This Constitution, and the Laws of the United States which shall be made in Pursuance thereof; and all Treaties made, or which shall be made, under the Authority of the United States, shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby, any Thing in the Constitution or Laws of any State to the Contrary notwithstanding." Art. VI, cl. 2. 15 At the time Takahashi was decided, federal law "permitted Japanese and certain other non-white racial groups to enter and reside in the country, but . . . made them ineligible for United States citizenship." 334 U.S., at 412, 68 S.Ct., at 1139 (footnote omitted). 16 Justice REHNQUIST, in dissent, suggests that the italicized language should not be interpreted literally. Post, at 28-29. Rather, he suggests, the language can only be understood as explaining three prior Court cases that Takahashi cited in a footnote immediately after the italicized language. 334 U.S., at 419, n. 6, 68 S.Ct., at 1142, n. 6, citing Truax v. Raich, 239 U.S. 33, 36 S.Ct. 7, 60 L.Ed. 131 (1915), Chy Lung v. Freeman, 92 U.S. 275, 280, 23 L.Ed. 550 (1876), and Hines v. Davidowitz, 312 U.S. 52, 65-68, 61 S.Ct. 399, 403-404, 85 L.Ed. 581 (1941). According to Justice REHNQUIST, "in each of these cases, the Court found either a clear encroachment on exclusive federal power to admit aliens into the country or a clear conflict with a specific congressional purpose." Post, at 29. Justice REHNQUIST thus concludes that the language in Takahashi does not mean what it says; instead it means that absent a clear encroachment on exclusive federal power or clear conflict with a federal statute, the States are free to treat aliens as they will. Justice REHNQUIST is wrong. If the language were read in the manner suggested by the dissent, it would fail to explain Takahashi itself: The California statute at issue in Takahashi, denying certain lawful aliens the right to obtain commercial fishing licenses from the State, presented neither "a clear encroachment on exclusive federal power to admit aliens" nor "a clear conflict with a specific congressional purpose." Justice REHNQUIST's non literal interpretation of the Takahashi holding is simply wishful thinking on his part. While pre-emption played a significant role in the Court's analysis in Takahashi, the actual basis for invalidation of the California statute was apparently the Equal Protection Clause of the Constitution. Commentators have noted, however, that many of the Court's decisions concerning alienage classifications, such as Takahashi, are better explained in pre-emption than in equal protection terms. See, e.g., Perry, Modern Equal Protection: A Conceptualization and Appraisal, 79 Colum.L.Rev. 1023, 1060-1065 (1979); Note, The Equal Treatment of Aliens: Preemption or Equal Protection?, 31 Stan.L.Rev. 1069 (1979). 17 Our cases do recognize, however, that a State, in the course of defining its political community, may, in appropriate circumstances, limit the participation of noncitizens in the States' political and governmental functions. See, e.g., Cabell v. Chavez-Salido, 454 U.S. 432, 102 S.Ct. 735, 70 L.Ed.2d 677 (1982); Ambach v. Norwick, 441 U.S. 68, 72-75, 99 S.Ct. 1589, 1592-1593, 60 L.Ed.2d 49 (1979); Foley v. Connelie, 435 U.S. 291, 295-296, 98 S.Ct. 1067, 1070, 55 L.Ed.2d 287 (1978); Sugarman v. Dougall, 413 U.S. 634, 646-649, 93 S.Ct. 2842, 2849-2851, 37 L.Ed.2d 853 (1973). 18 In De Canas, we considered whether a California statute making it unlawful in some circumstances to employ illegal aliens was invalid under the Supremacy Clause. We upheld the statute. Justice REHNQUIST's dissent in the present case suggests that the pre-emption claim was rejected in De Canas because "the Court found no strong evidence that Congress intended to pre-empt" the State's action. Post, at 31. Justice REHNQUIST has misread De Canas. We rejected the pre-emption claim not because of an absence of congressional intent to pre-empt, but because Congress intended that the States be allowed, "to the extent consistent with federal law, [to] regulate the employment of illegal aliens." 424 U.S., at 361, 96 S.Ct., at 939. 19 Immigrant aliens are subject to stricter qualitative tests than nonimmigrant aliens. See E. Harper, Immigration Laws of the United States 228 (3d ed. 1975). And whereas there are no quantitative restrictions on the admission of nonimmigrant aliens, there are, with a few exceptions, quota limitations for immigrant aliens. See 8 U.S.C. § 1151(a) (1976 ed., Supp.IV); Harper, supra, at 228. As we noted in Elkins v. Moreno : "Congress defined nonimmigrant classes to provide for the needs of international diplomacy, tourism, and commerce, each of which requires that aliens be admitted to the United States from time to time and all of which would be hampered if every alien entering the United States were subject to a quota and to the more strict entry conditions placed on immigrant aliens." 435 U.S., at 665, 98 S.Ct., at 1349 (footnote omitted). 20 See, e.g., 8 U.S.C. § 1101(a)(15)(B) (temporary visitors for pleasure or business); § 1101(a)(15)(C) (aliens in transit); § 1101(a)(15)(F) (foreign students); § 1101(a)(15)(H) (temporary workers). 21 Among the similar agreements pertaining to other international organizations are the following: Articles of Agreement of the International Finance Corporation, Art. VI, § 9(b), [1956] 7 U.S.T. 2197, 2216, T.I.A.S. No. 3620 (1955) ("No tax shall be levied on or in respect of salaries and emoluments paid by the Corporation to . . . employees of the Corporation who are not local citizens, local subjects, or other local nationals"); Articles of Agreement of the International Development Association, Art. VIII, § 9(b), [1960] 11 U.S.T. 2284, 2306, T.I.A.S. No. 4607 (1960) ("No tax shall be levied on or in respect of salaries and emoluments paid by the Association to . . . employees of the Association who are not local citizens, local subjects, or other local nationals"); Convention on the Settlement of Investment Disputes Between States and Nationals of Other States, Art. 24, § 2, [1966] 17 U.S.T. 1270, 1279, T.I.A.S. No. 6090 (1965) ("Except in the case of local nationals, no tax shall be levied . . . on or in respect of salaries, expense allowances or other emoluments paid by the Centre to officials or employees of the Secretariat"); Articles of Agreement of the International Monetary Fund, Art. IX, § 9(b), 60 Stat. 1414, T.I.A.S. No. 1501 (1945) ("No tax shall be levied on or in respect of salaries and emoluments paid by the Fund to . . . employees of the Fund who are not local citizens, local subjects, or other local nationals"). 22 And by virtue of Md.Ann.Code, Art. 81, § 280(a) (1980), this group of G-4 visaholders is able to shield organizational income from Maryland income tax. 23 Petitioners point out that the international banks for which the named respondents' parents work provide reimbursement for the difference between in-state and out-of-state tuition. Certainly, this fact does not assist—but undermines—petitioners' argument. Such reimbursements only add to the employment costs of the international organizations, thereby frustrating the federal intention of benefiting the international organizations. 24 Some members of the class represented by the respondents derive their state tax exemption not from a treaty or international agreement, but from the combination of federal and state statutes. See supra, at 15-16, and n. 22. As to these G-4 aliens, it is true, as the dissent notes, post, at 34-35, that the Federal Government has not precluded the collection of a state income tax that is imposed on domiciliaries of the State. But even with respect to this group of G-4 aliens, the Federal Government has taken the affirmative steps of permitting the establishment of domicile and of providing federal income tax exemption on organizational salaries. This special status afforded by the Federal Government is, in our view, inconsistent with the University of Maryland's discriminatory denial of in-state status to G-4 aliens who are domiciled in the State. 25 It is important to note that this case does not involve, and we express no views regarding, a State's imposition of a burden on all individuals sharing a common relevant characteristic, of whom only some are aliens. 26 Petitioners note, however, that whereas the District Court's 1976 order was based solely on due process grounds, the District Court, on remand, held the in-state policy as it operated during the period following the clarifying resolution invalid on two different grounds—equal protection and pre-emption. In our view, this fact is of little moment. Just as a respondent is entitled to defend in this Court a judgment on grounds different from those relied on by the court below, e.g., Colautti v. Franklin, 439 U.S. 379, 397, n. 16, 99 S.Ct. 675, 686, n. 16, 58 L.Ed.2d 596 (1979), respondents in this case were entitled, following our remand, to support a reaffirmance of the earlier order on grounds previously urged but not relied on. 27 Even if we were to assume that the judgment of the District Court was indeed vacated, we could not say that the terms of the University's waiver of sovereign immunity—that the District Court's order be "finally affirmed on appeal"—would not be satisfied. Petitioners have not prevailed on the merits in a single court, despite the numerous decisions that this litigation has prompted. By its original order, the District Court held that the University's in-state policy was invalid insofar as it discriminated against G-4 aliens. Today, we reaffirm that conclusion. 1 The Justice opines that "[i]f the Court has eschewed strict scrutiny in the 'political process' [alienage-equal protection] cases, it may be because the Court is becoming uncomfortable with the categorization of aliens as a suspect class." Post, at 42, n. 12. 2 Among other things, the Court noted in Sugarman that the State may exclude aliens from governmental positions "that go to the heart of representative government," in an attempt " 'to preserve the basic conception of a political community.' " 413 U.S., at 647, 93 S.Ct., at 2850, quoting Dunn v. Blumstein, 405 U.S. 330, 344, 92 S.Ct. 995, 1004, 31 L.Ed.2d 274 (1972). The Sugarman Court thus recognized the "State's historical power to exclude aliens from participation in its democratic political institutions." 413 U.S., at 648, 93 S.Ct., at 2850. This makes Justice REHNQUIST's analysis particularly perplexing; his discussion appears to suggest that Sugarman —decided in 1973 somehow undercut the analysis of Hampton v. Mow Sun Wong, 426 U.S. 88, 96 S.Ct. 1895, 48 L.Ed.2d 495 (1976). See post, at 40. 3 Justice REHNQUIST suggests that alienage classifications involving political interests are subjected to a lesser standard of review because "the strength of the State's interest is great when it seeks to exclude aliens from its political processes." Post, at 41, n. 12. This suggestion is inaccurate. Such classifications are permissible because the Court has recognized that they are likely to be based on meaningful distinctions: alienage "is a relevant ground for determining membership in the political community." Cabell v. Chavez-Salido, 454 U.S. 432, 438, 102 S.Ct. 735, 739, 70 L.Ed.2d 677 (1982). 1 The state courts in De Canas v. Bica, 424 U.S. 351, 96 S.Ct. 933, 47 L.Ed.2d 43 (1976), had not addressed the question in light of their determination that Congress had completely barred state action in the field of employment of illegal aliens. Consequently, this Court also deferred consideration of the issue. Id., at 363, 96 S.Ct., at 940. 2 As the Court obligatorily notes, ante, at 12, n. 17, but promptly ignores, our decisions in Foley v. Connelie, 435 U.S. 291, 98 S.Ct. 1067, 55 L.Ed.2d 287 (1978); Ambach v. Norwick, 441 U.S. 68, 99 S.Ct. 1589, 60 L.Ed.2d 28 (1979); and Cabell v. Chavez-Salido, 454 U.S. 432, 102 S.Ct. 735, 70 L.Ed.2d 677 (1982), all upheld state laws that expressly discriminated against lawfully admitted resident aliens. Such decisions would not have been possible if the mere fact that a law discriminated against aliens placed it in irreconcilable conflict with federal power over immigration. 3 As I have always understood the Supremacy Clause, if a state law is inconsistent with federal law, the state law is unenforceable. The inconsistency is made no less fatal because the State has a rational basis for, or a compelling interest in, its actions. Under the majority's formulation, a state law that arguably discriminates against aliens conflicts with federal law, and unless further modifications of the pre-emption doctrine are in the offing, that will be the end of the matter. 4 "[A] host of constitutional and statutory provisions rest on the premise that a legitimate distinction between citizens and aliens may justify attributes and benefits for one class not accorded to the other; and the class of aliens is itself a heterogenous multitude of persons with a wide-ranging variety of ties to this country." Mathews v. Diaz, 426 U.S. 67, 78-79, 96 S.Ct. 1883, 1890-1891, 48 L.Ed.2d 478 (1976) (footnote omitted). 5 Section 6 of the original bill, H.R. 4489, read as follows: "International organizations shall be exempt from all property taxes imposed by, or under the authority of, any act of Congress, including such acts as are applicable solely to the District of Columbia or the Territories; and shall be entitled to the same exemptions and immunities from State or local taxes as is the United States Government." 91 Cong.Rec. 10867 (1945). 6 Section 6 is codified at 22 U.S.C. § 288c and now reads: "International organizations shall be exempt from all property taxes imposed by, or under the authority of, any Act of Congress, including such Acts as are applicable solely to the District of Columbia or the Territories." 7 Thus, sponsors of the legislation in the House assured their colleagues that the bill would not admit such employees as immigrants. In addition, the following exchange occurred: "MR. RANKIN. This bill does not interfere with State laws in any way? "MR. ROBERTSON of Virginia. None whatever." 91 Cong.Rec. 10866 (1945). In the Senate, Senator Taft explained that his Committee had deleted the proposed exemption contained in § 6 because it "felt that that was wholly beyond the power of Congress." 91 Cong.Rec. 12432 (1945). 8 G-4 visaholders residing in Maryland who are relieved of federal taxes under the Internal Revenue Code have also been exempted from Maryland taxes by operation of state law. Maryland's tax code provides that, with certain exceptions not relevant here, the net income taxable under state law is the taxpayer's federal adjusted gross income. Md.Ann.Code, Art. 81, § 280(a) (1980). By operation of 26 U.S.C. § 893, that amount will not include wages paid by an international organization. The State's decision indirectly to relieve class members of state taxes on their salaries of course provides no basis for pre-emption of the State's tuition policy under the Supremacy Clause. 9 The District Court, which concluded that the State's tuition policy interfered with Congress' exclusive control over immigration, nevertheless rejected the argument that the policy conflicted with the treaties and agreements relieving respondents of liability for income taxes. "In this case it is apparent that there is no 'clear conflict' between the policies in question. The University's Policy seeks to confer certain economic benefits on individuals closely affiliated with the State of Maryland. The mere fact that one of the factors which is considered in determining eligibility for this benefit is whether or not the applicant's income is taxed by Maryland does not necessarily imply that the policy conflicts with the tax policies contained in the relevant international agreements. The 'conflict' between these policies, in and of itself, is too attenuated to warrant invalidating the University's Policy." 489 F.Supp. 658, 667 (Md.1980). 10 As petitioners explain, tuition and fee charges do not pay the full cost of a university education at the University of Maryland. In fiscal year 1981, for example, the University received appropriations from general fund revenues in the amount of $164 million. Brief for Petitioners 29, n. 23. Nearly half of general fund revenues are provided by the State's income tax. Ibid. The State, therefore, subsidizes the cost of education at the University. The amount of the subsidy, of course, is considerably greater for students who are eligible for in-state tuition. Since residents of the State normally pay income tax, and thereby indirectly contribute to the subsidy, it is not unreasonable for the State to accord such persons a reduced tuition. By charging respondents out-of-state tuition, the University is merely asking them to pay their fair share of the cost of state-supported education. 11 "This standard reduces to a minimum the likelihood that the federal judiciary will judge state policies in terms of the individual notions and predilections of its own members, and until recently it has been followed in all kinds of 'equal protection' cases." Harper v. Virginia Board of Elections, 383 U.S. 663, 681-682, 86 S.Ct. 1079, 1089, 16 L.Ed.2d 169 (1966) (Harlan, J., dissenting). 12 As suggested earlier, we have affirmed "the general principle that some state functions are so bound up with the operation of the State as a governmental entity as to permit the exclusion from those functions of all persons who have not become part of the process of self-government." Ambach v. Norwick, 441 U.S., at 73-74, 99 S.Ct., at 1592-1594. "[I]n those areas the State's exclusion of aliens need not 'clear the high hurdle of "strict scrutiny," because [that] would "obliterate all the distinctions between citizens and aliens, and thus depreciate the historic value of citizenship." ' Foley v. Connelie, 435 U.S., at 295 [98 S.Ct., at 1070 (citation omitted)." Cabell v. Chavez-Salido, 454 U.S., at 439, 102 S.Ct., at 739 (footnote omitted). The Court has recognized that the strength of the State's interest is great when it seeks to exclude aliens from its political processes, but selection of the appropriate level of "scrutiny" traditionally has depended, not on the nature of the State's interest, but on the nature of the burdened class. If the Court has eschewed strict scrutiny in the "political process" cases, it may be because the Court is becoming uncomfortable with the categorization of aliens as a suspect class. 13 That judgment was shared by the Framers of the Fourteenth Amendment. Indeed, the first clause of the first section of that Amendment confirms the importance of citizenship by defining the means of obtaining it in a way that encompassed the freed slaves: "All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and the State wherein they reside." Citizenship is also a concept fundamental to structures and processes established elsewhere in the Constitution: "The distinction between citizens and aliens, though ordinarily irrelevant to private activity, is fundamental to the definition and government of a State. The Constitution itself refers to the distinction no less than 11 times, see Sugarman v. Dougall, supra [413 U.S.], at 651-652 [93 S.Ct., at 2862] (REHNQUIST, J., dissenting), indicating that the status of citizenship was meant to have significance in the structure of our government. The assumption of that status, whether by birth or naturalization, denotes an association with the polity which, in a democratic republic, exercises the powers of governance." Ambach v. Norwick, supra, at 75, 99 S.Ct., at 1593. Justice BLACKMUN has chosen to respond to this portion of the dissent, but misunderstands my point. I have observed that the political powerlessness of aliens is the result of state-created classifications which this Court has upheld as constitutional. One may nevertheless conclude, as Justice BLACKMUN does, that the political powerlessness of aliens is still a reason for applying strict scrutiny to alienage classifications. My point, to which Justice BLACKMUN's concurrence is unresponsive, is that a classification which is constitutionally relevant to many important state purposes should not be considered "suspect." It is beside the point to recognize that alienage may be irrelevant for some other purposes. Were this consideration conclusive, all state classifications would be considered "suspect" under the Equal Protection Clause because every classification is relevant to some purposes and irrelevant to others. 14 The State's written policy, effective since 1975, reads in part as follows: "1. It is the policy of the University of Maryland to grant in-state status for admission, tuition and charge-differential purposes to United States citizens, and to immigrant aliens lawfully admitted for permanent residence in accordance with the laws of the United States, in the following cases: "a. Where a student is financially dependent upon a parent, parents, or spouse domiciled in Maryland for at least six consecutive months prior to the last day available for registration for the forthcoming semester. "b. Where a student is financially independent for at least the preceding twelve months, and provided the student has maintained his domicile in Maryland for at least six consecutive months immediately prior to the last day available for registration for the forthcoming semester. "c. Where a student is the spouse or a dependent child of a full-time employee of the University. "d. Where a student who is a member of the Armed Forces of the United States is stationed on active duty in Maryland for at least six consecutive months immediately prior to the last day available for registration for the forthcoming semester, unless such student has been assigned for educational purposes to attend the University of Maryland. "e. Where a student is a full-time employee of the University of Maryland. "2. It is the policy of the University of Maryland to attribute out-of-state status for admission, tuition, and charge-differential purposes in all other cases." App. to Pet. for Cert. 167a-168a. 15 For example, in Nyquist, the Court stated: "Resident aliens are obligated to pay their full share of the taxes that support the assistance programs. There thus is no real unfairness in allowing resident aliens an equal right to participate in programs to which they contribute on an equal basis." 432 U.S., at 12, 97 S.Ct., at 2126. 16 In addition, nonimmigrant aliens are not required to register for military service. See 50 U.S.C.A. App. § 453(a) (1976 ed., Supp.V); 32 CFR § 1611.2 (1980). 17 Other nonresident aliens whose tax liability is not the subject of a treaty or special law such as the IOIA are subject to taxation only on income received from sources within the United States at a maximum rate of 30%. 26 U.S.C. § 871(a)(1). 18 Respondents, citing Nyquist v. Mauclet, 432 U.S., at 9, 97 S.Ct., at 2125, argue that strict scrutiny applies even when the State discriminates only against a certain subclass of aliens rather than all aliens. In Nyquist, the State argued that its law limiting financial assistance for higher education to citizens and resident aliens who declared their intention to seek citizenship was not a classification on the basis of alienage. Rather, it distinguished between aliens who intended to become citizens and those who did not. The Court rejected this argument, noting that the statute was "directed at aliens and that only aliens [were] harmed by it." Ibid. In this case, however, the State also denies in-state tuition to certain resident citizens, as well as to G-4 visaholders. Moreover, even if the State denied in-state tuition to G-4 visaholders alone, strict scrutiny would not be called for. As argued in the text, G-4 visaholders and other nonimmigrant aliens, unlike permanent resident aliens who were the subject of discrimination in Nyquist, are not so similarly situated to citizens as to render distinctions between such aliens and citizens "suspect." 19 As respondents note, G-4 visaholders do pay state taxes other than the income tax. State and local property taxes, however, do not enter the general funds of the State and thus do not support the operation of the University. Brief for Petitioners 29, n. 23. In any event, "a State does not violate the Equal Protection Clause merely because the classifications made by its laws are imperfect." Dandridge v. Williams, 397 U.S. 471, 485, 90 S.Ct. 1153, 1161, 25 L.Ed.2d 491 (1970). Respondents' exemption from the income tax sufficiently distinguishes them from citizens and other aliens who do pay such taxes, and therefore contribute a greater portion of their incomes to support the University, that the State's decision to require higher tuition payments is certainly rational.

12

458 U.S. 176 102 S.Ct. 3034 73 L.Ed.2d 690 BOARD OF EDUCATION OF the HENDRICK HUDSON CENTRAL SCHOOL DISTRICT BD. OF ED., WESTCHESTER COUNTY, et al., Petitionersv.Amy ROWLEY, by her parents and natural guardians, Clifford and Nancy Rowley etc. No. 80-1002. Argued March 23, 1982. Decided June 28, 1982. Syllabus The Education of the Handicapped Act (Act) provides federal money to assist state and local agencies in educating handicapped children. To qualify for federal assistance, a State must demonstrate, through a detailed plan submitted for federal approval, that it has in effect a policy that assures all handicapped children the right to a "free appropriate public education," which policy must be tailored to the unique needs of the handicapped child by means of an "individualized educational program" (IEP). The IEP must be prepared (and reviewed at least annually) by school officials with participation by the child's parents or guardian. The Act also requires that a participating State provide specified administrative procedures by which the child's parents or guardian may challenge any change in the evaluation and education of the child. Any party aggrieved by the state administrative decisions is authorized to bring a civil action in either a state court or a federal district court. Respondents—a child with only minimal residual hearing who had been furnished by school authorities with a special hearing aid for use in the classroom and who was to receive additional instruction from tutors, and the child's parents—filed suit in Federal District Court to review New York administrative proceedings that had upheld the school administrators' denial of the parents' request that the child also be provided a qualified sign-language interpreter in all of her academic classes. Entering judgment for respondents, the District Court found that although the child performed better than the average child in her class and was advancing easily from grade to grade, she was not performing as well academically as she would without her handicap. Because of this disparity between the child's achievement and her potential, the court held that she was not receiving a "free appropriate public education," which the court defined as "an opportunity to achieve [her] full potential commensurate with the opportunity provided to other children." The Court of Appeals affirmed. Held: 1. The Act's requirement of a "free appropriate public education" is satisfied when the State provides personalized instruction with sufficient support services to permit the handicapped child to benefit educationally from that instruction. Such instruction and services must be provided at public expense, must meet the State's educational standards, must approximate grade levels used in the State's regular education, and must comport with the child's IEP, as formulated in accordance with the Act's requirements. If the child is being educated in regular classrooms, as here, the IEP should be reasonably calculated to enable the child to achieve passing marks and advance from grade to grade. Pp. 187-204. (a) This interpretation is supported by the definitions contained in the Act, as well as by other provisions imposing procedural requirements and setting forth statutory findings and priorities for States to follow in extending educational services to handicapped children. The Act's language contains no express substantive standard prescribing the level of education to be accorded handicapped children. Pp. 187-190. (b) The Act's legislative history shows that Congress sought to make public education available to handicapped children, but did not intend to impose upon the States any greater substantive educational standard than is necessary to make such access to public education meaningful. The Act's intent was more to open the door of public education to handicapped children by means of specialized educational services than to guarantee any particular substantive level of education once inside. Pp. 191-197. (c) While Congress sought to provide assistance to the States in carrying out their constitutional responsibilities to provide equal protection of the laws, it did not intend to achieve strict equality of opportunity or services for handicapped and nonhandicapped children, but rather sought primarily to identify and evaluate handicapped children, and to provide them with access to a free public education. The Act does not require a State to maximize the potential of each handicapped child commensurate with the opportunity provided nonhandicapped children. Pp. 198-200. 2. In suits brought under the Act's judicial-review provisions, a court must first determine whether the State has complied with the statutory procedures, and must then determine whether the individualized program developed through such procedures is reasonably calculated to enable the child to receive educational benefits. If these requirements are met, the State has complied with the obligations imposed by Congress and the courts can require no more. Pp. 204-208. (a) Although the judicial-review provisions do not limit courts to ensuring that States have complied with the Act's procedural requirements, the Act's emphasis on procedural safeguards demonstrates the legislative conviction that adequate compliance with prescribed procedures will in most cases assure much, if not all, of what Congress wished in the way of substantive content in an IEP. Pp. 204-207. (b) The courts must be careful to avoid imposing their view of preferable educational methods upon the States. Once a court determines that the Act's requirements have been met, questions of methodology are for resolution by the States. Pp. 207-208. 3. Entrusting a child's education to state and local agencies does not leave the child without protection. As demonstrated by this case, parents and guardians will not lack ardor in seeking to ensure that handicapped children receive all of the benefits to which they are entitled by the Act. Pp.208-209 4. The Act does not require the provision of a sign-language interpreter here. Neither of the courts below found that there had been a failure to comply with the Act's procedures, and the findings of neither court will support a conclusion that the child's educational program failed to comply with the substantive requirements of the Act. Pp. 209-210. 632 F.2d 945 (2d Cir.), reversed and remanded. Raymond G. Kuntz, Bedford Village, N. Y., for petitioners. Michael A. Chatoff, Floral Park, N. Y., for respondent. Elliott Schulder, Washington, D. C., for the United States as amicus curiae by special leave of Court. Justice REHNQUIST delivered the opinion of the Court. 1 This case presents a question of statutory interpretation. Petitioners contend that the Court of Appeals and the District Court misconstrued the requirements imposed by Congress upon States which receive federal funds under the Education of the Handicapped Act. We agree and reverse the judgment of the Court of Appeals. 2 * The Education of the Handicapped Act (Act), 84 Stat. 175, as amended, 20 U.S.C. § 1401 et seq. (1976 ed. and Supp.IV), provides federal money to assist state and local agencies in educating handicapped children, and conditions such funding upon a State's compliance with extensive goals and procedures. The Act represents an ambitious federal effort to promote the education of handicapped children, and was passed in response to Congress' perception that a majority of handicapped children in the United States "were either totally excluded from schools or [were] sitting idly in regular classrooms awaiting the time when they were old enough to 'drop out.' " H.R.Rep.No. 94-332, p. 2 (1975) (H.R.Rep.). The Act's evolution and major provisions shed light on the question of statutory interpretation which is at the heart of this case. 3 Congress first addressed the problem of educating the handicapped in 1966 when it amended the Elementary and Secondary Education Act of 1965 to establish a grant program "for the purpose of assisting the States in the initiation, expansion, and improvement of programs and projects . . . for the education of handicapped children." Pub.L. 89-750, § 161, 80 Stat. 1204. That program was repealed in 1970 by the Education of the Handicapped Act, Pub.L. 91-230, 84 Stat. 175, Part B of which established a grant program similar in purpose to the repealed legislation. Neither the 1966 nor the 1970 legislation contained specific guidelines for state use of the grant money; both were aimed primarily at stimulating the States to develop educational resources and to train personnel for educating the handicapped.1 4 Dissatisfied with the progress being made under these earlier enactments, and spurred by two District Court decisions holding that handicapped children should be given access to a public education,2 Congress in 1974 greatly increased federal funding for education of the handicapped and for the first time required recipient States to adopt "a goal of providing full educational opportunities to all handicapped children." Pub.L. 93-380, 88 Stat. 579, 583 (1974 statute). The 1974 statute was recognized as an interim measure only, adopted "in order to give the Congress an additional year in which to study what if any additional Federal assistance [was] required to enable the States to meet the needs of handicapped children." H.R.Rep., at 4. The ensuing year of study produced the Education for All Handicapped Children Act of 1975. 5 In order to qualify for federal financial assistance under the Act, a State must demonstrate that it "has in effect a policy that assures all handicapped children the right to a free appropriate public education." 20 U.S.C. § 1412(1). That policy must be reflected in a state plan submitted to and approved by the Secretary of Education,3 § 1413, which describes in detail the goals, programs, and timetables under which the State intends to educate handicapped children within its borders. §§ 1412, 1413. States receiving money under the Act must provide education to the handicapped by priority, first "to handicapped children who are not receiving an education" and second "to handicapped children . . . with the most severe handicaps who are receiving an inadequate education," § 1412(3), and "to the maximum extent appropriate" must educate handicapped children "with children who are not handicapped." § 1412(5).4 The Act broadly defines "handicapped children" to include "mentally retarded, hard of hearing, deaf, speech impaired, visually handicapped, seriously emotionally disturbed, orthopedically impaired, [and] other health impaired children, [and] children with specific learning disabilities." § 1401(1).5 6 The "free appropriate public education" required by the Act is tailored to the unique needs of the handicapped child by means of an "individualized educational program" (IEP). § 1401(18). The IEP, which is prepared at a meeting between a qualified representative of the local educational agency, the child's teacher, the child's parents or guardian, and, where appropriate, the child, consists of a written document containing 7 "(A) a statement of the present levels of educational performance of such child, (B) a statement of annual goals, including short-term instructional objectives, (C) a statement of the specific educational services to be provided to such child, and the extent to which such child will be able to participate in regular educational programs, (D) the projected date for initiation and anticipated duration of such services, and (E) appropriate objective criteria and evaluation procedures and schedules for determining, on at least an annual basis, whether instructional objectives are being achieved." § 1401(19). 8 Local or regional educational agencies must review, and where appropriate revise, each child's IEP at least annually. § 1414(a)(5). See also § 1413(a)(11). 9 In addition to the state plan and the IEP already described, the Act imposes extensive procedural requirements upon States receiving federal funds under its provisions. Parents or guardians of handicapped children must be notified of any proposed change in "the identification, evaluation, or educational placement of the child or the provision of a free appropriate public education to such child," and must be permitted to bring a complaint about "any matter relating to" such evaluation and education. §§ 1415(b)(1)(D) and (E).6 Complaints brought by parents or guardians must be resolved at "an impartial due process hearing," and appeal to the state educational agency must be provided if the initial hearing is held at the local or regional level. §§ 1415(b)(2) and (c).7 Thereafter, "[a]ny party aggrieved by the findings and decision" of the state administrative hearing has "the right to bring a civil action with respect to the complaint . . . in any State court of competent jurisdiction or in a district court of the United States without regard to the amount in controversy." § 1415(e)(2). 10 Thus, although the Act leaves to the States the primary responsibility for developing and executing educational programs for handicapped children, it imposes significant requirements to be followed in the discharge of that responsibility. Compliance is assured by provisions permitting the withholding of federal funds upon determination that a participating state or local agency has failed to satisfy the requirements of the Act, §§ 1414(b)(2)(A), 1416, and by the provision for judicial review. At present, all States except New Mexico receive federal funds under the portions of the Act at issue today. Brief for United States as Amicus Curiae 2, n. 2. II 11 This case arose in connection with the education of Amy Rowley, a deaf student at the Furnace Woods School in the Hendrick Hudson Central School District, Peekskill, N.Y. Amy has minimal residual hearing and is an excellent lipreader. During the year before she began attending Furnace Woods, a meeting between her parents and school administrators resulted in a decision to place her in a regular kindergarten class in order to determine what supplemental services would be necessary to her education. Several members of the school administration prepared for Amy's arrival by attending a course in sign-language interpretation, and a teletype machine was installed in the principal's office to facilitate communication with her parents who are also deaf. At the end of the trial period it was determined that Amy should remain in the kindergarten class, but that she should be provided with an FM hearing aid which would amplify words spoken into a wireless receiver by the teacher or fellow students during certain classroom activities. Amy successfully completed her kindergarten year. 12 As required by the Act, an IEP was prepared for Amy during the fall of her first-grade year. The IEP provided that Amy should be educated in a regular classroom at Furnace Woods, should continue to use the FM hearing aid, and should receive instruction from a tutor for the deaf for one hour each day and from a speech therapist for three hours each week. The Rowleys agreed with parts of the IEP, but insisted that Amy also be provided a qualified sign-language interpreter in all her academic classes in lieu of the assistance proposed in other parts of the IEP. Such an interpreter had been placed in Amy's kindergarten class for a 2-week experimental period, but the interpreter had reported that Amy did not need his services at that time. The school administrators likewise concluded that Amy did not need such an interpreter in her first-grade classroom. They reached this conclusion after consulting the school district's Committee on the Handicapped, which had received expert evidence from Amy's parents on the importance of a sign-language interpreter, received testimony from Amy's teacher and other persons familiar with her academic and social progress, and visited a class for the deaf. 13 When their request for an interpreter was denied, the Rowleys demanded and received a hearing before an independent examiner. After receiving evidence from both sides, the examiner agreed with the administrators' determination that an interpreter was not necessary because "Amy was achieving educationally, academically, and socially" without such assistance. App. to Pet. for Cert. F-22. The examiner's decision was affirmed on appeal by the New York Commissioner of Education on the basis of substantial evidence in the record. Id., at E-4. Pursuant to the Act's provision for judicial review, the Rowleys then brought an action in the United States District Court for the Southern District of New York, claiming that the administrators' denial of the sign-language interpreter constituted a denial of the "free appropriate public education" guaranteed by the Act. 14 The District Court found that Amy "is a remarkably well-adjusted child" who interacts and communicates well with her classmates and has "developed an extraordinary rapport" with her teachers. 483 F.Supp. 528, 531 (1980). It also found that "she performs better than the average child in her class and is advancing easily from grade to grade," id., at 534, but "that she understands considerably less of what goes on in class than she could if she were not deaf" and thus "is not learning as much, or performing as well academically, as she would without her handicap," id., at 532. This disparity between Amy's achievement and her potential led the court to decide that she was not receiving a "free appropriate public education," which the court defined as "an opportunity to achieve [her] full potential commensurate with the opportunity provided to other children." Id., at 534. According to the District Court, such a standard "requires that the potential of the handicapped child be measured and compared to his or her performance, and that the resulting differential or 'shortfall' be compared to the shortfall experienced by nonhandicapped children." Ibid. The District Court's definition arose from its assumption that the responsibility for "giv[ing] content to the requirement of an 'appropriate education' " had "been left entirely to the [federal] courts and the hearing officers." Id., at 533.8 15 A divided panel of the United States Court of Appeals for the Second Circuit affirmed. The Court of Appeals "agree[d] with the [D]istrict [C]ourt's conclusions of law," and held that its "findings of fact [were] not clearly erroneous." 632 F.2d 945, 947 (1980). 16 We granted certiorari to review the lower courts' interpretation of the Act. 454 U.S. 961, 102 S.Ct. 500, 70 L.Ed.2d 376 (1981). Such review requires us to consider two questions: What is meant by the Act's requirement of a "free appropriate public education"? And what is the role of state and federal courts in exercising the review granted by 20 U.S.C. § 1415? We consider these questions separately.9 III A. 17 This is the first case in which this Court has been called upon to interpret any provision of the Act. As noted previously, the District Court and the Court of Appeals concluded that "[t]he Act itself does not define 'appropriate education,' " 483 F.Supp., at 533, but leaves "to the courts and the hearing officers" the responsibility of "giv[ing] content to the requirement of an 'appropriate education.' " Ibid. See also 632 F.2d, at 947. Petitioners contend that the definition of the phrase "free appropriate public education" used by the courts below overlooks the definition of that phrase actually found in the Act. Respondents agree that the Act defines "free appropriate public education," but contend that the statutory definition is not "functional" and thus "offers judges no guidance in their consideration of controversies involving 'the identification, evaluation, or educational placement of the child or the provision of a free appropriate public education.' " Brief for Respondents 28. The United States, appearing as amicus curiae on behalf of respondents, states that "[a]lthough the Act includes definitions of a 'free appropriate public education' and other related terms, the statutory definitions do not adequately explain what is meant by 'appropriate.' " Brief for United States as Amicus Curiae 13. 18 We are loath to conclude that Congress failed to offer any assistance in defining the meaning of the principal substantive phrase used in the Act. It is beyond dispute that, contrary to the conclusions of the courts below, the Act does expressly define "free appropriate public education": "The term 'free appropriate public education' means special education and related services which (A) have been provided at public expense, under public supervision and direction, and without charge, (B) meet the standards of the State educational agency, (C) include an appropriate preschool, elementary, or secondary school education in the State involved, and (D) are provided in conformity with the individualized education program required under section 1414(a)(5) of this title." § 1401(18) (emphasis added). 19 "Special education," as referred to in this definition, means "specially designed instruction, at no cost to parents or guardians, to meet the unique needs of a handicapped child, including classroom instruction, instruction in physical education, home instruction, and instruction in hospitals and institutions." § 1401(16). "Related services" are defined as "transportation, and such developmental, corrective, and other supportive services . . . as may be required to assist a handicapped child to benefit from special education." § 1401(17).10 20 Like many statutory definitions, this one tends toward the cryptic rather than the comprehensive, but that is scarcely a reason for abandoning the quest for legislative intent. Whether or not the definition is a "functional" one, as respondents contend it is not, it is the principal tool which Congress has given us for parsing the critical phrase of the Act. We think more must be made of it than either respondents or the United States seems willing to admit. 21 According to the definitions contained in the Act, a "free appropriate public education" consists of educational instruction specially designed to meet the unique needs of the handicapped child, supported by such services as are necessary to permit the child "to benefit" from the instruction. Almost as a checklist for adequacy under the Act, the definition also requires that such instruction and services be provided at public expense and under public supervision, meet the State's educational standards, approximate the grade levels used in the State's regular education, and comport with the child's IEP. Thus, if personalized instruction is being provided with sufficient supportive services to permit the child to benefit from the instruction, and the other items on the definitional checklist are satisfied, the child is receiving a "free appropriate public education" as defined by the Act. 22 Other portions of the statute also shed light upon congressional intent. Congress found that of the roughly eight million handicapped children in the United States at the time of enactment, one million were "excluded entirely from the public school system" and more than half were receiving an inappropriate education. 89 Stat. 774, note following § 1401. In addition, as mentioned in Part I, the Act requires States to extend educational services first to those children who are receiving no education and second to those children who are receiving an "inadequate education." § 1412(3). When these express statutory findings and priorities are read together with the Act's extensive procedural requirements and its definition of "free appropriate public education," the face of the statute evinces a congressional intent to bring previously excluded handicapped children into the public education systems of the States and to require the States to adopt procedures which would result in individualized consideration of and instruction for each child. 23 Noticeably absent from the language of the statute is any substantive standard prescribing the level of education to be accorded handicapped children. Certainly the language of the statute contains no requirement like the one imposed by the lower courts—that States maximize the potential of handicapped children "commensurate with the opportunity provided to other children." 483 F.Supp., at 534. That standard was expounded by the District Court without reference to the statutory definitions or even to the legislative history of the Act. Although we find the statutory definition of "free appropriate public education" to be helpful in our interpretation of the Act, there remains the question of whether the legislative history indicates a congressional intent that such education meet some additional substantive standard. For an answer, we turn to that history.11 B 24 (i) 25 As suggested in Part I, federal support for education of the handicapped is a fairly recent development. Before passage of the Act some States had passed laws to improve the educational services afforded handicapped children,12 but many of these children were excluded completely from any form of public education or were left to fend for themselves in classrooms designed for education of their nonhandicapped peers. As previously noted, the House Report begins by emphasizing this exclusion and misplacement, noting that millions of handicapped children "were either totally excluded from schools or [were] sitting idly in regular classrooms awaiting the time when they were old enough to 'drop out.' " H.R.Rep., at 2. See also S.Rep., at 8. One of the Act's two principal sponsors in the Senate urged its passage in similar terms: 26 "While much progress has been made in the last few years, we can take no solace in that progress until all handicapped children are, in fact, receiving an education. The most recent statistics provided by the Bureau of Education for the Handicapped estimate that . . . 1.75 million handicapped children do not receive any educational services, and 2.5 million handicapped children are not receiving an appropriate education." 121 Cong.Rec. 19486 (1975) (remarks of Sen. Williams). 27 This concern, stressed repeatedly throughout the legislative history,13 confirms the impression conveyed by the language of the statute: By passing the Act, Congress sought primarily to make public education available to handicapped children. But in seeking to provide such access to public education, Congress did not impose upon the States any greater substantive educational standard than would be necessary to make such access meaningful. Indeed, Congress expressly "recognize[d] that in many instances the process of providing special education and related services to handicapped children is not guaranteed to produce any particular outcome." S.Rep., at 11, U.S.Code Cong. & Admin.News 1975, p. 1435. Thus, the intent of the Act was more to open the door of public education to handicapped children on appropriate terms than to guarantee any particular level of education once inside. 28 Both the House and the Senate Reports attribute the impetus for the Act and its predecessors to two federal-court judgments rendered in 1971 and 1972. As the Senate Report states, passage of the Act "followed a series of landmark court cases establishing in law the right to education for all handicapped children." S.Rep., at 6, U.S.Code Cong. & Admin.News 1975, p. 1430.14 The first case, Pennsylvania Assn. for Retarded Children v. Commonwealth, 334 F.Supp. 1257 (Ed Pa.1971) and 343 F.Supp. 279 (1972) (PARC ), was a suit on behalf of retarded children challenging the constitutionality of a Pennsylvania statute which acted to exclude them from public education and training. The case ended in a consent decree which enjoined the State from "deny[ing] to any mentally retarded child access to a free public program of education and training." 334 F.Supp., at 1258 (emphasis added). 29 PARC was followed by Mills v. Board of Education of District of Columbia, 348 F.Supp. 866 (D.C.1972), a case in which the plaintiff handicapped children had been excluded from the District of Columbia public schools. The court's judgment, quoted in S.Rep., at 6, provided that 30 "no [handicapped] child eligible for a publicly supported education in the District of Columbia public schools shall be excluded from a regular school assignment by a Rule, policy, or practice of the Board of Education of the District of Columbia or its agents unless such child is provided (a) adequate alternative educational services suited to the child's needs, which may include special education or tuition grants, and (b) a constitutionally adequate prior hearing and periodic review of the child's status, progress, and the adequacy of any educational alternative." 348 F.Supp., at 878 (emphasis added). 31 Mills and PARC both held that handicapped children must be given access to an adequate, publicly supported education. Neither case purports to require any particular substantive level of education.15 Rather, like the language of the Act, the cases set forth extensive procedures to be followed in formulating personalized educational programs for handicapped children. See 348 F.Supp., at 878-883; 334 F.Supp., at 1258-1267.16 The fact that both PARC and Mills are discussed at length in the legislative Reports17 suggests that the principles which they established are the principles which, to a significant extent, guided the drafters of the Act. Indeed, immediately after discussing these cases the Senate Report describes the 1974 statute as having "incorporated the major principles of the right to education cases." S.Rep., at 8, U.S.Code Cong. & Admin.News 1975, p. 1432. Those principles in turn became the basis of the Act, which itself was designed to effectuate the purposes of the 1974 statute. H.R.Rep., at 5.18 32 That the Act imposes no clear obligation upon recipient States beyond the requirement that handicapped children receive some form of specialized education is perhaps best demonstrated by the fact that Congress, in explaining the need for the Act, equated an "appropriate education" to the receipt of some specialized educational services. The Senate Report states: "[T]he most recent statistics provided by the Bureau of Education for the Handicapped estimate that of the more than 8 million children . . . with handicapping conditions requiring special education and related services, only 3.9 million such children are receiving an appropriate education." S.Rep., at 8, U.S.Code Cong. & Admin.News 1975, p. 1432.19 This statement, which reveals Congress' view that 3.9 million handicapped children were "receiving an appropriate education" in 1975, is followed immediately in the Senate Report by a table showing that 3.9 million handicapped children were "served" in 1975 and a slightly larger number were "unserved." A similar statement and table appear in the House Report. H.R.Rep., at 11-12. 33 It is evident from the legislative history that the characterization of handicapped children as "served" referred to children who were receiving some form of specialized educational services from the States, and that the characterization of children as "unserved" referred to those who were receiving no specialized educational services. For example, a letter sent to the United States Commissioner of Education by the House Committee on Education and Labor, signed by two key sponsors of the Act in the House, asked the Commissioner to identify the number of handicapped "children served" in each State. The letter asked for statistics on the number of children "being served" in various types of "special education program[s]" and the number of children who were not "receiving educational services." Hearings on S. 6 before the Subcommittee on the Handicapped of the Senate Committee on Labor and Public Welfare, 94th Cong., 1st Sess., 205-207 (1975). Similarly, Senator Randolph, one of the Act's principal sponsors in the Senate, noted that roughly one-half of the handicapped children in the United States "are receiving special educational services." Id., at 1.20 By characterizing the 3.9 million handicapped children who were "served" as children who were "receiving an appropriate education," the Senate and House Reports unmistakably disclose Congress' perception of the type of education required by the Act: an "appropriate education" is provided when personalized educational services are provided.21 (ii) 34 Respondents contend that "the goal of the Act is to provide each handicapped child with an equal educational opportunity." Brief for Respondents 35. We think, however, that the requirement that a State provide specialized educational services to handicapped children generates no additional requirement that the services so provided be sufficient to maximize each child's potential "commensurate with the opportunity provided other children." Respondents and the United States correctly note that Congress sought "to provide assistance to the States in carrying out their responsibilities under . . . the Constitution of the United States to provide equal protection of the laws." S.Rep., at 13, U.S.Code Cong. & Admin.News 1975, p. 1437.22 But we do not think that such statements imply a congressional intent to achieve strict equality of opportunity or services. 35 The educational opportunities provided by our public school systems undoubtedly differ from student to student, depending upon a myriad of factors that might affect a particular student's ability to assimilate information presented in the classroom. The requirement that States provide "equal" educational opportunities would thus seem to present an entirely unworkable standard requiring impossible measurements and comparisons. Similarly, furnishing handicapped children with only such services as are available to nonhandicapped children would in all probability fall short of the statutory requirement of "free appropriate public education"; to require, on the other hand, the furnishing of every special service necessary to maximize each handicapped child's potential is, we think, further than Congress intended to go. Thus to speak in terms of "equal" services in one instance gives less than what is required by the Act and in another instance more. The theme of the Act is "free appropriate public education," a phrase which is too complex to be captured by the word "equal" whether one is speaking of opportunities or services. 36 The legislative conception of the requirements of equal protection was undoubtedly informed by the two District Court decisions referred to above. But cases such as Mills and PARC held simply that handicapped children may not be excluded entirely from public education. In Mills, the District Court said: 37 "If sufficient funds are not available to finance all of the services and programs that are needed and desirable in the system then the available funds must be expended equitably in such a manner that no child is entirely excluded from a publicly supported education consistent with his needs and ability to benefit therefrom." 348 F.Supp., at 876. 38 The PARC court used similar language, saying "[i]t is the commonwealth's obligation to place each mentally retarded child in a free, public program of education and training appropriate to the child's capacity. . . ." 334 F.Supp., at 1260. The right of access to free public education enunciated by these cases is significantly different from any notion of absolute equality of opportunity regardless of capacity. To the extent that Congress might have looked further than these cases which are mentioned in the legislative history, at the time of enactment of the Act this Court had held at least twice that the Equal Protection Clause of the Fourteenth Amendment does not require States to expend equal financial resources on the education of each child. San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973); McInnis v. Shapiro, 293 F.Supp. 327 (ND Ill.1968), aff'd sub nom. McInnis v. Ogilvie, 394 U.S. 322, 89 S.Ct. 1197, 22 L.Ed.2d 308 (1969). 39 In explaining the need for federal legislation, the House Report noted that "no congressional legislation has required a precise guarantee for handicapped children, i.e. a basic floor of opportunity that would bring into compliance all school districts with the constitutional right of equal protection with respect to handicapped children." H.R.Rep., at 14. Assuming that the Act was designed to fill the need identified in the House Report—that is, to provide a "basic floor of opportunity" consistent with equal protection—neither the Act nor its history persuasively demonstrates that Congress thought that equal protection required anything more than equal access. Therefore, Congress' desire to provide specialized educational services, even in furtherance of "equality," cannot be read as imposing any particular substantive educational standard upon the States. 40 The District Court and the Court of Appeals thus erred when they held that the Act requires New York to maximize the potential of each handicapped child commensurate with the opportunity provided nonhandicapped children. Desirable though that goal might be, it is not the standard that Congress imposed upon States which receive funding under the Act. Rather, Congress sought primarily to identify and evaluate handicapped children, and to provide them with access to a free public education. 41 (iii) 42 Implicit in the congressional purpose of providing access to a "free appropriate public education" is the requirement that the education to which access is provided be sufficient to confer some educational benefit upon the handicapped child. It would do little good for Congress to spend millions of dollars in providing access to a public education only to have the handicapped child receive no benefit from that education. The statutory definition of "free appropriate public education," in addition to requiring that States provide each child with "specially designed instruction," expressly requires the provision of "such . . . supportive services . . . as may be required to assist a handicapped child to benefit from special education." § 1401(17) (emphasis added). We therefore conclude that the "basic floor of opportunity" provided by the Act consists of access to specialized instruction and related services which are individually designed to provide educational benefit to the handicapped child.23 43 The determination of when handicapped children are receiving sufficient educational benefits to satisfy the requirements of the Act presents a more difficult problem. The Act requires participating States to educate a wide spectrum of handicapped children, from the marginally hearing-impaired to the profoundly retarded and palsied. It is clear that the benefits obtainable by children at one end of the spectrum will differ dramatically from those obtainable by children at the other end, with infinite variations in between. One child may have little difficulty competing successfully in an academic setting with nonhandicapped children while another child may encounter great difficulty in acquiring even the most basic of self-maintenance skills. We do not attempt today to establish any one test for determining the adequacy of educational benefits conferred upon all children covered by the Act. Because in this case we are presented with a handicapped child who is receiving substantial specialized instruction and related services, and who is performing above average in the regular classrooms of a public school system, we confine our analysis to that situation. 44 The Act requires participating States to educate handicapped children with nonhandicapped children whenever possible.24 When that "mainstreaming" preference of the Act has been met and a child is being educated in the regular classrooms of a public school system, the system itself monitors the educational progress of the child. Regular examinations are administered, grades are awarded, and yearly advancement to higher grade levels is permitted for those children who attain an adequate knowledge of the course material. The grading and advancement system thus constitutes an important factor in determining educational benefit. Children who graduate from our public school systems are considered by our society to have been "educated" at least to the grade level they have completed, and access to an "education" for handicapped children is precisely what Congress sought to provide in the Act.25 C 45 When the language of the Act and its legislative history are considered together, the requirements imposed by Congress become tolerably clear. Insofar as a State is required to provide a handicapped child with a "free appropriate public education," we hold that it satisfies this requirement by providing personalized instruction with sufficient support services to permit the child to benefit educationally from that instruction. Such instruction and services must be provided at public expense, must meet the State's educational standards, must approximate the grade levels used in the State's regular education, and must comport with the child's IEP. In addition, the IEP, and therefore the personalized instruction, should be formulated in accordance with the requirements of the Act and, if the child is being educated in the regular classrooms of the public education system, should be reasonably calculated to enable the child to achieve passing marks and advance from grade to grade.26 IV A. 46 As mentioned in Part I, the Act permits "[a]ny party aggrieved by the findings and decision" of the state administrative hearings "to bring a civil action" in "any State court of competent jurisdiction or in a district court of the United States without regard to the amount in controversy." § 1415(e)(2). The complaint, and therefore the civil action, may concern "any matter relating to the identification, evaluation, or educational placement of the child, or the provision of a free appropriate public education to such child." § 1415(b)(1)(E). In reviewing the complaint, the Act provides that a court "shall receive the record of the [state] administrative proceedings, shall hear additional evidence at the request of a party, and, basing its decision on the preponderance of the evidence, shall grant such relief as the court determines is appropriate." § 1415(e)(2). 47 The parties disagree sharply over the meaning of these provisions, petitioners contending that courts are given only limited authority to review for state compliance with the Act's procedural requirements and no power to review the substance of the state program, and respondents contending that the Act requires courts to exercise de novo review over state educational decisions and policies. We find petitioners' contention unpersuasive, for Congress expressly rejected provisions that would have so severely restricted the role of reviewing courts. In substituting the current language of the statute for language that would have made state administrative findings conclusive if supported by substantial evidence, the Conference Committee explained that courts were to make "independent decision[s] based on a preponderance of the evidence." S.Conf.Rep.No.94-455, p. 50 (1975), U.S.Code Cong. & Admin.News 1975, p. 1503. See also 121 Cong.Rec. 37416 (1975) (remarks of Sen. Williams). 48 But although we find that this grant of authority is broader than claimed by petitioners, we think the fact that it is found in § 1415, which is entitled "Procedural safeguards," is not without significance. When the elaborate and highly specific procedural safeguards embodied in § 1415 are contrasted with the general and somewhat imprecise substantive admonitions contained in the Act, we think that the importance Congress attached to these procedural safeguards cannot be gainsaid. It seems to us no exaggeration to say that Congress placed every bit as much emphasis upon compliance with procedures giving parents and guardians a large measure of participation at every stage of the administrative process, see, e.g., §§ 1415(a)-(d), as it did upon the measurement of the resulting IEP against a substantive standard. We think that the congressional emphasis upon full participation of concerned parties throughout the development of the IEP, as well as the requirements that state and local plans be submitted to the Secretary for approval, demonstrates the legislative conviction that adequate compliance with the procedures prescribed would in most cases assure much if not all of what Congress wished in the way of substantive content in an IEP. 49 Thus the provision that a reviewing court base its decision on the "preponderance of the evidence" is by no means an invitation to the courts to substitute their own notions of sound educational policy for those of the school authorities which they review. The very importance which Congress has attached to compliance with certain procedures in the preparation of an IEP would be frustrated if a court were permitted simply to set state decisions at nought. The fact that § 1415(e) requires that the reviewing court "receive the records of the [state] administrative proceedings" carries with it the implied requirement that due weight shall be given to these proceedings. And we find nothing in the Act to suggest that merely because Congress was rather sketchy in establishing substantive requirements, as opposed to procedural requirements for the preparation of an IEP, it intended that reviewing courts should have a free hand to impose substantive standards of review which cannot be derived from the Act itself. In short, the statutory authorization to grant "such relief as the court determines is appropriate" cannot be read without reference to the obligations, largely procedural in nature, which are imposed upon recipient States by Congress. 50 Therefore, a court's inquiry in suits brought under § 1415(e)(2) is twofold. First, has the State complied with the procedures set forth in the Act?27 And second, is the individualized educational program developed through the Act's procedures reasonably calculated to enable the child to receive educational benefits?28 If these requirements are met, the State has complied with the obligations imposed by Congress and the courts can require no more. B 51 In assuring that the requirements of the Act have been met, courts must be careful to avoid imposing their view of preferable educational methods upon the States.29 The primary responsibility for formulating the education to be accorded a handicapped child, and for choosing the educational method most suitable to the child's needs, was left by the Act to state and local educational agencies in cooperation with the parents or guardian of the child. The Act expressly charges States with the responsibility of "acquiring and disseminating to teachers and administrators of programs for handicapped children significant information derived from educational research, demonstration, and similar projects, and [of] adopting, where appropriate, promising educational practices and materials." § 1413(a)(3). In the face of such a clear statutory directive, it seems highly unlikely that Congress intended courts to overturn a State's choice of appropriate educational theories in a proceeding conducted pursuant to § 1415(e)(2).30 52 We previously have cautioned that courts lack the "specialized knowledge and experience" necessary to resolve "persistent and difficult questions of educational policy." San Antonio Independent School Dist. v. Rodriguez, 411 U.S., at 42, 93 S.Ct., at 1301. We think that Congress shared that view when it passed the Act. As already demonstrated, Congress' intention was not that the Act displace the primacy of States in the field of education, but that States receive funds to assist them in extending their educational systems to the handicapped. Therefore, once a court determines that the requirements of the Act have been met, questions of methodology are for resolution by the States. V 53 Entrusting a child's education to state and local agencies does not leave the child without protection. Congress sought to protect individual children by providing for parental involvement in the development of state plans and policies, supra, at 182-183, and n. 6, and in the formulation of the child's individual educational program. As the Senate Report states: 54 "The Committee recognizes that in many instances the process of providing special education and related services to handicapped children is not guaranteed to produce any particular outcome. By changing the language [of the provision relating to individualized educational programs] to emphasize the process of parent and child involvement and to provide a written record of reasonable expectations, the Committee intends to clarify that such individualized planning conferences are a way to provide parent involvement and protection to assure that appropriate services are provided to a handicapped child." S.Rep., at 11-12, U.S.Code Cong. & Admin.News 1975, p. 1435. 55 See also S.Conf.Rep.No.94-445, p. 30 (1975); 34 CFR § 300.345 (1981). As this very case demonstrates, parents and guardians will not lack ardor in seeking to ensure that handicapped children receive all of the benefits to which they are entitled by the Act.31 VI 56 Applying these principles to the facts of this case, we conclude that the Court of Appeals erred in affirming the decision of the District Court. Neither the District Court nor the Court of Appeals found that petitioners had failed to comply with the procedures of the Act, and the findings of neither court would support a conclusion that Amy's educational program failed to comply with the substantive requirements of the Act. On the contrary, the District Court found that the "evidence firmly establishes that Amy is receiving an 'adequate' education, since she performs better than the average child in her class and is advancing easily from grade to grade." 483 F.Supp., at 534. In light of this finding, and of the fact that Amy was receiving personalized instruction and related services calculated by the Furnace Woods school administrators to meet her educational needs, the lower courts should not have concluded that the Act requires the provision of a sign-language interpreter. Accordingly, the decision of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.32 57 So ordered. 58 Justice BLACKMUN, concurring in the judgment. 59 Although I reach the same result as the Court does today, I read the legislative history and goals of the Education of the Handicapped Act differently. Congress unambiguously stated that it intended to "take a more active role under its responsibility for equal protection of the laws to guarantee that handicapped children are provided equal educational opportunity." S.Rep.No.94-168, p. 9 (1975), U.S.Code Cong. & Admin.News 1975, p. 1433 (emphasis added). See also 20 U.S.C. § 1412(2)(A)(i) (requiring States to establish plans with the "goal of providing full educational opportunity to all handicapped children"). 60 As I have observed before, "[i]t seems plain to me that Congress, in enacting [this statute], intended to do more than merely set out politically self-serving but essentially meaningless language about what the [handicapped] deserve at the hands of state . . . authorities." Pennhurst State School v. Halderman, 451 U.S. 1, 32, 101 S.Ct. 1531, 1547, 67 L.Ed.2d 694 (1981) (opinion concurring in part and concurring in the judgment). The clarity of the legislative intent convinces me that the relevant question here is not, as the Court says, whether Amy Rowley's individualized education program was "reasonably calculated to enable [her] to receive educational benefits," ante, at 207, measured in part by whether or not she "achieve[s] passing marks and advance[s] from grade to grade," ante, at 204. Rather, the question is whether Amy's program, viewed as a whole, offered her an opportunity to understand and participate in the classroom that was substantially equal to that given her nonhandicapped classmates. This is a standard predicated on equal educational opportunity and equal access to the educational process, rather than upon Amy's achievement of any particular educational outcome. 61 In answering this question, I believe that the District Court and the Court of Appeals should have given greater deference than they did to the findings of the School District's impartial hearing officer and the State's Commissioner of Education, both of whom sustained petitioners' refusal to add a sign-language interpreter to Amy's individualized education program. Cf. 20 U.S.C. § 1415(e)(2) (requiring reviewing court to "receive the records of the administrative proceedings" before granting relief). I would suggest further that those courts focused too narrowly on the presence or absence of a particular service—a sign-language interpreter—rather than on the total package of services furnished to Amy by the School Board. 62 As the Court demonstrates, ante, at 184-185, petitioner Board has provided Amy Rowley considerably more than "a teacher with a loud voice." See post, at 215 (dissenting opinion). By concentrating on whether Amy was "learning as much, or performing as well academically, as she would without her handicap," 483 F.Supp. 528, 532 (S.D.N.Y.1980), the District Court and the Court of Appeals paid too little attention to whether, on the entire record, respondent's individualized education program offered her an educational opportunity substantially equal to that provided her nonhandicapped classmates. Because I believe that standard has been satisfied here, I agree that the judgment of the Court of Appeals should be reversed. 63 Justice WHITE, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 64 In order to reach its result in this case, the majority opinion contradicts itself, the language of the statute, and the legislative history. Both the majority's standard for a "free appropriate education" and its standard for judicial review disregard congressional intent. 65 * The majority first turns its attention to the meaning of a "free appropriate public education." The Act provides: 66 "The term 'free appropriate public education' means special education and related services which (A) have been provided at public expense, under public supervision and direction, and without charge, (B) meet the standards of the State educational agency, (C) include an appropriate preschool, elementary, or secondary school education in the State involved, and (D) are provided in conformity with the individualized education program required under section 1414(a)(5) of this title." 20 U.S.C. § 1401(18). 67 The majority reads this statutory language as establishing a congressional intent limited to bringing "previously excluded handicapped children into the public education systems of the States and [requiring] the States to adopt procedures which would result in individualized consideration of and instruction for each child." Ante, at 189. In its attempt to constrict the definition of "appropriate" and the thrust of the Act, the majority opinion states: "Noticeably absent from the language of the statute is any substantive standard prescribing the level of education to be accorded handicapped children. Certainly the language of the statute contains no requirement like the one imposed by the lower courts—that States maximize the potential of handicapped children 'commensurate with the opportunity provided to other children.' " Ante, at 189-190, quoting 483 F.Supp. 528, 534 (SDNY 1980). 68 I agree that the language of the Act does not contain a substantive standard beyond requiring that the education offered must be "appropriate." However, if there are limits not evident from the face of the statute on what may be considered an "appropriate education," they must be found in the purpose of the statute or its legislative history. The Act itself announces it will provide a "full educational opportunity to all handicapped children." 20 U.S.C. § 1412(2)(A) (emphasis added). This goal is repeated throughout the legislative history, in statements too frequent to be " 'passing references and isolated phrases.' "1 Ante, at 204, n. 26, quoting Department of State v. Washington Post Co., 456 U.S. 596, 600, 102 S.Ct. 1957, 1960, 72 L.Ed.2d 358 (1982). These statements elucidate the meaning of "appropriate." According to the Senate Report, for example, the Act does "guarantee that handicapped children are provided equal educational opportunity." S.Rep.No.94-168, p. 9 (1975), U.S.Code Cong. & Admin.News 1975, p. 1433 (emphasis added). This promise appears throughout the legislative history. See 121 Cong.Rec. 19482-19483 (1975) (remarks of Sen. Randolph); id., at 19504 (Sen. Humphrey); id., at 19505 (Sen Beall); id., at 23704 (Rep. Brademas); id., at 25538 (Rep. Cornell); id., at 25540 (Rep. Grassley); id., at 37025 (Rep. Perkins); id., at 37030 (Rep. Mink); id., at 37412 (Sen. Taft); id., at 37413 (Sen. Williams); id., at 37418-37419 (Sen. Cranston); id., at 37419-37420 (Sen. Beall). Indeed, at times the purpose of the Act was described as tailoring each handicapped child's educational plan to enable the child "to achieve his or her maximum potential." H.R.Rep.No.94-332, pp. 13, 19 (1975); see 121 Cong.Rec. 23709 (1975). Senator Stafford, one of the sponsors of the Act, declared: "We can all agree that education [given a handicapped child] should be equivalent, at least, to the one those children who are not handicapped receive." Id., at 19483. The legislative history thus directly supports the conclusion that the Act intends to give handicapped children an educational opportunity commensurate with that given other children. 69 The majority opinion announces a different substantive standard, that "Congress did not impose upon the States any greater substantive educational standard than would be necessary to make such access meaningful." Ante, at 192. While "meaningful" is no more enlightening than "appropriate," the Court purports to clarify itself. Because Amy was provided with some specialized instruction from which she obtained some benefit and because she passed from grade to grade, she was receiving a meaningful and therefore appropriate education.2 70 This falls far short of what the Act intended. The Act details as specifically as possible the kind of specialized education each handicapped child must receive. It would apparently satisfy the Court's standard of "access to specialized instruction and related services which are individually designed to provide educational benefit to the handicapped child," ante, at 201, for a deaf child such as Amy to be given a teacher with a loud voice, for she would benefit from that service. The Act requires more. It defines "special education" to mean "specifically designed instruction, at no cost to parents or guardians, to meet the unique needs of a handicapped child. . . ." § 1401(16) (emphasis added).3 Providing a teacher with a loud voice would not meet Amy's needs and would not satisfy the Act. The basic floor of opportunity is instead, as the courts below recognized, intended to eliminate the effects of the handicap, at least to the extent that the child will be given an equal opportunity to learn if that is reasonably possible. Amy Rowley, without a sign-language interpreter, comprehends less than half of what is said in the classroom—less than half of what normal children comprehend. This is hardly an equal opportunity to learn, even if Amy makes passing grades. 71 Despite its reliance on the use of "appropriate" in the definition of the Act, the majority opinion speculates that "Congress used the word as much to describe the settings in which handicapped children should be educated as to prescribe the substantive content or supportive services of their education." Ante, at 197, n. 21. Of course, the word "appropriate" can be applied in many ways; at times in the Act, Congress used it to recommend mainstreaming handicapped children; at other points, it used the word to refer to the content of the individualized education. The issue before us is what standard the word "appropriate" incorporates when it is used to modify "education." The answer given by the Court is not a satisfactory one. II 72 The Court's discussion of the standard for judicial review is as flawed as its discussion of a "free appropriate public education." According to the Court, a court can ask only whether the State has "complied with the procedures set forth in the Act" and whether the individualized education program is "reasonably calculated to enable the child to receive educational benefits." Ante, at 206, 207. Both the language of the Act and the legislative history, however, demonstrate that Congress intended the courts to conduct a far more searching inquiry. 73 The majority assigns major significance to the review provision's being found in a section entitled "Procedural safeguards." But where else would a provision for judicial review belong? The majority does acknowledge that the current language, specifying that a court "shall receive the records of the administrative proceedings, shall hear additional evidence at the request of a party, and, basing its decision on the preponderance of the evidence, shall grant such relief as the court determines is appropriate," § 1415(e)(2), was substituted at Conference for language that would have restricted the role of the reviewing court much more sharply. It is clear enough to me that Congress decided to reduce substantially judicial deference to state administrative decisions. 74 The legislative history shows that judicial review is not limited to procedural matters and that the state educational agencies are given first, but not final, responsibility for the content of a handicapped child's education. The Conference Committee directs courts to make an "independent decision." S.Conf.Rep.No.94-455, p. 50 (1975). The deliberate change in the review provision is an unusually clear indication that Congress intended courts to undertake substantive review instead of relying on the conclusions of the state agency. 75 On the floor of the Senate, Senator Williams, the chief sponsor of the bill, Committee Chairman, and floor manager responsible for the legislation in the Senate, emphasized the breadth of the review provisions at both the administrative and judicial levels: 76 "Any parent or guardian may present a complaint concerning any matter regarding the identification, evaluation, or educational placement of the child or the provision of a free appropriate public education to such child. In this regard, Mr. President, I would like to stress that the language referring to 'free appropriate education' has been adopted to make clear that a complaint may involve matters such as questions respecting a child's individualized education program, questions of whether special education and related services are being provided without charge to the parents or guardians, questions relating to whether the services provided a child meet the standards of the State education agency, or any other question within the scope of the definition of 'free appropriate public education.' In addition, it should be clear that a parent or guardian may present a complaint alleging that a State or local education agency has refused to provide services to which a child may be entitled or alleging that the State or local educational agency has erroneously classified a child as a handicapped child when, in fact, that child is not a handicapped child." 121 Cong.Rec. 37415 (1975) (emphasis added). 77 There is no doubt that the state agency itself must make substantive decisions. The legislative history reveals that the courts are to consider, de novo, the same issues. Senator Williams explicitly stated that the civil action permitted under the Act encompasses all matters related to the original complaint. Id., at 37416. 78 Thus, the Court's limitations on judicial review have no support in either the language of the Act or the legislative history. Congress did not envision that inquiry would end if a showing is made that the child is receiving passing marks and is advancing from grade to grade. Instead, it intended to permit a full and searching inquiry into any aspect of a handicapped child's education. The Court's standard, for example, would not permit a challenge to part of the IEP; the legislative history demonstrates beyond doubt that Congress intended such challenges to be possible, even if the plan as developed is reasonably calculated to give the child some benefits. 79 Parents can challenge the IEP for failing to supply the special education and related services needed by the individual handicapped child. That is what the Rowleys did. As the Government observes, "courts called upon to review the content of an IEP, in accordance with 20 U.S.C. [§] 1415(e) inevitably are required to make a judgment, on the basis of the evidence presented, concerning whether the educational methods proposed by the local school district are 'appropriate' for the handicapped child involved." Brief for United States as Amicus Curiae 13. The courts below, as they were required by the Act, did precisely that. 80 Under the judicial review provisions of the Act, neither the District Court nor the Court of Appeals was bound by the State's construction of what an "appropriate" education means in general or by what the state authorities considered to be an appropriate education for Amy Rowley. Because the standard of the courts below seems to me to reflect the congressional purpose and because their factual findings are not clearly erroneous, I respectfully dissent. 1 See S.Rep.No. 94-168, p. 5 (1975) (S.Rep.); H.R.Rep., at 2-3, U.S.Code Cong. & Admin.News 1975, p. 1425. 2 Two cases, Mills v. Board of Education of District of Columbia, 348 F.Supp. 866 (D.C.1972), and Pennsylvania Assn. for Retarded Children v. Commonwealth, 334 F.Supp. 1257 (ED Pa.1971) and 343 F.Supp. 279 (1972), were later identified as the most prominent of the cases contributing to Congress' enactment of the Act and the statutes which preceded it. H.R.Rep., at 3-4. Both decisions are discussed in Part III of this opinion. 3 All functions of the Commissioner of Education, formerly an officer in the Department of Health, Education, and Welfare, were transferred to the Secretary of Education in 1979 when Congress passed the Department of Education Organization Act, 20 U.S.C. § 3401 et seq. (1976 ed., Supp.IV). See 20 U.S.C. § 3441(a)(1) (1976 ed., Supp.IV). 4 Despite this preference for "mainstreaming" handicapped children—educating them with nonhandicapped children—Congress recognized that regular classrooms simply would not be a suitable setting for the education of many handicapped children. The Act expressly acknowledges that "the nature or severity of the handicap [may be] such that education in regular classes with the use of supplementary aids and services cannot be achieved satisfactorily." § 1412(5). The Act thus provides for the education of some handicapped children in separate classes or institutional settings. See ibid.; § 1413(a)(4). 5 In addition to covering a wide variety of handicapping conditions, the Act requires special educational services for children "regardless of the severity of their handicap." §§ 1412(2)(C), 1414(a)(1)(A). 6 The requirements that parents be permitted to file complaints regarding their child's education, and be present when the child's IEP is formulated, represent only two examples of Congress' effort to maximize parental involvement in the education of each handicapped child. In addition, the Act requires that parents be permitted "to examine all relevant records with respect to the identification, evaluation, and educational placement of the child, and . . . to obtain an independent educational evaluation of the child." § 1415(b)(1)(A). See also §§ 1412(4), 1414(a)(4). State educational policies and the state plan submitted to the Secretary of Education must be formulated in "consultation with individuals involved in or concerned with the education of handicapped children, including handicapped individuals and parents or guardians of handicapped children." § 1412(7). See also § 1412(2)(E). Local agencies, which receive funds under the Act by applying to the state agency, must submit applications which assure that they have developed procedures for "the participation and consultation of the parents or guardian[s] of [handicapped] children" in local educational programs, § 1414(a)(1)(C)(iii), and the application itself, along with "all pertinent documents related to such application," must be made "available to parents, guardians, and other members of the general public." § 1414(a)(4). 7 "Any party" to a state or local administrative hearing must "be accorded (1) the right to be accompanied and advised by counsel and by individuals with special knowledge or training with respect to the problems of handicapped children, (2) the right to present evidence and confront, cross examine, and compel the attendance of witnesses, (3) the right to a written or electronic verbatim record of such hearing, and (4) the right to written findings of fact and decisions." § 1415(d). 8 For reasons that are not revealed in the record, the District Court concluded that "[t]he Act itself does not define 'appropriate education.' " 483 F.Supp., at 533. In fact, the Act expressly defines the phrase "free appropriate public education," see § 1401(18), to which the District Court was referring. See 483 F.Supp., at 533. After overlooking the statutory definition, the District Court sought guidance not from regulations interpreting the Act, but from regulations promulgated under § 504 of the Rehabilitation Act. See 483 F.Supp., at 533, citing 45 CFR § 84.33(b). 9 The IEP which respondents challenged in the District Court was created for the 1978-1979 school year. Petitioners contend that the District Court erred in reviewing that IEP after the school year had ended and before the school administrators were able to develop another IEP for subsequent years. We disagree. Judicial review invariably takes more than nine months to complete, not to mention the time consumed during the preceding state administrative hearings. The District Court thus correctly ruled that it retained jurisdiction to grant relief because the alleged deficiencies in the IEP were capable of repetition as to the parties before it yet evading review. 483 F.Supp. 536, 538 (1980). See Murphy v. Hunt, 455 U.S. 478, 482, 102 S.Ct. 1181, 1183, 71 L.Ed.2d 353 (1982); Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 348, 46 L.Ed.2d 350 (1975). 10 Examples of "related services" identified in the Act are "speech pathology and audiology, psychological services, physical and occupational therapy, recreation, and medical and counseling services, except that such medical services shall be for diagnostic and evaluation purposes only." § 1401(17). 11 The dissent, finding that "the standard of the courts below seems . . . to reflect the congressional purpose" of the Act, post, at 218, concludes that our answer to this question "is not a satisfactory one." Post, at 216. Presumably, the dissent also agrees with the District Court's conclusion that "it has been left entirely to the courts and the hearing officers to give content to the requirement of an 'appropriate education.' " 483 F.Supp., at 533. It thus seems that the dissent would give the courts carte blanche to impose upon the States whatever burden their various judgments indicate should be imposed. Indeed, the dissent clearly characterizes the requirement of an "appropriate education" as open-ended, noting that "if there are limits not evident from the face of the statute on what may be considered an 'appropriate education,' they must be found in the purpose of the statute or its legislative history." Post, at 213. Not only are we unable to find any suggestion from the face of the statute that the requirement of an "appropriate education" was to be limitless, but we also view the dissent's approach as contrary to the fundamental proposition that Congress, when exercising its spending power, can impose no burden upon the States unless it does so unambiguously. See infra, at 204, n. 26. No one can doubt that this would have been an easier case if Congress had seen fit to provide a more comprehensive statutory definition of the phrase "free appropriate public education." But Congress did not do so, and "our problem is to construe what Congress has written. After all, Congress expresses its purpose by words. It is for us to ascertain—neither to add nor to subtract, neither to delete nor to distort." 62 Cases of Jam v. United States, 340 U.S. 593, 596, 71 S.Ct. 515, 518, 95 L.Ed. 566 (1951). We would be less than faithful to our obligation to construe what Congress has written if in this case we were to disregard the statutory language and legislative history of the Act by concluding that Congress had imposed upon the States a burden of unspecified proportions and weight, to be revealed only through case-by-case adjudication in the courts. 12 See H.R.Rep., at 10; Note, The Education of All Handicapped Children Act of 1975, 10 U.Mich.J.L.Ref. 110, 119 (1976). 13 See, e.g., 121 Cong.Rec. 19494 (1975) (remarks of Sen. Javits) ("all too often, our handicapped citizens have been denied the opportunity to receive an adequate education"); id., at 19502 (remarks of Sen. Cranston) (millions of handicapped "children . . . are largely excluded from the educational opportunities that we give to our other children"); id., at 23708 (remarks of Rep. Mink) ("handicapped children . . . are denied access to public schools because of a lack of trained personnel"). 14 Similarly, the Senate Report states that it was an "[i]ncreased awareness of the educational needs of handicapped children and landmark court decisions establishing the right to education for handicapped children [that] pointed to the necessity of an expanded federal fiscal role." S.Rep., at 5, U.S.Code Cong. & Admin.News 1975, p. 1429. See also H.R.Rep., at 2-3. 15 The only substantive standard which can be implied from these cases comports with the standard implicit in the Act. PARC states that each child must receive "access to a free public program of education and training appropriate to his learning capacities," 334 F.Supp., at 1258 (emphasis added), and that further state action is required when it appears that "the needs of the mentally retarded child are not being adequately served," id., at 1266. (Emphasis added.) Mills also speaks in terms of "adequate" educational services, 348 F.Supp., at 878, and sets a realistic standard of providing some educational services to each child when every need cannot be met. "If sufficient funds are not available to finance all of the services and programs that are needed and desirable in the system then the available funds must be expended equitably in such a manner that no child is entirely excluded from a publicly supported education consistent with his needs and ability to benefit therefrom. The inadequacies of the District of Columbia Public School System whether occasioned by insufficient funding or administrative inefficiency, certainly cannot be permitted to bear more heavily on the 'exceptional' or handicapped child than on the normal child." Id., at 876. 16 Like the Act, PARC required the State to "identify, locate, [and] evaluate" handicapped children, 334 F.Supp., at 1267, to create for each child an individual educational program, id., at 1265, and to hold a hearing "on any change in educational assignment," id., at 1266. Mills also required the preparation of an individual educational program for each child. In addition, Mills permitted the child's parents to inspect records relevant to the child's education, to obtain an independent educational evaluation of the child, to object to the IEP and receive a hearing before an independent hearing officer, to be represented by counsel at the hearing, and to have the right to confront and cross-examine adverse witnesses, all of which are also permitted by the Act. 348 F.Supp., at 879-881. Like the Act, Mills also required that the education of handicapped children be conducted pursuant to an overall plan prepared by the District of Columbia, and established a policy of educating handicapped children with nonhandicapped children whenever possible. Ibid. 17 See S.Rep., at 6-7; H.R.Rep., at 3-4. 18 The 1974 statute "incorporated the major principles of the right to education cases," by "add[ing] important new provisions to the Education of the Handicapped Act which require the States to: establish a goal of providing full educational opportunities to all handicapped children; provide procedures for insuring that handicapped children and their parents or guardians are guaranteed procedural safeguards in decisions regarding identification, evaluation, and educational placement of handicapped children; establish procedures to insure that, to the maximum extent appropriate, handicapped children . . . are educated with children who are not handicapped; . . . and, establish procedures to insure that testing and evaluation materials and procedures utilized for the purposes of classification and placement of handicapped children will be selected and administered so as not to be racially or culturally discriminatory." S.Rep., at 8, U.S.Code Cong. & Admin.News 1975, p. 1432. The House Report explains that the Act simply incorporated these purposes of the 1974 statute: the Act was intended "primarily to amend . . . the Education of the Handicapped Act in order to provide permanent authorization and a comprehensive mechanism which will insure that those provisions enacted during the 93rd Congress [the 1974 statute] will result in maximum benefits for handicapped children and their families." H.R.Rep., at 5. Thus, the 1974 statute's purpose of providing handicapped children access to a public education became the purpose of the Act. 19 These statistics appear repeatedly throughout the legislative history of the Act, demonstrating a virtual consensus among legislators that 3.9 million handicapped children were receiving an appropriate education in 1975. See, e.g., 121 Cong.Rec. 19486 (1975) (remarks of Sen. Williams); id., at 19504 (remarks of Sen. Schweicker); id., at 23702 (remarks of Rep. Madden); ibid. (remarks of Rep. Brademas); id., at 23709) (remarks of Rep. Minish); id., at 37024 (remarks of Rep. Brademas); id., at 37027 (remarks of Rep. Gude); id., at 37417 (remarks of Sen. Javits); id., at 37420 (remarks of Sen. Hathaway). 20 Senator Randolph stated: "[O]nly 55 percent of the school-aged handicapped children and 22 percent of the pre-school-aged handicapped children are receiving special educational services." Hearings on S. 6 before the Subcommittee on the Handicapped of the Senate Committee on Labor and Public Welfare, 94th Cong., 1st Sess., 1 (1975). Although the figures differ slightly in various parts of the legislative history, the general thrust of congressional calculations was that roughly one-half of the handicapped children in the United States were not receiving specialized educational services, and thus were not "served." See, e.g., 121 Cong.Rec. 19494 (1975) (remarks of Sen. Javits) ("only 50 percent of the Nation's handicapped children received proper education services"); id., at 19504 (remarks of Sen. Humphrey) ("[a]lmost 3 million handicapped children, while in school, receive none of the special services that they require in order to make education a meaningful experience"); id., at 23706 (remarks of Rep. Quie) ("only 55 percent [of handicapped children] were receiving a public education"); id., at 23709 (remarks of Rep. Biaggi) ("[o]ver 3 million [handicapped] children in this country are receiving either below par education or none at all"). Statements similar to those appearing in the text, which equate "served" as it appears in the Senate Report to "receiving special educational services," appear throughout the legislative history. See, e.g., id., at 19492 (remarks of Sen. Williams); id., at 19494 (remarks of Sen. Javits); id., at 19496 (remarks of Sen. Stone); id., at 19504-19505 (remarks of Sen. Humphrey); id., at 23703 (remarks of Rep. Brademas); Hearings on H.R. 7217 before the Subcommittee on Select Education of the House Committee on Education and Labor, 94th Cong., 1st Sess., 91, 150, 153 (1975); Hearings on H.R. 4199 before the Select Subcommittee on Education of the House Committee on Education and Labor, 93d Cong., 1st Sess., 130, 139 (1973). See also 34 CFR § 300.343 (1981). 21 In seeking to read more into the Act than its language or legislative history will permit, the United States focuses upon the word "appropriate," arguing that "the statutory definitions do not adequately explain what [it means]." Brief for United States as Amicus Curiae 13. Whatever Congress meant by an "appropriate" education, it is clear that it did not mean a potential-maximizing education. The term as used in reference to educating the handicapped appears to have originated in the PARC decision, where the District Court required that handicapped children be provided with "education and training appropriate to [their] learning capacities." 334 F.Supp., at 1258. The word appears again in the Mills decision, the District Court at one point referring to the need for "an appropriate educational program," 348 F.Supp., at 879, and at another point speaking of a "suitable publicly-supported education," id., at 878. Both cases also refer to the need for an "adequate" education. See 334 F.Supp., at 1266; 348 F.Supp., at 878. The use of "appropriate" in the language of the Act, although by no means definitive, suggests that Congress used the word as much to describe the settings in which handicapped children should be educated as to prescribe the substantive content or supportive services of their education. For example, § 1412(5) requires that handicapped children be educated in classrooms with nonhandicapped children "to the maximum extent appropriate." Similarly, § 1401(19) provides that, "whenever appropriate," handicapped children should attend and participate in the meeting at which their IEP is drafted. In addition, the definition of "free appropriate public education" itself states that instruction given handicapped children should be at an "appropriate preschool, elementary, or secondary school" level. § 1401(18)(C). The Act's use of the word "appropriate" thus seems to reflect Congress' recognition that some settings simply are not suitable environments for the participation of some handicapped children. At the very least, these statutory uses of the word refute the contention that Congress used "appropriate" as a term of art which concisely expresses the standard found by the lower courts. 22 See also 121 Cong.Rec. 19492 (1975) (remarks of Sen. Williams); id., at 19504 (remarks of Sen. Humphrey). 23 This view is supported by the congressional intention, frequently expressed in the legislative history, that handicapped children be enabled to achieve a reasonable degree of self-sufficiency. After referring to statistics showing that many handicapped children were excluded from public education, the Senate Report states: "The long range implications of these statistics are that public agencies and taxpayers will spend billions of dollars over the lifetimes of these individuals to maintain such persons as dependents and in a minimally acceptable lifestyle. With proper education services, many would be able to become productive citizens, contributing to society instead of being forced to remain burdens. Others, through such services, would increase their independence, thus reducing their dependence on society." S.Rep., at 9, U.S.Code Cong. & Admin.News 1975, p. 1433. See also H.R.Rep., at 11. Similarly, one of the principal Senate sponsors of the Act stated that "providing appropriate educational services now means that many of these individuals will be able to become a contributing part of our society, and they will not have to depend on subsistence payments from public funds." 121 Cong.Rec. 19492 (1975) (remarks of Sen. Williams). See also id., at 25541 (remarks of Rep. Harkin); id., at 37024-37025 (remarks of Rep. Brademas); id., at 37027 (remarks of Rep. Gude); id., at 37410 (remarks of Sen. Randolph); id., at 37416 (remarks of Sen. Williams). The desire to provide handicapped children with an attainable degree of personal independence obviously anticipated that state educational programs would confer educational benefits upon such children. But at the same time, the goal of achieving some degree of self-sufficiency in most cases is a good deal more modest than the potential-maximizing goal adopted by the lower courts. Despite its frequent mention, we cannot conclude, as did the dissent in the Court of Appeals, that self-sufficiency was itself the substantive standard which Congress imposed upon the States. Because many mildly handicapped children will achieve self-sufficiency without state assistance while personal independence for the severely handicapped may be an unreachable goal, "self-sufficiency" as a substantive standard is at once an inadequate protection and an overly demanding requirement. We thus view these references in the legislative history as evidence of Congress' intention that the services provided handicapped children be educationally beneficial, whatever the nature or severity of their handicap. 24 Title 20 U.S.C. § 1412(5) requires that participating States establish "procedures to assure that, to the maximum extent appropriate, handicapped children, including children in public or private institutions or other care facilities, are educated with children who are not handicapped, and that special classes, separate schooling, or other removal of handicapped children from the regular educational environment occurs only when the nature or severity of the handicap is such that education in regular classes with the use of supplementary aids and services cannot be achieved satisfactorily." 25 We do not hold today that every handicapped child who is advancing from grade to grade in a regular public school system is automatically receiving a "free appropriate public education." In this case, however, we find Amy's academic progress, when considered with the special services and professional consideration accorded by the Furnace Woods school administrators, to be dispositive. 26 In defending the decisions of the District Court and the Court of Appeals, respondents and the United States rely upon isolated statements in the legislative history concerning the achievement of maximum potential, see H.R.Rep., at 13, as support for their contention that Congress intended to impose greater substantive requirements than we have found. These statements, however, are too thin a reed on which to base an interpretation of the Act which disregards both its language and the balance of its legislative history. "Passing references and isolated phrases are not controlling when analyzing a legislative history." Department of State v. Washington Post Co., 456 U.S. 595, 600, 102 S.Ct. 1957, 1960, 72 L.Ed.2d 358 (1982). Moreover, even were we to agree that these statements evince a congressional intent to maximize each child's potential, we could not hold that Congress had successfully imposed that burden upon the States. "[L]egislation enacted pursuant to the spending power is much in the nature of a contract: in return for federal funds, the States agree to comply with federally imposed conditions. The legitimacy of Congress' power to legislate under the spending power thus rests on whether the State voluntarily and knowingly accepts the terms of the 'contract.' . . . Accordingly, if Congress intends to impose a condition on the grant of federal moneys, it must do so unambiguously." Pennhurst State School v. Halderman, 451 U.S. 1, 17, 101 S.Ct. 1531, 1539-40, 67 L.Ed.2d 694 (1981) (footnote omitted). As already demonstrated, the Act and its history impose no requirements on the States like those imposed by the District Court and the Court of Appeals. A fortiori Congress has not done so unambiguously, as required in the valid exercise of its spending power. 27 This inquiry will require a court not only to satisfy itself that the State has adopted the state plan, policies, and assurances required by the Act, but also to determine that the State has created an IEP for the child in question which conforms with the requirements of § 1401(19). 28 When the handicapped child is being educated in the regular classrooms of a public school system, the achievement of passing marks and advancement from grade to grade will be one important factor in determining educational benefit. See Part III, supra. 29 In this case, for example, both the state hearing officer and the District Court were presented with evidence as to the best method for educating the deaf, a question long debated among scholars. See Large, Special Problems of the Deaf Under the Education for All Handicapped Children Act of 1975, 58 Wash.U.L.Q. 213, 229 (1980). The District Court accepted the testimony of respondents' experts that there was "a trend supported by studies showing the greater degree of success of students brought up in deaf households using [the method of communication used by the Rowleys]." 483 F.Supp., at 535. 30 It is clear that Congress was aware of the States' traditional role in the formulation and execution of educational policy. "Historically, the States have had the primary responsibility for the education of children at the elementary and secondary level." 121 Cong.Rec. 19498 (1975) (remarks of Sen. Dole). See also Epperson v. Arkansas, 393 U.S. 97, 104, 89 S.Ct. 266, 270, 21 L.Ed.2d 228 (1968) ("By and large, public education in our Nation is committed to the control of state and local authorities"). 31 In addition to providing for extensive parental involvement in the formulation of state and local policies, as well as the preparation of individual educational programs, the Act ensures that States will receive the advice of experts in the field of educating handicapped children. As a condition for receiving federal funds under the Act, States must create "an advisory panel, appointed by the Governor or any other official authorized under State law to make such appointments, composed of individuals involved in or concerned with the education of handicapped children, including handicapped individuals, teachers, parents or guardians of handicapped children, State and local education officials, and administrators of programs for handicapped children, which (A) advises the State educational agency of unmet needs within the State in the education of handicapped children, [and] (B) comments publicly on any rules or regulations proposed for issuance by the State regarding the education of handicapped children." § 1413(a)(12). 32 Because the District Court declined to reach respondents' contention that petitioners had failed to comply with the Act's procedural requirements in developing Amy's IEP, 483 F.Supp., at 533, n. 8, the case must be remanded for further proceedings consistent with this opinion. 1 The Court's opinion relies heavily on the statement, which occurs throughout the legislative history, that, at the time of enactment, one million of the roughly eight million handicapped children in the United States were excluded entirely from the public school system and more than half were receiving an inappropriate education. See, e.g., ante, at 189,195, 196-197, n. 20. But this statement was often linked to statements urging equal educational opportunity. See, e.g., 121 Cong.Rec. 19502 (1975) (remarks of Sen. Cranston); id., at 23702 (remarks of Rep. Brademas). That is, Congress wanted not only to bring handicapped children into the schoolhouse, but also to benefit them once they had entered. 2 As further support for its conclusion, the majority opinion turns to Pennsylvania Assn. for Retarded Children v. Commonwealth, 334 F.Supp. 1257 (ED Pa.1971), 343 F.Supp. 279 (1972) (PARC ), and Mills v. Board of Education of District of Columbia, 348 F.Supp. 866 (D.C.1972). That these decisions served as an impetus for the Act does not, however, establish them as the limits of the Act. In any case, the very language that the majority quotes from Mills, ante, at 193,199, sets a standard not of some education, but of educational opportunity equal to that of non-handicapped children. Indeed, Mills, relying on decisions since called into question by this Court's opinion in San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973), states: "In Hobson v. Hansen, [269 F.Supp. 401 (D.C.1967) ] Judge Wright found that denying poor public school children educational opportunity equal to that available to more affluent public school children was violative of the Due Process Clause of the Fifth Amendment. A fortiori, the defendants' conduct here, denying plaintiffs and their class not just an equal publicly supported education but all publicly supported education while providing such education to other children, is violative of the Due Process Clause." 348 F.Supp., at 875. Whatever the effect of Rodriguez on the validity of this reasoning, the statement exposes the majority's mischaracterization of the opinion and thus of the assumptions of the legislature that passed the Act. 3 "Related services" are "transportation, and such developmental, corrective, and other supportive services . . . as may be required to assist a handicapped child to benefit from special education." § 1401(17).

12

458 U.S. 276 102 S.Ct. 3477 73 L.Ed.2d 764 73 L.Ed.2d 1364 State of SOUTH DAKOTA, plaintiff,v.State of NEBRASKA, and Robert J. Foley and Phyllis K. Foley, as joint tenants and their survivor, and Helen Isaak, as executrix of the estate of Otto Isaak, deceased, Intervenors No. 72 Supreme Court of the United States June 28, 1982 1 On bill of complaint. 2 The Report of the Special Master is received and ordered filed. 3 This matter comes before this Court on Stipulation by each of the parties hereto: the State of South Dakota, by and through its Attorney General, Mark Meierhenry; the State of Nebraska, by and through its Attorney General, Paul L. Douglas; the Intervenors, by and through themselves and their attorney, Everett A. Bogue. FINDINGS AND DECREE IT IS THE FINDING OF THIS COURT THAT: 4 1. The parties have concluded that it is in the best interest of each of them to avoid litigation and multiple exercises of sovereignty and jurisdiction, encourage the optimum beneficial use of the river, its facilities and its waters, and remove all causes of controversy between said parties with respect to the location of the boundary between the States and, therefore, to settle and to terminate this dispute by agreement and compromise and submission of the boundary between the State of South Dakota and the State of Nebraska with respect to Rush Island in the future to a Joint State Boundary Commission. 5 2. In the furtherance of the best interests of each of the parties, it is agreed that that land known as "Elk/Rush Island," subject to this lawsuit, is now, and has been, within the boundary of the State of Nebraska and subject to its jurisdiction. 6 3. The State of South Dakota has agreed to cede to the Intervenors any right or title it may have in the subject property known as "Elk/Rush Island," more fully described in Appendix A attached to the Stipulation; and the State of South Dakota specifically recognizes the Judgment of the District Court of Cedar County, Nebraska, dated November 7, 1958, which quieted title to the above identified land in Clyde Gill and others, predecessors of the Intervenors, which action can be found in the records of Cedar County District Court of Nebraska, Case No. 5628, Docket 24, Page 13. 7 4. The State of South Dakota has further agreed to dismiss an action to quiet title filed in the Circuit Court of Yankton, South Dakota, regarding this subject property. 8 5. The States of Nebraska and South Dakota have agreed to submit the determination of future boundary changes with regard to "Elk/Rush Island" but not to the title to the premises therein, to a Joint State Boundary Commission appointed by the elected officials of these respective States for the Commission's determination from the date of that determination forth. 9 IT IS, THEREFORE, ORDERED, ADJUDGED and DECREED that pursuant to the above Stipulations and Findings: 10 1. That the land known as "Elk/Rush Island," the subject of this litigation, be and the same is within the boundary of the State of Nebraska and subject to its jurisdiction; 11 2. That the State of South Dakota hereby cedes to the Intervenors any right or title it may have to the subject property known as "Elk/Rush Island" as more fully described in Appendix A attached hereto; 12 3. That the States of Nebraska and South Dakota will submit the determination of the future boundary changes, if any, with regard, to "Elk/Rush Island," but not to the title of the premises thereto, to a Joint State Boundary Commission appointed by the elected officials of these respective States for the Commission's determination from the date of that determination forth; 13 4. That the State of South Dakota and the State of Nebraska will proceed to effectuate any and all other requirements agreed to in the Stipulation to the extent that conditions and circumstances permit; and 14 5. Each party shall bear its own costs. APPENDIX A 15 Rush Island in the Missouri River, said island being a part of Sections 5, 6 and 7, Township 33, Range 1, East, Cedar County, Nebraska, according to the original Government Survey and which is more fully and particularly described in the Surveyor's record No. 5, Page 70 as follows: Commencing at the Section corner of Sections 7 and 8, in Township 33, Range 1, East, Cedar County, Nebraska, as is located by the survey recorded in Surveyor's Record, Volume 4, page 29 of the records of Cedar County, Nebraska, thence running due north from this corner 5060 feet to the point of beginning, thence running 710 feet north 81 degrees east, thence 805 feet north 68 degrees east, thence 942 feet north 59 degrees east, thence 326 feet north 88 degrees east, thence 874 feet north 84 degrees east, thence 362 feet south 54 degrees east, thence 305 feet north 58 degrees east, thence 1140 feet north 65 degrees east, thence 1315 feet north 5 degrees east, thence 805 feet north 60 degrees west, thence 1800 feet north 69 degrees west, thence 1225 feet north 72 degrees west, thence 2080 feet north 79 degrees west, thence 1390 feet north 57 degrees west, thence 1160 feet north 86 degrees west, thence 582 feet south 84 degrees west, thence 1090 feet south 55 degrees west, thence 1980 feet south 89 degrees west, thence 688 feet south 47 degrees west, thence 1400 feet south 20 degrees west, thence 1415 feet south 55 degrees west, thence 865 feet south 78 degrees east, thence 588 feet south 87 degrees east, thence 2220 feet south 72 degrees east, thence 2530 feet south 78 degrees east, thence 1890 feet south 62 degrees east, thence 425 feet south 71 degrees east to the point of beginning, containing 994 acres more or less, all in Township 33, Range 1 East in Cedar County, Nebraska, and that intervenors are the record owners thereof.

1011

458 U.S. 259 102 S.Ct. 3079 73 L.Ed.2d 750 MICHIGANv.Lamont Charles THOMAS. No. 81-593. June 28, 1982. PER CURIAM. 1 While respondent was the front-seat passenger in an automobile, the car was stopped for failing to signal a left turn. As two police officers approached the vehicle, they saw respondent bend forward so that his head was at or below the level of the dashboard. The officers then observed an open bottle of malt liquor standing upright on the floorboard between respondent's feet, and placed respondent under arrest for possession of open intoxicants in a motor vehicle. The 14-year-old driver was issued a citation for not having a driver's license. Respondent claimed ownership of the car. 2 Respondent and the driver were taken to the patrol car, and a truck was called to tow respondent's automobile. One of the officers searched the vehicle, pursuant to a departmental policy that impounded vehicles be searched prior to being towed. The officer found two bags of marihuana in the unlocked glove compartment. The second officer then searched the car more thoroughly, checking under the front seat, under the dashboard, and inside the locked trunk. Opening the air vents under the dashboard, the officer discovered a loaded, .38-caliber revolver inside. 3 Respondent was convicted of possession of a concealed weapon. He moved for a new trial, contending that the revolver was taken from his car pursuant to an illegal search and seizure; the trial court denied the motion. 4 The Michigan Court of Appeals reversed, holding that the warrantless search of respondent's automobile violated the Fourth Amendment. 106 Mich.App. 601, 308 N.W.2d 170 (1981). The court acknowledged that in South Dakota v. Opperman, 428 U.S. 364, 96 S.Ct. 3092, 49 L.Ed.2d 1000 (1976), this Court upheld the validity of warrantless inventory searches of impounded motor vehicles. Moreover, the court found that, since respondent had been placed under arrest and the other occupant of the car was too young to legally drive, it was proper for the officers to impound the vehicle and to conduct an inventory search prior to its being towed. However, in the view of the Court of Appeals, the search conducted in this case was "unreasonable in scope," because it extended to the air vents which, unlike the glove compartment or the trunk, were not a likely place for the storage of valuables or personal possessions. 106 Mich.App., at 606, 308 N.W.2d, at 172. 5 The Court of Appeals also rejected the State's contention that the scope of the inventory search was properly expanded when the officers discovered contraband in the glove compartment. The court concluded that, because both the car and its occupants were already in police custody, there were no "exigent circumstances" justifying a warrantless search for contraband.1 6 We reverse. In Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970), we held that when police officers have probable cause to believe there is contraband inside an automobile that has been stopped on the road, the officers may conduct a warrantless search of the vehicle, even after it has been impounded and is in police custody. We firmly reiterated this holding in Texas v. White, 423 U.S. 67, 96 S.Ct. 304, 46 L.Ed.2d 209 (1975). See also United States v. Ross, 456 U.S. 798, 807, n.9, 102 S.Ct. 2157, 2163, n.9, 72 L.Ed.2d 572 (1982). It is thus clear that the justification to conduct such a warrantless search does not vanish once the car has been immobilized; nor does it depend upon a reviewing court's assessment of the likelihood in each particular case that the car would have been driven away, or that its contents would have been tampered with, during the period required for the police to obtain a warrant.2 See ibid. 7 Here, the Court of Appeals recognized that the officers were justified in conducting an inventory search of the car's glove compartment, which led to the discovery of contraband. Without attempting to refute the State's contention that this discovery gave the officers probable cause to believe there was contraband elsewhere in the vehicle, the Court of Appeals held that the absence of "exigent circumstances" precluded a warrantless search. This holding is plainly inconsistent with our decisions in Chambers and Texas v. White. 8 The petition for certiorari and the motion of respondent to proceed in forma pauperis are granted, the judgment of the Michigan Court of Appeals is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion. 9 It is so ordered. 10 Justice BRENNAN and Justice MARSHALL would grant the petition for a writ of certiorari and set the case for oral argument. 1 The Court of Appeals did not directly address the State's contention that the discovery of marihuana in the glove compartment provided probable cause to believe there was contraband hidden elsewhere in the vehicle. However, the court apparently assumed that the officers possessed information sufficient to support issuance of a warrant to search the automobile; the court's holding was that the officers were required to obtain such a warrant, and could not search on the basis of probable cause alone. See 106 Mich.App., at 606-608, 308 N.W.2d, at 172-173. 2 Even were some demonstrable "exigency" a necessary predicate to such a search, we would find somewhat curious the Court of Appeals' conclusion that no "exigent circumstances" were present in this case. Unlike the searches involved in Chambers v. Maroney, 399 U.S. 42, 90 S.Ct. 1975, 26 L.Ed.2d 419 (1970), and Texas v. White, 423 U.S. 67, 96 S.Ct. 304, 46 L.Ed.2d 209 (1975) which were conducted at the station house—the search at issue here was conducted on the roadside, before the car had been towed. As pointed out by Judge Deneweth, in dissent, "there was a clear possibility that the occupants of the vehicle could have had unknown confederates who would return to remove the secreted contraband." 106 Mich.App., at 609, 308 N.W.2d, at 174.

01

458 U.S. 263 102 S.Ct. 3081 73 L.Ed.2d 754 UNITED STATESv.HOLLYWOOD MOTOR CAR COMPANY, INC., et al. No. 81-1144. June 28, 1982. PER CURIAM. 1 Respondents, originally indicted in the Eastern District of Kentucky on two counts for violations of 18 U.S.C. §§ 371 and 545, succeeded in obtaining a change of venue to the Central District of California. In the latter District, the Government secured a superseding indictment charging four new substantive counts of making false statements to customs officers in violation of 18 U.S.C. § 542, in addition to the two original counts. The Government then obtained a voluntary dismissal of the original conspiracy count and two of the false-statement counts. Respondents moved to dismiss the remaining counts on the ground that the superseding indictment manifested prosecutorial vindictiveness and therefore ran afoul of the rule announced in Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974). The District Court denied respondents' motion, but stayed the commencement of trial to permit an appeal. The Court of Appeals for the Ninth Circuit held "that the denial of a motion to dismiss based on the ground of vindictive prosecution is immediately appealable as a final decision under 28 U.S.C. § 1291." 646 F.2d 384, 386 (9th Cir. 1981).1 In reaching this holding the Court of Appeals relied on its prior decisions in United States v. Burt, 619 F.2d 831 (1980), and United States v. Griffin, 617 F.2d 1342, cert. denied, 449 U.S. 863, 101 S.Ct. 167, 66 L.Ed.2d 80 (1980). Reaching the merits, the court held that respondents had established a case of prosecutorial vindictiveness requiring dismissal of the superseding indictment. The United States then sought review in this Court. 2 We do not reach the question of prosecutorial vindictiveness, for we hold that the Court of Appeals was without jurisdiction under 28 U.S.C. § 1291 to review the District Court's interlocutory order refusing to dismiss the indictment. Congress has limited the jurisdiction of the Courts of Appeals to "final decisions of the district courts." 28 U.S.C. § 1291. This Court has long held that the policy of Congress embodied in this statute is inimical to piecemeal appellate review of trial court decisions which do not terminate the litigation, and that this policy is at its strongest in the field of criminal law: 3 "The general principle of federal appellate jurisdiction, derived from the common law and enacted by the First Congress, requires that review of nisi prius proceedings await their termination by final judgment. . . . This insistence on finality and prohibition of piecemeal review discourage undue litigiousness and leaden-footed administration of justice, particularly damaging to the conduct of criminal cases. See Cobbledick v. United States, 309 U.S. 323, 324-326 [60 S.Ct. 540, 541, 84 L.Ed. 783]." DiBella v. United States, 369 U.S. 121, 124, 82 S.Ct. 654, 656, 7 L.Ed.2d 614 (7). 4 This Court has interpreted the jurisdictional statute to permit departures from the rule of finality in only a limited category of cases falling within the "collateral order" exception delineated in Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 545-547, 69 S.Ct. 1221, 1225-26, 93 L.Ed. 1528 (1949). Such orders "must conclusively determine the disputed question, resolve an important issue completely separate from the merits of the action, and be effectively unreviewable on appeal from a final judgment." Coopers & Lybrand v. Livesay, 437 U.S. 463, 468, 98 S.Ct. 2454, 2458, 57 L.Ed.2d 351 (1978). In criminal cases, we have adhered to the collateral order exception to the rule of finality on three occasions: present in each of these cases were factors noticeably lacking in the instant appeal. 5 In Stack v. Boyle, 342 U.S. 1, 72 S.Ct. 1, 96 L.Ed. 3 (1951), the Court held that an order denying a motion to reduce bail could be reviewed before trial. Writing separately, Justice Jackson (the author of Cohen) recognized that "an order fixing bail can be reviewed without halting the main trial—its issues are entirely independent of the issues to be tried—and unless it can be reviewed before sentence, it never can be reviewed at all." 342 U.S., at 12, 72 S.Ct., at 7. In Abney v. United States, 431 U.S. 651, 97 S.Ct. 2034, 52 L.Ed.2d 651 (1977), we permitted interlocutory appeal of an order denying a pretrial motion to dismiss an indictment on double jeopardy grounds. Perhaps most important among the relevant factors, we recognized that "the rights conferred on a criminal accused by the Double Jeopardy Clause would be significantly undermined if appellate review of double jeopardy claims were postponed until after conviction and sentence." Id., at 660, 97 S.Ct., at 2040. One right guaranteed by the Double Jeopardy Clause was the right not to be tried twice for the same offense. "[I]f a criminal defendant is to avoid exposure to double jeopardy and thereby enjoy the full protection of the Clause, his double jeopardy challenge to the indictment must be reviewable before that subsequent exposure occurs." Id., at 662, 97 S.Ct., at 2041 (emphasis in original). Finally, in Helstoski v. Meanor, 442 U.S. 500, 99 S.Ct. 2445, 61 L.Ed.2d 30 (1979), we held that a United States Congressman could have taken an interlocutory appeal in a criminal case to assert the immunity conferred upon him by the Speech or Debate Clause of the Constitution. Crucial to the holding was our view that the Speech or Debate Clause protected Congressmen " 'not only from the consequences of litigation's results but also from the burden of defending themselves.' " Id., at 508, 99 S.Ct., at 2449, quoting Dombrowski v. Eastland, 387 U.S. 82, 85, 87 S.Ct. 1425, 1427, 18 L.Ed.2d 577 (1967). The right protected by the Clause would have been lost if the appeal had been postponed. 6 Each of these cases, in addition to satisfying the other requirements of Cohen, involved "an asserted right the legal and practical value of which would be destroyed if it were not vindicated before trial." United States v. MacDonald, 435 U.S. 850, 860, 98 S.Ct. 1547, 1552, 56 L.Ed.2d 18 (1978). Our holding in United States v. MacDonald, underscores the significance of this feature. The issue in MacDonald was whether a defendant could appeal, prior to trial, a District Court's order denying his motion to dismiss the indictment because of an alleged violation of his Sixth Amendment right to a speedy trial. In concluding that such an appeal was not authorized by 28 U.S.C. § 1291, we noted: 7 "There perhaps is some superficial attraction in the argument that the right to a speedy trial . . . must be vindicated before trial in order to insure that no nonspeedy trial is ever held. Both doctrinally and pragmatically, however, this argument fails. Unlike the protection afforded by the Double Jeopardy Clause, the Speedy Trial Clause does not, either on its face or according to the decisions of this Court, encompass a 'right not to be tried' which must be upheld prior to trial if it is to be enjoyed at all. It is the delay before trial, not the trial itself, that offends against the constitutional guarantee of a speedy trial. . . . Proceeding with the trial does not cause or compound the deprivation already suffered." Id., at 860-861, 98 S.Ct., at 1552-53. 8 Respondents assert that their claim of prosecutorial vindictiveness, based on the modification of the original indictment in retaliation for their exercise of a right to move for change of venue, is analogous to the three instances in which we have allowed appeal in criminal cases under the collateral order doctrine. But we think that their claim is more analogous to the speedy trial claim which we held unreviewable under the collateral order doctrine in United States v. MacDonald, supra. We think that it particularly fails the third part of the test for Cohen appeals articulated in Coopers & Lybrand, supra, that the claim "be effectively unreviewable on appeal from a final judgment." 9 Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974), on which respondents base the merits of their claim of vindictive prosecution, was an application of the principles announced in North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969), to conduct on the part of a prosecutor. In Perry the defendant had been convicted in a state court of limited jurisdiction, and had exercised his statutory right to a de novo appeal to the state court of general jurisdiction. Prior to the commencement of the latter trial, the prosecutor obtained an indictment charging the defendant with crimes more severe than those for which he was initially convicted. Because the facts suggested "a realistic likelihood of 'vindictiveness,' " 417 U.S., at 27, 94 S.Ct., at 2102, we deemed it necessary to apply the "prophylactic rule of Pearce," id., at 26, 94 S.Ct., at 2102, in order to discourage retaliation by the State for the defendant's exercise of his procedural right. 10 Although there is language in the Perry opinion suggesting that the defendant possessed a "right not to be haled into court at all" upon the more serious charge, id., at 30, 94 S.Ct., at 2104, it is clear that the Court was not using this language to indicate that he was entitled to be free of any retrial whatever. We stated in Perry that "[w]hile the Due Process Clause of the Fourteenth Amendment bars trial of Perry on the felony assault charges in the Superior Court, North Carolina is wholly free to conduct a trial de novo in the Superior Court on the original misdemeanor assault charge." Id., at 31, n. 8, 94 S.Ct., at 2104 n. 8. The defendant in Perry was fully protected by post-conviction relief, leading to a new trial free of the taint of vindictiveness. 11 Obviously, it is wholly desirable to correct prior to trial any substantive errors noticed at that time. It is equally evident that when relief must await postconviction proceedings, the defendant is subjected to the burden of defending himself at trial, even though the presence of errors might require reversal of his conviction and possibly a second trial. Nevertheless, reversal of the conviction and, where the Double Jeopardy Clause does not dictate otherwise, the provision of a new trial free of prejudicial error normally are adequate means of vindicating the constitutional rights of the accused.2 As we noted in United States v. MacDonald, 435 U.S., at 860, n. 7, 98 S.Ct., at 1552, n. 7: 12 "Admittedly, there is value—to all but the most unusual litigant—in triumphing before trial, rather than after it, regardless of the substance of the winning claim. But this truism is not to be confused with the quite distinct proposition that certain claims (because of the substance of the rights entailed, rather than the advantage to a litigant in winning his claim sooner) should be resolved before trial." 13 Even when the vindication of the defendant's rights requires dismissal of charges altogether, the conditions justifying an interlocutory appeal are not necessarily satisfied. In MacDonald, for example, we declined to permit a defendant whose speedy trial motion had been denied before trial to obtain interlocutory appellate review, despite our recognition that "an accused who does successfully establish a speedy trial claim before trial will not be tried." Id., at 861, n. 8, 98 S.Ct., at 1553, n. 8. The nature of the speedy trial right was such that "[p]roceeding with the trial does not cause or compound the deprivation already suffered." Id., at 861, 98 S.Ct., at 1553. This holding reflects the crucial distinction between a right not to be tried and a right whose remedy requires the dismissal of charges. See id., at 860, n. 7, 98 S.Ct., at 1552, n. 7. The former necessarily falls into the category of rights that can be enjoyed only if vindicated prior to trial. The latter does not. 14 The right asserted by respondents is simply not one that must be upheld prior to trial if it is to be enjoyed at all.3 As noted in MacDonald, supra, there is a superficial plausibility to the contention that any claim, particularly a constitutional claim, that would be dispositive of the entire case if decided favorably to a criminal defendant, should be decided as quickly as possible in the course of the litigation. But if such a principle were to be applied, questions as to the constitutionality of the statutes authorizing the prosecution and doubtless numerous other questions would fall under such a definition, and the policy against piecemeal appeals in criminal cases would be swallowed by ever-multiplying exceptions. It is only a narrow group of claims which meet the test of being "effectively unreviewable on appeal from a final judgment," and the claim of prosecutorial vindictiveness is, we hold, not one of them. 15 The petition for certiorari is granted, and the judgment of the Court of Appeals for the Ninth Circuit is reversed, with instructions to that court to dismiss the appeal. 16 It is so ordered. 17 Justice BLACKMUN, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 18 Today, the Court carries its recent penchant for summary decision to a new extreme. The substantial and controversial question raised here—whether an order denying a motion to dismiss based on prosecutorial vindictiveness is appealable before trial—does not, in my view, lend itself to summary treatment. Nevertheless, the Court has decided this important question of appellate jurisdiction without briefing or argument, even though it was not briefed or argued before the District Court or the Court of Appeals and was not the subject of any lower court opinion. I dissent. 19 * The Court, it seems to me, has shown a disturbing tendency of late to dispose of difficult cases by summary per curiam reversals. I must assume that this tendency is prompted, at least in part, by the growing pressures of the Court's calendar and an ill-conceived conviction that we must stay abreast of the increasing workload whatever the costs may be. I regret this pattern, for I think it demeans the Court and its work and surely tends to lessen the quality of its legal product. 20 Summary action is particularly unfortunate in this case, for the Court directs that respondents' appeal be dismissed on an issue that was not raised by the Government until its petition for rehearing in the Court of Appeals. Indeed, for more than a year until the Court of Appeals ruled on the merits in favor of respondents—the Government affirmatively represented to that court that it "ha[d] jurisdiction to review prior to trial a District Court's denial of a defendant's motion to dismiss for vindictive prosecution." United States' Emergency Motion for Summary Affirmance of District Court, reprinted in App. to Brief in Opposition 5a. As a result, the jurisdictional question was not briefed or argued before the Court of Appeals or the District Court. 21 Respondents' opposition to the Government's petition for certiorari understandably focuses on arguments for denying certiorari—in particular, the Government's failure to raise the jurisdictional issue in a more timely fashion. Coupled with the Government's litigation strategy, the Court's summary disposition therefore deprives respondents of their "day in court" in a singularly inappropriate manner. II 22 Additionally, I do not find today's ruling so clearly compelled as to warrant summary treatment, especially when the Solicitor General, contrary to his frequent practice, has not suggested summary reversal in his petition for certiorari. In my view, the issue is important enough to be briefed and argued fully in at least one court. 23 Certainly, the Court's disposition is not mandated by our precedents. As the Court explains, Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949), and Abney v. United States, 431 U.S. 651, 97 S.Ct. 2034, 52 L.Ed.2d 651 (1977), interpret 28 U.S.C. § 1291 as authorizing interlocutory appeals when a three-part standard has been met. The Court fails, however, to mention the first two requirements, presumably because they are satisfied here: that the denial of a defendant's motion to dismiss an indictment on vindictive prosecution grounds "constitute[s] a complete, formal, and, in the trial court, final rejection of a criminal defendant's [vindictive prosecution] claim," 431 U.S., at 659, 97 S.Ct., at 2040, and that it resolves an issue that is "collateral to, and separable from, the principal issue at the accused's impending criminal trial, i.e., whether or not the accused is guilty of the offense charged," ibid. Unlike the speedy trial claim before the Court in United States v. MacDonald, 435 U.S. 850, 98 S.Ct. 1547, 56 L.Ed.2d 18 (1978), which depended on an assessment of the extent to which delay had prejudiced the defense—an assessment that could only be "speculative" prior to trial, id., at 858, 98 S.Ct., at 1551—an allegation that the prosecutor impermissibly increased the charges in response to the defendant's exercise of a legal right may be evaluated before trial, for all the facts relevant to such a claim are fully available. 24 The Court properly suggests that the third requirement set out in Cohen and Abney is the most difficult to apply in this context. See ante, at 267. But, unlike the Court, I find force in respondents' contention that the right they seek to vindicate, in contrast to the right at issue in MacDonald, is "a 'right not to be tried' which must be upheld prior to trial if it is to be enjoyed at all." 435 U.S., at 861, 98 S.Ct., at 1553. In MacDonald, it was "the delay before trial, not the trial itself, that offend[ed] against the constitutional guarantee of a speedy trial." Ibid. When the defendant has been the victim of vindictive prosecution, however, the Court has said that he may "not to be haled into court at all upon the [increased] charge." Blackledge v. Perry, 417 U.S. 21, 30, 94 S.Ct. 2098, 2104, 40 L.Ed.2d 628 (1974) (emphasis added). And we have analogized that right to the right protected by the Double Jeopardy Clause, see id., at 31, which must, of course, be vindicated prior to trial if it is to be protected at all.1 25 Moreover, postconviction review may not suffice to remedy the chilling effect the vindictive prosecution doctrine is designed to prevent. The Court repeatedly has declined to hold that the Due Process Clause forbids only prosecutorial action taken with an actual retaliatory motive. Rather, it has emphasized that " 'since the fear of such vindictiveness may unconstitutionally deter a defendant's exercise of [his] right[s] . . ., due process also requires that a defendant be freed of apprehension of such a retaliatory motivation. . . .' " Id., at 28, 94 S.Ct., at 2102 (emphasis added), quoting North Carolina v. Pearce, 395 U.S. 711, 725, 89 S.Ct. 2072, 2080, 23 L.Ed.2d 656 (1969); see also United States v. Goodwin, 457 U.S. 368, 375-376, 102 S.Ct. 2485, 2490, 73 L.Ed.2d 74 (1982). Even if a defendant is convinced that his vindictive prosecution claim ultimately will prevail on postconviction review, the increased burdens attending a trial on enhanced charges may deter him from exercising his legal rights. This may well not be a case, then, where "reversal of the conviction and . . . a new trial free of prejudicial error" are "adequate means of vindicating the constitutional rights of the accused." Ante, at 268. 26 In addition to extending our precedents, the Court goes well beyond the rulings of the two Courts of Appeals it characterizes as "directly conflict[ing]" with the decision below. Ante, at 264, n. 1. In United States v. Brizendine, 212 U.S.App.D.C. 169, 180, 659 F.2d 215, 226 (1981), the United States Court of Appeals for the District of Columbia Circuit refused to permit interlocutory appeals involving "claims of due process violations arising from the plea bargaining process." That court, however, expressly refused to consider the appealability of the denial of a motion to dismiss "in the Blackledge situation . . .," where the defendant's claim "turned on a purely legal issue . . ." and challenged increased charges filed in response to his exercise of a legal right. Id., at 175, 659 F.2d, at 221. Similarly, in United States v. Gregory, 656 F.2d 1132, 1136 (1981), the United States Court of Appeals for the Fifth Circuit did not foreclose the possibility that some claims of prosecutorial vindictiveness could be the subject of interlocutory appeal.2 27 I cannot conclude that the Court's ruling is compelled by our prior cases or is even consistent with the clear weight of authority in the Courts of Appeals. I would grant the petition for certiorari and set the case for oral argument. 1 The rule in the Ninth Circuit directly conflicts with that adopted by the Courts of Appeals for the District of Columbia and Fifth Circuits. Those courts have held that claims of prosecutorial vindictiveness may not be aired in interlocutory appeals. United States v. Brizendine, 212 U.S.App.D.C. 169, 659 F.2d 215 (1981); United States v. Gregory, 656 F.2d 1132 (CA 5 1981). 2 Noting that "encouragement of delay is fatal to the vindication of the criminal law," this Court has observed that "[b]earing the discomfiture and cost of a prosecution for crime even by an innocent person is one of the painful obligations of citizenship." Cobbledick v. United States, 309 U.S. 323, 325, 60 S.Ct. 540, 541, 84 L.Ed. 783 (1940). Thus, in the run of cases, "[t]he correctness of a trial court's rejection even of a constitutional claim made by the accused in the process of prosecution must await his conviction before its reconsideration by an appellate tribunal." Id., at 325-326, 60 S.Ct., at 541-42. This limitation imposed by Congress must be observed even when the defendant's claim is considered meritorious. See United States v. MacDonald, 435 U.S. 850, 857-858, n. 6, 98 S.Ct. 1547, 1551, n. 6, 56 L.Ed. 2d 18 (1978). 3 By holding that the right asserted by respondents is not one that "will have been lost, probably irreparably," Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 546, 69 S.Ct. 1221, 1225, 93 L.Ed. 1528 (1949), if appeal is postponed until the rendition of judgment, we imply no view on the merit of respondents' claim or on the more general question whether the Due Process Clause is violated by the addition of charges prior to trial. We merely hold that denial of a motion to dismiss an indictment when the motion is premised on a claim of prosecutorial vindictiveness is not a collateral order that can be appealed prior to judgment under § 1291. 1 See Abney v. United States, 431 U.S. 651, 659, 97 S.Ct. 2034, 2040, 52 L.Ed.2d 651 (1977) (noting that the defendant in Blackledge v. Perry was "contesting the very authority of the Government to hale him into court to face trial on the charge against him"); Blackledge v. Perry, 417 U.S., at 30-31, 94 S.Ct., at 2103-2104 (observing that the defendant's vindictive prosecution claim "went to the very power of the State to bring the defendant into court to answer the charge brought against him"; that, when the prosecutor acts vindictively in enhancing charges, the State is "simply precluded by the Due Process Clause from calling upon the respondent to answer to the more serious charge . . ."; that "[t]he very initiation of the proceedings against [a defendant subjected to vindictive prosecution] operated to deny him due process of law"; and that "North Carolina simply could not permissibly require Perry to answer to the felony charge"). 2 Other Courts of Appeals have not restricted interlocutory appeals to the three instances in which this Court has applied the collateral-order doctrine in criminal cases, see ante, at 265-266. See United States v. Venable, 585 F.2d 71, 74-75 (CA3 1978) (denial of motion to bar retrial on collateral estoppel grounds is immediately appealable); United States v. Alessi, 536 F.2d 978, 980-981 (CA2 1976) (denial of motion to dismiss indictment that allegedly violated prior plea bargain may be the subject of an interlocutory appeal).

89

458 U.S. 354 102 S.Ct. 3128 73 L.Ed.2d 819 F. W. WOOLWORTH CO., Appellantv.TAXATION AND REVENUE DEPARTMENT OF the STATE OF NEW MEXICO. No. 80-1745. Argued April 19, 1982. Decided June 29, 1982. Rehearing Denied Oct. 18, 1982. See 459 U.S. 961, 103 S.Ct. 274. Syllabus Appellant's principal place of business and commercial domicile are in New York, but it engages in chainstore retailing throughout the United States. Under its income tax laws, New Mexico distinguishes between "business" income, which it apportions between it and other States, and "nonbusiness" income, which it generally allocates to a single State on the basis of commercial domicile. Appellant reported its dividend income from four of its foreign subsidiaries, which engage in chainstore retailing in foreign countries, as "nonbusiness" income, none of which was to be allocated to New Mexico. Similarly, appellant did not report as New Mexico "business" income a sum, commonly known as "gross-up," that it never actually received from its foreign subsidiaries but that the Federal Government (for purposes of calculating appellant's federal foreign tax credit) deemed it to have received. On audit, appellee determined that appellant should have included in its apportionable New Mexico income both the dividends and the gross-up figure. Appellant's protest was denied, but appellee's decision was reversed by the New Mexico Court of Appeals. However, the New Mexico Supreme Court in turn reversed, holding that both the dividends and the gross-up figure were apportionable New Mexico income. Held : 1. New Mexico's tax on a portion of the dividends received by appellant from its foreign subsidiaries fails to meet established due process standards. Pp. 362-372. (a) The linchpin of apportionability for state income taxation of an interstate enterprise is the "unitary-business principle." Appellant—as owner of all of the stock of three of its subsidiaries and a majority interest in the fourth—potentially has the authority to operate these companies as integrated divisions of a single unitary business. But the potential to operate a company as part of a unitary business is not dispositive when, as here, the dividend income from the subsidiaries in fact is derived from unrelated business activity of the subsidiaries, each of which operates a discrete business enterprise. ASARCO Inc. v. Idaho State Tax Comm'n, 307 U.S. 458, 102 S.Ct. 3103, 73 L.Ed.2d 787; Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 100 S.Ct. 1223, 63 L.Ed.2d 510. P. 362. (b) For due process purposes, the income attributed to a State must be rationally related to values connected with the taxing State. This limitation is not satisfied merely because the nondomiciliary parent corporation derives some economic benefit from its ownership of stock in another corporation. Pp. 363-364. (c) None of the factors relevant to a State's right to tax dividends from foreign subsidiaries exists in this case. The record shows that appellant's and its subsidiaries' operations such as store site selection, advertising, accounting, purchasing, warehousing, and personnel training—were not functionally integrated. And except for the type of occasional oversight—with respect to capital structure, major debt, and dividends—that any parent gives to an investment in a subsidiary, there was little or no integration of business activities or centralization of management. Thus, the subsidiaries were not a part of a "unitary business." Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 100 S.Ct. 2109, 65 L.Ed.2d 66, distinguished. Pp. 364-372. 2. New Mexico's efforts to tax the "gross-up" income also contravenes the Due Process Clause. The "fictitious" gross-up figure is treated for federal foreign tax credit purposes as a dividend in the same manner as a dividend actually received by the domestic corporation from a foreign corporation. In this case the foreign tax credit arose from the taxation by foreign nations of appellant's foreign subsidiaries that had no unitary business relationship with New Mexico. Pp.372-373 95 N.M. 519, 624 P.2d 28, reversed. William L. Goldman, Washington, D. C., for appellant. Sarah E. Bennett, Santa Fe, N. M., for appellee. Justice POWELL delivered the opinion of the Court. 1 The question is whether the Due Process Clause permits New Mexico to tax a portion of dividends that appellant F. W. Woolworth Co. received from foreign subsidiaries that do no business in New Mexico. We also must decide whether New Mexico may include within Woolworth's apportionable New Mexico income a sum, commonly known as "gross-up," that Woolworth calculated in order to claim a foreign tax credit on its federal income tax. 2 * Woolworth's principal place of business and commercial domicile are in New York. It engages in retail business through chains of stores located in the United States, Puerto Rico, and the Virgin Islands. It sells a wide spectrum of merchandise, including dry goods, hardware, small appliances, confections, packaged goods, and fountain items. In the fiscal year ending January 31, 1977, Woolworth's gross domestic sales totalled approximately $2.5 billion, with New Mexico sales amounting to approximately $13 million—or about 0.5% of the gross figure. App. 57a. 3 Woolworth owns four foreign subsidiaries of relevance to this suit. Three are wholly owned: F. W. Woolworth GmbH, in Germany; F. W. Woolworth, Ltd., in Canada; and F. W. Woolworth, S. A. de C. V. Mexico. F. W. Woolworth Co., Ltd., is an English corporation of which Woolworth owns 52.7%, with the remainder held and traded publicly. These four corporations also engage in chainstore retailing.1 Together they paid Woolworth approximately $39.9 million in dividends during the fiscal year in question. 4 New Mexico adopted a version of the Uniform Division of Income for Tax Purposes Act in 1965, N.M.Stat.Ann. §§ 7-4-1—7-4-21 (1981), and joined the Multistate Tax Compact in 1967. §§ 7-5-1 7-5-7 (1981). See ASARCO Inc. v. Idaho State Tax Comm'n, 458 U.S. 307, 311-312, 102 S.Ct. 3103, 3107, 73 L.Ed.2d 787; United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S. 452, 98 S.Ct. 799, 54 L.Ed.2d 682 (1978). Consequently the State distinguishes between "business" income,2 which it apportions between it and other States for tax purposes,3 and "nonbusiness" income,4 which it generally allocates to a single State on the basis of commercial domicile.5 Woolworth reported its dividend income of $39.9 million from its German, Canadian, Mexican, and English subsidiaries as "nonbusiness" income, none of which was to be allocated to New Mexico. Woolworth also treated as "nonbusiness" income a $1.6 million gain from a hedging transaction in British pounds. This transaction was undertaken for the purpose of insuring the payment of the British subsidiary's dividend against currency fluctuations. See App. 52a-54a. Similarly, Woolworth did not report as New Mexico "business" income $25.5 million of "gross-up" that it never actually received but that the Federal Government (for purposes of calculating Woolworth's federal foreign tax credit pursuant to 26 U.S.C. §§ 78, 901(a), and 902(a)) deemed Woolworth to have received from its foreign subsidiaries.6 5 On audit, the New Mexico Taxation and Revenue Department determined that, under state law, Woolworth should have included in its apportionable New Mexico income the dividends from its four foreign subsidiaries, the foreign exchange gain, and the $25.5 million gross-up figure. These additions increased Woolworth's apportioned New Mexico income from $84,622 to $401,518. App. 69a. The Department denied Woolworth's protest,7 but this decision was reversed on appeal by the New Mexico Court of Appeals. F. W. Woolworth Co. v. Bureau of Revenue, 95 N.M. 542, 624 P.2d 51 (1979). 6 As a matter of state law, the Court of Appeals excluded from apportionable New Mexico income Woolworth's receipt of the dividends at issue. The court stated that "[t]here is no indication that the income from Woolworth's long-standing investments [in its subsidiaries] was used either in taxpayer's unitary domestic business or in its business conducted in New Mexico. . . ." Id., at 545, 624 P.2d, at 54. With respect to the gross-up issue, the Court of Appeals said that the State's "rigid insistence" on inclusion of this amount "is a refusal to recognize an obviously fictitious income figure, made artificial by the federal reporting requirements for a specific purpose. . . ." Id., at 543-544, 624 P.2d, at 52-53. The court said that " '[g]ross-up' in fact represents income to taxpayer's foreign subsidiaries [that] is paid out in taxes to foreign governments," id., at 544, 624 P.2d, at 53, and not income in fact to the parent. The court thus likewise excluded this sum from Woolworth's apportionable New Mexico income.8 7 The New Mexico Supreme Court reversed over one dissent. 95 N.M. 519, 624 P.2d 28 (1981). On the question whether Woolworth's receipt of dividends from its subsidiaries constituted apportionable New Mexico income, the court observed that, "[r]egrettably, it needs to be said that the State did a very poor job of inquiring into and developing the facts in this case." Id., at 524, 624 P.2d, at 33. The court nonetheless found substantial evidence to support the findings that the subsidiaries' dividend payments met the State's statutory test for inclusion in Woolworth's apportionable New Mexico income. On the constitutional issue, the court identified the "key question" after our decision in Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 100 S.Ct. 1223, 63 L.Ed.2d 510 (1980), as "whether those dividends were income earned in a unitary business." 95 N.M., at 528, 624 P.2d, at 37. The court stated: 8 "The [dividend] income [from Woolworth's subsidiaries] is obviously related to the mutual activities of the parent and its affiliates. The control over the subsidiaries, the interdependence, the history of the relationships, the placing of the [dividend] money in [Woolworth's] general operating account, all point to functional integration and reveal an underlying unitary business for our purposes here." Id., at 529, 624 P.2d, at 38. 9 Respecting the State's inclusion of Woolworth's federal gross-up figure as apportionable state income, the court "deem[ed] it unnecessary to delve into all the intricacies of the federal laws and regulations," but found it sufficient "to say that, since Woolworth decided to use the gross-up option, the income taxes paid by Woolworth's foreign subsidiaries to foreign governments must be deemed to be received as dividends . . . ." Id., at 521-522, 624 P.2d, at 30-31. "Admittedly, the fictitious gross-up, which the state claims is 'business income' and which Woolworth deliberately acceded to, does not fit the ordinary definition of 'income' . . . ." Id., at 522, 624 P.2d, at 31. Nevertheless, the court noted that there was no claim and no lower court finding that Woolworth did not "obtain an economic benefit from the gross-up procedure here." Id., at 523, 624 P.2d, at 32. The court consequently rejected Woolworth's statutory and constitutional challenges to the State's inclusion of the federal gross-up figure in Woolworth's apportionable New Mexico business income.9 II 10 This case was argued in tandem with ASARCO Inc. v. Idaho State Tax Comm'n, 458 U.S. 307, 102 S.Ct. 3103, 73 L.Ed.2d 787 which also involved dividends and gains from foreign subsidiaries. We have reiterated today in ASARCO that " '[t]he "linchpin of apportionability" for state income taxation of an interstate enterprise is the "unitary-business principle." ' " 458 U.S., at 319, 102 S.Ct., at 3110, quoting Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 223, 100 S.Ct. 2109, 2120, 65 L.Ed.2d 66 (1980), in turn quoting Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, at 439, 100 S.Ct., at 1232. 11 Woolworth owns all the stock of three of its dividend payors and a 52.7% majority interest in the fourth. As a result, Woolworth (at least with respect to the three wholly owned companies) elects all of the subsidiaries' directors. It potentially has the authority to operate these companies as integrated divisions of a single unitary business. Our decision in ASARCO makes clear, however, that the potential to operate a company as part of a unitary business is not dispositive when, looking at "the 'underlying economic realities of a unitary business,' " the dividend income from the subsidiaries in fact is "derive[d] from 'unrelated business activity' which constitutes a 'discrete business enterprise.' " Exxon, supra, at 223-224, 100 S.Ct., at 2120, quoting Mobil, supra, at 441, 442, 439, 100 S.Ct., at 1232, 1233, 1234. See ASARCO, 458 U.S., at 322-323, 102 S.Ct., at 3112-3113 (holding that a 52.7%-owned subsidiary is not part of its parent's unitary business). 12 * The State Supreme Court in important part analyzed this case under a different legal standard. After stating that the existence of a unitary business relationship was the "key question," the court proceeded to resolve this question largely by emphasizing the potentials of the relationship between Woolworth and its subsidiaries: 13 "The possession of large assets by subsidiaries is a business advantage of great value to the parent; 'it may give credit which will result in more economical business methods; it may give a standing which shall facilitate purchases; it may enable the corporation to enlarge the field of its activities and in many ways give it business standing and prestige.' Flint v. Stone Tracy Co., 220 U.S. 107, 166 [31 S.Ct. 342, 355, 55 L.Ed. 389] (1911)." 95 N.M., at 529, 624 P.2d, at 38. 14 This reliance on the Flint case was error. Flint upheld a federal excise tax levied on corporate income.10 The States, of course, are subject to limitations on their taxation powers that do not apply to the Federal Government. As relevant here, "the income attributed to [a] State for tax purposes must be rationally related to 'values connected with the taxing State.' Norfolk & Western R. Co. v. State Tax Comm'n, 390 U.S. 317, 325 [88 S.Ct. 995, 1000, 19 L.Ed.2d 1201]." Moorman Mfg. Co. v. Bair, 437 U.S. 267, 273, 98 S.Ct. 2340, 2344, 57 L.Ed.2d 197 (1978). The state court's reasoning would trivialize this due process limitation by holding it satisfied if the income in question "adds to the riches of the corporation . . . ." Wallace v. Hines, 253 U.S. 66, 70, 40 S.Ct. 435, 437, 64 L.Ed. 782 (1920). Income, from whatever source, always is a "business advantage" to a corporation. Our cases demand more. In particular, they specify that the proper inquiry looks to "the underlying unity or diversity of business enterprise," Mobil, supra, at 440, 100 S.Ct., at 1233, not to whether the nondomiciliary parent derives some economic benefit—as it virtually always will—from its ownership of stock in another corporation. See ASARCO, 458 U.S., at 325-329, 102 S.Ct., at 3114-3115.11 B 15 In Mobil we emphasized, as relevant to the right of a State to tax dividends from foreign subsidiaries, the question whether "contributions to income [of the subsidiaries] result[ed] from functional integration, centralization of management, and economies of scale." 445 U.S., at 438, 100 S.Ct., at 1232. If such "factors of profitability" arising "from the operation of the business as a whole" exist and evidence the operation of a unitary business, a State can gain a justification for its tax consideration of value that has no other connection with that State. Ibid. We turn now to consider the extent, if any, to which these factors exist in this case. 16 There was little functional integration. Woolworth's subsidiaries engaged exclusively in the business of retailing—the purchase of wholesale goods for resale to final consumers. This type of business differs significantly from the "highly integrated business" of locating, processing, and marketing a resource (such as petroleum) that we previously have found to constitute a unitary business. Exxon, 447 U.S., at 224, 100 S.Ct., at 2120. See also id., at 226, 100 S.Ct., at 2121 (describing "a unitary stream of income, of which the income derived from internal transfers of raw materials from exploration and production to refining is a part"); Mobil, 445 U.S., at 428, 100 S.Ct., at 1227. Consistent with this distinction, the evidence in this case is that no phase of any subsidiary's business was integrated with the parent's. With respect to "who makes the decision for seeing to the merchandise, [store] site selection, advertising and accounting control," the undisputed testimony stated that "[e]ach subsidiary performs these functions autonomously and independently of the parent company." App. 12a.12 "Each subsidiary has a complete accounting department and a financial staff." Id., at 14a. Each had its own outside counsel. App. to Juris. Statement 34a. It further appears that Woolworth engaged in no centralized purchasing, manufacturing, or warehousing of merchandise.13 The parent had no central personnel training school for its foreign subsidiaries. Ibid. And each subsidiary was responsible for obtaining its own financing from sources other than the parent.14 In sum, the record is persuasive that Woolworth's operations were not functionally integrated with its subsidiaries. 17 We now consider the extent to which there was centralization of management or achievement of other economies of scale. It appears that each subsidiary operated as a distinct business enterprise at the level of fulltime management. With one possible exception,15 none of the subsidiaries' officers during the year in question was a current or former employee of the parent. Ibid. The testimony was that the subsidiaries "figure that their operations are independent, autonomous." App. 13a. Woolworth did not "rotate personnel or train personnel to operate stores in those countries. There is no exchange of personnel." Ibid. There was no "training program that is central to transmit the Woolworth idea of merchandising[,] such as it may be[,] to the foreign subsidiaries." Id., at 15a. The subsidiaries "proceed . . . with their own programs, either formal or informal. They develop their own managers and instruct them in their methods of operation." Ibid. 18 This management decentralization was reflected in the fact that each subsidiary possessed autonomy to determine its own policies respecting its primary activity—retailing. According to the hearing examiner: 19 "Each of the four subsidiaries are responsible for determining the size and location of retail stores, the market conditions in their own territory and the mix of items to be sold. The German subsidiary emphasizes soft goods such as dresses and coats. It sells no food. The English subsidiary operates restaurants in its stores and also operates supermarkets. Each subsidiary attempts to cater to local tastes and needs. The inventory of each subsidiary consists, in large part, of home country produced items. This purchase-at-home practice is consistent with the policy of the taxpayer. A number of inventory items are purchased from the Orient or other places but there is no evidence that the subsidiaries purchase, or are required to purchase, inventory items from any particular source." App. to Juris. Statement 33a-34a. 20 Importantly, the Department's hearing examiner found that Woolworth had "no department or section, as such, devoted to overseeing the foreign subsidiary operations." Id., at 34a.16 Neither the parent corporation nor any of the subsidiaries consolidates its tax return with any of the other companies. App. 37a-38a. The tax manager for Woolworth stated that he did not review the subsidiaries' tax returns or consult with them on decisions affecting taxes. Id., at 14a. There was no "policy of the parent that all of the managers of all the operations get together periodically to discuss the overall Woolworth operations." Id., at 35a.17 21 There were some managerial links. Woolworth maintained one or several common directors with some of the subsidiaries.18 There also was irregular in-person19 and "frequent" mail, telephone, and teletype communication between the upper echelons of management of the parent and the subsidiaries.20 App. to Juris. Statement 34a. Decisions about major financial decisions, such as the amount of dividends to be paid by the subsidiaries and the creation of substantial debt, had to be approved by the parent.21 Id., at 35a. Woolworth's published financial statements, such as its annual reports, were prepared on a consolidated basis.22 Ibid. 22 We conclude, on the basis of undisputed facts, that the four subsidiaries in question are not a part of a unitary business under the principles articulated in Mobil and Exxon, and today reiterated in ASARCO. Except for the type of occasional oversight with respect to capital structure, major debt, and dividends—that any parent gives to an investment in a subsidiary, there is little or no integration of the business activities or centralization of the management of these five corporations. Woolworth has proved that its situation differs from that in Exxon, where the corporation's Coordination and Services Management office was found to provide for the asserted unitary business 23 "long-range planning for the company, maximization of overall company operations, development of financial policy and procedures, financing of corporate activities, maintenance of the accounting system, legal advice, public relations, labor relations, purchase and sale of raw crude oil and raw materials, and coordination between the refining and other operating functions 'so as to obtain an optimum short range operating program.' " 447 U.S., at 211, 100 S.Ct., at 2114. 24 In this case the parent company's operations are not interrelated with those of its subsidiaries so that one's "stable" operation is important to the other's "full utilization" of capacity. Id., at 218, 100 S.Ct., at 2117. See also id., at 225, 100 S.Ct., at 2121. The Woolworth parent did not provide "many essential corporate services" for the subsidiaries, and there was no "centralized purchasing office . . . whose obvious purpose was to increase overall corporate profits through bulk purchases and efficient allocation of supplies among retailers." Id., at 224, 100 S.Ct., at 2120.23 And it was not the case that "sales were facilitated through the use of a uniform credit card system, uniform packaging, brand names, and promotional displays, all run from the national headquarters." Ibid. See also Mobil, 445 U.S., at 428, 435, 100 S.Ct., at 1227, 1230.24 25 There is a critical distinction between a retail merchandising business as conducted by Woolworth and the type of multinational business—now so familiar—in which refined, processed, or manufactured products (or parts thereof) may be produced in one or more countries and marketed in various countries, often worldwide.25 In operations of this character there is a flow of international trade, often an interchange of personnel, and substantial mutual interdependence. The uncontradicted evidence demonstrates that Woolworth's international retail business is not comparable. There is no flow of international business. Nor is there any integration or unitary operation in the sense in which our cases consistently have used these terms. In Mobil, we recognized: 26 "[A]ll dividend income received by corporations operating in interstate commerce is [not] necessarily taxable in each State where that corporation does business. Where the business activities of the dividend payor have nothing to do with the activities of the recipient in the taxing State, due process considerations might well preclude apportionability, because there would be no underlying unitary business." Id., at 441-442, 100 S.Ct., at 1233-1234. 27 This is such a case. Each of the foreign subsidiaries at issue operates a "discrete business enterprise," Mobil, supra, at 439, 100 S.Ct., at 1232, with a notable absence of any "umbrella of centralized management and controlled interaction." Exxon, 447 U.S., at 224, 100 S.Ct., at 2120. New Mexico, in taxing a portion of dividends received from such enterprises, is attempting to reach "extraterritorial values," Mobil, supra, at 442, 100 S.Ct., at 1234, wholly unrelated to the business of the Woolworth stores in New Mexico. As a result, a "showing has been made that income unconnected with the unitary business has been used in the" levy of the New Mexico tax. Butler Bros. v. McColgan, 315 U.S. 501, 509, 62 S.Ct. 701, 705, 86 L.Ed. 991 (1942). We conclude that this tax does not bear the necessary relationship " 'to opportunities, benefits, or protection conferred or afforded by the taxing State. See Wisconsin v. J. C. Penney Co., 311 U.S. 435, 444 [61 S.Ct. 246, 249, 85 L.Ed. 267].' " Norfolk & Western R. Co. v. Missouri State Tax Comm'n, 390 U.S. 317, 325, n. 5, 88 S.Ct. 995, 1000, n. 5, 19 L.Ed.2d 1201 (1968), quoting Ott v. Mississippi Valley Barge Line Co., 336 U.S. 169, 174, 69 S.Ct. 432, 434, 93 L.Ed. 585 (1949). New Mexico's tax thus fails to meet established due process standards. III 28 We need not be detained by New Mexico's reaching out to tax "gross-up" amounts that even the Supreme Court of New Mexico recognized as "fictitious." 95 N.M., at 522, 624 P.2d, at 31. The gross-up computation is a figure that the Federal Government "deems" Woolworth to have received for purposes of part of Woolworth's federal foreign tax credit calculation. It "is treated [for this purpose] as a dividend in the same manner as a dividend actually received by the domestic corporation from a foreign corporation." H.R.Rep.No. 1447, 87th Cong., 2d Sess., A83 (1962). See also S.Rep.No. 1881, 87th Cong., 2d Sess., 227 (1962). In this case the foreign tax credit arose from the taxation by foreign nations of Woolworth foreign subsidiaries that had no unitary business relationship with New Mexico. New Mexico's effort to tax this income "deemed received"—with respect to which New Mexico contributed nothing—also must be held to contravene the Due Process Clause.26 IV 29 The judgment of the Supreme Court of New Mexico is reversed. 30 It is so ordered. 31 Justice O'CONNOR, with whom Justice BLACKMUN and Justice REHNQUIST join, dissenting. 32 The $39.9 million in dividend income at issue in this case was earned by four foreign subsidiaries of F. W. Woolworth Co.: F. W. Woolworth GmbH (Germany), F. W. Woolworth, Ltd. (Canada), F. W. Woolworth, S. A. de C. V. Mexico (Mexico), and F. W. Woolworth Co., Ltd. (England). F. W. Woolworth Co. wholly owned its German, Canadian, and Mexican subsidiaries, and had a 52.7% interest in its English subsidiary. During the tax year in question the subsidiaries apparently operated somewhat autonomously in their respective markets, but "mail, telephone, and teletype communication between the upper echelons of management of the parent and the subsidiaries" was " 'frequent.' " Ante, at 368 (footnote omitted) (quoting App. to Juris. Statement 34a). Moreover, "[d]ecisions about major financial decisions, such as the amount of dividends to be paid by the subsidiaries and the creation of substantial debt, had to be approved by the parent," and "Woolworth's published financial statements, such as its annual reports, were prepared on a consolidated basis." Ante, at 368-369 (citations and footnotes omitted). 33 These controlled subsidiaries, operating in geographically diverse markets in the same line of business as F. W. Woolworth itself, were simply not "unrelated,"1 "discrete business enterprise[s],"2 "hav[ing] nothing to do with the activities"3 of F. W. Woolworth in New Mexico. Because I disagree with the redefinition of the limits of a unitary business adopted today by the Court, and for the reasons expressed in my dissent in ASARCO Inc. v. Idaho State Tax Comm'n, 458 U.S. 307, 331, 102 S.Ct. 3103, 3117, 73 L.Ed.2d 787, which was argued in tandem with this case, I respectfully dissent. 1 The English subsidiary operates about 2,000 stores, App. 39a, the Canadian company about 500, id., at 24a, and the Mexican about 12. Id., at 28a. The record does not specify the number of stores the German company owns, but the company may be between the English and Canadian operations in size. 2 " '[B]usiness income' means income arising from transactions and activity in the regular course of the taxpayer's trade or business and includes income from tangible and intangible property if the acquisition, management and disposition of the property constitute integral parts of the taxpayer's regular trade or business operations . . . ." N.M.Stat.Ann. § 7-4-2(A) (1981). 3 "All business income shall be apportioned to this state by multiplying the income by a fraction, the numerator of which is the property factor plus the payroll factor plus the sales factor, and the denominator of which is three." N.M.Stat.Ann. § 7-4-10 (1981). "The property factor is a fraction, the numerator of which is the average value of the taxpayer's real and tangible personal property owned or rented and used in this state during the tax period and the denominator of which is the average value of all the taxpayer's real and tangible personal property owned or rented and used during the tax period." § 7-4-11. "The payroll factor is a fraction, the numerator of which is the total amount paid in this state during the tax period by the taxpayer for compensation, and the denominator of which is the total compensation paid everywhere during the tax period." § 7-4-14. "The sales factor is a fraction, the numerator of which is the total sales of the taxpayer in the state during the tax period, and the denominator of which is the total sales of the taxpayer everywhere during the tax period." § 7-4-16. 4 " '[N]onbusiness income' means all income other than business income." N.M.Stat.Ann. § 7-4-2(D) (1981). 5 "Rents and royalties from real or tangible personal property, capital gains, interest, dividends, or patent or copyright royalties, to the extent that they constitute nonbusiness income, shall be allocated as provided in Sections 6 through 9[7-4-6 to 7-4-9 NMSA 1978] of the Uniform Division of Income for Tax Purposes Act." N.M.Stat.Ann. § 7-4-5 (1981) (emphasis added). "Interest and dividends are allocable to this state if the taxpayer's commercial domicile is in this state." § 7-4-8. New Mexico defines "commercial domicile" as "the principal place from which the trade or business of the taxpayer is directed or managed." § 7-4-2(B). 6 "If a domestic corporation chooses to have the benefits of subpart A of part III of subchapter N (relating to foreign tax credit) for any taxable year, an amount equal to the taxes deemed to be paid by such corporation under section 902(a) (relating to credit for corporate stockholder in foreign corporation) . . . for such taxable year shall be treated for purposes of this title . . . as a dividend received by such domestic corporation from the foreign corporation." 26 U.S.C. § 78 (emphasis added). "If the taxpayer chooses to have the benefits of this subpart, the tax imposed by this chapter shall . . . be credited with the . . . taxes deemed to have been paid under sectio[n] 902. . . ." § 901(a). "For purposes of this subpart, a domestic corporation which owns at least 10 percent of the voting stock of a foreign corporation from which it receives dividends in any taxable year shall be deemed to have paid the same proportion of any income . . . taxes paid or deemed to be paid by such foreign corporation to any foreign country . . . on . . . the accumulated profits of such foreign corporation from which such dividends were paid, which the amount of such dividends (determined without regard to section 78) bears to the amount of such accumulated profits in excess of such income . . . taxes (other than those deemed paid)." § 902(a). Woolworth gives this example: "If a foreign subsidiary of a United States parent earns $100, pays foreign tax of $40, and pays a dividend of $30 out of its after-tax profits of $60, the deemed paid foreign tax credit of the parent under section 902(a) is 30/60 x $40, or $20. The parent includes $50 in dividend income (i.e., the actual dividend of $30 plus $20 of 'gross-up') and claims a foreign tax credit of $20 against the federal income tax on this income." Brief for Appellant 6, n. 4. See 26 CFR §§ 1.78-1, 1.902-1(h), (k) (1982); S.Rep.No. 1881, 87th Cong., 2d Sess., 222-228 (1962), U.S.Code Cong. & Admin.News 1962, p. 3304; H.R.Rep.No. 1447, 87th Cong., 2d Sess., A79-A84 (1962); 3 B. Bittker, Federal Taxation of Income, Estates and Gifts ¶ 69.2 (1981). 7 Only one witness—Woolworth's tax manager—appeared before the Department's hearing examiner. The State introduced as evidence Woolworth's tax return, a notice of assessment, its worksheets, Woolworth's protest, and tax regulations. Woolworth introduced a one-page diagram of its corporate structure. See App. B to Brief for Appellee. The testimony given before the examiner and referred to in this opinion is uncontroverted unless otherwise noted. The Department's decision and order did not mention Woolworth's foreign exchange gain. See App. to Juris. Statement 30a-38a. 8 The Court of Appeals did not refer to Woolworth's foreign exchange gain. 9 The New Mexico Supreme Court also ruled that the Court of Appeals had erred in holding that it was reasonable to conclude that Woolworth had simply passed the foreign dividends through its general treasury to its stockholders, without using the dividends for Woolworth's general corporate purposes. The Supreme Court ruled that the burden of proof was on the taxpayer. Consequently, the court held, the State's reasonable inference that Woolworth had used the foreign dividends for its own corporate purposes was supportable in the absence of evidence to the contrary. 95 N.M., at 529, 624 P.2d, at 38. The court further rejected Woolworth's claim that the apportionment formula should be adjusted if the dividend income were found to be apportionable. Id., at 529-530, 624 P.2d, at 38-39. The dissenting justice founded his position on "agree[ment] with the analysis of the Court of Appeals." Id., at 530, 624 P.2d, at 39. Neither the majority nor the dissenting opinion discussed Woolworth's foreign exchange gain. 10 The tax did not apply to a corporation's receipt of dividends from other companies subjected to the tax. 220 U.S., at 144-145, 31 S.Ct., at 346. 11 The hearing examiner and the New Mexico Supreme Court also thought it significant that Woolworth had commingled its dividends with its general funds and had used them for general corporate operating purposes. See 95 N.M., at 529, 624 P.2d, at 38; n. 9, supra. This analysis likewise subverts the unitary-business limitation. All dividend income—irrespective of whether it is generated by a "discrete business enterprise," Mobil, 445 U.S., at 439, 100 S.Ct., at 1232—would become part of a unitary business if the test were whether the corporation commingled dividends from other corporations, whether subsidiaries or not. 12 The testimony before the Department's hearing examiner, see n. 7, supra, focused primarily on the English and German subsidiaries. With respect to the Mexican and Canadian subsidiaries, the evidence was confined to the following: "Q. Now, I would like to[,] without repeating every question if I may[,] ask a summary question concerning the Canadian and the Mexican subsidiaries, we are talking about decisions concerning merchandise mix, site selection, advertising, accounting, training of personnel, and of those items[,] would you say that there is a similarity between the relationship of the U. S. parent to Canada as there is to the German and the English subsidiaries to the extent to which these things are decentralized? "A. Yes, there is a distinct similarity or philosophy involved in the ownership of these companies." App. 18a. The State did not undertake to controvert the implications of this statement, and neither of the courts below found any difference in the relationships between the parent and each of the four subsidiaries. We thus must assume that the relationship between the parent and the Mexican and Canadian subsidiaries paralleled that between the parent and the English and German subsidiaries in material respects. 13 The New Mexico Supreme Court did state that "[t]here is some flow back and forth of goods" between the parent and the subsidiaries. 95 N.M., at 524, 624 P.2d, at 33. It cited no evidence in support of this statement. Neither the Department's hearing examiner nor the Court of Appeals made such a finding. The testimony in the record was that there were not "any inter-company sales of inventory." App. 13a. The Woolworth witness also stated that, with respect to certain types of goods manufactured by other of the parent's subsidiaries, he lacked actual knowledge of whether there were intercompany sales. He continued that the idea was "inconceivable to me because of the autonomous operation of the company and the lack of coordination and other facets." Id., at 43a. Upon further questioning, the witness conceded that he did not "really know where [the English subsidiary's managers] buy their merchandise." But he affirmed his knowledge that the English company's sales to and purchases from the parent were "virtually nil." Id., at 44a. When questioned whether Woolworth utilized a central buying office for it and its subsidiaries, the witness replied that it did not. Id., at 14a. No other evidence indicated that the parent and the subsidiaries engaged in any joint manufacturing, purchasing, or warehousing functions. Nor did the New Mexico courts find otherwise. 14 Woolworth had no outstanding debts from its English subsidiary, App. to Juris. Statement 35a, and it had not reinvested its dividends in that company. App. 51a. The parent had reinvested dividends in the German company, id., at 52a, but the last additional capital contribution by the parent that the witness could recall, id., at 46a, was a $400,000 transfer made after the German company was demolished during World War II. Id., at 30a. See App. to Juris. Statement 35a. 15 The hearing examiner found that "[i]n the taxable year involved, none of the four subsidiarie[s'] officers were currently or formerly employees of the parent." Id., at 34a. Without explanation, he also later stated that "[a]t least one officer of the Canadian subsidiary [was] also an officer of the [parent]." Ibid. One officer from the Mexican subsidiary was a participant in Woolworth's profit-sharing plan. Woolworth paid the employee's share of this plan. Ibid. 16 Woolworth had one vice president "who is the liaison man with the smaller foreign subsidiaries," ibid., such as the Spanish and Mexican subsidiaries. App. 50a. The testimony was that this liaison man "from time to time . . . may have contact with the major subsidiaries . . . ." Ibid. 17 The witness replied that "[t]here hasn't been that" in response to the questions of whether "all of the managers of all of the operations" ever "get together and talk together about where the company is going or what it is doing or where it should be tomorrow or in ten years from today? Planning for future programs or for future expansion?" Id., at 35a. 18 The managing director of the English subsidiary sat as one of the parent's 15 directors, and the parent sent one of its officers to participate in meetings of the English board of directors. The state hearing examiner also found that "[a]t least one person who is on the Canadian subsidiary is also on the taxpayer's board." App. to Juris. Statement 34a. Cf. App. 41a (at least three members of the Canadian board of directors also sat on the parent's board). Although the Department's hearing examiner did not make findings in this connection, there was testimony that "some of the directors of Woolworth, the taxpayer, sit on the Board of the Mexican subsidiary . . . ." Id., at 29a. 19 "It would be on a rare occasion, once a year," that the "managing directors" of the foreign subsidiaries would come to New York. Id., at 34a. "It is only sporadically that the parent company is represented at the [English company's] Board meetings." Id., at 40a. Further, while Woolworth's chief executive officer and other officers would "on occasion" travel to confer with the subsidiaries' managers, it was "hard to discern any pattern of regular monthly visits or quarterly visits." Id., at 35 a. 20 "The exchange of information contacts by the subsidiaries is made by the Chief Executive of the company. The Director of Purchases of the parent company does not confer with the Director of Purchasing of the Canadian company or of the German company." Id., at 37a. 21 The testimony was that "the Canadian financial people have the right to finance their day-to-day operations and I doubt that they would be required . . . at the level of a million dollars to seek the permission of the parent company. If there was really substantial borrowings going on, I am sure the Canadians before borrowing . . . would obtain permission of the parent company." Id., at 26a. More generally, the witness stated that "[n]ormally, I would think that substantial borrowing, an unusual amount, would necessitate the checking with the principal stockholders or the Board of Directors." Id., at 28a. 22 The English subsidiary was not included in the consolidated statements. App. to Juris. Statement 35a. Cf. Keesling & Warren, The Unitary Concept in the Allocation of Income, 12 Hastings L.J. 42, 52 (1960) ("Central accounting, for instance, may result in some savings, but in most instances the amount is trifling in comparison with the income [involved]. Alone considered, it is too weak a connecting link to bind into one business, what would otherwise, from an operational standpoint, be considered separate businesses"). As noted, there was no centralized tax department and no consolidation of tax returns. In addition to the links set forth in the text, it is plain that the parent and the four subsidiaries all utilize the same general "F. W. Woolworth" corporate name. There is no record information on the significance of the use of this common name. Neither the Department nor the New Mexico Supreme Court gave any weight to use of a common corporate name when sustaining the tax at issue. 23 Cf. Butler Bros. v. McColgan, 315 U.S. 501, 508, 62 S.Ct. 701, 704, 86 L.Ed. 991 (1942). 24 In Mobil, the Court relied upon Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271, 45 S.Ct. 82, 69 L.Ed. 282 (1924): "A British corporation manufactured ale in Great Britain and sold some of it in New York. The corporation objected on due process grounds to New York's imposition of an apportioned franchise tax on the corporation's net income. The Court sustained the tax on the strength of its earlier decision in Underwood Typewriter Co. v. Chamberlain, [254 U.S. 113, 41 S.Ct. 45, 65 L.Ed. 165 (1920) ], where it had upheld a similar tax as applied to a business operating in several of our States. It ruled that the brewer carried on a unitary business, involving 'a series of transactions beginning with the manufacture in England and ending in sales in New York and other places'. . . . 266 U.S., at 282, 45 S.Ct., at 84." Mobil, 445 U.S., at 438, 100 S.Ct., at 1232. There is no comparable "series of transactions" in this case linking the retail merchandise business of Woolworth's foreign subsidiaries with its business in New Mexico. 25 See Hans Rees' Sons v. North Carolina ex rel. Maxwell, 283 U.S. 123, 133, 51 S.Ct. 385, 388, 75 L.Ed. 879 (1931) ("Undoubtedly, the enterprise of a corporation which manufactures and sells its manufactured product is ordinarily a unitary business, and all the factors in that enterprise are essential to the realization of profits"); Hellerstein, Recent Developments in State Tax Apportionment and the Circumscription of Unitary Business, 21 Nat.Tax J. 487, 496 (1968) ("Manufacturing or purchasing goods in one state and selling in another, and transportation and communication between the states are typical of cases considered unitary"); G. Altman & F. Keesling, Allocation of Income in State Taxation 101-102 (1946). 26 Woolworth challenges only New Mexico's tax treatment of its dividend and gross-up income. See Juris. Statement i; Brief for Appellant i. We therefore do not consider New Mexico's tax treatment of other items of Woolworth income. In particular, we do not pass upon the proper treatment of Woolworth's foreign exchange gain—a matter that was not considered below. See nn. 7-9, supra. 1 Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 439, 100 S.Ct. 1223, 1232, 63 L.Ed.2d 510 (1980). 2 Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 224, 100 S.Ct. 2109, 2120, 65 L.Ed.2d 66 (1980). 3 Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, 445 U.S., at 442, 100 S.Ct., at 1234.

78

458 U.S. 307 102 S.Ct. 3103 73 L.Ed.2d 787 ASARCO INCORPORATED, etc., Appellant,v.IDAHO STATE TAX COMMISSION. No. 80-2015. Argued April 19, 1982. Decided June 29, 1982. Rehearing Denied Oct. 18, 1982. See 459 U.S. 961, 103 S.Ct. 275. Syllabus Held : The State of Idaho may not constitutionally include within the taxable income of appellant nondomiciliary parent corporation doing some business (primarily silver mining) in the State a portion of intangible income (dividends, interest payments, and capital gains from the sale of stock) that appellant received from subsidiary corporations having no other connection with the State. Pp. 315-330. (a) As a general principle, a State may not tax value earned outside its borders. "[T]he linchpin of apportionability in the field of state income taxation is the unitary-business principle." Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 439, 100 S.Ct. 1223, 1232, 63 L.Ed.2d 510; Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 223, 100 S.Ct. 2109, 2120, 65 L.Ed.2d 66. Pp. 315-320. (b) Here, based on the findings in the state trial court and the undisputed facts, appellant succeeded in proving that no unitary business relationship existed between appellant and its subsidiaries. Pp. 320-324. (c) To have, as Idaho proposes, corporate purpose define unitary business—i.e., to consider intangible income as part of a unitary business if the intangible property (shares of stock) is "acquired, managed or disposed of for purposes relating or contributing to the taxpayer's business"—would destroy the concept of unitary business. Such a definition, which would permit nondomiciliary States to apportion and tax dividends "[w]here the business activities of the dividend payor have nothing to do with the activities of the recipient in the taxing State," Mobil Oil Corp., supra, 445 U.S., at 442, 100 S.Ct., at 1234, cannot be accepted consistently with recognized due process standards. While the dividend-paying subsidiaries in this case "ad[d] to the riches" of appellant, Wallace v. Hines, 253 U.S. 66, 70, 40 S.Ct. 435, 437, 64 L.Ed. 782 (1920), they are "discrete business enterprise[s]" that in "any business or economic sense" have "nothing to do with the activities" of appellant in Idaho. Mobil Oil Corp., supra, 445 U.S., at 439-442, 100 S.Ct., at 1232-1234. Therefore, there is no "rational relationship between [appellant's dividend] income attributed to the State and the intrastate values of the enterprise." Mobil Oil Corp., supra, at 437, 100 S.Ct., at 1231. The Due Process Clause bars Idaho's effort to levy upon income that is not properly within the reach of its taxing power. Pp. 325-329. (d) Under the same unitary-business standard applied to the dividend income in question, Idaho's attempt to tax the interest and capital gains income derived from its subsidiaries also violates the Due Process Clause. P. 329-330. 102 Idaho 38, 624 P.2d 946, reversed. George W. Beatty, Washington, D. C., for appellant. Theodore V. Spangler, Jr., Boise, Idaho, for appellee. Justice POWELL delivered the opinion of the Court. 1 The question is whether the State of Idaho constitutionally may include within the taxable income of a nondomiciliary parent corporation doing some business in Idaho a portion of intangible income—such as dividend and interest payments, as well as capital gains from the sale of stock—that the parent receives from subsidiary corporations having no other connection with the State. 2 * This case involves corporate income taxes that appellee Idaho State Tax Commission sought to levy on appellant ASARCO Inc. for the years 1968, 1969, and 1970. ASARCO is a corporation that mines, smelts, and refines in various States nonferrous metals such as copper, gold, silver, lead, and zinc. It is incorporated in New Jersey and maintains its headquarters and commercial domicile in New York. ASARCO's primary Idaho business is the operation of a silver mine. It also mines and sells other metals and operates the administrative office of its northwest mining division in Idaho. According to the appellee's tax calculations, approximately 2.5% of ASARCO's total business activities take place in Idaho. App. 59a, 67a, and 75a. 3 During the years in question, ASARCO received three types of intangible income of relevance to this suit.1 First, it collected dividends from five corporations in which it owned major interests: M. I. M. Holdings, Ltd.; General Cable Corp.; Revere Copper and Brass, Inc.; ASARCO Mexicana, S. A.; and Southern Peru Copper Corp.2 Second, ASARCO received interest income from three sources: from Revere's convertible debentures; from a note received in connection with a prior sale of Mexicana stock; and from a note received in connection with a sale of General Cable Stock. Third, ASARCO realized capital gains from the sale of General Cable and M. I. M. stock. 4 In 1965, Idaho adopted its version of the Uniform Division of Income for Tax Purposes Act (UDITPA).3 See Idaho Code § 63-3027 (1976 and Supp.1981); 7A U.L.A. 91 (1978). Under this statute, Idaho classifies corporate income from intangible property as either "business" or "nonbusiness" income. "Business" income is defined to include income from intangible property when "acquisition, management, or disposition [of the property] constitute[s] integral or necessary parts of the taxpayers' trade or business operations."4 Idaho apportions such "business" income according to a three-factor formula and includes this apportioned share of "business" income in the taxpayer's taxable Idaho income.5 "Nonbusiness" income, on the other hand, is defined as "all income other than business income." Idaho Code § 63-3027(a)(4) (Supp.1981). Idaho allocates intangible "nonbusiness" income entirely to the State of the corporation's commercial domicile instead of apportioning it among the States in which a corporate taxpayer owns property or carries on business.6 5 Idaho is a member of the Multistate Tax Compact, an interstate taxation agreement concerning state taxation of multistate businesses. The Compact established the Multistate Tax Commission, which is composed of the tax administrators from the member States.7 Article VIII of the Compact provides that any member State may request that the Commission perform an audit on its behalf. See United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S. 452, 457, 98 S.Ct. 799, 804, 54 L.Ed.2d 682 (1978) (upholding the Compact against a facial attack on Compact and Commerce Clauses and Fourteenth Amendment grounds). 6 In 1971, the Multistate Tax Commission audited ASARCO's tax returns for the years in question on behalf of six States, including Idaho. The auditor recommended adjusting ASARCO's tax computations in several respects. As accepted by the Idaho State Tax Commission and as relevant to the present dispute, the auditor first "unitized"—or treated as one single corporation—ASARCO and six of its wholly owned subsidiaries.8 As a consequence of unitization, the auditor combined ASARCO's income with that of these six subsidiaries and disregarded (as intracompany accounting transfers) the subsidiaries' dividend payments to ASARCO. Cf. United States Steel Corp. v. Multistate Tax Comm'n, supra, at 473, n. 25, 98 S.Ct., at 812, n. 25. The auditor listed five factors thought to justify unitizing treatment. First, ASARCO owned a majority (in fact, all) of the stock of each subsidiary. Second, "ASARCO, with its subsidiaries, conducts a vertically integrated non-ferrous metals operation. This is evidenced by the flow from the mines to the smelters to the refineries and ultimately to the sales made by the New York office." App. 88a. Third, "ASARCO and its subsidiaries have interlocking officers and directors, which enables ASARCO to control the major management decisions of each subsidiary." Ibid. Fourth, sales between the companies were numerous, making it "apparent . . . that the companies supplied markets to each other . . . ." Id., at 89a. And finally, various services were provided to the ASARCO group either by ASARCO or by subsidiaries specifically set up for such a purpose.9 The propriety of this treatment of the six wholly owned subsidiaries is not an issue before us. 7 The auditor found the situation to differ with respect to ASARCO's interest in M. I. M., General Cable, Revere, Mexicana, and Southern Peru. This judgment planted the seed of the current dispute. As to these five companies, the auditor determined that the links with ASARCO were not sufficient to justify unitary treatment. Nonetheless, he found that ASARCO's receipt of dividends from each of these did constitute "business" income to ASARCO. See n. 4, supra. The auditor similarly classified the interest and capital gains income at issue in this case. These categories of income also were added in ASARCO's total income to be apportioned among the various States in which ASARCO was subjected to an income tax. 8 The Idaho State Tax Commission adopted the auditor's adjustments in an unreported decision. App. to Juris. Statement 46a. In rejecting ASARCO's challenge to the auditor's unitized treatment of the six wholly owned corporations, see n. 8, supra, the Commission stated that it was "quite clear from the evidence produced at the hearing that [ASARCO's] business activities are so inter-related as to defy measurement by separate accounting. . . ." App. to Juris. Statement 49a-50a. The Commission likewise upheld the auditor's conclusion that the dividends presently at issue were properly treated as apportionable "business" income. It consequently assessed tax deficiencies against ASARCO of $92,471.88 for 1968, $111,292.44 for 1969, and $121,750.76 for 1970, plus interest. 9 On ASARCO's petition for review, the State District Court upheld the Commission's unitized treatment of the six subsidiaries in an unpublished opinion. The court, however, overruled the Commission's determination that the disputed dividends, interest, and capital gains constituted "business" income, on the reasoning that this income did not come from property or activities that were "an integral part of [ASARCO's] trade or business." Idaho Code § 63-3027(a)(1) (Supp.1981). In the court's view, "if the dividend income from other corporations is an integral part of the business of [ASARCO] . . . they should be unitized and all matters considered and[,] if they are not[,] . . . the income is not business income but is [nonapportionable] non business income." App. to Juris. Statement 37a. 10 The Commission, but not ASARCO, appealed to the Idaho Supreme Court. That court held that the trial court had erred by excluding from "business" income ASARCO's receipt of dividends, interest, and capital gains as a result of its owning stock in the five corporations.10 American Smelting & Refining Co. v. Idaho State Tax Comm'n, 99 Idaho 924, 935-937, 592 P.2d 39, 50-52 (1979). In response to ASARCO's constitutional arguments, the court decided that this tax treatment withstood attack under the Commerce and Due Process Clauses. We vacated and remanded the case for reconsideration in light of our decision in Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 100 S.Ct. 1223, 63 L.Ed.2d 510 (1980). ASARCO Inc. v. Idaho State Tax Comm'n, 445 U.S. 939, 100 S.Ct. 1333, 63 L.Ed.2d 773 (1980). The Idaho Supreme Court reinstated its previous opinion in a brief per curiam order on March 4, 1981. 102 Idaho 38, 624 P.2d 946. We noted probable jurisdiction, 454 U.S. 812, 102 S.Ct. 87, 70 L.Ed.2d 80 (1981), and we now reverse. II 11 As a general principle, a State may not tax value earned outside its borders. See, e.g., Connecticut General Life Ins. Co. v. Johnson, 303 U.S. 77, 80-81, 58 S.Ct. 436, 438, 82 L.Ed. 673 (1938).11 The broad inquiry in a case such as this, therefore, is "whether the taxing power exerted by the state bears fiscal relation to protection, opportunities and benefits given by the state. The simple but controlling question is whether the state has given anything for which it can ask return." Wisconsin v. J. C. Penney Co., 311 U.S. 435, 444, 61 S.Ct. 246, 249, 85 L.Ed. 267 (1940). 12 Our application of this general principle in this case is guided by two of our recent decisions. In Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, the taxpayer conducted "an integrated petroleum business," 445 U.S., at 428, 100 S.Ct., at 1227, that included international petroleum exploration, production, refining, transportation, distribution, and sale of petroleum, as well as related chemical and mining enterprises. Much of its business abroad was conducted through wholly or partly owned subsidiaries. The State of Vermont imposed a corporate income tax on that portion of Mobil's total income that the State attributed to Mobil's Vermont activity, which was confined to the wholesale and retail marketing of petroleum. The State sought to include within Mobil's apportionable Vermont income its receipt of dividends from its subsidiaries and affiliates that operated abroad. Mobil protested that the State could not properly apportion and tax this "foreign source" dividend income. 13 For present purposes, our analysis in Mobil began with the observation that Mobil's principal dividend payors were part of Mobil's integrated petroleum business. Although Mobil was "unwilling to concede the legal conclusion" that activities by these dividend payors formed part of Mobil's " 'unitary business,' " it "offered no evidence that would undermine the conclusion that most, if not all, of its subsidiaries and affiliates contribute[d] to [Mobil's] worldwide petroleum enterprise." Id., at 435, 100 S.Ct., at 1230. 14 The Court next stated that due process limitations on Vermont's attempted tax would be satisfied if there were "a 'minimal connection' between the interstate activities and the taxing State, and a rational relationship between the income attributed to the State and the intrastate values of the enterprise." Id., at 436-437, 100 S.Ct., at 1231, citing Moorman Mfg. Co. v. Bair, 437 U.S. 267, 272-273, 98 S.Ct. 2340, 2343-2344, 57 L.Ed.2d 197 (1978); National Bellas Hess, Inc. v. Illinois Dept. of Revenue, 386 U.S. 753, 756, 87 S.Ct. 1389, 1390, 18 L.Ed.2d 505 (1968); Norfolk & Western R. Co. v. Missouri Tax Comm'n, 390 U.S. 317, 325, 88 S.Ct. 995, 1000, 19 L.Ed.2d 1201 (1967). And we said that these limitations would not be contravened by state apportionment and taxation of income that were determined by geographic accounting to have arisen from a different State "so long as the intrastate and extrastate activities formed part of a single unitary business." 445 U.S., at 438, 100 S.Ct., at 1232 (emphasis added). 15 The Mobil Court explicated the limiting "unitary business" principle by observing that geographic accounting, in purporting to isolate income received in various States, "may fail to account for contributions to income resulting from functional integration, centralization of management, and economies of scale." Ibid. The fact that "these factors of profitability arise from the operation of the business as a whole," ibid., therefore could justify a State's otherwise impermissible inclusion of corporate income derived from corporate activities beyond the State's borders. The Court thus stated: 16 "[T]he linchpin of apportionability in the field of state income taxation is the unitary-business principle. In accord with this principle, what appellant must show, in order to establish that its dividend income is not subject to an apportioned tax in Vermont, is that the income was earned in the course of activities unrelated to the sale of petroleum products in that State. [Mobil] has made no effort to demonstrate that the foreign operations of its subsidiaries and affiliates are distinct in any business or economic sense from its petroleum sales activities in Vermont. Indeed, all indications in the record are to the contrary, since it appears that these foreign activities are part of [Mobil's] integrated petroleum enterprise. In the absence of any proof of discrete business enterprise, Vermont was entitled to conclude that the dividend income's foreign source did not destroy the requisite nexus with in-state activities." Id., at 439-440, 100 S.Ct., at 1232-1233 (emphasis added and footnote omitted). 17 We consequently rejected Mobil's constitutional challenge to Vermont's tax. In so doing, however, we cautioned that we did 18 "not mean to suggest that all dividend income received by corporations operating in interstate commerce is necessarily taxable in each State where that corporation does business. Where the business activities of the dividend payor have nothing to do with the activities of the recipient in the taxing State, due process considerations might well preclude apportionability, because there would be no underlying unitary business." Id., at 441-442, 100 S.Ct., at 1233-1234 (emphasis added). 19 We soon had occasion to reiterate these principles. Three months after Mobil, we decided Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 100 S.Ct. 2109, 65 L.Ed.2d 66 (1980). In Exxon, "a vertically integrated petroleum company," id., at 210, 100 S.Ct., at 2113, explored for, produced, refined, and marketed petroleum and related products. Although Exxon's activities in Wisconsin were confined to marketing, the State sought to apportion and tax Exxon's income from nonmarketing activities in the United States. 20 Exxon disputed the propriety of this treatment. The Wisconsin Tax Appeals Commission agreed with the objection on the basis of its conclusion that Exxon's "three main functional operating departments—Exploration and Production, Refining, and Marketing—were separate unitary businesses." Id., at 215, 100 S.Ct., at 2116 (emphasis added). The Commission found that the tax as applied " 'had the effect of imposing a tax on [Exxon's] exploration and on its refining net income, all of which was derived solely from operations outside the State of Wisconsin and which had no integral relationship to [Exxon's] marketing operations within Wisconsin.' " Ibid. On appeal, however, the Circuit Court for Dane County held that Exxon's three main functional operating departments were all part of a single unitary business. The Wisconsin Supreme Court agreed.12 21 In reviewing the case, this Court unanimously agreed with the State Commission and the two state courts that the decisive concept in the case was that of a unitary business. Significantly, we repeated Mobil § teaching that "[t]he 'linchpin of apportionability' for state income taxation of an interstate enterprise is the 'unitary-business principle.' " Id., at 223, 100 S.Ct., at 2120, quoting Mobil, 445 U.S., at 439, 100 S.Ct., at 1232. We also repeated that 22 "In order to exclude certain income from the apportionment formula, the company must prove that 'the income was earned in the course of activities unrelated to the sale of petroleum products in that State.' . . . The court looks to the 'underlying economic realities of a unitary business,' and the income must derive from 'unrelated business activity' which constitutes a 'discrete business enterprise,' 445 U.S., at 441, 442, 439 [, 100 S.Ct., at 1233, 1234, 1232.]" 447 U.S., at 223-224, 100 S.Ct., at 2120. 23 Examining the facts, the Court found that Exxon was "a highly integrated business which benefits from an umbrella of centralized management and controlled interaction." Id., at 224, 100 S.Ct., at 2120.13 We rejected the company's protest because "[w]e agree[d] with the Wisconsin Supreme Court that Exxon [was] such a unitary business and that Exxon has not carried its burden of showing that its functional departments are 'discrete business enterprises'. . . ." Ibid.14 III 24 In this case, ASARCO claims that it has succeeded, where the taxpayers in Mobil and Exxon failed, in proving that the dividend payors at issue are not part of its unitary business, but rather are "discrete business enterprises." 447 U.S., at 224, 100 S.Ct., at 2120. We must test this contention on the record before us. A. 25 The closest question is posed by ASARCO's receipt of dividends from Southern Peru. ASARCO is one of Southern Peru's four shareholders, holding 51.5% of its stock.15 Southern Peru produces smelted but unrefined "blister copper" in Peru, and sells 20-30% of its output to the Southern Peru Copper Sales Corp.16 The remainder of Southern Peru's output is sold under contracts to its shareholders in proportion to their ownership interests. Southern Peru sold about 35% of its output to ASARCO, App. 89a, at prices determined by reference to average representative trade prices quoted in a trade publication and over which the parties had no control.17 Id., at 125a-126a; 99 Idaho, at 928, 592 P.2d, at 43. 26 ASARCO's majority interest, if asserted, could enable it to control the management of Southern Peru. The Idaho State Tax Commission, however, found that Southern Peru's "remaining three shareholders, owning the remainder of the stock, refuse[d] to participate in [Southern Peru] unless assured that they would have a way to assure that management would not be completely dominated by ASARCO." App. to Juris. Statement 55. Consequently ASARCO entered a management agreement giving it the right to appoint 6 of Southern Peru's 13 directors. The other three shareholders also appointed six directors. Ibid. The thirteenth and final director is appointed by the joint action of either the shareholders or the first 12 directors. Ibid.; App. 121a. Southern Peru's bylaws provide that eight votes are required to pass any resolution, ibid., and its articles and bylaws can be changed only by unanimous consent of the four stockholders. 27 In its unreported opinion, the state trial court concluded that this management contract "insures that [ASARCO] will not be able to control [Southern Peru]." App. to Juris. Statement 43a. It likewise found that Southern Peru "operates independently of [ASARCO]." Id., at 42a. The court reached this conclusion after hearing testimony that ASARCO did not "control Southern Peru in any sense of that term," App. 121a, and that Southern Peru did not "seek direction or approval from ASARCO on major decisions." Id., at 124a. Idaho does not dispute any of these facts. In view of the findings and the undisputed facts, we conclude that ASARCO's Idaho silver mining and Southern Peru's autonomous business are insufficiently connected to permit the two companies to be classified as a unitary business. B 28 Under the principles of our decisions, the relationship of each of the other four subsidiaries to ASARCO falls far short of bringing any of them within its unitary business. M. I. M. Holdings engages in the mining, milling, smelting, and refining of copper, lead, zinc, and silver in Australia. The company also operates a lead and zinc refinery in England. During the years in question M. I. M. sold only about 1% of its output to ASARCO, for sums in the range of $0.2 to $2.2 million. Id., at 43a-47a. It appears that these sales were on the open market at prevailing market rates. ASARCO owns 52.7% of M. I. M.'s stock, and the rest is widely held. Although ASARCO has the control potential to manage M. I. M., no claim is made that it has done so.18 As an ASARCO executive explained, it never even elected a member of M. I. M.'s board: 29 "This company has been very successful in staffing the corporation with Australian people and [they have] been able to run this company by themselves and, therefore, in consequence of the nationalistic feeling which develops in most of such developing countries we have not exercised any right we might have to elect a director to the board of the company." Id., at 132a. 30 In addition to forgoing its right to elect directors, ASARCO similarly has taken no part in the selection of M. I. M.'s officers—a function of the board of directors. Nor do the two companies have any common directors or officers. Id., at 34a, 40a. The state trial court found that M. I. M. "operates entirely independently of and has minimal contact with" ASARCO. App. to Juris. Statement 43a. As the business relation also is nominal, it is clear that M. I. M. is merely an investment. See, e.g., Keesling & Warren, The Unitary Concept in the Allocation of Income, 12 Hastings L.J. 42, 52-53 (1960). 31 General Cable and Revere Copper, large publicly owned companies, fabricate metal products. Both are ASARCO customers.19 But ASARCO held only minority interests, owning approximately 34% of the outstanding common shares of each. The remaining shares—listed on the New York Stock Exchange—are widely held. App. 135a. The two companies occupy parallel positions with respect to ASARCO as a result of a 1961 Department of Justice antitrust suit against ASARCO. The suit was based on ASARCO's interests in each. In 1967, ASARCO consented to a decree that prohibited it from maintaining common officers in these companies, voting its stock in them, selling the companies copper at prices below those quoted to their competition, and from acquiring stock in any other copper fabricator. Id., at 96a. Neither Revere's nor General Cable's management seeks direction or approval from ASARCO on operational or other management decisions.20 Id., at 137a. 32 Mexicana mines and smelts lead and copper in Mexico. Originally it was a wholly owned subsidiary of ASARCO, but a change in Mexican law required ASARCO to divest itself of 51% of Mexicana's stock in 1965. This stock is now publicly held by Mexican nationals. The record does not reveal whether ASARCO and Mexicana have any common directors. The state trial court found, however, that Mexicana "operates independently of [ASARCO]," App. to Juris. Statement 43a, and the Idaho Supreme Court stated that "Mexicana does not seek approval from ASARCO concerning major policy decisions. . . ." 99 Idaho, at 929, 592 P.2d, at 44.21 C 33 Idaho does not dispute the foregoing facts. Neither does it question that a unitary business relationship between ASARCO and these subsidiaries is a necessary prerequisite to its taxation of the dividends at issue. E.g., Brief for Appellee 10 ("When income is earned from activities which are part of a unitary business conducted in several states, then the requirement that the income bear relation to the benefits and privileges conferred by the several states has been met") See also Tr. of Oral Arg. 25 ("[W]hen intangible assets such as, for example, shares of stock, are found to be a part of a taxpayer's own unitary business, . . . there is no logical or constitutional reason why the income from those same intangibles should be treated any differently than any other business income that that taxpayer might earn"). Rather the State urges that we expand the concept of a "unitary business" to cover the facts of this case. 34 Idaho's proposal is that corporate purpose should define unitary business. It argues that intangible income should be considered a part of a unitary business if the intangible property (the shares of stock) is "acquired, managed or disposed of for purposes relating or contributing to the taxpayer's business." Brief for Appellee 4. See also Tr. of Oral Arg. 25 (urging that income from intangible property be considered part of a unitary business when the intangibles "contribute to or relate to or are some way in furtherance of the taxpayer's own trade or business"). Idaho asserts that "[i]t is this integration—i.e., between the business use of the intangible asset (the shares of stock) and ASARCO's mining, smelting, and refining business—which makes the income part of the unitary business." Brief for Appellee 4. 35 This definition of unitary business would destroy the concept. The business of a corporation requires that it earn money to continue operations and to provide a return on its invested capital. Consequently all of its operations, including any investment made, in some sense can be said to be "for purposes related to or contributing to the [corporation's] business." When pressed to its logical limit, this conception of the "unitary business" limitation becomes no limitation at all. When less ambitious interpretations are employed, the result is simply arbitrary.22 36 We cannot accept, consistently with recognized due process standards, a definition of "unitary business" that would permit nondomiciliary States to apportion and tax dividends "[w]here the business activities of the dividend payor have nothing to do with the activities of the recipient in the taxing State. . . ."23 Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S., at 442, 100 S.Ct., at 1234. In such a situation, it is not true that "the state has given anything for which it can ask return." Wisconsin v. J. C. Penney Co., 311 U.S., at 444, 61 S.Ct., at 249. 37 Justice Holmes stated long ago that "the possession of bonds secured by mortgages of lands in other States, or of a land-grant in another State or of other property that adds to the riches of the corporation but does not affect the [taxing State's] part of the [business] is no sufficient ground for the increase of the tax whatever it may be . . . ." Wallace v. Hines, 253 U.S. 66, 69-70, 40 S.Ct. 435, 437, 64 L.Ed. 782 (1920). In this case, it is plain that the five dividend-paying subsidiaries "add to the riches" of ASARCO. But it is also true that they are "discrete business enterprise[s]" that—in "any business or economic sense"—have "nothing to do with the activities" of ASARCO in Idaho. Mobil, supra, 445 U.S., at 439-442, 100 S.Ct., at 1232-1234. Therefore there is no "rational relationship between the [ASARCO dividend] income attributed to the State and the intrastate values of the enterprise. Moorman Mfg. Co. v. Bair, 437 U.S. 267, 272-273, 98 S.Ct. 2340, 2343-2344, 57 L.Ed.2d 197 (1978)." Mobil, supra, 445 U.S., at 437, 100 S.Ct., at 1231. Idaho's attempt to tax a portion of these dividends can be viewed as "a mere effort to reach profits earned elsewhere under the guise of legitimate taxation." Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271, 283, 45 S.Ct. 82, 84, 69 L.Ed. 282 (1924). The Due Process Clause bars such an effort to levy upon income that is not properly "within the reach of [Idaho's] taxing power." Connecticut General Life Ins. Co. v. Johnson, 303 U.S., at 80, 58 S.Ct., at 438.24 IV 38 In addition to the disputed dividend income, Idaho also has sought to tax certain ASARCO interest and capital gains income. The interest income arose from a note ASARCO received from its sale of Mexicana stock and from a Revere convertible debenture, as well as in connection with ASARCO's 1970 disposition of its General Cable stock. See n. 21, supra. The General Cable stock sale also generated capital gains for ASARCO, as did ASARCO's sale of a portion of its stock in M. I. M. 39 Idaho and ASARCO agree that interest and capital gains income derived from these companies should be treated in the same manner as the dividend income.25 Brief for Appellant 27; Brief for Appellee 21. Cf. 99 Idaho, at 937, 592 P.2d, at 52 ("In our view the same standard applies to the question whether gains from the sale of stock are business income as applies to the question whether dividends from the stock are business income"). We also agree. "One must look principally at the underlying activity, not at the form of investment, to determine the propriety of apportionability." Mobil, 445 U.S., at 440, 100 S.Ct., at 1233. Changing the form of the income "works no change in the underlying economic realities of [whether] a unitary business [exists], and accordingly it ought not to affect the apportionability of income the parent receives." Id., at 441, 100 S.Ct., at 1233. We therefore hold that Idaho's attempt to tax this income also violated the Due Process Clause. V 40 For the reasons stated, the judgment of the Supreme Court of Idaho is 41 Reversed. Chief Justice BURGER, concurring. 42 I join the Court's opinions in both No. 80-2015 and No. 80-1745 in reliance on the Court's express statement that the Court's holdings do not preclude future congressional action in this area. 458 U.S. 307, 328, n.23, 102 S.Ct. 3103, 3115, n.23, 73 L.Ed.2d 787. 43 Justice O'CONNOR, with whom Justice BLACKMUN and Justice REHNQUIST join, dissenting. 44 The Court today declares that the Due Process Clause of the Constitution forbids a State to tax a proportionate share of the investment income of a nondomiciliary corporation doing business within its borders. In so doing, the Court groundlessly strikes down the eminently reasonable assertion of Idaho's taxing power at issue in this case. Far more dismaying, however, is that the Court's reliance on the Due Process Clause may deprive Congress of the authority necessary to rationalize the joint taxation of interstate commerce by the 50 States. 45 Today, the taxpayer wins. Yet in the end, today's decision may prove to be a loss for all concerned—interstate businesses themselves, which the Commerce Clause guarantees the opportunity to serve the country's needs unimpeded by a parochial hodgepodge of overlapping and conflicting tax levies; the Nation, which demands a prosperous interstate market; and the States, which deserve fair return for the advantages they afford interstate enterprise. For while this Court has the authority to invalidate a specific state tax, only Congress has both the ability to canvass the myriad facts and factors relevant to interstate taxation and the power to shape a nationwide system that would guarantee the States fair revenues and offer interstate businesses freedom from strangulation by multiple paperwork and tax burdens. Unfortunately, by apparently stripping Congress of the authority to do the job, the Court delays the day when a uniform system responsive to the needs of all can be fashioned. 46 The Court has strayed "beyond the extremely limited restrictions that the Constitution places" on the taxing power of the States, "inject[ed itself] in a merely negative way into the delicate processes of fiscal policy-making," and regrettably "imprison[ed] the taxing power of the states within formulas that are not compelled by the Constitution." Wisconsin v. J. C. Penney Co., 311 U.S. 435, 445, 61 S.Ct. 246, 250, 85 L.Ed. 267 (1940). I respectfully dissent. 47 * "Taxes," as Justice Holmes once observed, "are what we pay for civilized society." Compania General de Tabacos de Filipinas v. Collector of Internal Revenue, 275 U.S. 87, 100, 48 S.Ct. 100, 105, 72 L.Ed. 177 (1927) (dissenting opinion). A natural corollary of this proposition is that the Due Process Clause permits state taxation if "the state has given anything for which it can ask return." Wisconsin v. J. C. Penney Co., 311 U.S., at 444, 61 S.Ct., at 249. A State thus "is free to pursue its own fiscal policies, unembarrassed by the Constitution," if it "exert[s] its power in relation to opportunities which it has given, to protection which it has afforded, [or] to benefits which it has conferred by the fact of being an orderly, civilized society." Ibid. 48 In applying this fundamental principle to businesses that derive income from more than one State, we repeatedly have declared that a state tax passes constitutional muster unless the taxpayer can show that there is not even "a 'minimal connection' between [its] interstate activities and the taxing State," or a merely "rational relationship between the income attributed to the State and the intrastate values of [its] enterprise." Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 436-437, 100 S.Ct. 1223, 1231, 63 L.Ed.2d 510 (1980) (quoting Moorman Mfg. Co. v. Bair, 437 U.S. 267, 272-273, 98 S.Ct. 2340, 2343-2344, 57 L.Ed.2d 197 (1978)). As the present case demonstrates, however, this oft-repeated formula is more easily stated than applied when a State attempts to tax the net income of an enterprise doing business in many jurisdictions. 49 The principal difficulty arises because a multijurisdictional business is "an organic system," Wallace v. Hines, 253 U.S. 66, 69, 40 S.Ct. 435, 437, 64 L.Ed. 782 (1920) (Holmes, J.), whose income cannot sensibly be reduced to the sum of the hypothetical incomes of distinct component parts, each wrenched from the unitary whole and conceptually confined to operations within a single State.1 With this understanding in mind, for more than half a century we have held that a State is not constitutionally required to tax only that slice of an interstate enterprise operating physically within the State. See, e.g., Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271, 45 S.Ct. 82, 69 L.Ed. 282 (1924); Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 100 S.Ct. 2109, 65 L.Ed.2d 66 (1980). Instead, as we stated only two Terms ago, "[i]t has long been settled that 'the entire net income of a corporation, generated by interstate as well as intrastate activities, may be fairly apportioned among the States for tax purposes by formulas utilizing in-state aspects of interstate affairs,' " id., 447 U.S., at 219, 100 S.Ct., at 2118 (quoting Northwestern States Portland Cement Co. v. Minnesota, 358 U.S. 450, 460, 79 S.Ct. 357, 363, 3 L.Ed.2d 421 (1959)), provided only that in each case the resulting tax liability is not " 'out of all appropriate proportion to the business transacted' " in the taxing State itself, 447 U.S., at 220, 100 S.Ct., at 2118 (quoting Hans Rees' Sons, Inc. v. North Carolina ex rel. Maxwell, 283 U.S. 123, 135, 51 S.Ct. 385, 389, 75 L.Ed. 879 (1931)). 50 In short, the "linchpin" of apportioned state taxation is the concept of an organic, unitary business. Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, 445 U.S., at 439, 100 S.Ct., at 1232. The constitutionality of a state tax levied on extraterritorial business operations thus turns on whether the out-of-state business activity can be characterized as a separate business with no in-state contacts or whether instead it is a part of a unitary enterprise doing business in the State. In the case before us, the Court first errs when it attempts to determine whether or not ASARCO's investments were part of ASARCO's unitary nonferrous metals business. II 51 ASARCO realized capital gains, dividends, and interest income from its ownership of securities issued by five foreign subsidiaries. The issue for the Court is whether that income was earned by ASARCO's unitary nonferrous metals business, and therefore was subject to Idaho's taxes, or instead was earned by a separate investment business unrelated to ASARCO's operations in Idaho, and therefore was constitutionally exempt from taxation by that State. As always, of course, the State's taxation of the company's income is presumptively constitutional. To overcome that presumption, ASARCO has the " 'distinct burden of showing by "clear and cogent evidence" ' " that Idaho's scheme " 'results in extraterritorial values being taxed.' " Exxon Corp. v. Wisconsin Dept. of Revenue, supra, 447 U.S., at 221, 100 S.Ct., at 2119 (quoting Butler Bros. v. McColgan, 315 U.S. 501, 507, 62 S.Ct. 701, 703, 86 L.Ed. 991 (1942), in turn quoting Norfolk & Western R. Co. v. North Carolina ex rel. Maxwell, 297 U.S. 682, 688, 56 S.Ct. 625, 628, 80 L.Ed. 977 (1936)). 52 According to the Court, ASARCO has met this burden by showing that during the relevant tax years its holdings in the five subsidiaries were passive investments not functionally integrated with ASARCO's nonferrous metals business. On this basis, the Court concludes that ASARCO's holdings were, in effect, part of a separate investment business having too little to do with ASARCO's unitary nonferrous metals business to support apportioned taxation. 53 Both common sense and business reality dictate a different result. ASARCO, far from showing that its investment holdings were part of an "unrelated,"2 "discrete business enterprise,"3 "hav[ing] nothing to do with the activities"4 of its unitary nonferrous metals business, has failed in at least three ways to bear its "distinct burden" of demonstrating that Idaho's tax was unconstitutionally levied. 54 First, even accepting, arguendo, the Court's conclusion that the contested income was derived from passive investments, ASARCO has failed to show that its investment decisionmaking was segregated from its nonferrous metals business. ASARCO cannot deny that the subsidiary companies in which it invested were participants in the nonferrous metals industry, the very industry in which ASARCO played a major operational role. As the Court acknowledges, ASARCO "mine[d], smelt[ed], and refine[d] . . . nonferrous metals such as copper, gold, silver, lead, and zinc," ante, at 309, while one of its subsidiaries "engaged in the mining, milling, smelting, and refining of nonferrous metals," ante, at 309-310, n. 2, another engaged "in the same general line of business" as did "ASARCO in the United States," ibid., a third "mine[d] and smelt[ed] copper," ante, at 3106, n. 2, and the last two were important "ASARCO customers," ante, at 323, fabricating, respectively, cables and copper wares, ante, at 309, n. 2. In short, ASARCO invested not in "unrelated business[es]," such as hotel chains and breweries, but in companies participating in the nonferrous metals markets. Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S., at 224, 100 S.Ct., at 2120 (quoting Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S., at 442, 100 S.Ct., at 1234). 55 ASARCO invested in these nonferrous metals companies with a well-founded confidence that few other investors could muster, since much of what it had learned in operating its own nonferrous metals business must have been invaluable in evaluating the prospects for other companies engaged in similar businesses and markets worldwide. Put another way, it would have been a perverse act of self-denial for ASARCO to ignore its intimate knowledge of world markets, refining and smelting technology, mining operations, and geological reserves when it decided whether and how to invest in the five companies of concern here. Thus, the investment decisions ASARCO made regarding the securities of these five participants in the nonferrous metals markets undoubtedly depended heavily on ASARCO's knowledge of its own business. 56 In fact, during the course of this litigation, ASARCO has admitted as much. In the trial court, ASARCO's vice president and comptroller answered questions put to him by ASARCO's counsel as follows: 57 "Q. Now, ASARCO has investments in a lot of mining companies, does it not? 58 "A. Yes. 59 "Q. Can you tell the court why ASARCO makes investments of this nature? 60 "A. The ASARCO management utilizes its expertizes [sic] in channeling funds into these areas where they are knowledgeable and—most knowledgeable—and hopefully accrue [sic] to the benefit of ASARCO stockholders. 61 "Q. So that essentially ASARCO feels that it can use its expertizes [sic] to produce the maximum returns for its shareholders? 62 "A. That is correct. 63 "Q. And whereas with other purchases I suppose ASARCO feels that the shareholders can make the investments themselves as well as ASARCO can? 64 "A. Right." Id., at 81-82. 65 Moreover, in its brief to the Supreme Court of Idaho ASARCO flatly stated that "[i]t invests its shareholders' money in businesses in which it has expertise and distributes the investment return in the form of dividends to its shareholders." Record 521. 66 In sum, far from showing by "clear and cogent evidence" that its investment decisions regarding other nonferrous metal suppliers and users were segregated from the resources of information and expertise developed in its own nonferrous metals business, ASARCO itself provided evidence that its investment decisionmaking was part of an indivisible, unitary nonferrous metals business. This alone warrants affirming the Idaho Supreme Court's due process ruling. B 67 Second, again assuming, arguendo, that the contested investments were in fact passive, ASARCO has failed to show that its holdings were divorced from its management of the financial requirements of its nonferrous metals business. For all we know, ASARCO's investments were triggered by its need to obtain a return on idle financial resources accumulated for the future operation of its own primary business. 68 ASARCO does not, and could not, contend that all its investment income is per se beyond the taxing power of the nondomiciliary States in which it operates. Rather, it concedes that the Due Process Clause permits Idaho to tax, on an apportioned basis, the income ASARCO earned on short-term investments of its working capital.5 After all, an appropriate amount of liquid working capital is necessary to the day-to-day operation of a business, and any return earned from its temporary investment is a byproduct of the operation of the business. ASARCO thus admits that Idaho could tax a portion of the income realized from an investment in, say, short-term commercial paper, even though the underlying operations of the issuing companies were far less related to ASARCO's nonferrous metals business than the operations of the five subsidiaries at issue here. 69 The interim investment of retained earnings prior to their commitment to a major corporate project, however, merely recapitulates on a grander scale the short term investment of working capital prior to its commitment to the daily financial needs of the company. Just as companies prefer to maintain a cushion of working capital rather than resort to the short-term capital markets on an hourly basis for the money necessary to operate their businesses, many enterprises prefer to acquire the capital necessary for the expansion and replacement of plants and equipment by creating long-term funds, rather than resort to the vagaries of the capital markets. In order to prevent the accumulating capital from sitting idle, such funds are usually invested in financial assets with a degree of liquidity appropriate to the money's intended ultimate use.6 Any return ASARCO earned on such investments plainly would be functionally related to the conduct of its nonferrous metals business and, therefore, taxable by Idaho on an apportioned basis as unitary business income. 70 Such investment of idle funds, after all, mirrors the borrowing of funds a company lacks. Undoubtedly, ASARCO would be quick to assert that any long-term borrowing recorded on the liability side of its balance sheet is an integral part of its unitary business justifying the deduction of interest expense in the computation of apportionable net income. If so, ASARCO cannot contend that the long-term investments recorded on the asset side of its balance sheet are automatically separate from its unitary business, thereby justifying the exclusion of the revenues received from apportionable net income. The same principles apply whether the money is going in or coming out. 71 Thus, because investments of ASARCO's working capital are functionally integrated with its unitary nonferrous metals business, and because ASARCO failed to show by "clear and cogent evidence" the facts necessary to distinguish, on a principled basis, its investments in the securities of the five subsidiaries at issue here,7 the Idaho Supreme Court correctly concluded that apportioned taxation of ASARCO's contested investment income does not violate the Due Process Clause. C 72 Finally, the Court errs even in its fundamental determination that ASARCO's holdings were passive investments unrelated to ASARCO's operational business. In fact, the disputed investments actively contributed to ASARCO's nonferrous metals business. 73 To begin with, ASARCO had effective operational control of at least three of the five subsidiaries. ASARCO's commanding 52.7% interest in M. I. M. Holdings, Ltd., uncontestably gave it full control of that company. Although ASARCO did not wield quite the same power over Southern Peru Copper, its 51.5% interest nonetheless gave it unilateral veto power over all corporate decisions, including those supported unanimously by all other shareholders.8 Finally, in the case of ASARCO Mexicana, the record discloses only that ASARCO had been forced to sell 51% of its initial 100% interest to Mexican nationals, retaining a 49% interest for itself. ASARCO has made no showing that it is not the principal investor in Mexicana and thus able to control the company. In sum, ASARCO undoubtedly was the dominant factor in at least three of the five subsidiaries under consideration here, with the power to use them to advantage in its nonferrous metals business.9 The Court, however, minimizes the significance of this control, emphasizing that ASARCO did not openly and aggressively assert its control during the tax years in question and concluding that ASARCO's subsidiaries did not contribute to its nonferrous metals business. 74 The Court's result is hard to understand in view of our decision just two years ago in Exxon Corp. v. Wisconsin Dept. of Revenue. In summarizing our result in Exxon, we asserted that the "important link" establishing the unity of Exxon's business "most clearly" was based on two factors. 447 U.S., at 224, 100 S.Ct., at 2120. 75 First, we noted that " 'placing individual segments under one corporate entity . . . provide[s] greater profits stability' " because " 'nonparallel and nonmutual economic factors which may affect one department may be offset by the factors existing in another department.' " Id., at 225, 100 S.Ct., at 2121 (quoting the testimony of an Exxon senior vice president). ASARCO's ownership of subsidiaries doing business in precisely ASARCO's line of work in two different geographical markets, M. I. M. Holdings in Australia and ASARCO Mexicana in Mexico, undoubtedly provided exactly that sort of advantage; economic conditions in Australia and Mexico do not track those in the United States, so that when the nonferrous metals business is in the doldrums in one country, it may be prospering in another. But, unlike the Exxon Court, today's Court is blind to the significance to the "profits stability" of ASARCO's nonferrous metals business of its subsidiaries in unrelated geographical markets. 76 Second, in Exxon we noted that the vertical relationship between the various departments in Exxon's business provided both "an assured supply of raw materials" and an "assured and stable outlet for products" so that Exxon could "minimiz[e]" the "risk of disruptions" "due to [the] supply and demand imbalances that may occur from time to time." Ibid. The Exxon Court's recognition of the business importance of captive suppliers and customers merely confirmed our earlier decision in Mobil Oil, in which we affirmed Vermont's apportioned taxation of the more than $115 million in dividend income Mobil had received from its 10% interest in the Arabian American Oil Co., 445 U.S., at 457, n. 10, 100 S.Ct., at 1241, n. 10 (STEVENS, J., dissenting). Mobil's 10% investment, apart from providing handsome dividends, apparently had helped to assure Mobil of supplies of crude oil for its petroleum business. By contrast, the Court today inexplicably invalidates Idaho's taxation of ASARCO's dividend income from its fivefold greater 51.5% interest in Southern Peru Copper Corp., an investment that evidently helped to assure ASARCO of supplies of unrefined copper, since 35% of the entire copper output of Southern Peru was sold to ASARCO.10 77 Apparently, the Court no longer believes it significant that the subsidiaries in which a parent has major holdings "minimiz[e]" the "risk of disruptions" "due to [the] supply and demand imbalances that may occur from time to time,"Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S., at 225, 100 S.Ct., at 2121, by providing "assured suppl[ies]" and "stable outlet[s]," ibid., unless the subsidiaries are actively managed on a day-to-day basis. The Court evidently would find that ASARCO's subsidiaries were part of ASARCO's unitary business only if ASARCO experienced a "supply and demand imbalanc[e]" sufficiently severe to force it to exercise day-to-day control of its captive subsidiaries. In this regard, the Court's position is akin to the view that a paid-up fire insurance policy is a worthless asset unless smoke is in the air. 78 In sum, despite ASARCO's failure on each of the three counts just discussed to bear its "distinct burden" of showing that its investments are unrelated to its nonferrous metals business, the Court rules that Idaho cannot tax the investment income at issue here. In so doing, the Court unwisely substitutes for the multifaceted analysis used to determine whether the businesses in Mobil Oil and Exxon were unitary the oversimplified test of active operational control. The result is that the Court has ignored business advantages to ASARCO more than sufficient to establish that its holdings in its subsidiaries were part of its unitary business. In consequence, the Court wrongly concludes that ASARCO has borne the "distinct burden" of showing that its holdings in the five affected subsidiaries are not functionally related to the income of its operational nonferrous metals business.11 79 Trying to justify that result, the Court suggests that for it to hold otherwise "would destroy the concept" of a unitary business by expanding the idea until it "becomes no limitation at all" on the power of the States to tax. Ante, at 326. In actuality, the Court's decision today shrinks the concept beyond all recognition. Thus it is the Court's holding, not Idaho's tax, that menaces the unitary-business principle. The "linchpin" is loose from the axle. III 80 As a natural consequence of its decision that Idaho cannot tax ASARCO's investment income, the Court simultaneously, if implicitly, rules out taxation of the disputed income by any other nondomiciliary State in which ASARCO conducts its nonferrous metals business, absent a special connection between the would-be taxing State and ASARCO's investments.12 By the process of elimination, then, the Court's holding provides a partial answer to the question of which State or States the Constitution permits to tax this income. The answer the Court gives, however, demonstrates how ill-advised is the course on which it embarks. By its analysis, the Court leaves open three possible choices regarding which States, if any, may tax ASARCO's contested income. Each of these possibilities suffers from crippling defects, pointing to the conclusion that the Court errs in prohibiting apportioned taxation of investment income by nondomiciliary States. 81 First, there is the disturbing possibility that no State could satisfy the requirements of the Due Process Clause as interpreted today by the Court, so that the contested income would be, in the words of state tax administrators, "nowhere income."13 If so, today's holding casts a deep shadow on the ability of the States to tax their fair share of the corporate income they help to produce by providing an "orderly, civilized society." Even more disturbing, given such an interpretation, the Court's decision endangers even federal taxation of passive investment income, since the Federal Government's contacts with the income at issue here obviously cannot exceed the sum of the contacts of the various States. Presumably, the Court's opinion should not be read as erecting so high a hurdle to state and federal taxation. B 82 Second, there is the possibility that only a domiciliary State or States could tax the disputed income. In Mobil Oil, the Court stated that "[t]axation by apportionment and taxation by allocation to a single situs are theoretically incommensurate, and if the latter method is constitutionally preferred, a tax based on the former cannot be sustained." 445 U.S., at 444-445, 100 S.Ct., at 1235 (emphasis added). If so, the converse may also be true: if taxation by apportionment is constitutionally condemned, taxation by allocation to a single situs may be constitutionally preferred. The Court's decision today thus could be read as broadly hinting that a domiciliary State enjoys a preference of constitutional dimension justifying its—and only its—taxation of income such as that derived from ASARCO's investments. 83 Perhaps such a preference could find some blessing in tradition, but certainly not in logic or in the recent opinions of this Court. In Mobil Oil itself, the Court declared: 84 "We find no adequate justification . . . for such a preference. Although a fictionalized situs for intangible property sometimes has been invoked to avoid multiple taxation of ownership, there is nothing talismanic about the concepts of 'business situs' or 'commercial domicile' that automatically renders those concepts applicable when taxation of income from intangibles is at issue. The Court has observed that the maxim mobilia sequuntur personam, upon which these fictions of situs are based, 'states a rule without disclosing the reasons for it.' The Court also has recognized that 'the reason for a single place of taxation no longer obtains' when the taxpayer's activities with respect to the intangible property involve relations with more than one jurisdiction. . . . Moreover, cases upholding allocation to a single situs for property tax purposes have distinguished income tax situations where the apportionment principle prevails." Id., at 445, 100 S.Ct., at 1235 (citations omitted). 85 The Court thus made clear only two years ago that a State of domicile cannot expect automatically to meet the due process requirements for the taxation of investment income. As with a nondomiciliary State, a domiciliary State may tax investment income only if it confers benefits on or affords protection to the investment activity. Mere assertion of the arbitrary legal fiction that intangible property is located at its owner's domicile no longer suffices to repel a reluctant taxpayer's due process attack. 86 The principal functional basis on which this Court has justified taxation by the commercial domicile, moreover, actually supports the fully apportioned taxation of investment income that today's decision rules out, rather than taxation by allocation to a single situs. In Wheeling Steel Corp. v. Fox, 298 U.S. 193, 56 S.Ct. 773, 80 L.Ed. 1143 (1936), for example, we sustain an ad valorem tax on accounts receivable and bank deposits levied by the State in which the taxpayer maintained "the actual seat of its corporate government,"id., at 212, 56 S.Ct., at 778, for the reason that the intangibles at issue had become " 'integral parts of some local business,' " id., at 210, 56 S.Ct., at 777 (quoting Farmers Loan & Trust Co. v. Minnesota, 280 U.S. 204, 213, 50 S.Ct. 98, 101, 74 L.Ed. 371 (1930)). Thus, other than the arbitrary fiction that intangible property is "located" at the domicile of its owner, the underlying jurisdictional basis for taxation at the commercial domicile is grounded in the fact that intangibles are an "integral part" of the business. This justification supports the principle of apportionment rather than allocation solely to the single domiciliary State. After all, if intangibles are an "integral part" of the unitary business in the domiciliary State, they also are related to the business of the corporation elsewhere. It hardly makes sense to allocate income to the commercial domicile on the theory that business activity at the commercial domicile promotes the unitary business everywhere, and then to ignore those connections and to disregard the claims of the other States in which the unitary business operates. See Dexter, Taxation of Income from Intangibles of Multistate-Multinational Corporations, 29 Vand.L.Rev. 401, 416 (1976). 87 In short, unless the Court is prepared to abandon the unitary-business principle as applied to investment income and to read into the Constitution the arbitrary legal fiction that intangibles are situated at the domicile of their owner, the Court will be unable to sustain a domiciliary State's allocation of all passive investment income to itself against due process attack. C 88 We thus arrive at the only remaining possibility. The Court's holding today, taken with past decisions, may imply that ASARCO's investments must be treated as though ASARCO were not only running its nonferrous metals business but also running as another, separate business a sort of mutual fund or holding company specializing in the worldwide nonferrous metals industry. The income from this fictitious separate business would then be taxable on an apportioned basis by those States in which the business was carried out, just as ASARCO's unitary nonferrous metals business could be taxed on an apportioned basis by those States in which that business is conducted. 89 If so, the Constitution apparently requires that a very small tail be permitted to wag a very big dog. For in the case of companies like ASARCO with tens or hundreds of millions of dollars of dividend income generated by a handful of long-term investments, vast differences in state revenues may turn on whether the quarterly dividend checks sent from "passive" subsidiaries are sent to a clerk in a company office in one State rather than another. Surely it is highly anomalous that the Due Process Clause should require the dividend income of a farflung interstate business selectively to be attributed solely to the State or two in which a few minimal securities management functions are carried out, rather than apportioned among all the States whose "civilized society" has made the income-generating wealth of the larger enterprise possible. 90 Moreover, if such a requirement were judicially imposed it would create potentially staggering practical difficulties for taxpayers, state tax administrators, and, ultimately, the courts. For despite the Court's easy conclusion today that ASARCO's supposedly discrete investment business is distinct from ASARCO's operational nonferrous metals business, it is unlikely in practice that the two could be so readily disentangled. Imagine, for example, that the dividend checks were received and the management decisions regarding ASARCO's investments were made at ASARCO's corporate headquarters in one State, while the expertise and information relied on to make those decisions were drawn from corporate sources in many States. In apportioning the income of this purportedly separate investment business among the States, the question inescapably would arise as to what limits the Constitution places on how little of the taxable values at ASARCO's headquarters the expertise-and information-producing States could allocate to ASARCO's investment business as opposed to the theoretically distinct operational nonferrous metals business. Stating the question suffices to show that it reintroduces just the sort of insoluble problem of dividing businesses that the unitary-business principle was designed to avoid. Thus, if the Court does not abandon the separate-business theory that it endorses today, it merely will have substituted the vexing constitutional problem of how to apportion businesses for today's problem of how to apportion taxes. 91 In sum, the Court has erred. Without a well-founded constitutional mandate, it has straitjacketed the States' ability to develop fair systems of apportionment, prematurely ending the evolutionary process begun by the Uniform Division of Income for Tax Purposes Act and the Multistate Tax Commission. By limiting the apportionment concept by restrictions not found anywhere in the Constitution, moreover, the Court has committed itself to a path leading to more constitutional problems and greater involvement by this Court in the intricacies of interstate taxation. IV 92 The Court's error, moreover, is compounded by its decision to invoke the Due Process Clause as the source of its authority, despite the ready availability of the Commerce Clause.14 For unlike a Commerce Clause ruling which is susceptible to repair by Congress, today's due process decision may be beyond Congress' power to correct. 93 This constitutional shortsightedness overlooks the fact that Congress, not this Court, holds the ultimate responsibility for maintaining a healthy system of interstate commerce. Moreover, it is Congress, not this Court, which has the institutional tools to deal with these complex problems. Congress itself is only too aware of the limitations under which the judiciary operates when it attempts to deal with the knotty problems of state taxation of multistate enterprises within a federal system. As the Special Subcommittee on State Taxation of the Committee on the Judiciary of the House of Representatives bluntly put it: 94 "[T]he courts have over the years attempted to resolve the numerous and complex problems [of state taxation of interstate commerce] brought before them. . . . [T]heir decisions on State taxation leave much to be desired both in individual cases and as a body of law. The reason for this inadequacy is completely unrelated to the ability or diligence of a particular court or of any particular judge. The inadequacy is entirely institutional. 95 "The problem arises from the fact that a court deals in absolutes, and in this area an absolute decision in either direction is not likely to be satisfactory. . . . [T]he court is substantially handicapped by its inability to explore fully the nature, the extent, and the impact of the burdens created [by state taxes]. 96 "The inherent inadequacy of the judicial process to achieve a full accommodation of the competing demands of the States for taxes and of the national interest in unhindered commerce, is perhaps nowhere more clear than in the apportionment of income for tax purposes. . . . In the typical case of this kind, the tax of only one of the States would be before the court . . . . The court has only the records of the cases before it with only such information as may be necessary to state the facts and consequences in those cases. On this basis, is the court in a position to choose between [competing approaches], striking down all formulas containing one or the other? If so, what standard should it use in deciding which one? Given the adversary system of litigation, how does the court obtain the necessary data on economic burdens and revenue consequences? . . . 97 "Difficulties also arise from the limitations of the judicial process in prescribing what constitutes adequate jurisdiction to tax. As a question of due process the court can do no more than decide on conceptual grounds . . . whether the quality of the relationship in the case before it is sufficient to sustain the imposition of the tax. . . . 98 ". . . It is no better suited to devise and prescribe general rules setting the optimum level of jurisdiction than it is to impose a uniform apportionment formula. For example, the judicial process does not lend itself to a determination of what level of nexus would strike the most equitable balance between the demands of the States for revenue and the probable burdens of compliance." H.R.Rep.No.1480, 88th Cong., 2d Sess., 11-12 (1964). 99 Nor is Congress alone in recognizing the limitations of the judiciary in this field. Many Justices of this Court have acknowledged "the weakness of the judicial process in these tax questions where the total problem . . . reaches us only in installments." Northwest Airlines, Inc. v. Minnesota, 322 U.S. 292, 307, 64 S.Ct. 950, 957, 88 L.Ed. 1283 (1944) (Jackson, J., concurring). The Court itself has said: "To introduce a new doctrine of tax apportionment . . . is not merely to indulge in constitutional innovation. It is to introduce practical dislocation into the established taxing systems of the States. . . . [C]ertainly we ought not to embarrass the future by judicial answers which at best can deal only in a truncated way with problems sufficiently difficult even for legislative statesmanship." Id., at 299-300, 64 S.Ct., at 954 (opinion of the Court). Surely in a case such as the one before us, Congress, unconfined by "the narrow scope of judicial proceedings," Pennsylvania v. Wheeling & Belmont Bridge Co., 13 How. 518, 592, 14 L.Ed. 249 (1852) (Taney, C. J., dissenting), is in a better position "in the exercise of its plenary constitutional control over interstate commerce, not only [to] consider whether such a tax as now under scrutiny is consistent with the best interests of our national economy, but . . . also on the basis of full exploration of the many aspects of a complicated problem [to] devise a national policy fair alike to the States and our Union." McCarroll v. Dixie Greyhound Lines, Inc., 309 U.S. 176, 189, 60 S.Ct. 504, 510, 84 L.Ed. 683 (1940) (Black, J., dissenting). But it is just this sort of congressional action which today's due process decision appears to preclude. This Court should not so confidently pre-empt the Congress. V 100 In sum, the Court has focused its attention solely on the question whether ASARCO's interests in its subsidiaries represented active investments and concludes they did not. The Court then permits this initial erroneous result to derail its analysis. Instead of continuing, the Court fails to consider the possibility that ASARCO's investment decisionmaking was not segregated from its operational nonferrous metals business; fails to consider the possibility that ASARCO's investments were simply an interim use of long-term funds accumulated for ultimate use elsewhere in the business; fails to consider the possibility that ruling out apportioned taxation of income earned from intangibles may imply that such income is "nowhere income"; fails to consider the possibility that its ruling may be inconsistent with the unitary-business principle because it suggests that income from intangibles may be taxed only by a domiciliary State; and fails to consider the possibility that it may be as difficult to apportion a business as to apportion income for constitutional purposes. Finally, and most distressingly, the Court fails to consider its own limitations and Congress' constitutional prerogatives. Had the Court given the intricate questions presented by this case the attention they deserve, it might have reached a different result. I respectfully dissent. 1 ASARCO also received other intangible income, but the proper tax treatment of that income is not at issue in this case. 2 M. I. M. Holdings, Ltd., is a publicly owned corporation engaged in the mining, milling, smelting, and refining of nonferrous metals in Australia and England. ASARCO owned about 53% of M. I. M.'s stock during the period in question. General Cable Corp. and Revere Copper and Brass, Inc., are publicly owned companies that respectively fabricate cables and manufacture copper wares. ASARCO owned about 34% of the stock of each. ASARCO Mexicana, S. A., engages in Mexico in the same general line of business as does ASARCO in the United States. ASARCO owned 49% of Mexicana. Southern Peru Copper Corp. mines and smelts copper in Peru. ASARCO owned about 51.5% of Southern Peru during the time at issue. 3 The UDITPA is a tax allocation system approved in 1957 by the National Conference of Commissioners on Uniform State Laws and by the American Bar Association. See 7A U.L.A. 91 (1978). At least 23 States have adopted substantially all of the UDITPA to date. See id., at 10 (Supp.1982); Brief for State of Illinois as Amicus Curiae 1-2. The UDITPA has been adopted as Article IV of the Multistate Tax Compact. See United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S. 452, 457-458, n. 6, 98 S.Ct. 799, 804-805, n. 6, 54 L.Ed.2d 682 (1978). 4 Idaho Code § 63-3027(a)(1) (Supp.1981). The complete definition provides that " '[b]usiness income' means income arising from transactions and activity in the regular course of the taxpayers' trade or business and includes income from the acquisition, management, or disposition of tangible and intangible property when such acquisition, management, or disposition constitute[s] integral or necessary parts of the taxpayers' trade or business operations. Gains or losses and dividend and interest income from stock and securities of any foreign or domestic corporation shall be presumed to be income from intangible property, the acquisition, management, or disposition of which constitute an integral part of the taxpayers' trade or business; such presumption may only be overcome by clear and convincing evidence to the contrary." Ibid. (emphasis added). See UDITPA, § 1(a), 7A U.L.A. 93 (1978). 5 "All business income shall be apportioned to this state by multiplying the income by a fraction, the numerator of which is the property factor plus the payroll factor plus the sales factor, and the denominator of which is three (3)." Idaho Code § 63-3027(i) (Supp.1981). "The property factor is a fraction, the numerator of which is the average value of the taxpayer's real and tangible personal property owned or rented and used in this state during the tax period and the denominator of which is the average value of all the taxpayer's real and tangible personal property owned or rented and used during the tax period." § 63-3027(j). "The payroll factor is a fraction, the numerator of which is the total amount paid in this state during the tax period by the taxpayer for compensation, and the denominator of which is the total compensation paid everywhere during the tax period." § 63-3027(m). "The sales factor is a fraction, the numerator of which is the total sales of the taxpayer in this state during the tax period, and the denominator of which is the total sales of the taxpayer everywhere during the tax period." § 63-3027(o ). 6 "Capital gains and losses from sales of intangible personal property are allocable to this state if the taxpayer's commercial domicile is in this state, unless such gains and losses constitute business income as defined in this section." Idaho Code § 63-3027(f)(3) (Supp.1981). "Interest and dividends are allocable to this state if the taxpayer's commercial domicile is in this state unless such interest or dividends constitute business income as defined in this section." § 63-3027(g). Idaho defines "commercial domicile" as "the principal place from which the trade or business of the taxpayer is directed or managed." § 63-3027(a)(2). 7 Presently 19 States and the District of Columbia have joined the Compact as full members. Eleven States have joined as associate members. Brief for Multistate Tax Commission and Participating States as Amici Curiae 2. 8 Idaho law provides that "two . . . or more corporations the voting stock of which is more than fifty percent . . . owned directly or indirectly by a common owner or owners may, when necessary to accurately reflect income, be considered a single corporation." Idaho Code § 63-3027(s) (Supp.1981). The six unitized subsidiaries are Federated Metals of Canada; ASARCO Mercantile Co.; Enthone, Inc.; International Mining Co.; Lone Star Lead Construction Corp.; and Northern Peru Mining Corp. The auditor also stated that Southern Peru Copper Corp. and Southern Peru Copper Sales Corp. "were deemed unitary and were combined only for those states in which ownership of less than 80% presents no problem." App. 88a. See Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 435, n. 12, 100 S.Ct. 1223, 1230, n. 12, 63 L.Ed.2d 510 (1980). These two corporations were not "deemed unitary" in Idaho. 9 The auditor noted that ASARCO Mercantile bought and sold equipment for the subsidiaries and that International Metals acted as ASARCO's foreign sales agent. He further observed that exploration, research and development, insurance procurement, and tax preparation were performed jointly for most or all of these companies. Finally, he stated that ASARCO's audit staff examined the operations of these subsidiaries to enable "ASARCO to know whether the subsidiaries are operating along the lines set down by its management." App. 90. 10 The Idaho Supreme Court also ruled that ASARCO's receipt of certain rents and royalties, as well as its receipt of dividends from Compania American Smelting, S. A., constituted apportionable "business" income. It further upheld the trial court's "nonbusiness" income classification with respect to dividends received by ASARCO from Lake Asbestos of Quebec, ASARCO International Corp., Hecla Mining Co., Kennecott Copper Co., Phelps-Dodge, and United Park City Mines. American Smelting & Refining Co. v. Idaho State Tax Comm'n, 99 Idaho 924, 935-937, 592 P.2d 39, 50-52 (1979). These rulings are not disputed here. 11 See Rudolph, State Taxation of Interstate Business: The Unitary Business Concept and Affiliated Corporate Groups, 25 Tax L.Rev. 171, 181 (1970). 12 That court examined Exxon's organizational structure and business operations. It found that Exxon's full utilization of its production and refining functions were dependent on its marketing system, which encompassed Exxon's Wisconsin activities. The court therefore agreed with the Circuit Court for Dane County that Exxon's marketing in Wisconsin was an integral part of one unitary business and, consequently, that its total corporate income derived from the United States was subject to Wisconsin's apportioned taxation. See 447 U.S., at 217-219, 100 S.Ct., at 2117-2118. 13 It was noted that Exxon's Coordination and Services Management office "provided many essential corporate services for the entire company, including the coordination of the refining and other operational functions 'to obtain an optimum short range operating program.' . . . Many of the items sold by appellant in Wisconsin were obtained through a centralized purchasing office in Houston whose obvious purpose was to increase overall corporate profits through bulk purchases and efficient allocation of supplies among retailers. . . . Even the gasoline sold in Wisconsin was available only because of an exchange agreement with another company arranged by the Supply Department, part of Coordination and Services Management, and the Refining Department. Similarly, sales were facilitated through the use of a uniform credit card system, uniform packaging, brand names, and promotional displays, all run from the national headquarters." Id., at 224, 100 S.Ct., at 2120. 14 The unitary-business principle applied in Mobil and Exxon is not new. It has been a familiar concept in our tax cases for over 60 years. See United States Steel Corp. v. Multistate Tax Comm'n, 434 U.S., at 473 n. 25, 474, n. 26, 98 S.Ct., at 812, n. 25, 813, n. 26; General Motors Corp. v. Washington, 377 U.S. 436, 439, 84 S.Ct. 1564, 1567, 12 L.Ed.2d 430 (1964); Northwestern States Portland Cement Co. v. Minnesota, 358 U.S. 450, 460, 79 S.Ct. 357, 363, 3 L.Ed.2d 421 (1959); Butler Bros. v. McColgan, 315 U.S. 501, 508, 62 S.Ct. 701, 704, 86 L.Ed. 991 (1942); Ford Motor Co. v. Beauchamp, 308 U.S. 331, 336, 60 S.Ct. 273, 276, 84 L.Ed. 304 (1939); Norfolk & Western R. Co. v. North Carolina, ex rel. Maxwell, 297 U.S. 682, 684, 56 S.Ct. 625, 626, 80 L.Ed. 977 (1936); Hans Rees' Sons, Inc. v. North Carolina, ex rel. Maxwell, 283 U.S. 123, 132-133, 51 S.Ct. 385, 388, 75 L.Ed. 879 (1931); Bass, Ratcliff & Gretton, Ltd. v. State Tax Comm'n, 266 U.S. 271, 282, 45 S.Ct. 82, 84, 69 L.Ed. 282 (1924). Cf. Burnet v. Aluminum Goods Mfg. Co., 287 U.S. 544, 550, 53 S.Ct. 227, 230, 77 L.Ed. 484 (1933); Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113, 120-121, 41 S.Ct. 45, 46-47, 65 L.Ed. 165 (1920); Wallace v. Hines, 253 U.S. 66, 69, 40 S.Ct. 435, 437, 64 L.Ed. 782 (1920); Fargo v. Hart, 193 U.S. 490, 499-500, 24 S.Ct. 498, 499-500, 48 L.Ed. 761 (1904); Adams Express Co. v. Ohio State Auditor, 165 U.S. 194, 221-222, 17 S.Ct. 305, 309, 41 L.Ed. 683 (1897); Adams Express Co. v. Ohio State Auditor, 166 U.S. 185, 219, 222, 223-224, 17 S.Ct. 604, 606, 607, 41 L.Ed. 965 (1897). A review of our cases before Mobil made plain that "[f]ormulary apportionment, which takes into account the entire business income of a multistate business in determining the income taxable by a particular state, is constitutionally permissible only in the case of a unitary business." Rudolph, supra n. 11, at 183-184. 15 The other shareholders are Phelps Dodge, 16%; Newmont Mining, 10.25%; and Cerro Corp., 22.25%. App. 86a. Either these large companies or their parents were traded on the New York Stock Exchange at the time in question. 16 Southern Peru Copper Sales Corp. in turn sells the copper to European customers. These European sales are handled in this manner to preserve Southern Peru's favorable federal tax status as a Western Hemisphere Trading Corporation. Southern Peru Copper Sales Corp.'s stock is owned in the same manner as is Southern Peru's. Unlike Southern Peru, however, Southern Peru Copper Sales Corp. has no employees of its own. Sales are generally transacted by ASARCO's New York office, for which that office earns a sales commission. Ibid. 17 These sales ranged between $44 and $65 million for the years in question. Id., at 89a. There was evidence that ASARCO could replace this output contract "[w]ithin a short time" if it were lost, and that loss of ASARCO's ownership in Southern Peru would not cause the loss of the output contract. Id., at 128a. Southern Peru has a "staff that you'd expect a major corporation to have." Id., at 122a. ASARCO provided Southern Peru with purchasing service outside of Peru, traffic service for its exports and imports outside of Peru, and preparation service for its United States tax return. Id., at 123a-124a. The contract for ASARCO's purchasing services provided for payment of this service on the basis of a fixed fee plus a commission based on the dollar volume of purchases. Id., at 124a. ASARCO received separate "negotiated fair fee[s]" for its tax and traffic services. Ibid. Southern Peru has its own purchasing, traffic, and tax departments in Peru. Id., at 123a. 18 M. I. M. did use an ASARCO melting furnace patent for which it pays a price "the same that would be paid by any other company using it." Id., at 133a. 19 For the years in question, Revere's purchases averaged 3-4% of ASARCO's sales and totalled from $17 to $29 million. Id., at 27a. Revere in turn sold ASARCO from 1 to 2% of its total output, which totalled $4-$6 million. Id., at 43a-47a. General Cable's purchases accounted for approximately 6% of ASARCO's sales and ranged from $31 to $47 million. Id., at 27a. ASARCO's purchases from General Cable averaged 0.1% of General Cable's total sales and ranged between $0.3 and $0.5 million. Id., at 43a-47a. 20 Both Revere and General Cable utilized ASARCO's stock transfer department on a contract basis, and Revere licensed one patent from ASARCO for a "fair price." Id., at 136a. In 1970, ASARCO was compelled, apparently by the Department of Justice, to divest itself of all its General Cable stock. Id., at 95a. 21 ASARCO sold Mexicana none of its output in 1968 and insignificant amounts (totalling $24,169 and $14,902) in 1969 and 1970. Id., at 27a. Mexicana apparently did not sell ASARCO any of its output during the time in question. Id., at 43a-47a. For a commission, ASARCO does act as a contract sales agent for Mexicana in the United States. Id., at 131a. This contract would continue if ASARCO lost its investment interest in Mexicana. Ibid. ASARCO also provides technical services to Mexicana for a fee. Id., at 130a; 99 Idaho, at 929, 592 P.2d, at 44. The dissent's perception of some of the facts differs substantially from the record. It speculates that ASARCO's unitary-business experience "must" have aided ASARCO's stock investments, post, at 336, despite the undisputed trial court finding that ASARCO's stock investments were "not integral to nor a necessary part of [ASARCO's] business operation . . . ." App. to Juris. Statement 44a. See also id., at 43a. It maintains that "[f]or all we know"—ASARCO's stock investments were interim uses of idle funds "accumulated for the future operation of [ASARCO's] own primary business." Post, at 337. The trial court, however, found that ASARCO "has never been required to utilize its stock as security for borrowing of working capital, acquiring stock or securities in other companies or to support any bond issues." App. to Juris. Statement 41a. Moreover, ASARCO was found to have "sufficient cash flow from mining to provide operating capital for all mining operations without reliance upon cash flow from . . . income from intangibles." Ibid. The dissent also describes the five companies as "captive suppliers and customers. . . ." Post, at 342. This description is at odds with the undisputed facts. See supra, at 320-324. 22 Cf. Keesling & Warren, California's Uniform Division of Income for Tax Purposes Act, Part I, 15 UCLA L.Rev. 156, 172 (1967). Such arbitrariness is evident in this case. As previously noted, see n. 10, supra, ASARCO received dividend income from Lake Asbestos of Quebec. Lake Asbestos is a wholly owned subsidiary of ASARCO that is engaged in extracting and processing asbestos fibers in Canada. App. 86a. Its selling agent is ASARCO International, another wholly owned ASARCO subsidiary. Ibid. The Idaho Supreme Court found that "Lake Asbestos has its own staff, but ASARCO provides important management services. Lake Asbestos seeks ASARCO's direction and approval on major policy decisions." 99 Idaho, at 929, 592 P.2d, at 44. Lake Asbestos consequently appears in many respects to be more likely to qualify as part of ASARCO's unitary business than does any of the five corporations involved in this case. Yet Idaho now agrees, as the courts below held, that on this record Lake Asbestos is not part of such a business. See Tr. of Oral Arg. 39. ASARCO also received approximately $250,000 a year in dividends from Kennecott Copper Corp., and approximately $300,000 a year in dividends from Phelps Dodge. App. 52a. The record describes the business of these two corporations as the mining, smelting, refining, and sale of copper. Id., at 50a. ASARCO made sales in the range of $3-$5 million a year to Kennecott, and $24,000-$800,000 a year to Phelps Dodge. Id., at 27a. Each in turn made some (albeit "minimal") sales to ASARCO. Id., at 43a-47a. In light of this information, we have no basis for concluding that ASARCO did not "acquire" stock in Kennecott and Phelps Dodge "for purposes relating or contributing to the taxpayer's business." Brief for Appellee 4. ASARCO's management scarcely would make such an admission. Yet Idaho makes no claim that Kennecott and Phelps Dodge are part of ASARCO's unitary business. Justice O'CONNOR's dissent views the Court's decision as "prohibiting apportioned taxation of investment income by nondomiciliary states." Post, at 3124, et seq. This reflects a serious misunderstanding of our decision today and the cases on which we rely. The case we follow primarily is Mobil. It sustained the taxation of investment income after applying enunciated principles carefully to the facts of the case. In this case we have applied the same principles but have reached a different result because the facts differ in critical respects. As we have said elsewhere, the application of the unitary-business principle requires in each case a careful examination both of the way in which the corporate enterprise is structured and operates, and of the relationship with the taxing State. 23 The dissent, argues that our reliance on the Due Process Clause is inappropriate. It also says that our holding that Idaho has exceeded its jurisdiction to tax somehow "strip[s] Congress of the authority" to authorize or regulate state taxation. Post, at 331. See also post, at 349-350, 353. In analyzing the validity of Idaho's tax, we follow long-established precedent in relying on the Due Process Clause of the Fourteenth Amendment. See, e.g., Exxon, 447 U.S., at 219, 221-225, 226, 227, 100 S.Ct., at 2118, 2119-2121, 2122; Mobil, 445 U.S., at 436-442, 100 S.Ct., at 1231-1234; Butler Bros. v. McColgan, 315 U.S., at 507, 508, 62 S.Ct., at 704; Underwood Typewriter Co. v. Chamberlain, 254 U.S., at 120-121, 41 S.Ct., at 46-47. In view of our decision on due process, it is unnecessary to address appellant's Commerce Clause argument. In any event, it is elementary that the "States . . . are subject to limitations on their taxation powers that do not apply to the Federal Government." F. W. Woolworth Co. v. Taxation & Revenue Dept., 458 U.S. 363, 102 S.Ct. 3128, 73 L.Ed.2d 819. Cf. Insurance Corp. of Ireland, Ltd. v. Compagnie des Bauxites de Guinee, 456 U.S. 694, 713, 102 S.Ct. 2099, 2110, 72 L.Ed.2d 492 (1982) (POWELL, J., concurring in judgment). The question of federal authority to legislate in this area—whether to lay taxes or to delegate such power—is not presented in this case, and we imply no view as to it. 24 The dissenting opinion reflects profound—though unexpressed—dissatisfaction with the unitary-business principle, even though it was firmly established by more than a half a dozen decisions of this Court prior to Mobil and Exxon. See n. 14, supra. The dissent purports to rely on these recent cases, and yet its basic arguments—in practical effect—would seriously undermine their force as precedents. It relies primarily on considerations quite different from those identified as controlling in Mobil and Exxon. The dissent does not deny that ASARCO's subsidiaries were discrete business enterprises; rather it submits that they were engaged "in the same general line of business." Post, at 335. It notes—though the relevance is not obvious—that the management of ASARCO had special knowledge of the types of business engaged in by these subsidiaries. Post, at 336-337. The dissent also perceives a relationship between Idaho and the investment income simply because ASARCO has the use in its business of income from whatever source it may be derived. Post, at 337-339. Finally, it emphasizes the limited amount of open market buying and selling of products between ASARCO and the companies in which it has invested funds. See Parts III-A and III-B, supra. In Mobil, in applying the unitary-business principle, the Court stated that the question is whether "the [investment] income was earned in the course of activities unrelated to the sale of petroleum products" in the State seeking to tax the income. Our decision went against Mobil Oil Corp. because we found that its "foreign activities [were] part of appellant's integrated petroleum enterprise," and because the subsidiaries in question were not shown to operate as "discrete business enterprise[s]." 445 U.S., at 439, 100 S.Ct., at 1232. In this case, in sharp contrast, the record establishes that each of the three partial subsidiaries in question operated a "discrete business enterprise" having "nothing to do with the activities of [ASARCO] in the taxing State." Id., at 442, 100 S.Ct., at 1234. As we recognize in this opinion, these cases are decided on their facts in light of established general principles. The most comprehensive discussion of the factors that are relevant is contained in our recent decisions in Mobil and Exxon. In both of those cases, that we follow today, the States prevailed because it was clear that the corporations operated unitary businesses with a continuous flow and interchange of common products. ASARCO has proved that these essential factors are wholly absent in this case. It is late in the day to confuse this important area of state tax law by rewriting the standards of Mobil and Exxon. 25 ASARCO has not challenged Idaho's taxation of interest and capital gains income other than that attributable to the five corporations discussed in the text. Brief for Appellant 26. We do not intimate views on such matters. 1 For example, if the expenses incurred by a vertically integrated business in manufacturing a product in State A are deducted from the revenues generated by marketing the product in State B, the resulting net income cannot reasonably be allocated either entirely to State A or entirely to State B. See Underwood Typewriter Co. v. Chamberlain, 254 U.S. 113, 120-121, 41 S.Ct. 45, 46-47, 65 L.Ed. 165 (1920). Even when state-by-state allocation superficially seems feasible, as in the case of a horizontally integrated, but loosely knit chain of retail stores, "economies of scale" resulting from "functional integration" and "centralization of management," Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S. 425, 438, 100 S.Ct. 1223, 1232, 63 L.Ed.2d 510 (1980), may augment the net income earned by the whole to a value greater than the sum of what could be earned by each of the parts operating separately. See Butler Bros. v. McColgan, 315 U.S. 501, 508-509, 62 S.Ct. 701, 704, 86 L.Ed. 991 (1942). When this happens, the additional increment in income resulting from the combination cannot be allocated to any single one of the parts on other than an arbitrary basis. See, e.g., Adams Express Co. v. Ohio State Auditor, 165 U.S. 194, 221, 17 S.Ct. 305, 309, 41 L.Ed. 683 (1897) (a unitary enterprise has a "value resulting from the combination of the means by which the business [is] carried on [which] exist[s] . . . throughout the entire domain of [its] operation"). The net income of an organic, unitary business, in short, is indivisible. 2 Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, 445 U.S., at 439, 100 S.Ct., at 1232. 3 Exxon Corp. v. Wisconsin Dept. of Revenue, 447 U.S. 207, 224, 100 S.Ct. 2109, 2120, 65 L.Ed.2d 66 (1980). 4 Mobil Oil Corp. v. Commissioner of Taxes of Vermont, supra, 445 U.S., at 442, 100 S.Ct., at 1234. 5 ASARCO states that it "has not contested Idaho's right to treat interest income from temporary deposits of [its] working capital funds as apportionable business income derived in the ordinary course of [its] Unitary Business activities." Brief for Appellant 26. See also Tr. of Oral Arg. 6-7. 6 This process is analogous to the leasing of idle physical assets until they are needed in the business. By analogy to ASARCO's investment of its working capital in short-term securities, for example, a company might lease time on one of its computers to outsiders on an hourly basis in order to keep the computer fully occupied during slack periods. Similarly, in analogy to the interim investment of retained earnings, a company might lease for a term of years the areas of its office buildings into which it intends ultimately to expand. It could hardly be claimed that by so doing the company had opened up a separate and unrelated leasing business. To the contrary, the income from such leases would be functionally related to the company's unitary business and, therefore, taxable on an apportioned basis by a State in which the company did business. Accordingly, it is hard to see why a company that rents out idle money rather than idle physical assets should be treated differently under the Due Process Clause, and harder still to see why the Constitution would treat short-term investments of working capital differently than longer-term investments of retained earnings. 7 Of course, had ASARCO attempted to do so, it might have been able to make such a showing. ASARCO did argue, and the trial court found, that it had "never been required to utilize its stock as security for borrowing of working capital, acquiring stock or securities in other companies or to support any bond issues." Record 326. 8 Under the management agreement between ASARCO and the other three shareholders of Southern Peru, ASARCO has the right to appoint 6 of the company's 13 directors, the other three shareholders together have the right to appoint another 6, and the 13th and final director is appointed by the first 12 directors or by joint action of all the shareholders. Southern Peru's bylaws, which can be changed only by unanimous consent, provide that eight votes are needed to pass any resolution. App. to Juris. Statement 55a; App. 121a. 9 ASARCO also dominated Revere Copper and General Cable. While ASARCO owned only 34% interests in each company, the remaining shares of each were "widely held," ante, at 324, so that ordinarily ASARCO would be assured control of each company and almost never face effective opposition to its wishes. Prior to the tax years in question, however, the Justice Department brought an antitrust suit against ASARCO related to its holdings in General Cable and Revere Copper. The suit culminated in a consent decree issued in 1967 forbidding ASARCO to sell to General Cable and Revere Copper at prices lower than ASARCO sold to other customers and preventing ASARCO from voting its stock in either company. In 1970, the last of the three tax years at issue here, the decree was modified and "ASARCO was compelled . . . to divest itself of all its General Cable stock." Ante, at 324, n. 20. ASARCO argues that the 1967 consent decree, which was issued before the tax years of concern here, prevented General Cable and Revere Copper from being a part of ASARCO's unitary business. To the contrary, however, the antitrust suit and consent decree strongly suggest that General Cable and Revere Copper were entangled in ASARCO's unitary business, perhaps unlawfully so. On the present record, ASARCO has not borne the burden of showing that the 1967 decree severed enough of the connections between ASARCO and the two subsidiaries to transform ASARCO's holdings from active components of its unitary business to passive investments. At least with respect to General Cable, the modification in 1970 (the last tax year in issue here) of the consent decree so that ASARCO was required to sell its holdings in the company indicates the contrary, since divestiture would scarcely be necessary if the business of the two companies had been unrelated. Moreover, the mere fact that ASARCO traded with General Cable and Revere Copper at market rates does not compel the conclusion that ASARCO's investments in those companies were part of its unitary business. In Exxon, for example, we concluded that the fact that "wholesale market values" were assigned "to interdepartmental transfers of products and supplies," 447 U.S., at 225, 100 S.Ct., at 2121, did not undermine the unitary nature of Exxon's business. Although I believe that ASARCO's showing was insufficient to establish that General Cable and Revere Copper were not under its control for purposes of determining the constitutional limits of Idaho's taxation, for simplicity in what follows I assume only that ASARCO was a major investor in these two subsidiaries. 10 Just as inexplicably, the Court reverses Idaho's apportioned tax on ASARCO's dividend income from its 34% interests in Revere Copper & Brass and in General Cable Corp., companies that were "major customers" of ASARCO buying tens of millions of dollars of goods from ASARCO each year. 11 The Court suggests that my "perception of some of the facts" necessary to reach this conclusion "differs substantially from the record." Ante, at 325, n. 21. In fact, however, my view of the facts differs from neither the record nor, I think, that of the Court. As should be apparent, my disagreement with the Court is based, not on the facts, but on the constitutional significance to be given those facts. 12 The Court is careful not to extend the reach of its holding to domiciliary States. It states the question presented as "whether the State of Idaho constitutionally may include within the taxable income of a non domiciliary parent corporation . . . a portion of intangible income . . . that the parent receives from subsidiary corporations having no other connection with the State." Ante, at 308-309 (emphasis added). As its holding, the Court asserts that it "cannot accept . . . a definition of 'unitary business' that would permit non domiciliary States to apportion and tax dividends '[w]here the business activities of the dividend payor have nothing to do with the activities of the recipient in the taxing State. . . .' " Ante, at 327 (citation omitted, emphasis added). 13 See Dexter, Taxation of Income from Intangibles of Multistate-Multinational Corporations, 29 Vand.L.Rev. 401, 403 (1976). 14 This Court's authority to invalidate state legislation because it interferes with interstate commerce is inferred from the Constitution's grant to Congress of the authority to regulate interstate commerce. Art. I, § 8, cl. 3. For this reason, Congress may "confe[r] upon the States an ability to restrict the flow of interstate commerce that they would not otherwise enjoy." Lewis v. BT Investment Managers, Inc., 447 U.S. 27, 44, 100 S.Ct. 2009, 2019, 64 L.Ed.2d 702 (1980) (citations omitted). Consistent with this principle, it has long been established that Congress generally has the power to "overrule" a decision of this Court invalidating state legislation on Commerce Clause grounds. Compare Leisy v. Hardin, 135 U.S. 100, 10 S.Ct. 681, 34 L.Ed. 128 (1890), with In re Rahrer, 140 U.S. 545, 11 S.Ct. 865, 35 L.Ed. 572 (1891). By contrast, Congress generally cannot waive a ruling of this Court decided under the Due Process Clause. Accordingly, this Court's "threshold" for invalidating state legislation should be considerably higher under the Due Process Clause than under the Commerce Clause. In the present case, the Court could have reached its result by relying on the Commerce Clause. Our cases establish that analysis of the validity of state taxation under the Commerce Clause is similar to analysis under the Due Process Clause. The test under the Commerce Clause is whether the tax " 'is applied to an activity with a substantial nexus with the taxing State, is fairly apportioned, does not discriminate against interstate commerce, and is fairly related to the services provided by the State.' " Mobil Oil Corp. v. Commissioner of Taxes of Vermont, 445 U.S., at 443, 100 S.Ct., at 1234, quoting Complete Auto Transit, Inc. v. Brady, 430 U.S. 274, 279, 97 S.Ct. 1076, 1079, 51 L.Ed.2d 326 (1977). In his dissent in Mobil Oil, Justice STEVENS explained how a violation of the Due Process Clause could also be a violation of the Commerce Clause: "[I]f, in a particular case, use of an allocation formula has the effect of taxing income earned by an interstate entity outside the State, it could alternatively be said to have the effect of taxing the income earned by that entity inside the State at a rate higher than that used for a comparable, wholly intrastate business, a discrimination that violates the Commerce Clause." 445 U.S., at 452, n. 4, 100 S.Ct., at 1239, n. 4.

78

458 U.S. 279 102 S.Ct. 3088 73 L.Ed.2d 767 William Archie WILLIAMS, Petitionerv.UNITED STATES. No. 80-2116. Argued April 20, 1982. Decided June 29, 1982. Syllabus Title 18 U.S.C. § 1014 makes it a crime to "knowingly mak[e] any false statement or report," or "willfully overvalu[e] any land, property or security," for the purpose of influencing the action of described financial institutions (including federally insured banks) "upon any application, advance, . . . commitment, or loan." Petitioner engaged in a series of transactions seemingly amounting to a case of "check kiting" between his accounts in federally insured banks, first drawing a check far in excess of his account balance in one bank and depositing it in his account in the other, and then reversing the process between his accounts. Petitioner was convicted in Federal District Court of violating § 1014, and the Court of Appeals affirmed. Held : Petitioner's conduct in depositing "bad checks" in federally insured banks is not proscribed by § 1014. Pp. 284-290. (a) Petitioner's actions did not involve the making of a "false statement." Technically speaking, a check is not a factual assertion at all, and therefore cannot be characterized as "true" or "false." Similarly, petitioner's conduct cannot be regarded as "overvalu[ing]" property or a security. In a literal sense, the face amounts of the checks were their "values." To interpret § 1014 as meaning that a drawer of a check has made a "false" statement whenever he has insufficient funds in his account at the moment the check is presented would "sligh[t] the wording of the statute" United States v. Enmons, 410 U.S. 396, 399, 93 S.Ct. 1007, 1009, 35 L.Ed.2d 379, and would render a wide range of unremarkable conduct violative of federal law. When § 1014 was enacted, federal action was not necessary to interdict the deposit of bad checks, for fraudulent checking activities already were addressed in comprehensive fashion by state law. Pp. 284-287. (b) The legislative history does not support the proposition that § 1014 was designed to have general application to the passing of worthless checks, and does not demand that the statute be read as applicable to anything other than representations made in connection with conventional loan or related transactions. A narrow interpretation of § 1014 is consistent with the usual approach of lenity in the construction of criminal statutes. Pp. 288-290. 639 F.2d 1311 (C.A.5 1981), reversed and remanded. Nickolas P. Chilivis, Atlanta, Ga., for petitioner. Richard G. Wilkins, Washington, D. C., for respondent, pro hac vice, by special leave of Court. Justice BLACKMUN delivered the opinion of the Court. 1 In this case we must decide whether the deposit of a "bad check" in a federally insured bank is proscribed by 18 U.S.C. § 1014. 2 * In 1975, petitioner William Archie Williams purchased a controlling interest in the Pelican State Bank in Pelican, La., and appointed himself president. The bank's deposits were insured by the Federal Deposit Insurance Corporation. 3 Among the services the bank provided its customers at the time of petitioner's purchase was access to a "dummy account," used to cover checks drawn by depositors who had insufficient funds in their individual accounts. Any such check was processed through the dummy account and paid from the bank's general assets. The check was then held until the customer covered it by a deposit to his own account, at which time the held check was posted to the customer's account and the dummy account was credited accordingly. As president of the bank, petitioner enjoyed virtually unlimited use of the dummy account, and by May 2, 1978, his personal overdrafts amounted to $58,055.44, approximately half the total then covered by the account. 4 On May 8, 1978, federal and state examiners arrived at the Pelican Bank to conduct an audit. That same day, petitioner embarked on a series of transactions that seemingly amounted to a case of "check kiting."1 He began by opening a checking account with a deposit of $4,649.97 at the federally insured Winn State Bank and Trust Company in Winnfield, La. The next day, petitioner drew a check on his new Winn account for $58,500—a sum far in excess of the amount actually on deposit at the Winn Bank—and deposited it in his Pelican account. Pelican credited his account with the face value of the check, at the same time deducting from petitioner's account the $58,055.44 total of his checks that previously had been cleared through the dummy account. At the close of business on May 9, then, petitioner had a balance of $452.89 at the Pelican Bank. 5 On May 10, petitioner wrote a $60,000 check on his Pelican account—again, a sum far in excess of the account balance—and deposited it in his Winn account. The Winn Bank immediately credited the $60,000 to petitioner's account there, and Pelican cleared the check through its dummy account when it was presented for payment on May 11. The Winn Bank routinely paid petitioner's May 9 check for $58,500 when it cleared on May 12. 6 Petitioner next attempted to balance his Pelican account by depositing a $65,000 check drawn on his account at yet another institution, the Sabine State Bank in Many, La. Unfortunately, the balance in petitioner's Sabine account at the time was only $1,204.81. The Sabine Bank therefore refused payment when Pelican presented the check on May 17. On May 23, petitioner settled his Pelican account by depositing at the Pelican Bank a $65,000 money order obtained with the proceeds from a real estate mortgage loan. 7 The bank examiners, meanwhile, had been following petitioner's activities with considerable interest. Their scrutiny ultimately led to petitioner's indictment, in the United States District Court for the Western District of Louisiana, on two counts of violating 18 U.S.C. § 1014.2 That provision makes it a crime to 8 "knowingly mak[e] any false statement or report, or willfully overvalu[e] any land, property or security, for the purpose of influencing in any way the action of [certain enumerated financial institutions, among them banks whose deposits are insured by the Federal Deposit Insurance Corporation], upon any application, advance, discount, purchase, purchase agreement, repurchase agreement, commitment, or loan. . . ." 9 The first of the counts under § 1014 was directed at the May 9, 1978, check drawn on the Winn Bank, and charged that petitioner "did knowingly and willfully overvalue . . . a security, that is a check . . . for the purpose of influencing the Pelican State Bank, . . . a bank the deposits of which are insured by the Federal Deposit Insurance Corporation, upon an advance of money and extension of credit." The other § 1014 count used virtually identical language to indict petitioner for depositing in his Winn account the May 10 check drawn on the Pelican Bank. App. 3-4.3 10 At petitioner's trial the court charged the jury that "[a] check is a security for purposes of Section 1014." The court then explained that "[t]he Government charges that Mr. Williams was involved in check-kiting—a scheme whereby false credit is obtained by the exchange and passing of worthless checks between two or more banks." Id., at 36. To convict petitioner, the court continued, the jury had to find as to each count that "the defendant . . . did knowingly and willfully make a false statement of a material fact," that the statement "influence[d] the decision of the [bank] officers or employees," and that "the defendant made the false statement with fraudulent intent to influence the [bank] to extend credit to the defendant." Id., at 37-38. "The crucial question in check-kiting," the court concluded, "is whether the defendant intended to write checks which he could not reasonably expect to cover and thereby defraud the bank, or whether he was genuinely involved in the process of depositing funds and then making legitimate withdrawals against them." Id., at 38. The jury convicted petitioner on both counts, and he was sentenced to six months' incarceration on the second § 1014 count. For the first § 1014 count he was placed on five years' probation, to begin upon his release from confinement. App. 39.4 11 Among other things, petitioner argued on appeal that the indictment did not state a violation of § 1014. The Court of Appeals rejected this contention, however, concluding that petitioner's actions "constitute classic incidents of check kiting." 639 F.2d 1311, 1319 (C.A.5 1981). In line with its earlier decision in United States v. Payne, 602 F.2d 1215 (C.A.5 1979), cert. denied, 445 U.S. 903, 100 S.Ct. 1079, 63 L.Ed.2d 319 (1980), the court found such action proscribed by the statute. 12 We granted certiorari, limited to Questions 3 and 4 presented by the petition, in order to resolve a conflict concerning the reach of § 1014.5 454 U.S. 1030, 102 S.Ct. 565, 70 L.Ed.2d 473 and 454 U.S. 1096, 102 S.Ct. 668, 70 L.Ed.2d 637 (1981). II 13 To obtain a conviction under § 1014, the Government must establish two propositions: it must demonstrate (1) that the defendant made a "false statement or report," or "willfully overvalue[d] any land, property or security" and (2) that he did so "for the purpose of influencing in any way the action of [a described financial institution] upon any application, advance, . . . commitment, or loan." We conclude that petitioner's convictions under § 1014 cannot stand, because the Government has failed to meet the first of these burdens. A. 14 Although petitioner deposited several checks that were not supported by sufficient funds, that course of conduct did not involve the making of a "false statement," for a simple reason: technically speaking, a check is not a factual assertion at all, and therefore cannot be characterized as "true" or "false." Petitioner's bank checks served only to direct the drawee banks to pay the face amounts to the bearer, while committing petitioner to make good the obligations if the banks dishonored the drafts. Each check did not, in terms, make any representation as to the state of petitioner's bank balance. As defined in the Uniform Commercial Code, 2 U.L.A. 17 (1977), a check is simply "a draft drawn on a bank and payable on demand," § 3-104(2)(b), which "contain[s] an unconditional promise or order to pay a sum certain in money," § 3-104(1)(b). As such, "[t]he drawer engages that upon dishonor of the draft and any necessary notice of dishonor or protest he will pay the amount of the draft to the holder." § 3-413(2), 2 U.L.A. 424 (1977). The Code also makes clear, however, that "[a] check or other draft does not of itself operate as an assignment of any funds in the hands of the drawee available for its payment, and the drawee is not liable on the instrument until he accepts it." § 3-409(1), 2 U.L.A. 408 (1977). Louisiana, the site of petitioner's unfortunate banking career, embraces verbatim each of these definitions. See La.Rev.Stat.Ann. §§ 10:3-104, 10:3-409, 10:3-413 (West Supp.1982).6 15 For similar reasons, we conclude that petitioner's actions cannot be regarded as "overvalu[ing]" property or security. Even assuming that petitioner's checks were property or security as defined by § 1014, the value legally placed upon them was the value of petitioner's obligation; as defined by Louisiana law, that is the only meaning actually attributable to a bank check. See La.Rev.St.Ann., §§ 10:3-409(1), 10:3-413(2) (West) (Supp.1981). In a literal sense, then, the face amounts of the checks were their "values." 16 The foregoing description of bank checks is concededly a technical one, and the Government therefore argues with some force that a drawer is generally understood to represent that he "currently has funds on deposit sufficient to cover the face value of the check." Brief for United States 19. See United States v. Payne, 602 F.2d, at 1218. If the drawer has insufficient funds in his account at the moment the check is presented, the Government continues, he effectively has made a "false statement" to the recipient. While this broader reading of § 1014 is plausible, we are not persuaded that it is the preferable or intended one. It "slights the wording of the statute," United States v. Enmons, 410 U.S. 396, 399, 93 S.Ct. 1007, 1009, 35 L.Ed.2d 379 (1973), for, as we have noted, a check is literally not a "statement" at all. In any event, whatever the general understanding of a check's function, "false statement" is not a term that, in common usage, is often applied to characterize "bad checks." And, when interpreting a criminal statute that does not explicitly reach the conduct in question, we are reluctant to base an expansive reading on inferences drawn from subjective and variable "understandings."7 17 Equally as important, the Government's interpretation of § 1014 would make a surprisingly broad range of unremarkable conduct a violation of federal law. While the Court of Appeals addressed itself only to check kiting, its ruling has wider implications: it means that any check, knowingly supported by insufficient funds, deposited in a federally insured bank could give rise to criminal liability, whether or not the drawer had an intent to defraud. Under the Court of Appeals' approach, the violation of § 1014 is not the scheme to pass a number of bad checks; it is the presentation of one false statement—that is, one check that at the moment of deposit is not supported by sufficient funds—to a federally insured bank. The United States acknowledged as much at oral argument. Tr. of Oral Arg. 40. Indeed, each individual count of the indictment in this case stated only that petitioner knowingly had deposited a single check that was supported by insufficient funds, not that he had engaged in an extended scheme to obtain credit fraudulently.8 18 Yet, if Congress really set out to enact a national bad check law in § 1014, it did so with a peculiar choice of language and in an unusually backhanded manner. Federal action was not necessary to interdict the deposit of bad checks, for, as Congress surely knew, fraudulent checking activities already were addressed in comprehensive fashion by state law. See Comment, Insufficient Funds Checks in the Criminal Area: Elements, Issues, and Proposals, 38 Mo.L.Rev. 432 (1973). Absent support in the legislative history for the proposition that § 1014 was "designed to have general application to the passing of worthless checks," United States v. Krown, 675 F.2d 46, 50 (C.A.2 1982), we are not prepared to hold petitioner's conduct proscribed by that particular statute.9 B 19 In the 1948 codification of Title 18 of the United States Code, 62 Stat. 683, § 1014 reduced 13 existing statutes, which criminalized fraudulent practices directed at a variety of financial and credit institutions, to a single section. See 18 U.S.C. § 1014, Historical and Revision Notes. Of the originally enumerated institutions,10 only two—the Reconstruction Finance Corporation, see 15 U.S.C. § 616(a) (1946 ed.), and the Federal Reserve Banks, see 12 U.S.C. § 596 (1946 ed.)—performed duties other than the making of farm and home loans, and neither of those two organizations accepted checks for deposit from private customers. See United States v. Sabatino, 485 F.2d 540, 548 (C.A.2 1973), cert. denied, 415 U.S. 948, 94 S.Ct. 1469, 39 L.Ed.2d 563 (1974); United States v. Edwards, 455 F.Supp. 1354, 1357 (M.D.Pa.1978). It is evident, then, that bad checks were not among the "false statements" or "overvalued property" originally addressed by the statute. While Congress has added and subtracted certain institutions to and from the list covered by § 1014 over the intervening years, no changes have been made in the type of transactions proscribed by the provision. 20 The legislative history does not demand a broader reading of the statute. The amendments adding institutions to § 1014's list attracted little attention in Congress and were dealt with summarily; at no point was it suggested that the statute should be applicable to anything other than representations made in connection with conventional loan or related transactions. In 1964, for example, when Congress, by Pub.L.88-353, § 5, 78 Stat. 269, added Federal Credit Unions to the statutory list, § 1014 was described as barring "false statements or willful overvaluations in connection with applications, loans, and the like." S.Rep.No.1078, 88th Cong., 2d Sess., 1 (1964), U.S.Code Cong. & Admin.News 1964, pp. 2519, 2520. Thus, the Senate Committee on Banking and Currency declared that § 1014 "is designed primarily to apply to borrowers from Federal agencies or federally chartered organizations."11 Id., at 4, U.S.Code Cong. & Admin.News 1964, 2522. Similarly, the first of two 1970 amendments, which added state-chartered credit unions to the statutory list, Pub.L.91-468, § 7, 84 Stat. 1017, was characterized simply as "relating to false statements in loan and credit applications." H.R.Rep.No.91-1457, p. 21 (1970), U.S.Code Cong. & Admin.News 1970, pp. 4166, 4187. 21 A second 1970 amendment, Pub.L.91-609, § 915, 84 Stat. 1815, added banks insured by the Federal Deposit Insurance Corporation, Federal Home Loan Banks, and institutions insured by the Federal Savings and Loan Insurance Corporation, for the first time listing institutions that engaged in commercial checking.12 But there was no contemporaneous congressional recognition of the substantial expansion of federal criminal jurisdiction that would attend the proscription of bad checks. To the contrary, the Reports accompanying the amendment stated simply that the addition "would describe more explicitly the institutions which are covered by 18 U.S.C. § 1014, which provides penalties for making false statements or reports in connection with loans or other similar transactions." H.R.Rep.No.91-1556, p. 35 (1970). See H.R.Conf.Rep.No.91-1784, p. 66 (1970), U.S.Code Cong. & Admin.News 1970, pp. 5582, 5652. Congressional debate was directed only at the addition of federally insured savings and loan institutions, which was said to "mak[e] it a Federal crime to submit false data to an insured savings and loan on the true value of a property on which a mortgage is to be granted." 116 Cong.Rec. 42633 (1970) (remarks of Rep. Sullivan). 22 Given this background—a statute that is not unambiguous in its terms and that if applied here would render a wide range of conduct violative of federal law, a legislative history that fails to evidence congressional awareness of the statute's claimed scope, and a subject matter that traditionally has been regulated by state law—we believe that a narrow interpretation of § 1014 would be consistent with our usual approach to the construction of criminal statutes. The Court has emphasized that " 'when choice has to be made between two readings of what conduct Congress has made a crime, it is appropriate, before we choose the harsher alternative, to require that Congress should have spoken in language that is clear and definite.' " United States v. Bass, 404 U.S. 336, 347, 92 S.Ct. 515, 522, 30 L.Ed.2d 488 (1971), quoting United States v. Universal C.I.T. Credit Corp., 344 U.S. 218, 221-222, 73 S.Ct. 227, 229, 97 L.Ed. 260 (1952).13 To be sure, the rule of lenity does not give courts license to disregard otherwise applicable enactments. But in a case such as this one, where both readings of § 1014 are plausible, "it would require statutory language much more explicit than that before us here to lead to the conclusion that Congress intended to put the Federal Government in the business of policing the" deposit of bad checks. United States v. Enmons, 410 U.S., at 411, 93 S.Ct., at 1015. 23 The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. 24 It is so ordered. Justice WHITE, with whom Justice BRENNAN joins, dissenting. 25 The majority reverses petitioner's conviction under 18 U.S.C. § 1014 on the grounds that the Government has not shown that he made a "false statement or report" or "willfully overvalue[d] any land, property or security." Ante, at 284. According to the majority, a check is not a statement; it is merely an order to the drawee bank to pay the face amount to the payee and a promise to pay the amount of the check upon notice of dishonor. Ante, at 284-285. Like Justice MARSHALL, I do not disagree with the majority that under the Uniform Commercial Code a check constitutes an order to the drawee bank and a promise to pay upon notice of dishonor. However, the fact that the Uniform Commercial Code describes a check in this manner does not mean that a check does not carry with it other representations, for the Code does not purport to contain an all-inclusive definition of a check. 26 It defies common sense and everyday practice to maintain, as the majority does, that a check carries with it no representation as to the drawer's account balance. No bank would give a customer immediate credit for a check drawn on another bank or reduce a check to cash if it did not believe that the check would be paid in the normal course of collection. It could be argued that petitioner did not make a false statement with respect to the May 10 check drawn on the Pelican Bank because he knew the bank would pay the check through its dummy account. However, petitioner does not contend that he had any such arrangement with the Winn Bank, and thus the May 9 check for $58,500 drawn on the Winn Bank, when his balance was $4,649.97, can fairly be said to constitute a false statement. In any event, a properly instructed jury surely found that Williams had made false representations with respect to each of the checks that were the subject of this indictment. 27 If the majority really means what it says in Part II-A of its opinion—that the Government failed to show that petitioner made a false statement or overvalued property or security—it is unnecessary to explore the legislative history of § 1014 or to apply the rule of lenity. On the other hand, if the majority reverses the Court of Appeals because it cannot conceive that Congress intended § 1014 to reach the conduct at issue because the area has long been regulated by state law, it is not necessary to employ the fiction that a check does not entail a representation that it will be paid in the normal course of business by the drawee bank. Because the majority opinion appears to me to rest on that fiction, I respectfully dissent. I also join Justice MARSHALL's dissenting opinion. 28 Justice MARSHALL, with whom THE CHIEF JUSTICE, Justice BRENNAN, and Justice WHITE join, dissenting. 29 The majority, after developing an overly technical "definition" of the meaning of a check—a definition which will come as quite a surprise to banks and businesses that accept checks in exchange for goods, services, or cash on the representation that the drawer has sufficient funds to cover the check—concludes that the question whether petitioner Williams' check-kiting scheme is covered by 18 U.S.C. § 1014 is ambiguous. The majority then applies its version of the rule of lenity, and decides that Williams cannot be convicted for violating this statute. Because I believe that the majority misapplies the rule of lenity, and because Williams' conduct is clearly prohibited by the statute, I respectfully dissent. 30 * Before addressing the application of § 1014 to Williams' conduct, I think that it is helpful to set forth clearly what is not involved here. This is not a case in which a defendant, through careless bookkeeping, wrote checks on accounts with insufficient funds. Nor is this a case in which a defendant wrote a check on an account containing insufficient funds with the good-faith intention to deposit in that account an amount that would cover the check before it cleared in the normal course of business. Rather, this case clearly involves fraudulent conduct. Petitioner Williams engaged in an intentional check-kiting scheme. He misled the first bank into honoring his worthless, or virtually worthless, check and extending him immediate credit. This extension of credit enabled him to "play the float" and cover that check by misleading another bank into extending him credit on an equally worthless check. In effect, Williams was able to obtain interest-free extensions of credit. Williams, who was a bank president, does not, nor can he, make any credible argument that he was unaware that his conduct was wrongful. With this in mind, I turn to the question whether Williams' conduct constitutes a violation of 18 U.S.C. § 1014. 31 Section 1014 is a comprehensive statute designed to protect the assets of federally insured lending institutions. The Government establishes a violation of this statute by proving that the defendant "knowingly [made] any false statement or . . . willfully overvalue[d] any . . . property or security, for the purpose of influencing in any way the action of [any federally insured bank] upon any . . . advance, . . . commitment, or loan." 18 U.S.C. § 1014 (emphasis added). Just last Term, we reiterated that "[i]n determining the scope of a statute, we look first to its language. If the statutory language is unambiguous, in the absence of a 'clearly expressed legislative intent to the contrary, that language must ordinarily be regarded as conclusive.' " United States v. Turkette, 452 U.S. 576, 580, 101 S.Ct. 2524, 2527, 69 L.Ed.2d 246 (1981) (quoting Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 108, 100 S.Ct. 2051, 2056, 64 L.Ed.2d 766 (1980)). In my view, the plain language of § 1014 covers the check-kiting scheme practiced by Williams, and nothing in the legislative history of the statute indicates that Congress intended to exclude this type of scheme from the coverage of the statute. A. 32 The language of § 1014 is sweeping. It embraces numerous entities in which the Federal Government has a financial interest. It proscribes, in the disjunctive, a wide variety of deceptive schemes that might impair the financial stability of these institutions. Cf. United States v. Naftalin, 441 U.S. 768, 774, 99 S.Ct. 2077, 2082, 60 L.Ed.2d 624 (1979) (disjunctive prohibitions intended to "cover additional kinds of illegalities not to narrow the reach of the prior sections"). The statute refers broadly to "any false statement or report," and to overvaluations of "any" property or security. The list of transactions to which the statute applies is equally expansive—it covers "any application, advance, discount, purchase, purchase agreement, repurchase agreement, commitment, or loan, or any change or extension of any of the same, by renewal, deferment of action or otherwise, or the acceptance, release, or substitution of security therefor." 18 U.S.C. § 1014. 33 The broad statutory language clearly evinces its legislative purpose—Congress hoped to protect federally insured institutions from losses stemming from false statements or misrepresentations that mislead the institutions into making financial commitments, advances, or loans. The statute was intended to be broad enough "to maintain the vitality of the FDIC insurance program . . . and 'to cover all undertakings which might subject the FDIC insured bank to risk of loss.' " United States v. Pinto, 646 F.2d 833, 838 (C.A.3) (quoting United States v. Stoddart, 574 F.2d 1050, 1053 (CA10 1978), "cert. denied, 454 U.S. 816, 102 S.Ct. 94, 70 L.Ed.2d 85 (1981)). This broad language does not lend itself to the restrictive interpretation endorsed by the Court today. Cf. United States v. Culbert, 435 U.S. 371, 98 S.Ct. 1112, 55 L.Ed.2d 349 (1978). 34 Nothing on the face of § 1014 "suggests a congressional intent to limit its coverage" to a particular kind of transaction. United States v. Culbert, supra, at 373, 98 S.Ct., at 1113. Check kiting, which threatens the assets of federally insured banks in precisely the same way as a misrepresentation in a loan application, should not be excluded from the reach of the statute simply because the terms of the statute and its legislative history do not specifically identify check kiting by name or precise description. This method of statutory construction was rejected recently in Harrison v. PPG Industries, Inc., 446 U.S. 578, 592, 100 S.Ct. 1889, 1897, 64 L.Ed.2d 525 (1980): 35 "[I]t would be a strange canon of statutory construction that would require Congress to state in committee reports or elsewhere in its deliberations that which is obvious on the face of a statute. In ascertaining the meaning of a statute, a court cannot, in the manner of Sherlock Holmes, pursue the theory of the dog that did not bark." 36 Unfortunately, in my view, the Court's approach to interpreting § 1014 comes dangerously close to the method we rejected in Harrison. Unless one accepts the Court's overly restrictive and technical "definition" of a check, check-kiting schemes clearly fall within the broad language of that statute. B 37 As the majority recognizes, a violation of § 1014 is established when the Government proves two elements: that the defendant either made a "false statement or report," or "willfully overvalue[d] any . . . property or security;" and that the defendant did so "for the purpose of influencing in any way the action of [a federally insured institution] upon any application, advance, . . . commitment, or loan." After recognizing this, however, the majority's analysis jumps the track. The majority concludes that when a drawer presents a kited check to a bank with the knowledge that he does not have sufficient funds, and with the intent not to cover that check with anything other than another virtually worthless kited check, he has not made "any false statement or report," or "willfully overvalue[d] any . . . property or security" within the meaning of the statute. In my view, neither of these conclusions withstands analysis. 38 (1) 39 The basis for the Court's conclusion that Williams did not make a "false statement or report" is concededly technical and "simple": "a check is not a factual assertion at all, and therefore cannot be characterized as 'true' or 'false.' " Ante, at 284. This argument proves too much: it would apply equally to material omissions or failures to disclose in connection with loan applications. However, the Courts of Appeals have held that the failure to disclose material information needed to avoid deception in connection with loan transactions covered by § 1014 constitutes a "false statement or report," and thus violates the statute. See, e.g., United States v. Greene, 578 F.2d 648, 657 (C.A.5 1978), cert. denied, 439 U.S. 1133, 99 S.Ct. 1056, 59 L.Ed.2d 96 (1979). I assume that the majority would not disagree with this analysis, which is based on established contract principles. I am at a loss as to why the majority does not apply the same analysis to the transactions at issue in this case. 40 The majority's description of a check as an " 'unconditional promise or order to pay a sum certain in money,' " ante, at 285 (quoting the Uniform Commercial Code § 3-104(1)(b), 2 U.L.A. 17 (1977)), is unexceptionable as a conclusory description of "black-letter" law. However, this oversimplified description fails to look behind the bare technical definition of a check. Moreover, this description is not at all inconsistent with the necessary implications that a check carries. "In giving a check, the drawer impliedly represents that he has on deposit with the drawee banks funds equivalent to the face amount of the check." F. Whitney, The Law of Modern Commercial Practices § 341 (2d ed. 1965).1 Despite the majority's equivocation on this point, those who write or accept checks in exchange for goods, services, or cash undoubtedly understand that this implicit representation has been made.2 A check is accepted with the expectation that it will be paid in the normal course of collection. A banker who knew that the drawer did not have funds on deposit would not credit the check to the drawer's account or reduce it to cash. Regardless of any contractual breach also involved in check kiting, a person who writes a series of checks knowing that there are no funds to cover them has made intentional false representations within the reach of § 1014. 41 Any other view, including that endorsed by the Court today, would interfere with the manner in which a major portion of commercial transactions are conducted in our society today. Williams was charged with, and the jury convicted him of, making a false representation (or, more precisely, a material omission) when he presented his check to the bank with the knowledge that he did not have sufficient funds to cover the check, and with the further intent not to cover that check before it cleared with anything other than another worthless kited check. See n. 2, supra. Therefore, his conviction under § 1014 should stand. 42 (2) 43 In addition to violating § 1014 by intentionally making a false statement to a federally insured bank for the purpose of obtaining credit, Williams also violated the statute for a separate and independent reason. Although Williams presented to the bank for immediate credit a check which on its face represented an amount exceeding $50,000, he well knew that in fact the check was virtually worthless. In so doing, he "willfully overvalue[d] . . . property or security" for the purpose of obtaining credit.3 The Court's rejection of the Government's argument with respect to this issue is startling in both its brevity and its concededly technical and "literal" interpretation of the legal value of a check which completely ignores the meaning attributed to checks in the real world. 44 The very essence of a check-kiting scheme is the successful overvaluation of a security or property which misleads a bank into issuing immediate credit on the assumption that the security or property is in fact valued at the amount represented on its face. A check-kiting scheme is successful only when the bank to which the check is presented assumes that the check is supported by adequate funds in the account upon which it is drawn, and that the face amount of the check is in fact its value. See supra, at 296-298; United States v. Payne, 602 F.2d 1215, 1217-1218 (C.A.5 1979). If the bank does not accept the valuation on the face of the check, and instead either inquires into the status of the account on which the check is drawn or waits until the check clears before paying the face amount of the check, the scheme will collapse. Of course, it would be more prudent for a bank to take such precautions just as it would be prudent for banks to inquire carefully into the accuracy of all representations made concerning the value of collateral pledged as security for conventional loans. However, this more prudent course is not always practicable. Moreover, the bank may not believe that such precautions are necessary where, as here, the person presenting the check is the president of another bank presumed to know the illegality, and the drastic adverse consequences to a bank, of a check-kiting scheme. In any event, a bank's failure to take all possible precautions does not bar prosecution under § 1014, which places the burden of avoiding false representations, at the risk of criminal prosecution, upon the person who seeks the funds of the federally insured bank. Section 1014 forbids a person seeking such funds to make "any" false statement or to "willfully overvalue" any security or property to obtain use of the bank's funds. A kited check is "willfully overvalued" within the meaning of the statute, just as worthless securities presented as collateral for a loan are "willfully overvalued." See United States v. Calandrella, 605 F.2d 236 (C.A.6), cert. denied sub nom. Kaye v. United States, 444 U.S. 991, 100 S.Ct. 522, 62 L.Ed.2d 420 (1979). 45 (3) 46 The Court does not question that the second element of a § 1014 violation—that Williams presented his kited check for the purpose of influencing the bank to extend him credit in the form of a loan or an advance—is satisfied in this case. Clearly, Williams' conduct was directed at misleading a bank into extending immediate credit. Indeed, the whole purpose of Williams' kiting scheme was to obtain an immediate extension of credit by depositing a check purportedly supported by adequate funds. The banks that extended funds on the basis of Williams' worthless, and not yet collected, checks made an "advance," a "loan," and a "commitment" within the ordinary meaning of these terms. See, e.g., United States v. Payne, supra, at 1218 (check kiting has effect of inducing a credit, a loan, or an advance); United States v. Street, 529 F.2d 226, 229 (C.A.6 1976) (check kiting is the obtaining of "forced credit"); J. White & R. Summers, Uniform Commercial Code 558 (2d ed. 1980); F. Whitney, supra n. 1, § 310, pp. 451-452. 47 If a worthless check is submitted to a bank for reasons other than to obtain an extension of credit, the conduct simply is not check kiting in the ordinary sense of the term, and would not fall within the prohibition of § 1014.4 However, if a properly instructed jury concludes that a worthless check was submitted in order to obtain immediate credit from a bank, there is no reason to regard the conduct as falling outside the reach of § 1014. The jury that convicted Williams was so instructed, see n. 2, supra, and found that Williams' conduct constituted a "false representation" designed to influence the banks into extending him immediate credit. C 48 The unambiguous language of § 1014 clearly proscribes conduct commonly referred to as check kiting. This language should be given effect in the absence of clear indications in the legislative history that Congress did not intend to proscribe this conduct. See United States v. Turkette, 452 U.S., at 580, 101 S.Ct., at 2527. There are no such indications in the legislative history. To the contrary, the legislative history makes clear that the statute was not limited to borrowers or to loan applications. See S.Rep.No.1078, 88th Cong., 2d Sess., 4 (1964); H.R.Conf.Rep.No.91-1784, 91st Cong., 2d Sess., p. 66 (1970), U.S.Code Cong. & Admin.News 1970, p. 5582. 49 The Court finds no indication that Congress intended to exclude check-kiting schemes from the scope of the statute. The Court's brief review of the legislative history to § 1014 does suggest that the primary purpose of the statute is to prohibit misrepresentations in connection with conventional loan applications. However, neither this fact, nor the fact that most convictions under the statute involve such transactions, compels the Court to ignore the broad language and purposes of the statute by interpreting it to cover only these transactions. In the past, we have consistently rejected the argument that a criminal statute must be given its narrowest meaning by limiting its scope to effectuate only its primary purpose. See, e.g., United States v. Turkette, supra; United States v. Naftalin, 441 U.S. 768, 99 S.Ct. 2077, 60 L.Ed.2d 624 (1979); United States v. Moore, 423 U.S. 122, 96 S.Ct. 335, 46 L.Ed.2d 333 (1975). II 50 In light of the broad protection Congress intended to accord federally insured institutions against fraudulent or deceptive conduct intended to mislead these institutions into extending credit and the broad, unrestricted statutory language embodied in § 1014, I marvel at the Court's method of interpreting this statute. Indeed, today's decision is utterly incompatible with a number of prior decisions of this Court in which we addressed similar arguments raised by persons convicted under broad federal statutes. See, e.g., United States v. Turkette, supra; Rubin v. United States, 449 U.S. 424, 101 S.Ct. 698, 66 L.Ed.2d 633 (1981); United States v. Naftalin, supra; United States v. Culbert, 435 U.S. 371, 98 S.Ct. 1112, 55 L.Ed.2d 349 (1978). In these decisions, we have consistently looked first to the statutory language to determine the scope and purpose of the statute. If it were evident from the face of the statute that the statute was written broadly in order to prohibit certain kinds of conduct which entail specific risks or dangers deemed by the legislators to be sufficiently unacceptable to warrant criminal sanction, we do not frustrate this purpose by distorting either the statutory language employed or the conduct of the accused in the name of the "rule of lenity." See, e.g., United States v. Turkette, supra; Rubin v. United States, supra. 51 In contrast with this established approach, the majority today interprets § 1014 without acknowledging the broad statutory language chosen by Congress. This error is compounded by the Court's failure to address the fact that this broad language was intended to proscribe, in generic and disjunctive terms, precisely the type of conduct of which Williams was found guilty—intentionally misleading the bank into extending him credit and to protect federally insured institutions from precisely the risk of loss to which Williams' conduct subjected them. Ignoring these factors, the majority begins its analysis by employing an oversimplified, concededly technical and literal interpretation of the "legal definition" of a check. In then observes that Congress never explicitly stated that it intended the statute to cover check-kiting schemes. It concludes that in the absence of such an express statement, the rule of lenity requires that the statute not cover these schemes. 52 The majority's approach to the question of statutory construction is a prime example of what this Court has time and again said the rule of lenity does not entail: 53 "The canon in favor of strict construction is not an inexorable command to override common sense and evident statutory purpose. It does not require magnified emphasis upon a single ambiguous word in order to give it a meaning contradictory to the fair import of the whole remaining language. As was said in United States v. Gaskin, 320 U.S. 527, 530, 64 S.Ct. 318, 319, 88 L.Ed. 287 the canon 'does not require distortion or nullification of the evident meaning and purpose of the legislation.' Nor does it demand that a statute be given the 'narrowest meaning'; it is satisfied if the words are given their fair meaning in accord with the manifest intent of the lawmakers." United States v. Brown, 333 U.S. 18, 25-26, 68 S.Ct. 376, 379-380, 92 L.Ed. 442 (1948) (quoted in United States v. Turkette, supra, at 588, n. 10, 101 S.Ct., at 2531, n. 10, and United States v. Moore, supra, at 145, 96 S.Ct., at 346). 54 If the broad language and evident purpose of the statute had been given effect, there would have been no need to parse the legislative history for affirmative evidence that Congress "demand[ed] a broader reading of the statute." Ante, at 288. Holding that § 1014 reaches check kiting does not produce an absurd result, render the statute internally contradictory, or diverge from legislative policy. To the contrary, Congress' policy, manifest in § 1014 and elsewhere throughout Title 18 of the United States Code, is that federal criminal sanctions are necessary to provide federally insured banking institutions with comprehensive protection against practices that cause risk of loss. The Court's construction of § 1014, on the other hand, results in a large loophole in the protection afforded these institutions by limiting the statute's application to formal loan transactions. After today's decision, a bank's protection against false statements intended to influence credit transactions depends not upon whether a misrepresentation was made in connection with a loan, advance, or commitment, but rather upon whether a court concluded that the transaction was "traditional" or that Congress specified that transaction by name in a committee report. 55 It is worth observing that in this case, none of the general justifications for applying the rule of lenity are present. In Huddleston v. United States, 415 U.S. 814, 831, 94 S.Ct. 1262, 1271, 39 L.Ed.2d 782 (1974), this Court explained that the rule of lenity "is rooted in the concern of the law for individual rights, and in the belief that fair warning should be accorded as to what conduct is criminal and punishable by deprivation of liberty or property." There is no question that Williams, a bank president, knew that his check-kiting scheme was wrongful. The majority's attempt to buttress its decision by arguing that check kiting has traditionally been regulated by the States, and that federal enforcement might interfere with this regulation, is completely unjustified.5 The Federal Government, which provides deposit insurance, has a paramount interest in safeguarding the financial integrity of federally insured banking institutions. The Courts of Appeals have been virtually unanimous in holding that check kiting is subject to federal prosecution under the mail and wire fraud statutes, see, e.g., United States v. Giordano, 489 F.2d 327 (C.A.2 1973); United States v. Constant, 501 F.2d 1284 (C.A.5 1974), cert. denied, 420 U.S. 910, 95 S.Ct. 830, 42 L.Ed.2d 840 (1975), and the majority apparently does not question these decisions. Therefore, a check-kiting prosecution under § 1014, which by its terms applies only to federally insured institutions, results in no new inroad upon state criminal jurisdiction. 56 Under the version of the rule of lenity adopted today, conduct which falls within the literal terms of a broad statute, which proscribes in disjunctive and generic terms the type of conduct at issue, and which is designed to protect against the very risk created by such conduct, escapes the reach of the statute unless Congress specifies that conduct by name in the statute or describes it in detail in the statute's legislative history. In order to find Williams' conduct outside the scope of § 1014, the majority ignores the function of a check in today's society. The rule of lenity has never been interpreted to require this kind of result. I am at a loss to explain why the Court adopts this approach today and consequently turns the rule of lenity on its head. Accordingly, I dissent. 1 As the Government explains, a check-kiting scheme typically works as follows: "The check kiter opens an account at Bank A with a nominal deposit. He then writes a check on that account for a large sum, such as $50,000. The check kiter then opens an account at Bank B and deposits the $50,000 check from Bank A in that account. At the time of deposit, the check is not supported by sufficient funds in the account at Bank A. However, Bank B, unaware of this fact, gives the check kiter immediate credit on his account at Bank B. During the several-day period that the check on Bank A is being processed for collection from that bank, the check kiter writes a $50,000 check on his account at Bank B and deposits it into his account at Bank A. At the time of the deposit of that check, Bank A gives the check kiter immediate credit on his account there, and on the basis of that grant of credit pays the original $50,000 check when it is presented for collection. "By repeating this scheme, or some variation of it, the check kiter can use the $50,000 credit originally given by Bank B as an interest-free loan for an extended period of time. In effect, the check kiter can take advantage of the several-day period required for the transmittal, processing, and payment of checks from accounts in different banks . . . ." Brief for United States 12-13. 2 Petitioner also was charged with—and thereafter convicted of—one count of misapplying bank funds, in violation of 18 U.S.C. § 656. The validity of that conviction, which was affirmed on appeal, is not before us. 3 Neither of the § 1014 counts of the indictment expressly charged petitioner with making a "false statement." The first count, however, did allege that he "presented said check for deposit at Pelican State Bank . . . and represented and caused to be represented to said bank that said check was of a value equal to the face amount of the check, when in truth and fact, as the [petitioner] then well knew, there were no sufficient funds in the account of W. A. Williams at the Winn State Bank and Trust Company, to cover said check." App. 3. Similar language was employed in the second § 1014 count. Id., at 4. 4 The sentence of probation also applied to petitioner's conviction for misapplication of bank funds. See n. 2, supra. 5 See United States v. Sher, 657 F.2d 28 (C.A.3 1981), cert. pending, No. 81-1047 (holding that § 1014 does not proscribe check kiting). Cf. United States v. Krown, 675 F.2d 46, 50 (C.A.2 1982) (noting the conflict). 6 Unlike many state statutes that do proscribe conduct such as that engaged in by petitioner, the federal scheme obviously does not in terms reach the deposit of checks that are supported by insufficient funds. See Comment, Insufficient Funds Checks in the Criminal Area: Elements, Issues, and Proposals, 38 Mo.L.Rev. 432 (1973). 7 That is particularly true where, as here, it is not immediately clear what "common understanding" would recognize as the implied representation of the act of depositing one's own check. The United States suggests that one who deposits a check represents that he "currently has funds on deposit sufficient to cover the face value." Brief for United States 19. But it would be equally plausible to suggest that many people understand a check to represent that the drawer will have sufficient funds deposited in his account by the time the check clears, or that the drawer will make good the face value of the draft if it is dishonored by the bank. We therefore find "common understanding" a particularly fragile foundation upon which to base an interpretation of § 1014. 8 Justice MARSHALL's dissent does not fully respond to this point. That opinion, like the Government's brief, emphasizes that petitioner's "conduct was wrongful," post, at 293, and deals only with § 1014's application to check kiting. See also post, at 294, 295, 299, 300, and 300. Indeed, the dissent seems to suggest that that statute would not reach the conduct of a defendant who "wrote a check on an account containing insufficient funds with the good-faith intention to deposit in that account an amount that would cover the check before it cleared in the normal course of business." Post, at 292. Accepting Justice MARSHALL's theory, however, would bring such conduct within the literal language of the statute, for a "false statement" would have been submitted with the hope of inducing a bank to "advance" funds. While the dissent attempts to avoid this by suggesting that there would be no violation of § 1014 absent an intent "to defraud," post, at 301, n. 4, the language of the statute imposes no such intent requirement. And as we emphasize above, we believe that the wording of § 1014 would be a peculiar choice of terms if Congress wished to proscribe such conduct. 9 Justice MARSHALL's dissent rests entirely on the proposition that petitioner's conduct falls within the "plain language" of § 1014. Post, at 293. See also post, at 301, 302, and 305-306. In our view, that literally is not true. And even if one looks to the "common understanding" so emphasized by Justice MARSHALL, post, at 296-298, the statute is at best ambiguous, for we doubt that the public typically describes bad checks as "false statements." 10 These included the Farmers' Home Corporation, the Federal Crop Insurance Corporation, Federal Reserve Banks, the Farm Credit Administration, Federal Credit Banks, the Federal Farm Mortgage Corporation, the National Agricultural Credit Corporation, Federal Home Loan Banks, the Home Owners' Loan Corporation, the Reconstruction Finance Corporation, and related institutions. See 7 U.S.C. §§ 1026(a), 1514(a) (1946 ed.); 12 U.S.C. §§ 596, 981, 1122, 1123, 1138d(a), 1248, 1312, 1313, 1441(a), 1467(a) (1946 ed.); 15 U.S.C. § 616(a) (1946 ed.). 11 The Committee added ambiguously that the statute "is not, however, limited by its terms to borrowers and would seem also to apply to others, including for example, officers and employees of the agencies and institutions named." S.Rep.No.1078, 88th Cong., 2d Sess., 4 (1964), U.S.Code Cong. & Admin.News 1964, 2522. 12 Also added to the list in 1970 were the Federal Deposit Insurance Corporation and the Federal Savings and Loan Insurance Corporation themselves, as well as the Administrator of the National Credit Union Administration. Pub.L.91-609, § 915, 84 Stat. 1815. 13 We therefore find it somewhat surprising that Justice MARSHALL's dissenting opinion takes us to task for noting the applicability of the rule of lenity to the interpretation of what we believe to be an ambiguous statute. 1 The Court's facile conclusion that Williams made no false statement or misrepresentation when he presented his check to a bank for immediate credit, knowing that the check was not supported by sufficient funds and that he was not going to cover the check before it cleared with anything other than another kited check, is contrary to the theory underlying most prosecutions under state bad check laws. These laws are not based upon the defendant's breach of a contractual promise that he will pay a sum certain upon demand, but upon the fact that in knowingly presenting a bad check the defendant has committed fraud and misrepresentation and can be punished for committing a crime. Brief for United States 20; Brief for Petitioner 28-29, and n. 17. See also F. Whitney, The Law of Modern Commercial Practices § 341 (2d ed. 1965). The Court attempts to avoid the obvious problem this fact presents to its method of statutory interpretation by stating that the federal statute does not apply "in terms" to check kiting, while some state laws do. See ante, at 285, n. 6. This reasoning is circular. The reason why § 1014 does not "in terms" reach a check-kiting scheme, while certain state laws do, is because the Court ipse dixit totally discredits the theory upon which the state laws are premised and refuses to read the terms of the statute in the only manner that is consistent with this theory. 2 The manner in which the Court manufactures "confusion" over the common understanding of a check is difficult to comprehend. See ante, at 286, n. 7. Most of it is totally irrelevant because each of the majority's "common understandings" of the meaning of a check are entirely consistent with prosecuting a check-kiting scheme under § 1014. The majority suggests that the "common understanding" of a check is only that sufficient funds will be present by the time the check clears or that the drawer will make good the payment of the face amount of the check if the bank refuses payment. Even if the majority is correct, prosecuting a check-kiting scheme under § 1014 would be justified because the jury found that Williams had intentionally acted inconsistently with each of these understandings. The jury was specifically instructed that it could not convict unless it found that Williams "made the false statement with fraudulent intent to influence the [bank] to extend [him] credit." App. 37. The judge added that a statement is "false" if it "relates to a material fact and is untrue and is then known to be untrue by the person making it." Id., at 38. The judge further instructed the jury that "[t]he crucial question in check-kiting is whether the defendant intended to write checks which he could not reasonably expect to cover and thereby defraud the bank, or whether he was genuinely involved in the process of depositing funds and then making legitimate withdrawals against them. Hence, proof that the checks were eventually paid might well be pertinent to defendant's initial intent, that is, whether he intended to deceive the bank." Ibid. Therefore, the jury was clearly instructed to acquit Williams if he had shared with the Court even its most lenient and unrealistic interpretation of the implied representation made when one presents a check. The jury had to find that Williams had given the bank the kited check with the express intent not to actually cover the check, but only to receive this extension of credit for as long as the check-kiting scheme continued. 3 Section 1014 applies to the willful overvaluation of "any . . . property, or security." Again, this element of the statute is cast in broad rather than restrictive terms. Congress plainly intended to proscribe the willful overvaluation of anything of value given to a lending institution. There is no suggestion that the broad generic terms "any . . . property or security" were meant to exclude items such as checks presented to obtain a temporary extension of credit. There is no reason to interpret this language to exclude checks. A check is plainly a form of property under even the majority's most restrictive definition—it is a demand to a drawee to pay a sum certain of money, which is backed by a promise of the drawer to make payment in the event of default. Furthermore, as evidenced by other provisions of Title 18, including the general definitional section, 18 U.S.C. § 8, a check is a type of "security." See, e.g., 18 U.S.C. § 2311. 4 The Court's fears that holding a check-kiting scheme to be covered by § 1014 would entail broad implications, see ante, at 286-287, are misguided. If there was no intent on the part of the check kiter to defraud the bank into extending credit, there would be no § 1014 violation. The fact that the Government brought separate counts for each check in the check-kiting scheme does not alter the fact that it was essential to conviction under the jury instructions for the jury to find that petitioner was involved in a check-kiting scheme intentionally designed to defraud the banks. 5 In Title 18, Congress has provided comprehensive criminal sanctions to protect federally insured institutions. See, e.g., 18 U.S.C. §§ 212, 213 (loans or gratuities offered to bank examiners by bank officials; acceptance of same by examiners); 18 U.S.C. § 493 (forging, counterfeiting, or passing bonds and obligations); 18 U.S.C. § 656 (theft from banks by bank examiners); 18 U.S.C. § 709 (1976 ed. and Supp.IV) (false advertising that bank deposits are insured by Federal Deposit Insurance Corporation). Congress has sought to protect fully the integrity of the federal insurance program, and the protection against check kiting afforded by § 1014 is consistent with this scheme. See, e.g., United States v. Bush, 599 F.2d 72, 75 (C.A.5 1979); United States v. Pinto, 646 F.2d 833, 838 (C.A.3), cert. denied, 454 U.S. 816, 102 S.Ct. 94, 70 L.Ed.2d 85 (1981); United States v. Stoddart, 574 F.2d 1050, 1053 (C.A.10 1978). Construing § 1014 to cover check kiting does not displace the authority of the States. Rather, it complements state law enforcement in an area where the federal interest is substantial. See United States v. Turkette, 452 U.S. 576, 586, n. 9, 101 S.Ct. 2524, 2530, n. 9, 69 L.Ed.2d 246 (1981) (interpreting the Racketeer Influenced and Corrupt Organizations statute) ("[T]he States remain free to exercise their police powers to the fullest constitutional extent in defining and prosecuting crimes within their respective jurisdictions. That some of those crimes may also constitute [violations of federal law], is no restriction on the separate administration of criminal justice by the States").

01

458 U.S. 375 102 S.Ct. 3141 73 L.Ed.2d 835 GENERAL BUILDING CONTRACTORS ASSOCIATION, INC., Petitionerv.PENNSYLVANIA et al. UNITED ENGINEERS & CONSTRUCTORS, INC., Petitioner, v. PENNSYLVANIA et al. CONTRACTORS ASSOCIATION OF EASTERN PENNSYLVANIA and United Contractors Association, Petitioners, v. PENNSYLVANIA et al. GLASGOW, INC., Petitioner, v. PENNSYLVANIA et al. BECHTEL POWER CORPORATION, Petitioner, v. PENNSYLVANIA et al. Nos. 81-280, 81-330 to 81-333. Argued March 3, 1982. Decided June 29, 1982. Syllabus Respondents—the Commonwealth of Pennsylvania and several black individuals representing a class of racial minorities who are skilled or seek work as operating engineers in the construction industry in Eastern Pennsylvania and Delaware—brought an action in Federal District Court under 42 U.S.C. § 1981, seeking to redress alleged racial discrimination in the operation of an exclusive hiring hall established in collective-bargaining contracts between the local union representing operating engineers and petitioner trade associations and construction industry employers. Respondents also alleged discrimination in the operation of an apprenticeship program established by the union and the trade associations and administered by the Joint Apprenticeship and Training Committee (JATC), half of whose members are appointed by the union and half by the trade associations. Named as defendants were the union and petitioners. The District Court found that although the hiring hall system was neutral on its face, the union in administering the system practiced a pattern of intentional discrimination, and the court also found similar discrimination in the JATC's administration of the apprenticeship program. On the basis of these findings, the court held that the union and the JATC had violated § 1981, and that, although petitioners as a class did not intentionally discriminate against minority workers and were not aware of the union's discriminatory practices, they were nevertheless liable under § 1981 for the purpose of imposing an injunctive remedy. The court reasoned that liability under § 1981 requires no proof of purposeful conduct on any of the defendants' part, but it was sufficient that the employers delegated the hiring procedure to the union and that the union, in effectuating this delegation, intentionally discriminated or, alternatively, produced a discriminatory impact. The court concluded that respondents had shown the requisite relationship among the employers, trade associations, and union to render applicable the theory of respondeat superior, thus making petitioners liable for the union's discriminatory acts. The Court of Appeals affirmed. Held: 1. Liability may not be imposed under § 1981 without proof of intentional discrimination. This conclusion is supported by the legislative history. The fact that the prohibitions of § 1981 encompass private as well as governmental action does not suggest that the statute reaches more than purposeful discrimination, whether public or private. Pp. 382-391. 2. The District Court was unable to find discriminatory intent on petitioners' part, and liability under § 1981 cannot be vicariously imposed on them based on the discriminatory conduct of the union or the JATC. Pp. 391-397. (a) There is no basis for holding petitioners liable under the doctrine of respondeat superior. The union, in operating the hiring hall, performed no function as the agent or servant of petitioner trade associations. Nor can the relationship between petitioner employers and the union be characterized as one between principal and agent without proof of a right to control the union's activities. Such a conception is alien to the fundamental assumption upon which the federal labor laws are structured and was not established by the evidence on which the District Court relied. And there is no evidence that an agency relationship existed between petitioners and the JATC. The fact that the employers fund the JATC does not render the JATC the employers' servant or agent, nor does the fact that the trade associations appoint half of the JATC's members infer a right of the associations to control the JATC. Pp. 391-395. (b) Nor is there any basis for holding petitioners liable on the ground that § 1981 imposes a "nondelegable duty" on them to see that discrimination does not occur in the selection of their work force. Section 1981 does no more than prohibit petitioners from intentionally depriving black workers of the rights enumerated in the statute, including the equal right to contract, and was not intended to make them guarantors of the workers' rights against third parties who would infringe them. Pp. 395-397. 3. The District Court had no inherent power under its traditional equitable authority to allocate to petitioners a portion of the costs of the remedial decree, absent a supportable finding of liability upon petitioners' part. Nor does the All Writs Act constitute an independent basis for the injunctive portions of the District Court's order running against petitioners. There was no need to treat petitioners as strangers to the suit and therefore to rely upon some extraordinary form of procedure or writ to bring them before the court, since they were named as defendants and litigated the issue of injunctive liability. Pp. 397-402. 648 F.2d 923, reversed and remanded. John J. McAleese, Jr., Bala Cynwyd, Pa., for petitioners in 81-330, et al. John G. Kester, Washington, D. C., for petitioners in 81-280, et al. Harold I. Goodman, Philadelphia, Pa., for respondents. Justice REHNQUIST delivered the opinion of the Court. 1 Respondents, the Commonwealth of Pennsylvania and the representatives of a class of racial minorities who are skilled or seek work as operating engineers in the construction industry in Eastern Pennsylvania and Delaware, commenced this action under a variety of federal statutes protecting civil rights, including 42 U.S.C. § 1981. The complaint sought to redress racial discrimination in the operation of an exclusive hiring hall established in contracts between Local 542 of the International Union of Operating Engineers and construction industry employers doing business within the Union's jurisdiction. Respondents also alleged discrimination in the operation of an apprenticeship program established by Local 542 and several construction trade associations. Named as defendants were Local 542, the trade associations, the organization charged with administering the trade's apprenticeship program, and a class of approximately 1,400 construction industry employers. Petitioners, the defendant contractors and trade associations, seek review of a judgment granting an injunction against them. The questions we resolve are whether liability under 42 U.S.C. § 1981 requires proof of discriminatory intent and whether, absent such proof, liability can nevertheless be imposed vicariously on the employers and trade associations for the discriminatory conduct of the Union. 2 * The hiring hall system that is the focus of this litigation originated in a collective-bargaining agreement negotiated in 1961 by Local 542 and four construction trade associations in the Philadelphia area, three of whom are petitioners in this Court.1 The agreement was concluded only after a 10-week strike prompted by the resistance of the trade associations to the Union's demand for an exclusive hiring hall.2 Under the terms of the agreement, the Union was to maintain lists of operating engineers, or would-be engineers, classified according to the extent of their recent construction experience. Signatory employers were contractually obligated to hire operating engineers only from among those referred by the Union from its current lists. Workers affiliated with the Union were barred from seeking work with those employers except through Union referrals. Thus, the collective-bargaining agreement effectively channeled all employment opportunities through the hiring hall. Since 1961 this requirement has been a constant feature of contracts negotiated with Local 542 by the trade associations, as well as of contracts signed with the Union by employers who were not represented by one of those associations in collective bargaining.3 3 Among the means of gaining access to the Union's referral lists is an apprenticeship program established in 1965 by Local 542 and the trade associations. The program, which involves classroom and field training, is administered by the Joint Apprenticeship and Training Committee (JATC), a body of trustees half of whom are appointed by the Union and half by the trade associations. While enrolled in the program, apprentices are referred by the Union for unskilled construction work. Graduates of the program become journeymen operating engineers and are referred for heavy equipment jobs. 4 This action was filed in 1971 by the Commonwealth of Pennsylvania and 12 black plaintiffs representing a proposed class of minority group members residing within the jurisdiction of Local 542. The complaint charged that the Union and the JATC had violated numerous state and federal laws prohibiting employment discrimination, including Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. (1976 ed. and Supp.IV), and 42 U.S.C. § 1981. The complaint alleged that these defendants had engaged in a pattern and practice of racial discrimination, by systematically denying access to the Union's referral lists, and by arbitrarily skewing referrals in favor of white workers, limiting most minority workers who did gain access to the hiring hall to jobs of short hours and low pay. The contractor employers and trade associations were also named as defendants, although the complaint did not allege a Title VII cause of action against them.4 5 The District Court divided the trial into two stages. See Pennsylvania v. Local 542, Int'l Union of Operating Engineers, 469 F.Supp. 329, 348 (ED Pa. 1978). The first stage, from which petitioners appeal, addressed issues of liability; assessment of damages was deferred to a second stage. For purposes of the first phase of the proceedings, the court certified a plaintiff class of minority operating engineers and would-be engineers, as well as a defendant class consisting of all trade associations and employers who had been parties to labor contracts with Local 542. A single employer, petitioner Glasgow, Inc., was certified to represent the defendant subclass of approximately 1,400 contractor employers.5 6 The District Court's opinion in the liability phase of the trial is lengthy. For our purposes, however, the relevant findings and conclusions can be summarized briefly. First, the court found that the hiring hall system established by collective bargaining was neutral on its face. Id., at 342. Indeed, after May 1, 1971, the contracts contained a provision expressly prohibiting employment discrimination on the basis of race, religion, color, or national origin. Id., at 340, and n. 6. But the court found that Local 542, in administering the system, "practiced a pattern of intentional discrimination and that union practices in the overall operation of a hiring hall for operating engineers created substantial racial disparities." Id., at 370. The court made similar findings regarding the JATC's administration of the job-training program. Id., at 384. On the basis of these findings, the District Court held that Local 542 and the JATC had violated Title VII, both because they intentionally discriminated and because they enforced practices that resulted in a disparate racial impact. Id., at 397-399.6 The court also interpreted 42 U.S.C. § 1981 to permit imposition of liability "on roughly the same basis as a Title VII claim," 469 F.Supp., at 401, and therefore concluded that the Union and the JATC had also violated § 1981. Id., at 399-401. 7 Turning to petitioners' liability under § 1981, the court found that the plaintiffs had failed to prove "that the associations or contractors viewed simply as a class were actually aware of the union discrimination," id., at 401, and had failed to show "intent to discriminate by the employers as a class," id., at 412. Nevertheless, the court held the employers and the associations liable under § 1981 for the purpose of imposing an injunctive remedy "as a result of their contractual relationship to and use of a hiring hall system which in practice effectuated intentional discrimination, whether or not the employers and associations knew or should have known [of the Union's conduct]." Id., at 401. The court reasoned that liability under § 1981 "requires no proof of purposeful conduct on the part of any of the defendants." Id., at 407. Instead, it was sufficient that "(1) the employers delegated an important aspect of their hiring procedure to the union; [and that] (2) the union, in effectuating the delegation, intentionally discriminated or, alternatively, produced a discriminatory impact." Id., at 412. "[P]laintiffs have shown that the requisite relationship exists among employers, associations, and union to render applicable the theory of respondeat superior, thus making employers and associations liable injunctively for the discriminatory acts of the union." Id., at 413.7 8 Following an appeal authorized by 28 U.S.C. § 1292(b), the Court of Appeals for the Third Circuit, sitting en banc, affirmed the judgment of liability against petitioners by an equally divided vote. 648 F.2d 923 (1981). We granted certiorari, 454 U.S. 939, 102 S.Ct. 473, 70 L.Ed.2d 246 (1981), and we now reverse. II 9 The District Court held that petitioners had violated 42 U.S.C. § 1981 notwithstanding its finding that, as a class, petitioners did not intentionally discriminate against minority workers and neither knew nor had reason to know of the Union's discriminatory practices. The first question we address, therefore, is whether liability may be imposed under § 1981 without proof of intentional discrimination.8 Title 42 U.S.C. § 1981 provides: 10 "All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and to no other." 11 We have traced the evolution of this statute and its companion, 42 U.S.C. § 1982,9 on more than one occasion, see, e.g., McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 287-296, 96 S.Ct. 2574, 2582-2586, 49 L.Ed.2d 493 (1976); Runyon v. McCrary, 427 U.S. 160, 168-170, 96 S.Ct. 2586, 2593-2594, 49 L.Ed.2d 415 (1976); Jones v. Alfred H. Mayer Co., 392 U.S. 409, 422-437, 88 S.Ct. 2186, 2194-2202, 20 L.Ed.2d 1189 (1968), and we will not repeat the narrative again except in broad outline. 12 The operative language of both laws apparently originated in § 1 of the Civil Rights Act of 1866, 14 Stat. 27, enacted by Congress shortly after ratification of the Thirteenth Amendment.10 "The legislative history of the 1866 Act clearly indicates that Congress intended to protect a limited category of rights, specifically defined in terms of racial equality." Georgia v. Rachel, 384 U.S. 780, 791, 86 S.Ct. 1783, 1789, 16 L.Ed.2d 925 (1966). The same Congress also passed the Joint Resolution that was later adopted as the Fourteenth Amendment. See Cong.Globe, 39th Cong., 1st Sess., 3148-3149, 3042 (1866). As we explained in Hurd v. Hodge, 334 U.S. 24, 32-33, 68 S.Ct. 847, 851-852, 92 L.Ed. 1187 (1948) (footnotes omitted): 13 "Frequent references to the Civil Rights Act are to be found in the record of the legislative debates on the adoption of the Amendment. It is clear that in many significant respects the statute and the Amendment were expressions of the same general congressional policy. Indeed, as the legislative debates reveal, one of the primary purposes of many members of Congress in supporting the adoption of the Fourteenth Amendment was to incorporate the guaranties of the Civil Rights Act of 1866 in the organic law of the land. Others supported the adoption of the Amendment in order to eliminate doubt as to the constitutional validity of the Civil Rights Act as applied to the States." 14 Following ratification of the Fourteenth Amendment, Congress passed what has come to be known as the Enforcement Act of 1870, 16 Stat. 140, pursuant to the power conferred by § 5 of the Amendment. Section 16 of that Act contains essentially the language that now appears in § 1981.11 Indeed, the present codification is derived from § 1977 of the Revised Statutes of 1874, which in turn codified verbatim § 16 of the 1870 Act. Section 16 differed from § 1 of the 1866 Act in at least two respects. First, where § 1 of the 1866 Act extended its guarantees to "citizens, of every race and color," § 16 of the 1870 Act—and § 1981—protects "all persons." See United States v. Wong Kim Ark, 169 U.S. 649, 675, 18 S.Ct. 456, 467, 42 L.Ed. 890 (1898). Second, the 1870 Act omitted language contained in the 1866 Act, and eventually codified as § 1982, guaranteeing property rights equivalent to those enjoyed by white citizens. Thus, "[a]lthough the 1866 Act rested only on the Thirteenth Amendment . . . and, indeed, was enacted before the Fourteenth Amendment was formally proposed, . . . the 1870 Act was passed pursuant to the Fourteenth, and changes in wording may have reflected the language of the Fourteenth Amendment." Tillman v. Wheaton-Haven Recreation Assn., 410 U.S. 431, 439-440, n. 11, 93 S.Ct. 1090, 1095, n. 11, 35 L.Ed.2d 403 (1973). See Runyon v. McCrary, supra, 427 U.S. at 168-170, n. 8, 96 S.Ct. at 2593-2594, n. 8. 15 In determining whether § 1981 reaches practices that merely result in a disproportionate impact on a particular class, or instead is limited to conduct motivated by a discriminatory purpose, we must be mindful of the "events and passions of the time" in which the law was forged. United States v. Price, 383 U.S. 787, 803, 86 S.Ct. 1152, 1161, 16 L.Ed.2d 267 (1966). The Civil War had ended in April 1865. The First Session of the Thirty-ninth Congress met on December 4, 1865, some six months after the preceding Congress had sent to the States the Thirteenth Amendment and just two weeks before the Secretary of State certified the Amendment's ratification. On January 5, 1866, Senator Trumbull introduced the bill that would become the 1866 Act.12 16 The principal object of the legislation was to eradicate the Black Codes, laws enacted by Southern legislatures imposing a range of civil disabilities on freedmen.13 Most of these laws embodied express racial classifications and although others, such as those penalizing vagrancy, were facially neutral, Congress plainly perceived all of them as consciously conceived methods of resurrecting the incidents of slavery.14 Senator Trumbull summarized the paramount aims of his bill: 17 "Since the abolition of slavery, the Legislatures which have assembled in the insurrectionary States have passed laws relating to the freedmen, and in nearly all the States they have discriminated against them. They deny them certain rights, subject them to severe penalties, and still impose upon them the very restrictions which were imposed upon them in consequence of the existence of slavery, and before it was abolished. The purpose of the bill under consideration is to destroy all these discriminations, and to carry into effect the [Thirteenth] amendment." Cong.Globe, 39th Cong., 1st Sess., 474 (1866). 18 Senator Trumbull emphasized: "This bill has nothing to do with the political rights or status of parties. It is confined exclusively to their civil rights, such rights as should appertain to every free man." Id., at 476 (emphasis in original). 19 Of course, this Court has found in the legislative history of the 1866 Act evidence that Congress sought to accomplish more than the destruction of state-imposed civil disabilities and discriminatory punishments. We have held that both § 1981 and § 1982 "prohibit all racial discrimination, whether or not under color of law, with respect to the rights enumerated therein." Jones v. Alfred H. Mayer Co., 392 U.S., at 436, 88 S.Ct., at 2201. See Johnson v. Railway Express Agency, Inc., 421 U.S. 454, 459-460, 95 S.Ct. 1716, 1719-1720, 44 L.Ed.2d 295 (1975); Runyon v. McCrary, 427 U.S., at 168, 96 S.Ct., at 2593. Nevertheless, the fact that the prohibitions of § 1981 encompass private as well as governmental action does not suggest that the statute reaches more than purposeful discrimination, whether public or private. Indeed, the relevant opinions are hostile to such an implication. Thus, although we held in Jones, supra, that § 1982 reaches private action, we explained that § 1 of the 1866 Act "was meant to prohibit all racially motivated deprivations of the rights enumerated in the statute." 392 U.S., at 426, 88 S.Ct., at 2196 (emphasis on "racially motivated" added). Similarly, in Runyon v. McCrary, supra, we stated that § 1981 would be violated "if a private offeror refuses to extend to a Negro, solely because he is a Negro, the same opportunity to enter into contracts as he extends to white offerees." 427 U.S., at 170-171, 96 S.Ct., at 2594. 20 The immediate evils with which the Thirty-ninth Congress was concerned simply did not include practices that were "neutral on their face, and even neutral in terms of intent," Griggs v. Duke Power Co., 401 U.S. 424, 430, 91 S.Ct. 849, 853, 28 L.Ed.2d 158 (1971), but that had the incidental effect of disadvantaging blacks to a greater degree than whites. Congress instead acted to protect the freedmen from intentional discrimination by those whose object was "to make their former slaves dependent serfs, victims of unjust laws, and debarred from all progress and elevation by organized social prejudices." Cong.Globe, 39th Cong., 1st Sess., 1839 (1866) (Rep. Clarke). See Memphis v. Greene, 451 U.S. 100, 131-135, 101 S.Ct. 1584, 1602-1604, 67 L.Ed.2d 769 (1981) (WHITE, J., concurring in judgment). The supporters of the bill repeatedly emphasized that the legislation was designed to eradicate blatant deprivations of civil rights, clearly fashioned with the purpose of oppressing the former slaves. To infer that Congress sought to accomplish more than this would require stronger evidence in the legislative record than we have been able to discern.15 21 Our conclusion that § 1981 reaches only purposeful discrimination is supported by one final observation about its legislative history. As noted earlier, the origins of the law can be traced to both the Civil Rights Act of 1866 and the Enforcement Act of 1870. Both of these laws, in turn, were legislative cousins of the Fourteenth Amendment. The 1866 Act represented Congress' first attempt to ensure equal rights for the freedmen following the formal abolition of slavery effected by the Thirteenth Amendment. As such, it constituted an initial blueprint of the Fourteenth Amendment, which Congress proposed in part as a means of "incorporat[ing] the guaranties of the Civil Rights Act of 1866 in the organic law of the land." Hurd v. Hodge, 334 U.S., at 32, 68 S.Ct., at 851.16 The 1870 Act, which contained the language that now appears in § 1981, was enacted as a means of enforcing the recently ratified Fourteenth Amendment. In light of the close connection between these Acts and the Amendment, it would be incongruous to construe the principal object of their successor, § 1981, in a manner markedly different from that of the Amendment itself.17 22 With respect to the latter, "official action will not be held unconstitutional solely because it results in a racially disproportionate impact," Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 264-265, 97 S.Ct. 555, 562-563, 50 L.Ed.2d 450 (1977). "[E]ven if a neutral law has a disproportionately adverse impact upon a racial minority, it is unconstitutional under the Equal Protection Clause only if that impact can be traced to a discriminatory purpose." Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 272, 99 S.Ct. 2282, 2292, 60 L.Ed.2d 870 (1979). See Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976). The same Congress that proposed the Fourteenth Amendment also passed the Civil Rights Act of 1866, and the ratification of that Amendment paved the way for the Enforcement Act of 1870. These measures were all products of the same milieu and were directed against the same evils. Although Congress might have charted a different course in enacting the predecessors to § 1981 than it did in proposing the Fourteenth Amendment, we have found no convincing evidence that it did so. 23 We conclude, therefore, that § 1981, like the Equal Protection Clause, can be violated only by purposeful discrimination. III 24 The District Court held petitioners liable under § 1981 notwithstanding its finding that the plaintiffs had failed to prove intent to discriminate on the part of the employers and associations as a class. In light of our holding that § 1981 can be violated only by intentional discrimination, the District Court's judgment can stand only if liability under § 1981 can properly rest on some ground other than the discriminatory motivation of the petitioners themselves. Both the District Court and respondents have relied on such grounds, but we find them unconvincing. A. 25 The District Court reasoned that liability could be vicariously imposed upon the employers and associations, based upon the intentional discrimination practiced by Local 542 in its operation of the hiring hall. The court's theory was that petitioners had delegated to the "union hiring hall" the authority to select workers as "the agent for two principals—the union and the contractors, with their respective associations." 469 F.Supp., at 411. Since the hiring hall came into existence only through the agreement of petitioners, and since the exclusive hiring hall was the means by which "the intentional discrimination of the union was able to work its way broadly into the common workforce of operating engineers," id., at 412, the court concluded that "[t]he acts of the union therefore justify imposition of responsibility upon those employers participating in the original delegation," ibid. The effect of this holding, as the court recognized, was to impose a "duty to see that discrimination does not take place in the selection of one's workforce," regardless of where the discrimination originates. Ibid. 26 As applied to the petitioner associations, the District Court's theory is flawed on its own terms. The doctrine of respondeat superior, as traditionally conceived and as understood by the District Court, see id., at 411, enables the imposition of liability on a principal for the tortious acts of his agent and, in the more common case, on the master for the wrongful acts of his servant. See Restatement (Second) of Agency §§ 215-216, 219 (1958) (Restatement); W. Prosser, Law of Torts §§ 69-70 (4th ed. 1971) (Prosser); W. Seavey, Law of Agency § 83 (1964) (Seavey). "Agency is the fiduciary relation which results from the manifestation of consent by one person to another that the other shall act on his behalf and subject to his control, and consent by the other so to act." Restatement § 1. A master-servant relationship is a form of agency in which the master employs the servant as "an agent to perform service in his affairs" and "controls or has the right to control the physical conduct of the other in the performance of the service." Id., § 2. See 2 F. Harper & F. James, Law of Torts § 26.6 (1956) (Harper & James). Local 542, in its operation of the hiring hall, simply performed no function as the agent or servant of the associations. The record demonstrates that the associations themselves do not hire operating engineers, and never have. Their primary purpose is to represent certain employers in contract negotiations with the Union. Even if the doctrine of respondeat superior were broadly applicable to suits based on § 1981, therefore, it would not support the imposition of liability on a defendant based on the acts of a party with whom it had no agency or employment relationship.18 27 We have similar difficulty in accepting the application of traditional respondeat superior doctrine to the class of contractor employers. In the run of cases, the relationship between an employer and the union that represents its employees simply cannot be accurately characterized as one between principal and agent or master and servant. Indeed, such a conception is alien to the fundamental assumptions upon which the federal labor laws are structured. 28 At the core of agency is a "fiduciary relation" arising from the "consent by one person to another that the other shall act on his behalf and subject to his control." Restatement § 1. Equally central to the master-servant relation is the master's control over or right to control the physical activities of the servant. See Restatement id., § 220; 2 Harper & James, § 26.3; Seavey, § 84, p. 142. See also Logue v. United States, 412 U.S. 521, 527, 93 S.Ct. 2215, 2219, 37 L.Ed.2d 121 (1973). The District Court found that the requirement of control was satisfied because "the employers retained power to oppose the union discrimination." 469 F.Supp., at 411, n. 61. However, the "power to oppose" the Union, even when the opposition is grounded in the terms of the collective-bargaining agreement, is not tantamount to a "right to control" the Union. See Lummus Co. v. NLRB, 119 U.S.App.D.C. 229, 236, 339 F.2d 728, 735 (1964).19 Indeed, a rule equating the two would convert every contractual relationship into an agency relationship, a result clearly unsupported by the common-law doctrines on which the District Court relied. 29 The District Court's assumptions about the relation between the Union and the class of employers with whom it has contracted also runs counter to the premises on which the federal labor laws have been constructed. While authorizing collective bargaining and providing means of enforcing the resultant contracts, the National Labor Relations Act expressly prohibits employers from compromising the independence of labor unions. See 49 Stat. 452, as amended, 29 U.S.C. § 158(a); 61 Stat. 157, as amended, 29 U.S.C. § 186. The entire process of collective bargaining is structured and regulated on the assumption that "[t]he parties even granting the modification of views that may come from a realization of economic interdependence—still proceed from contrary and to an extent antagonistic viewpoints and concepts of self-interest." NLRB v. Insurance Agents, 361 U.S. 477, 488, 80 S.Ct. 419, 426, 4 L.Ed.2d 454 (1960). See Vaca v. Sipes, 386 U.S. 171, 177, 87 S.Ct. 903, 909, 17 L.Ed.2d 842 (1967). We have no reason to doubt the validity of that assumption in the instant case. 30 Respondents also suggest that petitioners can be held vicariously liable for the discriminatory conduct of the JATC. They argue that the JATC is properly viewed as an agent of both Local 542 and the associations, emphasizing that half of the trustees charged with administering the JATC are appointed by the associations and that the JATC is wholly funded by mandatory contributions from the employers. We note initially that the District Court premised petitioners' liability not on the actions of the JATC, but on the discriminatory conduct of the Union. See 469 F.Supp., at 411-413. The record, therefore, contains no findings regarding the relationship between the JATC and petitioners, beyond those noted above, that might support application of respondeat superior. The facts emphasized by respondents, standing alone, are inadequate. That the employers fund the activities of the JATC does not render the JATC the employers' servant or agent any more than an independent contractor is rendered an agent simply because he is compensated by the principal for his services. The employers must also enjoy a right to control the activities of the JATC, and there is no record basis for believing that to be the case. Neither is a right of control inferable merely from the power of the associations to appoint half of the JATC's trustees. It is entirely possible that the trustees, once appointed, owe a fiduciary duty to the JATC and the apprentices enrolled in its programs, rather than to the entities that appointed them. Cf. NLRB v. Amax Coal Co., 453 U.S. 322, 101 S.Ct. 2789, 69 L.Ed.2d 72 (1981). On the assumption that respondeat superior applies to suits based on § 1981, there is no basis for holding either the employers or the associations liable under that doctrine without evidence that an agency relationship existed at the time the JATC committed the acts on which its own liability was premised. B 31 The District Court also justified its result by concluding that § 1981 imposes a "nondelegable duty" on petitioners "to see that discrimination does not take place in the selection of [their] workforce." 469 F.Supp., at 412.20 The concept of a nondelegable duty imposes upon the principal not merely an obligation to exercise care in his own activities, but to answer for the well-being of those persons to whom the duty runs. See Restatement § 214. The duty is not discharged by using care in delegating it to an independent contractor. Consequently, the doctrine creates an exception to the common-law rule that a principal normally will not be liable for the tortious conduct of an independent contractor. See 2 Harper & James, § 26.11, pp. 1405-1408; Prosser, § 70, p. 467, § 71, p. 470. So understood, a nondelegable duty is an affirmative obligation to ensure the protection of the person to whom the duty runs. 32 In a sense, to characterize such a duty as "nondelegable" is merely to restate the duty. Thus, in this litigation the question is not whether the employers and associations are free to delegate their duty to abide by § 1981, for whatever duty the statute imposes, they are bound to adhere to it. The question is what duty does § 1981 impose. More precisely, does § 1981 impose a duty to refrain from intentionally denying blacks the right to contract on the same basis as whites or does it impose an affirmative obligation to ensure that blacks enjoy such a right? The language of the statute does not speak in terms of duties. It merely declares specific rights held by "[a]ll persons within the jurisdiction of the United States." We are confident that the Thirty-ninth Congress meant to do no more than prohibit the employers and associations in these cases from intentionally depriving black workers of the rights enumerated in the statute, including the equal right to contract. It did not intend to make them the guarantors of the workers' rights as against third parties who would infringe them. Cf. Furnco Construction Corp. v. Waters, 438 U.S. 567, 577-578, 98 S.Ct. 2943, 2949-2950, 57 L.Ed.2d 957 (1978) (Title VII); Rizzo v. Goode, 423 U.S. 362, 376-377, 96 S.Ct. 598, 606-607, 46 L.Ed.2d 561 (1976) ( 42 U.S.C. § 1983). 33 Our earlier holding that § 1981 reaches only intentional discrimination virtually compels this conclusion. It would be anomalous to hold that § 1981 could be violated only by intentional discrimination and then to find this requirement satisfied by proof that the individual plaintiffs did not enjoy "the same right . . . to make and enforce contracts . . . as is enjoyed by white citizens" and that the defendants merely failed to ensure that the plaintiffs enjoyed employment opportunities equivalent to that of whites. Such a result would be particularly inappropriate in the case of the associations, who are not engaged in the construction business, do not employ operating engineers, and consequently did not delegate to the Union any hiring functions which they otherwise would have performed themselves. Neither the District Court nor respondents identify anything in the language or legislative history of the statute to support a contrary conclusion.21 IV 34 In a separate portion of their brief, respondents urge several independent bases for the issuance of an injunction against the petitioners and the allocation to them of a portion of the costs of the remedial decree. Respondents first assert that the court had inherent equitable power to allocate remedial costs among all the named defendants. They also rely on the All Writs Act, 28 U.S.C. § 1651(a), as an independent basis for the injunctive portions of the District Court's order running against petitioners. We shall deal with these contentions in turn. 35 The District Court in an opinion issued after judgment set forth the basis for its holding that "defendants held injunctively liable solely under a theory of vicarious responsibility are nevertheless liable for 'a share' of the costs under Rule 54(d)." Pennsylvania v. Local 542, Int'l Union of Operating Engineers, 507 F.Supp. 1146, 1152 (1980). The District Court framed the inquiry before it as whether a party held vicariously liable to an injunction, but not for damages, might nonetheless have a proportionate share of the costs assessed against it. While this may have been an entirely appropriate frame of reference for the District Court, following its holding that petitioners were vicariously liable and therefore subject to an injunction, it is obviously not the proper frame of reference for our discussion. For the reasons previously stated, we have concluded that petitioners were not properly subject to an injunction on any of the theories set forth by the District Court. The issue before us, therefore, is whether a party not subject to liability for violating the law may nonetheless be assessed a proportionate share of the costs of implementing a decree to assure nondiscriminatory practices on the part of another party which was properly enjoined. 36 We find respondent's arguments based on the traditional equitable authority of courts to be unpersuasive. In Milliken v. Bradley, 433 U.S. 267, 97 S.Ct. 2749, 53 L.Ed.2d 745 (1977), upon which respondents rely, and which we believe to be the case most closely in point, we expressly noted that the state petitioners had been found guilty of creating at least a portion of the constitutional violation which the order challenged in that case was designed to remedy. Id., at 281-282, 289, 97 S.Ct., at 2757-2758, 2761. Thus our holding there was consistent with our opinion in Hills v. Gautreaux, 425 U.S. 284, 96 S.Ct. 1538, 47 L.Ed.2d 792 (1976), where we explained the relationship between our holding in the first Milliken case, Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974), and our opinion in Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 91 S.Ct. 1267, 28 L.Ed.2d 554 (1971). We read these earlier decisions as recognizing "fundamental limitations on the remedial powers of the federal courts." 425 U.S., at 293, 96 S.Ct., at 1544. Those powers could be exercised only on the basis of a violation of the law and could extend no farther than required by the nature and the extent of that violation. Id., at 293-294, 96 S.Ct., at 1544. This principle, we held, was not one limited to school desegregation cases, but was instead "premised on a controlling principle governing the permissible scope of federal judicial power, a principle not limited to a school desegregation context." Id., at 294, n. 11, 96 S.Ct., at 1545, n. 11. 37 We think that the principle enunciated in these cases, transposed to the instant factual situation, offers no support for the imposition of injunctive relief against a party found not to have violated any substantive right of respondents. This is not to say that defendants in the position of petitioners might not, upon an appropriate evidentiary showing, be retained in the lawsuit and even subjected to such minor and ancillary provisions of an injunctive order as the District Court might find necessary to grant complete relief to respondents from the discrimination they suffered at the hands of the Union. See Zipes v. Trans World Airlines, Inc., 455 U.S. 385, 399-400, 102 S.Ct. 1127, 1135-1136, 71 L.Ed.2d 234 (1982). But that sort of minor and ancillary relief is not the same, and cannot be the same, as that awarded against a party found to have infringed the statutory rights of persons in the position of respondents. 38 The order of the District Court, insofar as it runs against petitioners, cannot be regarded as "minor" or "ancillary" in any proper sense of those terms. First, it imposes considerable burdens on the employers and associations. It directs the employers to meet detailed "minority utilization goals" in their hiring, keyed to the number of hours worked. App. to Pet. for Cert. in No. 81-280, p. 236. If they are unable to do so through referrals from Local 542, they are required to hire minority operating engineers who are not affiliated with the Union. Ibid. If the goals are still not satisfied, the employers must recruit and hire unskilled minority workers from the community and provide on-the-job training. Id., at 236-237. The employers are also obligated to make quarterly reports detailing the extent of their compliance with these directives. Id., at 241-242. Finally, the District Court imposed on the employers and the associations a share of the financial cost incidental to enforcement of the remedial decree as a whole. Id., at 252-254. See 507 F.Supp. 1146 (1980). According to petitioners, the expense of the decree in the first year of its 5-year life exceeded $200,000. See Brief for Petitioner in No. 81-280, p. 45, n. 77. 39 Absent a supportable finding of liability, we see no basis for requiring the employers or the associations to aid either in paying for the cost of the remedial program as a whole or in establishing and administering the training program. Nor is the imposition of minority hiring quotas directly upon petitioners the sort of remedy that may be imposed without regard to a finding of liability. If the Union and the JATC comply with the decree by training and referring minority workers, we see no reason to assume, absent supporting evidence, that the employers will not hire the minority workers referred pursuant to the collective-bargaining agreement, and employ them at wages and hours commensurate with those of nonminority workers. If experience proves otherwise, the District Court will then have more than sufficient grounds for including the employers within the scope of the remedial decree. 40 To the extent that the remedy properly imposed upon the Union and the JATC requires any adjustment in the collective-bargaining contract between petitioners and the Union, it is entirely appropriate for the District Court to fashion its injunctive remedy to so provide, and to have that remedy run against petitioners as well as the Union and the JATC. But the injunctive decree entered by the District Court as presently drawn treats petitioners as if they had been properly found liable for the Union's discrimination. A decree containing such provisions, we hold, is beyond the traditional equitable limitations upon the authority of a federal court to formulate such decrees. 41 Nor does the All Writs Act, 28 U.S.C. § 1651(a), support the extensive liability imposed upon petitioners by the District Court. The District Court did not rely upon this Act, and we think it completely wide of the mark in justifying the relief granted by the District Court. That Act was most recently considered by this Court in United States v. New York Telephone Co., 434 U.S. 159, 98 S.Ct. 364, 54 L.Ed.2d 376 (1977), where we said: "This Court has repeatedly recognized the power of a federal court to issue such commands under the All Writs Act as may be necessary or appropriate to effectuate and prevent the frustration of orders it has previously issued in its exercise of jurisdiction otherwise obtained. . . ." Id., at 172, 98 S.Ct., at 372. In New York Telephone, we held that the All Writs Act was available to require a third party to assist in the carrying out of a District Court order pertaining to the installation of pen registers, and in doing so we noted that "[t]he order provided that the Company be fully reimbursed at prevailing rates, and compliance with it required minimal effort on the part of the Company and no disruption to its operations." Id., at 175, 98 S.Ct., at 373. 42 An examination of our cases which have relied on the All Writs Act convinces us that respondents are simply barking up the wrong tree when they seek to support the injunctive order of the District Court against petitioners on the basis of the provisions of that Act. There was no need for the District Court to treat petitioners as strangers to this lawsuit, and therefore to rely upon some extraordinary form of process or writ to bring them before the court. Petitioners had been named as defendants by respondents in their complaint, and they litigated the injunctive liability phase of the action before the District Court. Petitioners were parties to the action in every sense of the word, and subject to the jurisdiction of the District Court both as to the imposition of liability and as to the framing of a remedial decree. The difficulty faced by respondents in supporting the decree of the District Court insofar as it grants affirmative relief and requires payment toward the cost of implementing the decree is not that petitioners would otherwise be strangers to the action. The difficulty lies instead with the fact that on the record before the District Court the petitioners could not properly be held liable to any sort of injunctive relief based on their own conduct. 43 Thus insofar as respondents' arguments for the imposition of remedial obligations upon petitioners rests upon the assumption that petitioners were properly found liable for the violation of respondents' rights to be free from discrimination, that assumption can no longer stand in view of the conclusions previously set forth in this opinion. Insofar as respondents' assertions are based on some authority of the District Court to impose the sort of obligations which it did upon petitioners even though petitioners could not be held liable on the record before the District Court, we hold that such obligations can be imposed neither under traditional equitable authority of the District Court nor under the All Writs Act.22 44 The judgment of the Court of Appeals is reversed, and the case is remanded for proceedings consistent with this opinion. 45 It is so ordered. 46 Justice O'CONNOR, with whom Justice BLACKMUN joins, concurring. 47 I concur in the Court's opinion today holding that a cause of action based on 42 U.S.C. § 1981 requires proof of intent to discriminate, that the employers cannot be held vicariously liable for the discrimination practiced by Local 542, and that § 1981 does not impose a "nondelegable duty" on the employers to insure that there is no discrimination in the Union's selection of the work force. I write separately, however, in order to state expressly one of the options open to the District Court on remand, and to elaborate on the Court's comments regarding the scope of the federal courts' equitable power to afford full relief. 48 * In determining that the petitioners cannot be held vicariously liable for the discriminatory conduct of the JATC, the Court is careful to note that its holding is based on the failure of the trial court to make "findings regarding the relationship between the JATC and petitioners . . . that might support application of respondeat superior." Ante, at 394.1 In particular, because the record contains no findings regarding whether the employers maintain some control over the activities of the JATC, either through the employer-appointed trustees or through other means, the doctrine of respondeat superior is simply inapplicable. 49 I would briefly note the limits of the Court's holding. Once this case has been remanded to the District Court, nothing in the Court's opinion prevents the respondents from litigating the question of the employers' liability under § 1981 by attempting to prove the traditional elements of respondeat superior. II 50 Regarding the scope of a federal court's equitable powers to afford full relief, I agree with the Court's holding that "a party not subject to liability for violating the law [may not] be assessed a proportionate share of the costs of implementing a decree to assure nondiscriminatory practices on the part of another party which was properly enjoined." Ante, at 398.2 I also agree with the Court's ancillary holding that the District Court may not require quarterly reports from the employers detailing their compliance with the court's ill-founded injunction. Of course, since the employers are not liable for general injunctive relief, such reports are unnecessary. 51 Under the appropriate circumstances, however, I believe other reports properly could be required of the employers, for example, to aid the court by charting the changes resulting from the injunction imposed on the Union and the JATC. Quite recently, in Zipes v. Trans World Airlines, Inc., 455 U.S. 385, 102 S.Ct. 1127, 71 L.Ed.2d 234 (1982), this Court held that § 706(g) of Title VII of the Civil Rights Act of 1964 authorizes a federal court to order retroactive seniority relief over the objections of a union that was not guilty of discrimination. The Court stated: 52 "Teamsters v. United States, 431 U.S. 324 [97 S.Ct. 1843, 52 L.Ed.2d 396] (1977), . . . makes it clear that once there has been a finding of discrimination by the employer, an award of retroactive seniority is appropriate even if there is no finding that the union has also illegally discriminated. In Teamsters, the parties agreed to a decree which provided that the District Court would decide 'whether any discriminatees should be awarded additional equitable relief such as retroactive seniority.' Id., at 331, n. 4 [97 S.Ct., at 1852, n. 4]. Although we held that the union had not violated Title VII by agreeing to and maintaining the seniority system, we nonetheless directed the union to remain in the litigation as a defendant so that full relief could be awarded the victims of the employer's post-Act discrimination. Id., at 356, n. 43 [97 S.Ct., at 1865, n. 43]." Id., at 400, 102 S.Ct., at 1136.3 53 As the Court acknowledges today, it is entirely possible that full relief cannot be granted without subjecting the petitioners to some incidental or ancillary provisions of the court's injunctive order. It is thus conceivable, for example, that quarterly reports providing employment statistics necessary for the court to ascertain whether its injunctive decree is being properly implemented could be ordered under the court's equitable powers to effectuate its decree. 54 Justice STEVENS, concurring in part and concurring in the judgment. 55 As I noted in my separate opinion in Runyon v. McCrary, 427 U.S. 160, 189, 96 S.Ct. 2586, 2603, 49 L.Ed.2d 415, the Congress that enacted § 1 of the Civil Rights Act of 1866 "intended only to guarantee all citizens the same legal capacity to make and enforce contracts, to obtain, own, and convey property, and to litigate and give evidence." Any violation of that guarantee—whether deliberate, negligent, or purely accidental—would, in my opinion, violate 42 U.S.C. § 1981. The statute itself contains no requirement that an intent to discriminate must be proved. 56 The Court has broadened the coverage of § 1981 far beyond the scope actually intended by its authors; in essence, the Court has converted a statutory guarantee of equal rights into a grant of equal opportunities. See Jones v. Alfred H. Mayer Co., 392 U.S. 409, 88 S.Ct. 2186, 20 L.Ed.2d 1189; Runyon v. McCrary, supra. Whether or not those decisions faithfully reflect the intent of Congress, the enlarged coverage of the statute "is now an important part of the fabric of our law." Runyon, supra, 427 U.S., at 190, 96 S.Ct., at 2604 (STEVENS, J., concurring). 57 Since I do not believe Congress intended § 1981 to have any application at all in the area of employment discrimination generally covered by Title VII of the Civil Rights Act of 1964, an analysis of the motives and intent of the Reconstruction Congress cannot be expected to tell us whether proof of intentional discrimination should be required in the judicially created portion of the statute's coverage. Since Congress required no such proof in the statute it actually enacted, a logician would be comfortable in concluding that no such proof should ever be required. Nevertheless, since that requirement tends to define the entire coverage of § 1981 in a way that better reflects the basic intent of Congress than would a contrary holding, I concur in the conclusion reached by the Court in Part II of its opinion insofar as it relates to the statutory protection of equal opportunity but, perhaps illogically, would reach a different conclusion in a case challenging a denial of a citizen's civil rights. 58 Accordingly, I join the Court's judgment and Parts III and IV of its opinion. 59 Justice MARSHALL, with whom Justice BRENNAN joins, dissenting. 60 Today the Court reaches out and decides that 42 U.S.C. § 1981 requires proof of an intent to discriminate—an issue that is not at all necessary to the disposition of these cases. Because I find no support for the majority's resolution of this issue, and because I disagree with its disposition of these cases even if proof of intent should ordinarily be required, I respectfully dissent. 61 * The question whether intent generally should be required in § 1981 actions is at most tangentially related to these cases. There was unquestionably intentional discrimination on the part of both the union (Local 542) and the Joint Apprenticeship and Training Committee (JATC), a body composed of officials from the union and the petitioner contracting associations, which jointly administered the apprenticeship and training program. As a result, the only question that the Court need address today is whether limited injunctive liability may be vicariously imposed upon an employer when the person or entity to whom it delegates a large portion of its hiring decisions intentionally discriminates on the basis of race. However, because the majority has chosen to reach first the more general question whether proof of intent is a prerequisite to recovery in a § 1981 action, I likewise will address this issue first. 62 Section 1981 provides in unqualified terms: 63 "All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens. . . ." 42 U.S.C. § 1981. 64 The plain language does not contain or suggest an intent requirement. A violation of § 1981 is not expressly conditioned on the motivation or intent of any person. The language focuses on the effects of discrimination on the protected class, and not on the intent of the person engaging in discriminatory conduct. Nothing in the statutory language implies that a right denied because of sheer insensitivity, or a pattern of conduct that disproportionately burdens the protected class of persons, is entitled to any less protection than one denied because of racial animus. 65 The Court attaches no significance to the broad and unqualified language of § 1981. Furthermore, the majority finds no support for its conclusion that intent should be required in the legislative history to § 1 of the 1866 Act, the precursor to § 1981. Instead, in the face of this unqualified language and the broad remedial purpose § 1981 was intended to serve, the majority assumes that Congress intended to restrict the scope of the statute to those situations in which racial animus can be proved on the ground that the legislative history contains no "convincing evidence" to the contrary. Ante, at 391. In my view, this approach to statutory construction is not only unsound, it is also contrary to our prior decisions, which have consistently given § 1981 as broad an interpretation as its language permits. See, e.g., McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 96 S.Ct. 2574, 49 L.Ed.2d 493 (1976); Runyon v. McCrary, 427 U.S. 160, 96 S.Ct. 2586, 49 L.Ed.2d 415 (1976); Johnson v. Railway Express Agency, Inc., 421 U.S. 454, 95 S.Ct. 1716, 44 L.Ed.2d 295 (1975); Tillman v. Wheaton-Haven Recreation Assn., 410 U.S. 431, 93 S.Ct. 1090, 35 L.Ed.2d 403 (1973); Sullivan v. Little Hunting Park, Inc., 396 U.S. 229, 90 S.Ct. 400, 24 L.Ed.2d 386 (1969); Jones v. Alfred H. Mayer Co., 392 U.S. 409, 88 S.Ct. 2186, 20 L.Ed.2d 1189 (1968). 66 The fallacy in the Court's approach is that, in construing § 1981 and its legislative history, the Court virtually ignores Congress' broad remedial purposes and our paramount national policy of eradicating racial discrimination and its pernicious effects. When viewed in this light, it is clear that proof of intentional discrimination should not be required in order to find a violation of § 1981. 67 Although the Thirty-ninth Congress that passed the Civil Rights Act of 1866 did not specifically address the question whether intent should be required, the conclusion is inescapable that the congressional leadership intended to effectuate "the result of a change from a centuries old social system based on involuntary labor, with all the notions of racial unsuitability for the performance of anything but menial labor under close supervision, to the free labor system." Croker v. Boeing Co., 662 F.2d 975, 1006 (CA3 1981) (Gibbons, J., with whom Higginbotham and Sloviter, JJ., joined, dissenting in part) (emphasis in original). When this Congress convened, the Thirteenth Amendment had been ratified, abolishing slavery as a legal status. However, it was clear that in reality, Negroes were hardly accorded the employment and other opportunities accorded white persons generally. Thus, this Congress undertook to provide in fact the rights and privileges that were available to Negroes in theory. See generally J. tenBroek, The Antislavery Origins of the Fourteenth Amendment 156-180 (1951) (discussing the intent of the Thirty-ninth Congress to ensure to Negroes the practical freedom and equality which was already present at law, to reach private, not merely governmental conduct, and to provide affirmative obligations on the government to protect Negroes from unequal treatment). Four separate but related measures were proposed in an effort to accomplish this purpose.1 68 In this general climate, the 1866 Civil Rights Act was not an isolated technical statute dealing with only a narrow subject. Instead, it was an integral part of a broad congressional scheme intended to work a major revolution in the prevailing social order.2 It is inconceivable that the Congress which enacted this statute would permit this purpose to be thwarted by excluding from the statute private action that concededly creates serious obstacles to the pursuit of job opportunities by Negroes solely because the aggrieved persons could not prove that the actors deliberately intended such a result. Even less conceivable is the notion, embraced by the Court's opinion today, that this Congress intended to absolve employers from even injunctive liability imposed as a result of intentional discrimination practiced by the persons to whom they had delegated their authority to hire employees. See infra, at 414-418. 69 The legislative history demonstrates that the Thirty-ninth Congress intended not merely to provide a remedy for preexisting rights, but to eradicate the "badges of slavery" that remained after the Civil War and the enactment of the Thirteenth Amendment. Congress was acutely aware of the difficulties that federal officials had encountered in effectuating the change from the system of slavery to a system of free labor even though the legal and constitutional groundwork for this change had already been laid. In the report that formed the working paper for the Joint Committee on Reconstruction and was of central importance to the deliberations of the Thirty-ninth Congress, General Schurz noted: 70 "That the result of the free labor experiment made under circumstances so extremely unfavorable should at once be a perfect success, no reasonable person would expect. Nevertheless, a large majority of the southern men with whom I came into contact announced their opinions with so positive an assurance as to produce the impression that their minds were fully made up. In at least nineteen cases of twenty the reply I received to my inquiry about their views on the new system was uniformly this: 'You cannot make the negro work without physical compulsion.' I heard this hundreds of times, heard it wherever I went, heard it in nearly the same words from so many different persons, that at last I came to the conclusion that this is the prevailing sentiment among the southern people. There are exceptions to the rule, but, as far as my information extends, far from enough to affect the rule. In the accompanying documents you will find an abundance of proof in support of this statement. There is hardly a paper relative to the negro question annexed to this report which does not, in some direct or indirect way, corroborate it." S.Exec.Doc.No. 2, 39th Cong., 1st Sess. (1865), reprinted in The Reconstruction Amendments' Debates 88 (Virginia Comm'n on Constitutional Government (1967). 71 Fully aware of this prevailing attitude, the leaders of Congress set about to enact legislation that would ensure to Negroes the opportunity to participate equally in the free labor system by providing an instrument by which they could strike down barriers to their participation, whether those barriers were erected with the conscious intent to exclude or with callous indifference to exclusionary effects. Congress knew that this attitude could manifest itself in a number of different ways and intended to protect Negro workers against not only flagrant, intentional discrimination, but also against more subtle forms of discrimination which might successfully camouflage the intent to oppress through facially neutral policies. Congressional awareness of the potential role that facially neutral measures might play in impeding the ability of Negroes to enjoy equal job opportunities is also reflected in the working paper which formed the basis for the 1866 Act. Addressing this problem, General Schurz stated: 72 "What particular shape the reactionary movement will assume it is at present unnecessary to inquire. There are a hundred ways of framing apprenticeship, vagrancy, or contract laws, which will serve the purpose. . . ." Id., at 92. 73 Unfortunately, this awareness seems utterly lacking in the Court's opinion today. In order to hold that § 1981 requires a showing of intent, the majority must assume that the rights guaranteed under § 1981—to make and enforce contracts on the same basis as white persons—can be adequately protected by limiting the statute to cases where the aggrieved person can prove intentional discrimination. In taking this extraordinarily naive view, the Court shuts its eyes to reality, ignoring the manner in which racial discrimination most often infects our society. Today, although flagrant examples of intentional discrimination still exist, discrimination more often occurs "on a more sophisticated and subtle level," the effects of which are often as cruel and "devastating as the most crude form of discrimination." Pennsylvania v. Local 542, Int'l Union of Operating Engineers, 469 F.Supp. 329, 337 (ED Pa.1978) (Higginbotham, Circuit Judge, sitting by designation).3 I think that Judge Higginbotham most accurately recognized this problem when he noted that "[t]he facts of the instant case . . . demonstrate the complexity and subtlety of the interrelationship of race, collective bargaining, craft unions, the employment process and that ultimate goal—real jobs." Ibid. He further noted that "[a]t the critical level of viable jobs and equal opportunities, there were intentional and persistent efforts to exclude and discourage most of the minorities who, but for their race, would have been considered for entry into the union and for the more lucrative jobs." Ibid. 74 Racial discrimination in all areas, and particularly in the areas of education and employment, is a devastating and reprehensible policy that must be vigilantly pursued and eliminated from our society: 75 "Racial discrimination can be the most virulent of strains that infect a society, and the illness in any society so infected can be quantified. Exposure to embarrassment, humiliation, and the denial of basic respect can and does cause psychological and physiological trauma to its victims. This disease must be recognized and vigorously eliminated wherever it occurs. But racial discrimination takes its most malevolent form when it occurs in employment, for prejudice here not only has an immediate economic effect, it has a fulminating integrant that perpetuates the pestilences of degraded housing, unsatisfactory neighborhood amenities, and unequal education." Croker v. Boeing Co., 662 F.2d, at 1002 (Aldisert, J., with whom Higginbotham, J., joined, dissenting in part). 76 The purposes behind § 1981, and the profound national policy of blotting out all vestiges of racial discrimination, are no less frustrated when equal opportunities are denied through cleverly masked or merely insensitive practices, where proof of actual intent is nearly impossible to obtain, than when instances of intentional discrimination escape unremedied. For this reason, I cannot accept the Court's glib and unrealistic view that requiring proof of intent in § 1981 actions does not frustrate that statute's purpose of protecting against the devastating effects of racial discrimination in employment. II 77 Even if I agreed with the Court that intent must be proved in a § 1981 action, I could not agree with its conclusion that the petitioner contracting associations should be immunized, even from injunctive liability, for the intentional discrimination practiced by the union hall to which they delegated a major portion of their hiring decisions. Under § 1981, minorities have an unqualified right to enter into employment contracts on the same basis as white persons. It is undisputed that in these cases, the respondent class was denied this right through intentional discrimination. The fact that the associations chose to delegate a large part of the hiring process to the local union hiring hall, which then engaged in intentional discrimination, does not alter the fact that respondents were denied the right to enter into employment contracts with the associations on the same basis as white persons. 78 At the very least, § 1981 imposes on employers the obligation to make employment decisions free from racial considerations. The hiring decisions made by the contracting associations in these cases were fraught with racial discrimination. Solely because of their race, hundreds of minority operating engineers were totally excluded from the industry and could not enter into employment contracts with any employer. Those minorities allowed into the industry suffered discrimination in referrals, and thus they too were denied the same right as white persons to contract with the contracting associations. Not one of the petitioner contracting associations has ever claimed, nor could they, that minorities had the same right as white operating engineers to contract for employment. 79 Instead, the contracting associations attempt to hide behind the veil of ignorance, shifting their responsibility under § 1981 to the very entity which they chose to assist them in making hiring decisions.4 The suggestion that an employer's responsibility under § 1981 depends upon its own choice of a hiring agent finds no support in the statute, nor does any other source of law authorize the circumvention of § 1981 that the contracting associations seek here. Their obligation to make employment contracts free from racial discrimination is a nondelegable one—it does not disappear when, as is often the case, the actual employer designates a particular agent to assist in the hiring process. In my view, the fact that the discriminating entity here is a union hiring hall, and not a person or corporation which has a traditional agent-principal relationship with the employer, does not alter this analysis. Cf. Morrison-Knudsen Co. v. NLRB, 275 F.2d 914 (CA2 1960) (per Swan, J.) (employer cannot escape liability for discrimination against nonunion members by the union hiring hall to which it turns over the task of supplying men for employment), cert. denied, 366 U.S. 909, 81 S.Ct. 1082, 6 L.Ed.2d 234 (1961). 80 The majority does not really analyze the question whether petitioners should be held injunctively liable because § 1981 imposes upon them a nondelegable duty. Instead the majority argues that, because it has held that § 1981 is intended only to reach intentional discrimination, the statute cannot make employers "guarantors of the workers' rights as against third parties who would infringe them." Ante, at 396. This argument does not withstand analysis. The majority does not assert that employers may escape liability under § 1981 by delegating their hiring decisions to a third-party agent. Indeed, in light of the importance attached to the rights § 1981 is intended to safeguard, the duty to abide by this statute must be nondelegable, as the majority apparently recognizes. Ante, at 396. Instead, the majority argues that because § 1981 imposes only the duty to refrain from intentional discrimination in hiring, it somehow automatically follows that this duty could not have been violated in this case. However, it was precisely this duty that was violated here. The District Court found, and this Court does not disagree, that the entity to whom the petitioner associations effectively delegated their hiring decisions intentionally discriminated against the respondent class on the basis of race in making these decisions. Even under the Court's own narrow view of the scope of the duty imposed by § 1981, then, the duty was unquestionably violated in these cases. 81 The majority obfuscates the issue by suggesting that the District Court imposed upon the contracting associations an obligation to seek out and eliminate discrimination by unrelated third parties wherever it may occur. In reality, the District Court did nothing more than impose limited injunctive liability upon the associations for violating their nondelegable duty under § 1981 when the union hiring hall, which effectively made hiring decisions for the associations, engaged in intentional discrimination on the basis of race in making these decisions. 82 By immunizing the employer from the injunctive relief necessary to remedy the intentional discrimination practiced by those through whom the employer makes its hiring decisions, the Court removes the person most necessary to accord full relief—the entity with whom the aggrieved persons will ultimately make a contract. I believe that the District Court appropriately rejected the petitioners' argument when it explained: "With intensity some employers urge that they agreed to the exclusive hiring hall system solely as a matter of economic survival at the end of a destructive ten week strike when the union would not compromise for any other hiring alternative. Yet economic pressures, however strong and harmful they might be, do not create immunity for employers, at least not in [the injunctive] liability phase." 469 F.Supp., at 338. 83 Section 1981 provides Negroes "the same right" to make contracts as white persons enjoy. In the present cases, this unqualified right was violated, and the violation is made no more palatable because the persons who actually made the hiring decisions and referrals, and not the employer itself, engaged in intentional discrimination.5 The devastating violation of their rights under § 1981 remains the same and will go at least partially unremedied when the person with whom the ultimate employment contract must be made is immunized from even injunctive relief. I cannot impute to the Congress which enacted § 1981 the intention to reach such an inequitable and nonsensical result. Accordingly, I must dissent. 1 The petitioner associations are the General Building Contractors Association (GBCA), the Contractors Association of Eastern Pennsylvania (CAEP), and the United Contractors Association (UCA). The fourth group, the Pennsylvania Excavating Contractors Association, was dissolved in 1972 after the commencement of this action. 2 A second strike occurred in 1963 when the contractor associations unsuccessfully sought to remove the hiring hall provision from the area collective-bargaining agreement. 3 The District Court found that "a vast majority of the employers are not and have not been active members of the defendant associations." Pennsylvania v. Local 542, Int'l Union of Operating Engineers, 469 F.Supp. 329, 342 (ED Pa.1978). Nevertheless, the court found that "the negotiations conducted by those bodies have established a standard to which the unaffiliated contractors may conform." Ibid. 4 The complaint did not assert a Title VII cause of action against petitioners because they were not named in the complaint filed by the plaintiffs with the Equal Employment Opportunity Commission, a precondition to suit in federal court. See Brief for Individual and Class Respondents 12, n. 18. 5 Certification of this class evidently was influenced by the District Court's conclusion that liability could be imposed without regard to individualized issues such as the intent or work-force statistics of the individual employers. See 469 F.Supp., at 384, 414. The court emphasized that the determination of liability in damages could require individualized proof; it therefore held out the possibility that the defendant class might be decertified in the second stage of the proceedings. Id., at 413, 415, 419-420. 6 The District Court's legal conclusions addressed only the liability of Local 542. The court explained: "Because of the JATC's participation in the overall intentional discrimination of the union, there is no need to discuss its legal liability separately. The JATC is liable as the union is liable." Id., at 401, n. 52. 7 The District Court absolved petitioners of liability under 42 U.S.C. § 1985(3) (1976 ed., Supp.IV), noting that "no per se or vicarious liability theory could be used to hold a class of employers liable for conspiracy to commit the discrimination practiced by the union." 469 F.Supp., at 413. Absent such a theory, the plaintiffs could not prevail because "there was no sufficient proof that as a class the employers agreed to violate equal protection rights or equal privileges and immunities." Ibid. Moreover, "[n]ot even acquiescence of the whole class of employers in the sense of a conscious toleration of the discrimination of the union ha[d] been shown." Ibid. In light of its disposition, the court found it unnecessary to address other causes of action alleged by the plaintiffs. See id., at 386, n. 43. 8 The District Court concluded, by analogy to Title VII, that a violation of § 1981 could be made out by "proof of disparate impact alone." Id., at 401. The court referred to Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), in which we held that Title VII forbids the use of employment tests that produce a disproportionate racial impact unless the employer shows "a manifest relationship to the employment in question," id., at 432, 91 S.Ct. at 854. See Teamsters v. United States, 431 U.S. 324, 335-336, 97 S.Ct. 1843, 1854, 52 L.Ed.2d 396 (1977). The District Court's holding on this issue is contrary to the holding of every Court of Appeals that has addressed the matter, including that of the Third Circuit in a subsequent case. See Guardians Assn. v. Civil Service Comm'n of New York City, 633 F.2d 232, 263-268 (CA2 1980), cert. granted, 454 U.S. 1140, 102 S.Ct. 997, 71 L.Ed.2d 291 (1982); Croker v. Boeing Co., 662 F.2d 975, 984-989 (CA3 1981) (en banc); Williams v. DeKalb County, 582 F.2d 2 (CA5 1978); Mescall v. Burrus, 603 F.2d 1266, 1269-1271 (CA7 1979); Craig v. County of Los Angeles, 626 F.2d 659, 668 (CA9 1980), cert. denied, 450 U.S. 919, 101 S.Ct. 1364, 67 L.Ed.2d 345 (1981); Chicano Police Officer's Assn. v. Stover, 552 F.2d 918, 920-921 (CA10 1977). Two other Circuits have approved a requirement of discriminatory intent in dicta. See Des Vergnes v. Seekonk Water Dist., 601 F.2d 9, 14 (CA1 1979); Detroit Police Officers' Assn. v. Young, 608 F.2d 671, 692 (CA6 1979), cert. denied, 452 U.S. 938, 101 S.Ct. 3079, 69 L.Ed.2d 951 (1981). See also Johnson v. Alexander, 572 F.2d 1219, 1223-1224 (CA8), cert. denied, 439 U.S. 986, 99 S.Ct. 576, 58 L.Ed.2d 658 (1978); Donnell v. General Motors Corp., 576 F.2d 1292, 1300 (CA8 1978). But see Kinsey v. First Regional Securities, Inc., 181 U.S.App.D.C. 207, 215, n. 22, 557 F.2d 830, 838, n. 22 (1977). 9 Section 1982 provides: "All citizens of the United States shall have the same right, in every State and Territory, as is enjoyed by white citizens thereof to inherit, purchase, lease, sell, hold, and convey real and personal property." 10 Section 1 of the Act of Apr. 9, 1866, read in part: "That all persons born in the United States and not subject to any foreign power, . . . are hereby declared to be citizens of the United States; and such citizens, of every race and color, without regard to any previous condition of slavery or involuntary servitude, . . . shall have the same right, in every State and Territory in the United States, to make and enforce contracts, to sue, be parties, and give evidence, to inherit, purchase, lease, sell, hold, and convey real and personal property, and to full and equal benefit of all laws and proceedings for the security of person and property, as is enjoyed by white citizens, and shall be subject to like punishment, pains, and penalties, and to none other, any law, statute, ordinance, regulation, or custom, to the contrary notwithstanding." 11 "That all persons within the jurisdiction of the United States shall have the same right in every State and Territory in the United States to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of person and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and none other, any law, statute, ordinance, regulation, or custom to the contrary notwithstanding. No tax or charge shall be imposed or enforced by any State upon any person immigrating thereto from a foreign country which is not imposed and enforced upon every person immigrating to such State from any other foreign country; and any law of any State in conflict with this provision is hereby declared null and void." 16 Stat. 144. Section 18 of the 1870 Act also re-enacted the 1866 Act and declared that § 16 "shall be enforced according to the provisions of said act." Ibid. 12 Cong. Globe, 39th Cong., 1st Sess., 129 (1866). 13 Discussion of the Black Codes occupied a central place in the congressional debates leading to enactment of the 1866 Act. See id., at 588-589 (remarks of Rep. Donnelly); 602 (Sen. Lane); 603 (Sen. Wilson); 1123-1124 (Rep. Cook); 1118-1119 (Rep. Wilson); 1151-1152, 1153 (Rep. Thayer); 1160 (Rep. Windom); 1785 (Sen. Stewart); 1833-1835 (Rep. Lawrence); 1838-1839 (Rep. Clarke). The Codes are described in E. McPherson, The Political History of the United States of America During the Period of Reconstruction 29-44 (1871). 14 See, e.g., Cong. Globe, 39th Cong., 1st Sess., supra, at 1124 (Rep. Cook); 1151-1152 (Rep. Thayer); 1159 (Rep. Windom); 1785 (Sen. Stewart); 1839 (Rep. Clarke). See also Memphis v. Greene, 451 U.S. 100, 131-135, 101 S.Ct. 1584, 1602-1604, 67 L.Ed.2d 769 (1981) (WHITE, J., concurring in judgment). 15 We attach significance to the fact that throughout much of the congressional debates, S.B. 61, which became the 1866 Act, contained an opening declaration that "there shall be no discrimination in civil rights or immunities among citizens of the United States in any State or Territory of the United States on account of race, color, or previous condition of slavery." See Cong. Globe, 39th Cong., 1st Sess., 474 (1866). This passage had occasioned controversy in both the Senate and the House because of the breadth of the phrase "civil rights and immunities." After the Senate had passed the bill and as debates in the House were drawing to a close, the bill's floor manager, Representative Wilson, introduced an amendment proposed by the House Judiciary Committee, of which he was also the Chairman. That amendment deleted the language quoted above and left the bill as it would read when ultimately enacted. See n. 10, supra. Representative Wilson explained that the broad language of the original bill could have been interpreted to encompass the right of suffrage and other political rights. "To obviate that difficulty and the difficulty growing out of any other construction beyond the specific rights named in the section, our amendment strikes out all of those general terms and leaves the bill with the rights specified in the section." Cong. Globe, 39th Cong., 1st Sess., supra, at 1367. See McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 292, n. 22, 96 S.Ct. 2574, 2584, n. 22, 49 L.Ed.2d 493 (1976). The deleted language, emphasized above, strongly suggests that Congress was primarily concerned with intentional discrimination. That the passage was removed in an effort to narrow the scope of the legislation sharply undercuts the view that the 1866 Act reflects broader concerns. 16 See, e.g., Cong. Globe, 39th Cong., 1st Sess., supra, at 1294 (Rep. Wilson); id. at 2465 (Rep. Thayer). 17 It is true that § 1981, because it is derived in part from the 1866 Act, has roots in the Thirteenth as well as the Fourteenth Amendment. Indeed, we relied on that heritage in holding that Congress could constitutionally enact § 1982, which is also traceable to the 1866 Act, without limiting its reach to "state action." See Jones v. Alfred H. Mayer Co., 392 U.S. 409, 438, 88 S.Ct. 2186, 2202, 20 L.Ed.2d 1189 (1968). As we have already intimated, however, the fact that Congress acted in the shadow of the Thirteenth Amendment does not demonstrate that Congress sought to eradicate more than purposeful discrimination when it passed the 1866 Act. For example, Congress also enacted 42 U.S.C. § 1985(3) (1976 ed., Supp.IV) in part to implement the commands of the Thirteenth Amendment. See Griffin v. Breckenridge, 403 U.S. 88, 104-105, 91 S.Ct. 1790, 1799, 29 L.Ed.2d 338 (1971). While holding that § 1985(3) does not require state action but also reaches private conspiracies, we have emphasized that a violation of the statute requires "some racial, or perhaps otherwise class-based, invidiously discriminatory animus behind the conspirators' action." Id., at 102, 91 S.Ct. at 1798. We need not decide whether the Thirteenth Amendment itself reaches practices with a disproportionate effect as well as those motivated by discriminatory purpose, or indeed whether it accomplished anything more than the abolition of slavery. See Memphis v. Greene, 451 U.S., at 125-126, 101 S.Ct., at 1598-1600. We conclude only that the existence of that Amendment, and the fact that it authorized Congress to enact legislation abolishing the "badges and incidents of slavery," Civil Rights Cases, 109 U.S. 3, 20, 3 S.Ct. 18, 27, 27 L.Ed. 835 (1883), do not evidence congressional intent to reach disparate effects in enacting § 1981. 18 In this case, the associations were held liable because they negotiated an agreement, fair on its face, which was later implemented by another party in a manner that was not only discriminatory but in violation of the agreement itself and in a manner of which the associations were neither aware nor had reason to be aware. Since the associations' only role was as agent for employers whose hiring would actually be governed by the agreement, the District Court's theory presumably would also permit the imposition of liability on the attorneys who actually conducted the contract negotiations. We are unaware of any authority supporting such an extended application of respondeat superior. 19 According to respondents, the District Court's conclusion that petitioners retained the power to control the hiring hall was a finding of fact that cannot be set aside unless clearly erroneous. We disagree. The District Court found that petitioners had the "power to oppose" the Union, a conclusion we do not question. Whether the power to oppose the Union is equivalent to a right of control sufficient to invoke the doctrine of respondeat superior is, however, a legal question to which we must devote our independent judgment. 20 The court relied on Restatement § 214: "A master or other principal who is under a duty to provide protection for or to have care used to protect others or their property and who confides the performance of such duty to a servant or other person is subject to liability to such others for harm caused to them by the failure of such agent to perform the duty." 21 Respondents also contend that petitioners can be held liable on the theory that the hiring hall was a "joint enterprise" involving petitioners as well as the Union. They point to language in the District Court's opinion holding that "the union hiring hall was the agent for two principals—the union and the contractors, with their respective associations." 469 F.Supp., at 411. Even this theory, however, requires, among other things, the existence of a mutual right of control as between the members of the enterprise. See Restatement § 491; 2 Harper & James, § 26.13, p. 1414. For reasons we have already stated, there is no record basis for finding that petitioners had a right to control Local 542 in its administration of the hiring hall. We also doubt the validity of the assumption that the hiring hall is a separate entity, except perhaps as a physical structure. The District Court did not find, and respondents do not assert, that the hiring hall has a separate juridical existence. Indeed, in discussing the operation of the hiring hall, the District Court made clear that it was imposing liability on the basis of the Union's conduct. As used in the court's opinion, the phrase "hiring hall" appears to be no more than a shorthand reference for the referral process administered on a day-to-day basis by the Union. 22 Petitioners have raised several objections to the District Court's certification of a defendant class. In light of our disposition, however, we find it unnecessary to reach these issues. It is evident from the District Court's opinion that certification of the defendant class was premised on theories of liability that made individualized questions irrelevant. See n. 5, supra. We have now rejected those theories, and we assume that the District Court will reconsider the issue of class certification in the event of a new trial to determine liability. Petitioners have also questioned the standing of respondent Commonwealth of Pennsylvania to act either on its own behalf or as parens patriae in this litigation. We need not reach this issue either. Petitioners have not challenged the standing of the other plaintiffs and, therefore, even if Pennsylvania lacks standing, the District Court possessed Art. III jurisdiction to entertain those common issues presented by all plaintiffs. See Watt v. Energy Action Educational Foundation, 454 U.S. 151, 160, 102 S.Ct. 205, 212, 70 L.Ed.2d 309 (1981); Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 264, n. 9, 97 S.Ct. 555, 562, n. 9, 50 L.Ed.2d 450 (1977). Petitioners note that Pennsylvania has sought attorney's fees in its own right, but our judgment has removed the basis for such an award against petitioners until such time as Pennsylvania can again assert status as a prevailing party. Until Pennsylvania obtains relief different from that sought by plaintiffs whose standing has not been questioned, we decline to address the Commonwealth's standing. 1 The only facts offered by the respondents supporting application of respondeat superior are that half of the trustees administering the JATC are appointed by the employer associations, and that the JATC is funded entirely by mandatory employer contributions. 2 In the present cases, the District Court ordered the three employer associations to pay 10% of the costs of remedial relief, and the employer, Glasgow, to pay 5%. Because the cost of relief to date has been approximately $200,000, the petitioners' share of the cost has been $70,000. 3 In support of this statement, the Court in Teamsters cited Rule 19(a)(1) of the Federal Rules of Civil Procedure, which requires a district court to join a person as a party if "in his absence complete relief cannot be accorded among those already parties." 1 These measures included the Civil Rights Act of 1866, passed over President Johnson's veto; the Freedman's Bureau bill, which would have created a federal agency to ensure that a free labor system in which Negroes had equal participation would in fact be accomplished, and which commanded a clear majority in Congress, but failed to pass over a Presidential veto; a constitutional amendment sponsored by Representative Bingham but not recommended; and the Fourteenth Amendment. 2 As the majority recognizes, ante, at 386-387, one of the principal changes Congress hoped to achieve was the elimination of the infamous Black Codes. These included state laws regulating the terms and conditions of employment. In many States, these oppressive laws were facially neutral, literally applying to all laborers without regard to race. The laws prohibited such conduct as refusing to perform work and disobeying an employer, or inducing an employee away from his employer, and many provided for forfeiture of wages if the employee did not fulfill the terms of his employment contract. Other Codes included vagrancy laws, which were vague and broad enough to encompass virtually all Negro adults, and many were facially neutral, applying to white persons as well as to Negroes. See Croker v. Boeing Co., 662 F.2d 975, 1004, n. 5 (CA3 1981) (Gibbons, J., dissenting in part) (citing E. McPherson, Political History of the United States of America During the Period of Reconstruction 30-44 (1871)). The Black Codes were constantly discussed during the debates over the Civil Rights Act of 1866, and Congress clearly intended that the Act would eliminate even those Codes which were facially neutral. See, e.g., Cong. Globe, 39th Cong., 1st Sess., 39-41, 118-125 (1865); id., at 1151-1160, 1838-1839 (1866). See also University of California Regents v. Bakke, 438 U.S. 265, 390-391, 98 S.Ct. 2733, 2799, 57 L.Ed.2d 750 (1978) (separate opinion of MARSHALL, J.). 3 When discussing the scope of the Fifteenth Amendment in 1939, Justice Frankfurter was sensitive to the subtle forms that racial discrimination often takes. Writing for the Court in Lane v. Wilson, 307 U.S. 268, 275, 59 S.Ct. 872, 876, 83 L.Ed. 1281, he stated: "The Amendment nullifies sophisticated as well as simple-minded modes of discrimination." Unfortunately, the Court no longer seems sensitive to this reality. 4 Although the District Court held that respondents had not proved that the contracting associations as a class had actual knowledge or had specifically approved of the intentional discrimination, it hardly found them totally blameless in this regard, and it found that the petitioner associations in particular were not innocent. One part of the proof of intentional discrimination by the hiring hall was the fact that Local 542 had intentionally overstated its percentage of minority members to the Federal Government in order to receive federal funds while maintaining an extraordinarily low actual minority percentage. With respect to the petitioner contracting associations, the District Court found: "Any argument that, because the union alone had primary access to the membership data, the [petitioner] contracting associations . . . were not at least reckless participants in this scheme, I find to be devoid of merit and patently incredible. . . . The prospect of deriving . . . an immediate and substantial financial benefit from the federal coffers allowed them to become willing parties to the scheme by capriciously certifying 'facts' in anticipation of the government's reliance on them: Having sought to enrich their members with substantial profits, it is now too late to cry innocence and cast the blame elsewhere. These were no innocent prognosticators who were misled by the union's scheme to give inaccurate information." Pennsylvania v. Local 542, Int'l Union of Operating Engineers, 469 F.Supp. 329, 345 (ED Pa. 1978). The District Court further found: "The fact is that the vast majority of individual contractors never hired a minority operating engineer; that the [petitioner associations] signed a statement, relevant to federal approval of the 'Affirmative Action Program' . . ., grossly exaggerating minority union membership; and that the gross disparity between the percentage of the minority representation in the labor pool and minority representation in the union along with a gross disparity in hours and wages of minorities as against the minority labor pool percentage is a matter of such broad scope that some or all of the contractors and associations might have had knowledge of it." Id., at 401 (emphasis added). 5 I agree with Justice O'CONNOR's observation that nothing in the Court's opinion prevents the District Court on remand from holding the petitioner associations liable for discrimination practiced by the JATC. Specifically, they may be held liable because the trustees administering the JATC are appointed by the petitioner associations, the JATC is funded by employer contributions, and the associations exercise control over the JATC's actions. I also agree with Justice O'CONNOR that the Court's opinion does not prevent the District Court from requiring petitioners to comply with incidental or ancillary provisions contained in its injunctive order.

12

458 U.S. 527 102 S.Ct. 3211 73 L.Ed.2d 948 Mary Ellen CRAWFORD, a Minor, etc., et al., Petitionersv.BOARD OF EDUCATION OF the CITY OF LOS ANGELES et al. No. 81-38. Argued March 22, 1982. Decided June 30, 1982. Syllabus In a California state-court action seeking desegregation of the schools in the Los Angeles Unified School District (District), the trial court, in 1970, found de jure segregation in violation of both the State and Federal Constitutions and ordered the District to prepare a desegregation plan. The California Supreme Court affirmed, but based its decision solely upon the Equal Protection Clause of the State Constitution, which bars de facto as well as de jure segregation. On remand, the trial court approved a desegregation plan that included substantial mandatory pupil reassignment and busing. While the trial court was considering alternative new plans in 1979, the voters of California ratified an amendment (Proposition I) to the State Constitution which provides that state courts shall not order mandatory pupil assignment or transportation unless a federal court "would be permitted under federal decisional law" to do so to remedy a violation of the Equal Protection Clause of the Fourteenth Amendment to the Federal Constitution. The trial court denied the District's request to halt all mandatory reassignment and busing, holding that Proposition I was not applicable in light of the court's 1970 finding of de jure segregation in violation of the Fourteenth Amendment. The court then ordered implementation of a revised plan that again included substantial mandatory pupil reassignment and busing. The California Court of Appeal reversed, concluding that the trial court's 1970 findings of fact would not support the conclusion that the District had violated the Federal Constitution through intentional segregation. The Court of Appeal also held that Proposition I was constitutional under the Fourteenth Amendment and barred that part of the plan requiring mandatory student reassignment and busing. Held: Proposition I does not violate the Fourteenth Amendment. Pp. 535-545. (a) This Court's decisions will not support the contention that once a State choses to do "more" than the Fourteenth Amendment requires, it may never recede. Such an interpretation of that Amendment would be destructive of a State's democratic processes and of its ability to experiment in dealing with the problems of a heterogeneous population. Proposition I does not embody, expressly or implicitly, a racial classification. The simple repeal or modification of desegregation or antidiscrimination laws, without more, does not embody a presumptively invalid racial classification. Pp. 535-540. (b) Proposition I cannot be characterized as something more than a mere repeal. Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616, distinguished. The State Constitution still places upon school boards a greater duty to desegregate than does the Fourteenth Amendment. Nor does Proposition I allocate governmental or judicial power on the basis of a discriminatory principle. A "dual court system"—one for the racial majority and one for the racial minority—is not established simply because civil rights remedies are different from those available in other areas. It was constitutional for the people of the State to determine that the Fourteenth Amendment's standard was more appropriate for California courts to apply in desegregation cases than the standard repealed by Proposition I. Pp. 540-542. (c) Even if it could be assumed that Proposition I had a disproportionate adverse effect on racial minorities, there is no reason to differ with the state appellate court's conclusion that Proposition I in fact was not enacted with a discriminatory purpose. The purposes of the Proposition—chief among them the educational benefits of neighborhood schooling—are legitimate, nondiscriminatory objectives, and the state court characterized the claim of discriminatory intent on the part of millions of voters as but "pure speculation." Pp. 543-545. 113 Cal.App.3d 633, 170 Cal.Rptr. 495, affirmed. Laurence H. Tribe, Cambridge, Mass., for petitioners. G. William Shea, Los Angeles, Cal., for respondents. Sol. Gen. Rex E. Lee, Washington, D. C., for the United States as amicus curiae by special leave of Court. Justice POWELL delivered the opinion of the Court. 1 An amendment to the California Constitution provides that state courts shall not order mandatory pupil assignment or transportation unless a federal court would do so to remedy a violation of the Equal Protection Clause of the Fourteenth Amendment of the United States Constitution. The question for our decision is whether this provision is itself in violation of the Fourteenth Amendment. 2 * This litigation began almost 20 years ago in 1963, when minority students attending school in the Los Angeles Unified School District (District) filed a class action in state court seeking desegregation of the District's schools.1 The case went to trial some five years later, and in 1970 the trial court issued an opinion finding that the District was substantially segregated in violation of the State and Federal Constitutions. The court ordered the District to prepare a desegregation plan for immediate use. App. 139. 3 On the District's appeal, the California Supreme Court affirmed, but on a different basis. Crawford v. Board of Education, 17 Cal.3d 280, 130 Cal.Rptr. 724, 551 P.2d 28 (1976). While the trial court had found de jure segregation in violation of the Fourteenth Amendment of the United States Constitution, see App. 117, 120-121, the California Supreme Court based its affirmance solely upon the Equal Protection Clause of the State Constitution.2 The court explained that under the California Constitution "state school boards . . . bear a constitutional obligation to take reasonable steps to alleviate segregation in the public schools, whether the segregation be de facto or de jure in origin." 17 Cal.3d, at 290, 130 Cal.Rptr., at 730, 551 P.2d, at 34. The court remanded to the trial court for preparation of a "reasonably feasible" plan for school desegregation. Id., at 310, 130 Cal.Rptr., at 744, 551 P.2d, at 48.3 4 On remand, the trial court rejected the District's mostly voluntary desegregation plan but ultimately approved a second plan that included substantial mandatory school reassignment and transportation—"busing"—on a racial and ethnic basis.4 The plan was put into effect in the fall of 1978, but after one year's experience, all parties to the litigation were dissatisfied. See 113 Cal.App.3d 633, 636, 170 Cal.Rptr. 495, 497 (1981). Although the plan continued in operation, the trial court began considering alternatives in October 1979. 5 In November 1979 the voters of the State of California ratified Proposition I, an amendment to the Due Process and Equal Protection Clauses of the State Constitution.5 Proposition I conforms the power of state courts to order busing to that exercised by the federal courts under the Fourteenth Amendment: 6 "[N]o court of this state may impose upon the State of California or any public entity, board, or official any obligation or responsibility with respect to the use of pupil school assignment or pupil transportation, (1) except to remedy a specific violation by such party that would also constitute a violation of the Equal Protection Clause of the 14th Amendment to the United States Constitution, and (2) unless a federal court would be permitted under federal decisional law to impose that obligation or responsibility upon such party to remedy the specific violation of the Equal Protection Clause. . . ."6 7 Following approval of Proposition I, the District asked the Superior Court to halt all mandatory reassignment and busing of pupils. App. 185. On May 19, 1980, the court denied the District's application. The court reasoned that Proposition I was of no effect in this case in light of the court's 1970 finding of de jure segregation by the District in violation of the Fourteenth Amendment. Shortly thereafter, the court ordered implementation of a revised desegregation plan, one that again substantially relied upon mandatory pupil reassignment and transportation.7 8 The California Court of Appeal reversed. 113 Cal.App.3d 633, 170 Cal.Rptr. 495 (1981). The court found that the trial court's 1970 findings of fact would not support the conclusion that the District had violated the Federal Constitution through intentional segregation.8 Thus, Proposition I was applicable to the trial court's desegregation plan and would bar that part of the plan requiring mandatory student reassignment and transportation. Moreover, the court concluded that Proposition I was constitutional under the Fourteenth Amendment. Id., at 654, 170 Cal.Rptr., at 509. The court found no obligation on the part of the State to retain a greater remedy at state law against racial segregation than was provided by the Federal Constitution. Ibid. The court rejected the claim that Proposition I was adopted with a discriminatory purpose. Id., at 654-655, 170 Cal.Rptr., at 509.9 9 Determining Proposition I to be applicable and constitutional, the Court of Appeal vacated the orders entered by the Superior Court. The California Supreme Court denied hearing. App. to Pet. for Cert. 73a.10 We granted certiorari. 454 U.S. 892, 102 S.Ct. 386, 70 L.Ed.2d 206 (1981). II 10 We agree with the California Court of Appeal in rejecting the contention that once a State chooses to do "more" than the Fourteenth Amendment requires, it may never recede.11 We reject an interpretation of the Fourteenth Amendment so destructive of a State's democratic processes and of its ability to experiment. This interpretation has no support in the decisions of this Court. 11 Proposition I does not inhibit enforcement of any federal law or constitutional requirement. Quite the contrary, by its plain language the Proposition seeks only to embrace the requirements of the Federal Constitution with respect to mandatory school assignments and transportation. It would be paradoxical to conclude that by adopting the Equal Protection Clause of the Fourteenth Amendment, the voters of the State thereby had violated it. Moreover, even after Proposition I, the California Constitution still imposes a greater duty of desegregation than does the Federal Constitution. The state courts of California continue to have an obligation under state law to order segregated school districts to use voluntary desegregation techniques, whether or not there has been a finding of intentional segregation. The school districts themselves retain a state-law obligation to take reasonably feasible steps to desegregate, and they remain free to adopt reassignment and busing plans to effectuate desegregation.12 12 Nonetheless, petitioners contend that Proposition I is unconstitutional on its face. They argue that Proposition I employs an "explicit racial classification" and imposes a "race-specific" burden on minorities seeking to vindicate state-created rights. By limiting the power of state courts to enforce the state-created right to desegregated schools, petitioners contend, Proposition I creates a "dual court system" that discriminates on the basis of race.13 They emphasize that other state-created rights may be vindicated by the state courts without limitation on remedies. Petitioners argue that the "dual court system" created by Proposition I is unconstitutional unless supported by a compelling state interest. 13 We would agree that if Proposition I employed a racial classification it would be unconstitutional unless necessary to further a compelling state interest. "A racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification." Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 272, 99 S.Ct. 2282, 2292, 60 L.Ed.2d 870 (1979). See McLaughlin v. Florida, 379 U.S. 184, 196, 85 S.Ct. 283, 290, 13 L.Ed.2d 222 (1964). But Proposition I does not embody a racial classification.14 It neither says nor implies that persons are to be treated differently on account of their race. It simply forbids state courts to order pupil school assignment or transportation in the absence of a Fourteenth Amendment violation. The benefit it seeks to confer—neighborhood schooling—is made available regardless of race in the discretion of school boards.15 Indeed, even if Proposition I had a racially discriminatory effect, in view of the demographic mix of the District it is not clear which race or races would be affected the most or in what way.16 In addition, this Court previously has held that even when a neutral law has a disproportionately adverse effect on a racial minority, the Fourteenth Amendment is violated only if a discriminatory purpose can be shown.17 14 Similarly, the Court has recognized that a distinction may exist between state action that discriminates on the basis of race and state action that addresses, in neutral fashion, race-related matters.18 This distinction is implicit in the Court's repeated statement that the Equal Protection Clause is not violated by the mere repeal of race-related legislation or policies that were not required by the Federal Constitution in the first place. In Dayton Bd. of Education v. Brinkman, 433 U.S. 406, 414, 97 S.Ct. 2766, 2772, 53 L.Ed.2d 851 (1977), we found that the school board's mere repudiation of an earlier resolution calling for desegregation did not violate the Fourteenth Amendment.19 In Reitman v. Mulkey, 387 U.S. 369, 376, 87 S.Ct. 1627, 1631, 18 L.Ed.2d 830 (1967), and again in Hunter v. Erickson, 393 U.S. 385, 390, n. 5, 89 S.Ct. 557, 560, n. 5, 21 L.Ed.2d 616 (1969), we were careful to note that the laws under review did more than "mere[ly] repeal" existing antidiscrimination legislation.20 In sum, the simple repeal or modification of desegregation or antidiscrimination laws, without more, never has been viewed as embodying a presumptively invalid racial classification.21 15 Were we to hold that the mere repeal of race-related legislation is unconstitutional, we would limit seriously the authority of States to deal with the problems of our heterogeneous population. States would be committed irrevocably to legislation that has proved unsuccessful or even harmful in practice. And certainly the purposes of the Fourteenth Amendment would not be advanced by an interpretation that discouraged the States from providing greater protection to racial minorities.22 Nor would the purposes of the Amendment be furthered by requiring the States to maintain legislation designed to ameliorate race relations or to protect racial minorities but which has produced just the opposite effects.23 Yet these would be the results of requiring a State to maintain legislation that has proved unworkable or harmful when the State was under no obligation to adopt the legislation in the first place. Moreover, and relevant to this case, we would not interpret the Fourteenth Amendment to require the people of a State to adhere to a judicial construction of their State Constitution when that Constitution itself vests final authority in the people. III 16 Petitioners seek to avoid the force of the foregoing considerations by arguing that Proposition I is not a "mere repeal." Relying primarily on the decision in Hunter v. Erickson, supra, they contend that Proposition I does not simply repeal a state-created right but fundamentally alters the judicial system so that "those seeking redress from racial isolation in violation of state law must be satisfied with less than full relief from a state court."24 We do not view Hunter as controlling here, nor are we persuaded by petitioners' characterization of Proposition I as something more than a mere repeal. 17 In Hunter the Akron city charter had been amended by the voters to provide that no ordinance regulating real estate on the basis of race, color, religion, or national origin could take effect until approved by a referendum. As a result of the charter amendment, a fair housing ordinance, adopted by the City Council at an earlier date, was no longer effective. In holding the charter amendment invalid under the Fourteenth Amendment, the Court held that the charter amendment was not a simple repeal of the fair housing ordinance. The amendment "not only suspended the operation of the existing ordinance forbidding housing discrimination, but also required the approval of the electors before any future [antidiscrimination] ordinance could take effect." 393 U.S., at 389-390, 89 S.Ct., at 559-560. Thus, whereas most ordinances regulating real property would take effect once enacted by the City Council, ordinances prohibiting racial discrimination in housing would be forced to clear an additional hurdle.25 As such, the charter amendment placed an impermissible, "special burde[n] on racial minorities within the governmental process." Id., at 391, 89 S.Ct., at 560-561.26 18 Hunter involved more than a "mere repeal" of the fair housing ordinance; persons seeking anti-discrimination housing laws presumptively racial minorities—were "singled out for mandatory referendums while no other group . . . face[d] that obstacle." James v. Valtierra, supra, 402 U.S. 137, 142, 91 S.Ct. 1331, 1334, 28 L.Ed.2d 678 (1971). By contrast, even on the assumption that racial minorities benefited from the busing required by state law, Proposition I is less than a "repeal" of the California Equal Protection Clause. As noted above, after Proposition I, the State Constitution still places upon school boards a greater duty to desegregate than does the Fourteenth Amendment. 19 Nor can it be said that Proposition I distorts the political process for racial reasons or that it allocates governmental or judicial power on the basis of a discriminatory principle. "The Constitution does not require things which are different in fact or opinion to be treated in law as though they were the same." Tigner v. Texas, 310 U.S. 141, 147, 60 S.Ct. 879, 882, 84 L.Ed. 1124 (1940). Remedies appropriate in one area of legislation may not be desirable in another. The remedies available for violation of the antitrust laws, for example, are different than those available for violation of the Civil Rights Acts. Yet a "dual court system"—one for the racial majority and one for the racial minority—is not established simply because civil rights remedies are different from those available in other areas.27 Surely it was constitutional for the California Supreme Court to caution that although "in some circumstances busing will be an appropriate and useful element in a desegregation plan," in other circumstances "its 'costs,' both in financial and educational terms, will render its use inadvisable." See n. 3, supra. It was equally constitutional for the people of the State to determine that the standard of the Fourteenth Amendment was more appropriate for California courts to apply in desegregation cases than the standard repealed by Proposition I.28 20 In short, having gone beyond the requirements of the Federal Constitution, the State was free to return in part to the standard prevailing generally throughout the United States. It could have conformed its law to the Federal Constitution in every respect. That it chose to pull back only in part, and by preserving a greater right to desegregation than exists under the Federal Constitution, most assuredly does not render the Proposition unconstitutional on its face. IV 21 The California Court of Appeal also rejected petitioners' claim that Proposition I, if facially valid, was nonetheless unconstitutional because enacted with a discriminatory purpose. The court reasoned that the purposes of the Proposition were well stated in the Proposition itself.29 Voters may have been motivated by any of these purposes, chief among them the educational benefits of neighborhood schooling. The court found that voters also may have considered that the extent of mandatory busing, authorized by state law, actually was aggravating rather than ameliorating the desegregation problem. See n. 1, supra. It characterized petitioners' claim of discriminatory intent on the part of millions of voters as but "pure speculation." 113 Cal.App.3d, at 655, 170 Cal.Rptr., at 509. 22 In Reitman v. Mulkey, 387 U.S. 369, 87 S.Ct. 1627, 18 L.Ed.2d 830 (1967), the Court considered the constitutionality of another California Proposition. In that case, the California Supreme Court had concluded that the Proposition was unconstitutional because it gave the State's approval to private racial discrimination. This Court agreed, deferring to the findings made by the California court. The Court noted that the California court was "armed . . . with the knowledge of the facts and circumstances concerning the passage and potential impact" of the Proposition and "familiar with the milieu in which that provision would operate." Id., at 378, 87 S.Ct., at 1633. Similarly, in this case, again involving the circumstances of passage and the potential impact of a Proposition adopted at a statewide election, we see no reason to differ with the conclusions of the state appellate court.30 23 Under decisions of this Court, a law neutral on its face still may be unconstitutional if motivated by a discriminatory purpose. In determining whether such a purpose was the motivating factor, the racially disproportionate effect of official action provides "an 'important starting point.' " Personnel Administrator of Massachusetts v. Feeney, 442 U.S., at 274, 99 S.Ct., at 2293, quoting Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 266, 97 S.Ct. 555, 563, 50 L.Ed.2d 450 (1977). 24 Proposition I in no way purports to limit the power of state courts to remedy the effects of intentional segregation with its accompanying stigma. The benefits of neighborhood schooling are racially neutral. This manifestly is true in Los Angeles where over 75% of the public school body is composed of groups viewed as racial minorities. See nn. 1 and 16, supra. Moreover, the Proposition simply removes one means of achieving the state-created right to desegregated education. School districts retain the obligation to alleviate segregation regardless of cause. And the state courts still may order desegregation measures other than pupil school assignment or pupil transportation.31 25 Even if we could assume that Proposition I had a disproportionate adverse effect on racial minorities, we see no reason to challenge the Court of Appeal's conclusion that the voters of the State were not motivated by a discriminatory purpose. See 113 Cal.App.3d, at 654-655, 170 Cal.Rptr., at 509. In this case the Proposition was approved by an overwhelming majority of the electorate.32 It received support from members of all races.33 The purposes of the Proposition are stated in its text and are legitimate, nondiscriminatory objectives. In these circumstances, we will not dispute the judgment of the Court of Appeal or impugn the motives of the State's electorate. 26 Accordingly the judgment of the California Court of Appeal is 27 Affirmed. 28 Justice BLACKMUN, with whom Justice BRENNAN joins, concurring. 29 While I join the opinion of the Court, I write separately to address what I believe are the critical distinctions between this case and Washington v. Seattle School District No. 1, 458 U.S. 457, 102 S.Ct. 3187, 73 L.Ed.2d 896. 30 The Court always has recognized that distortions of the political process have special implications for attempts to achieve equal protection of the laws. Thus the Court has found particularly pernicious those classifications that threaten the ability of minorities to involve themselves in the process of self-government, for if laws are not drawn within a "just framework," Hunter v. Erickson, 393 U.S. 385, 393, 89 S.Ct. 557, 562, 21 L.Ed.2d 616 (1969) (Harlan, J., concurring), it is unlikely that they will be drawn on just principles. 31 The Court's conclusion in Seattle followed inexorably from these considerations. In that case the statewide electorate reallocated decisionmaking authority to " 'mak[e] it more difficult for certain racial and religious minorities [than for other members of the community] to achieve legislation that is in their interest.' " Washington v. Seattle School District No. 1, supra, at 470, 102 S.Ct., at 3195 (emphasis in original), quoting Hunter v. Erickson, 393 U.S., at 395, 89 S.Ct., at 562 (Harlan, J., concurring). The Court found such a political structure impermissible, recognizing that if a class cannot participate effectively in the process by which those rights and remedies that order society are created, that class necessarily will be "relegated, by state fiat, in a most basic way to second-class status." Plyler v. Doe, 457 U.S. 202, 233, 102 S.Ct. 2382, 2403, 72 L.Ed.2d 786 (1982) (BLACKMUN, J., concurring). 32 In my view, something significantly different is involved in this case. State courts do not create the rights they enforce; those rights originate elsewhere—in the state legislature, in the State's political subdivisions, or in the state constitution itself. When one of those rights is repealed, and therefore is rendered unenforceable in the courts, that action hardly can be said to restructure the State's decisionmaking mechanism. While the California electorate may have made it more difficult to achieve desegregation when it enacted Proposition I, to my mind it did so not by working a structural change in the political process so much as by simply repealing the right to invoke a judicial busing remedy. Indeed, ruling for petitioners on a Hunter theory seemingly would mean that statutory affirmative-action or antidiscrimination programs never could be repealed, for a repeal of the enactment would mean that enforcement authority previously lodged in the state courts was being removed by another political entity. 33 In short, the people of California—the same "entity" that put in place the State Constitution, and created the enforceable obligation to desegregate—have made the desegregation obligation judicially unenforceable. The "political process or the decisionmaking mechanism used to address racially conscious legislation" has not been "singled out for peculiar and disadvantageous treatment," Washington v. Seattle School District No. 1, 458 U.S., at 458, 102 S.Ct., at 3203 (emphasis in original), for those political mechanisms that create and repeal the rights ultimately enforced by the courts were left entirely unaffected by Proposition I. And I cannot conclude that the repeal of a state-created right—or, analogously, the removal of the judiciary's ability to enforce that right—" 'curtail[s] the operation of those political processes ordinarily to be relied upon to protect minorities.' " Supra, at 486, 102 S.Ct., at 3203, quoting United States v. Carolene Products Co., 304 U.S. 144, 153, n. 4, 58 S.Ct. 778, 783-784, n. 4, 82 L.Ed. 1234 (1938). 34 Because I find Seattle distinguishable from this case, I join the opinion and judgment of the Court. 35 Justice MARSHALL, dissenting. 36 The Court today addresses two ballot measures, a state constitutional amendment, and a statutory initiative each of which is admittedly designed to substantially curtail, if not eliminate, the use of mandatory student assignment or transportation as a remedy for de facto segregation. In Washington v. Seattle School District No. 1, 458 U.S. 457, 102 S.Ct. 3187, 73 L.Ed.2d 896 (Seattle), the Court concludes that Washington's Initiative 350, which effectively prevents school boards from ordering mandatory school assignment in the absence of a finding of de jure segregation within the meaning of the Fourteenth Amendment, is unconstitutional because "it uses the racial nature of an issue to define the governmental decisionmaking structure, and thus imposes substantial and unique burdens on racial minorities." Seattle, supra, at 470, 102 S.Ct., at 3195. Inexplicably, the Court simultaneously concludes that California's Proposition I, which effectively prevents a state court from ordering the same mandatory remedies in the absence of a finding of de jure segregation, is constitutional because "having gone beyond the requirements of the Federal Constitution, the State was free to return in part to the standard prevailing generally throughout the United States." Ante, at 542. Because I fail to see how a fundamental redefinition of the governmental decisionmaking structure with respect to the same racial issue can be unconstitutional when the State seeks to remove the authority from local school boards, yet constitutional when the State attempts to achieve the same result by limiting the power of its courts, I must dissent from the Court's decision to uphold Proposition I. 37 * In order to understand fully the implications of the Court's action today, it is necessary to place the facts concerning the adoption of Proposition I in their proper context. Nearly two decades ago, a unanimous California Supreme Court declared that "[t]he segregation of school children into separate schools because of their race, even though the physical facilities and the methods and quality of instruction in the several schools may be equal, deprives the children of the minority group of equal opportunities for education and denies them equal protection and due process of the law." Jackson v. Pasadena City School District, 59 Cal.2d 876, 880, 31 Cal.Rptr. 606, 608-609, 382 P.2d 878, 880-881 (1963). Recognizing that the "right to an equal opportunity for education and the harmful consequences of segregation" do not differ according to the cause of racial isolation, the California Supreme Court declined to adopt the distinction between de facto and de jure segregation engrafted by this Court on the Fourteenth Amendment. Id., at 881, 31 Cal.Rptr., at 609-610, 382 P.2d, at 881-882. Instead, the court clearly held that "school boards [must] take steps, insofar as reasonably feasible, to alleviate racial imbalance in schools regardless of its cause." Id., at 881, 31 Cal.Rptr., at 610, 382 P.2d, at 882. 38 As the California Supreme Court subsequently explained, the duty established in Jackson does not require that "each school in a district . . . reflect the racial composition of the district as a whole." Crawford v. Board of Education, 17 Cal.3d 280, 302, 130 Cal.Rptr. 724, 738, 551 P.2d 28, 42 (1976) (Crawford I ). Rather, it is sufficient that school authorities "take reasonable and feasible steps to eliminate segregated schools, i.e., schools in which the minority student enrollment is so disproportionate as realistically to isolate minority students from other students and thus deprive minority students of an integrated educational experience." Id., at 303, 130 Cal.Rptr., at 739, 551 P.2d, at 43 (emphasis in original). Moreover, the California courts have made clear that the primary responsibility for implementing this state constitutional duty lies with local school boards. "[S]o long as a local school board initiates and implements reasonably feasible steps to alleviate school segregation in its district, and so long as such steps produce meaningful progress in the alleviation of such segregation, and its harmful consequences, . . . the judiciary should [not] intervene in the desegregation process." Id., at 305-306, 130 Cal.Rptr., at 741, 551 P.2d, at 45. If, however, a school board neglects or refuses to implement meaningful programs designed to bring about an end to racial isolation in the public schools, "the court is left with no alternative but to intervene to protect the constitutional rights of minority children." Id., at 307, 130 Cal.Rptr., at 741, 551 P.2d, at 45. When judicial intervention is necessary, the court "may exercise broad equitable powers in formulating and supervising a plan which the court finds will insure meaningful progress to alleviate the harmful consequences of school segregation in the district." Id., at 307, 130 Cal.Rptr., at 742, 551 P.2d, at 46. Moreover, "once a school board defaults in its constitutional task, the court, in devising a remedial order, is not precluded from requiring the busing of children as part of a reasonably feasible desegregation plan." Id., at 310, 130 Cal.Rptr., at 744, 551 P.2d, at 48. 39 Like so many other decisions protecting the rights of minorities, California's decision to eradicate the evils of segregation regardless of cause has not been a popular one. In the nearly two decades since the State Supreme Court's decision in Jackson, there have been repeated attempts to restrain school boards and courts from enforcing this constitutional guarantee by means of mandatory student transfers or assignments. In 1970, shortly after the San Francisco Unified School District voluntarily adopted a desegregation plan involving mandatory student assignment, the California Legislature enacted Education Code § 1009.5, Cal.Educ.Code Ann. § 1009.5, currently codified at Cal.Educ.Code Ann. § 35350 (West 1978), which provides that "[n]o governing board of a school district shall require any student or pupil to be transported for any purpose or for any reason without the written permission of the parent or guardian." In San Francisco Unified School District v. Johnson, 3 Cal.3d 937, 92 Cal.Rptr. 309, 479 P.2d 669 (1971), the California Supreme Court interpreted this provision only to bar a school district from compelling students, without parental consent, to use means of transportation furnished by the district. Construing the statute to prohibit nonconsensual assignment of students for the purpose of eradicating de jure or de facto segregation, the court concluded, would clearly violate both the State and the Federal Constitutions by "exorcising a method that in many circumstances is the sole and exclusive means of eliminating racial segregation in the schools." Id., at 943, 92 Cal.Rptr., at 311, 479 P.2d, at 671. 40 The very next year, opponents of mandatory student assignment for the purpose of achieving racial balance again attempted to eviscerate the state constitutional guarantee recognized in Jackson. Proposition 21, which was enacted by referendum in November 1972, stated that "[n]o public school student shall, because of his race, creed, or color, be assigned to or be required to attend a particular school." Predictably, the California Supreme Court struck down Proposition 21 "for the same reasons set forth by us in Johnson." Santa Barbara School District v. Superior Court, 13 Cal.3d 315, 324, 118 Cal.Rptr. 637, 645, 530 P.2d 605, 613 (1975). 41 Finally, in 1979, the people of California enacted Proposition I. That Proposition, like all of the previous initiatives, effectively deprived California courts of the ability to enforce the state constitutional guarantee that minority children will not attend racially isolated schools by use of what may be "the sole and exclusive means of eliminating racial segregation in the schools," San Francisco Unified School District v. Johnson, supra, at 943, 92 Cal.Rptr., at 311, 479 P.2d, at 671, mandatory student assignment and transfer. Unlike the earlier attempts to accomplish this objective, however, Proposition I does not purport to prevent mandatory assignments and transfers when such measures are predicated on a violation of the Federal Constitution. Therefore, the only question presented by this case is whether the fact that mandatory transfers may still be made to vindicate federal constitutional rights saves this initiative from the constitutional infirmity presented in the previous attempts to accomplish this same objective. In my view, the recitation of the obvious—that a state constitutional amendment does not override federal constitutional guarantees—cannot work to deprive minority children in California of their federally protected right to the equal protection of the laws. II A. 42 In Seattle, the Court exhaustively set out the relevant principles that control the present inquiry. We there found that a series of precedents, exemplified by Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616 (1969), and Lee v. Nyquist, 318 F.Supp. 710 (W.D.N.Y.1970) (three-judge court), summarily aff'd, 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105 (1971), establish that the Fourteenth Amendment prohibits a State from allocating "governmental power nonneutrally, by explicitly using the racial nature of a decision to determine the decisionmaking process." Seattle, 458 U.S., at 470, 102 S.Ct., at 3195 (emphasis in original). We concluded that "state action of this kind . . . 'places special burdens on racial minorities within the governmental process' . . . thereby 'making it more difficult for certain racial and religious minorities [than for other members of the community] to achieve legislation that is in their interest.' " Ibid. (emphasis in original), quoting Hunter v. Erickson, supra, at 391, 395, 89 S.Ct., at 560, 562 (Harlan, J., concurring). 43 It is therefore necessary to determine whether Proposition I works a "nonneutral" reallocation of governmental power on the basis of the racial nature of the decision. This determination is also informed by our decision in Seattle. In that case we were presented with a statewide initiative which effectively precluded local school boards from ordering mandatory student assignment or transfer except where required to remedy a constitutional violation. We concluded that the initiative violated the Fourteenth Amendment because it reallocated decisionmaking authority over racial issues from the local school board to a "new and remote level of government." Seattle, at 483, 102 S.Ct., at 3202. In reaching this conclusion, we specifically affirmed three principles that are particularly relevant to the present inquiry. 44 First, we rejected the State's argument that a statewide initiative prohibiting mandatory student assignment has no "racial overtones" simply because it does not mention the words "race" or "integration." Seattle, at 471, 102 S.Ct., at 3195. We noted that "[n]either the initiative's sponsors, nor the District Court, nor the Court of Appeals had any difficulty perceiving the racial nature of the issue settled by Initiative 350." Ibid. In light of its language and the history surrounding its adoption, we found it "beyond reasonable dispute . . . that the initiative was enacted ' "because of," not merely "in spite of," its adverse effects upon' busing for integration." Ibid., quoting Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 279, 99 S.Ct. 2282, 2296, 60 L.Ed.2d 870 (1979). Moreover, we rejected the Solicitor General's remarkable contention, a contention also pressed here, that "busing for integration . . . is not a peculiarly 'racial' issue at all." Seattle, at 471-472, 102 S.Ct., at 3196. While not discounting the value of an integrated education to non-minority students, we concluded that Lee v. Nyquist, supra, definitively established that "desegregation of the public schools . . . at bottom inures primarily to the benefit of the minority, and is designed for that purpose," thereby bringing it within the Hunter doctrine. Seattle, 458 U.S., at 472, 102 S.Ct., at 3196. 45 Second, the Seattle Court determined that Initiative 350 unconstitutionally reallocated power from local school boards to the state legislature or the statewide electorate. After the enactment of Initiative 350, local school boards continued to exercise considerable discretion over virtually all educational matters, including student assignment. Those seeking to eradicate de facto segregation, however, were forced to "surmount a considerably higher hurdle than persons seeking comparable legislative action," Seattle, at 474, 102 S.Ct., at 3197, for instead of seeking relief from the local school board, those pursuing this racial issue were forced to appeal to a different and more remote level of government. Just as in Hunter v. Erickson, supra, where those interested in enacting fair housing ordinances were compelled to gain the support of a majority of the electorate, we held that this reallocation of governmental power along racial lines offends the Equal Protection Clause. Our holding was not altered by the fact that those seeking to combat de facto segregation could still pursue their cause by petitioning local boards to enact voluntary measures or by seeking action from the state legislature. Nor were we persuaded by the argument that no transfer of power had occurred because the State was ultimately responsible for the educational policy of local school boards We found it sufficient that Initiative 350 had deprived those seeking to redress a racial harm of the right to seek a particularly effective form of redress from the level of government ordinarily empowered to grant the remedy. 46 Finally, the Court's decision in Seattle implicitly rejected the argument that state action that reallocates governmental power along racial lines can be immunized by the fact that it specifically leaves intact rights guaranteed by the Fourteenth Amendment. The fact that mandatory pupil reassignment was still available as a remedy for de jure segregation did not alter the conclusion that an unconstitutional reallocation of power had occurred with respect to those seeking to combat de facto racial isolation in the public schools. B 47 In my view, these principles inexorably lead to the conclusion that California's Proposition I works an unconstitutional reallocation of state power by depriving California courts of the ability to grant meaningful relief to those seeking to vindicate the State's guarantee against de facto segregation in the public schools. Despite Proposition I's apparent neutrality, it is "beyond reasonable dispute," Seattle, at 471, 102 S.Ct., at 3195, and the majority today concedes, that "court-ordered busing in excess of that required by the Fourteenth Amendment . . . prompted the initiation and probably the adoption of Proposition I." Ante, at 538, n.18 (emphasis in original).1 Because "minorities may consider busing for integration to be 'legislation that is in their interest,' " Seattle, at 474, 102 S.Ct., at 3197, quoting Hunter v. Erickson, 393 U.S., at 395, 89 S.Ct., at 562 (Harlan, J., concurring), Proposition I is sufficiently "racial" to invoke theHunter doctrine.2 48 Nor can there be any doubt that Proposition I works a substantial reallocation of state power. Prior to the enactment of Proposition I, those seeking to vindicate the rights enumerated by the California Supreme Court in Jackson v. Pasadena City School District, 59 Cal.2d 876, 31 Cal.Rptr. 606, 382 P.2d 878 (1963), just as those interested in attaining any other educational objective, followed a two-stage procedure. First, California's minority community could attempt to convince the local school board voluntarily to comply with its constitutional obligation to take reasonably feasible steps to eliminate racial isolation in the public schools. If the board was either unwilling or unable to carry out its constitutional duty, those seeking redress could petition the California state courts to require school officials to live up to their obligations. Busing could be required as part of a judicial remedial order. Crawford I, 17 Cal.3d, at 310, 130 Cal.Rptr., at 744, 551 P.2d, at 48. 49 Whereas Initiative 350 attempted to deny minority children the first step of this procedure, Proposition I eliminates by fiat the second stage: the ability of California courts to order meaningful compliance with the requirements of the State Constitution. After the adoption of Proposition I, the only method of enforcing against a recalcitrant school board the state constitutional duty to eliminate racial isolation is to petition either the state legislature or the electorate as a whole. Clearly, the rules of the game have been significantly changed for those attempting to vindicate this state constitutional right.3 50 The majority seeks to conceal the unmistakable effects of Proposition I by calling it a "mere repeal" of the State's earlier commitment to do " 'more' than the Fourteenth Amendment requires." Ante, at 3216. Although it is true that we have never held that the "mere repeal of an existing [anti-discrimination] ordinance violates the Fourteenth Amendment," Hunter v. Erickson, supra, at 390, n.5, 89 S.Ct., at 560, n.5, it is equally clear that the reallocation of governmental power created by Proposition I is not a "mere repeal" within the meaning of any of our prior decisions. 51 In Dayton Bd. of Education v. Brinkman, 433 U.S. 406, 97 S.Ct. 2766, 53 L.Ed.2d 851 (1977), the new members of the Dayton Board of Education repudiated a resolution drafted by their predecessors admitting the Board's role in the establishment of a segregated school system and calling for various remedial actions. In concluding that the Board was constitutionally permitted to withdraw its own prior mea culpa, this Court was careful to note that "[t]he Board had not acted to undo operative regulations affecting the assignment of pupils or other aspects of the management of school affairs." Id., at 413, 97 S.Ct., at 2772 (emphasis added). Therefore, the only time that this Court has squarely held that a "mere repeal" did not violate the Fourteenth Amendment, it was presented with a situation where a governmental entity rescinded its own prior statement of policy without affecting any existing educational policy. It is no surprise that such conduct passed constitutional muster. 52 By contrast, in Seattle, Hunter, and Reitman v. Mulkey, 387 U.S. 369, 87 S.Ct. 1627, 18 L.Ed.2d 830 (1967),4 the three times that this Court has explicitly rejected the argument that a proposed change constituted a "mere repeal" of an existing policy, the alleged rescission was accomplished by a governmental entity other than the entity that had taken the initial action, and resulted in a drastic alteration of the substantive effect of existing policy. This case falls squarely within this latter category. To be sure, the right to be free from racial isolation in the public schools remains unaffected by Proposition I. See ante, at 535-536; see, McKinny v. Oxnard Union High School District Board of Trustees, 31 Cal.3d 79, 92-93, 181 Cal.Rptr. 549, 556, 642 P.2d 460, 467 (1982). But Proposition I does repeal the power of the state court to enforce this existing constitutional guarantee through the use of mandatory pupil assignment and transfer. 53 The majority asserts that the Fourteenth Amendment does not "require the people of a State to adhere to a judicial construction of their State Constitution when that Constitution itself vests final authority in the people." Ante, at 540. A state court's authority to order appropriate remedies for state constitutional violations, however, is no more based on the "final authority" of the people than the power of the local Seattle School Board to make decisions regarding pupil assignment is premised on the State's ultimate control of the educational process. Rather, the authority of California courts to order mandatory student assignments in this context springs from the same source as the authority underlying other remedial measures adopted by state and federal courts in the absence of statutory authorization: the "courts power to provide equitable relief" to remedy a constitutional violation. Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 30, 91 S.Ct. 1267, 1283, 28 L.Ed.2d 554 (1971); Crawford I, 17 Cal.3d, at 307, 130 Cal.Rptr., at 742, 551 P.2d, at 46 ("a trial court may exercise broad equitable powers in formulating and supervising a plan which the court finds will insure meaningful progress to alleviate . . . school segregation"). Even assuming that the source of a court's power to remedy a constitutional violation can be traced back to "the people," the majority's conclusion that "the people" can therefore confer that remedial power on a discriminatory basis is plainly inconsistent with our prior decisions. InHunter v. Erickson, 393 U.S., at 392, 89 S.Ct., at 561, we struck down the referendum at issue even though the people of Akron, Ohio, undoubtedly retained "final authority" for all legislation. Similarly, in Seattle we concluded that the reallocation of power away from local school boards offended the Equal Protection Clause even though the State of Washington "is ultimately responsible for providing education within its borders." 458 U.S., at 477, 102 S.Ct., at 3199. The fact that this change was enacted through popular referendum, therefore, cannot immunize it from constitutional review. SeeLucas v. Colorado General Assembly, 377 U.S. 713, 736-737, 84 S.Ct. 1459, 1473-1474, 12 L.Ed.2d 632 (1964). 54 As in Seattle, Hunter, and Reitman, Proposition I's repeal of the state court's enforcement powers was the work of an independent governmental entity, and not of the state courts themselves. That this repeal drastically alters the substantive rights granted by existing policy is patently obvious from the facts of this litigation.5 By prohibiting California courts from ordering mandatory student assignment when necessary to eliminate racially isolated schools, Proposition I has placed an enormous barrier between minority children and the effective enjoyment of their constitutional rights, a barrier that is not placed in the path of those who seek to vindicate other rights granted by state law. This Court's precedents demonstrate that, absent a compelling state interest, which respondents have hardly demonstrated, such a discriminatory barrier cannot stand.6 55 The fact that California attempts to cloak its discrimination in the mantle of the Fourteenth Amendment does not alter this result. Although it might seem "paradoxical" to some Members of this Court that a referendum that adopts the wording of the Fourteenth Amendment might violate it, the paradox is specious. Because of the Supremacy Clause, Proposition I would have precisely the same legal effect if it contained no reference to the Fourteenth Amendment. The lesson of Seattle is that a State, in prohibiting conduct that is not required by the Fourteenth Amendment, may nonetheless create a discriminatory reallocation of governmental power that does violate equal protection. The fact that some less effective avenues remain open to those interested in mandatory student assignment to eliminate racial isolation, like the fact that the voters in Hunter conceivably might have enacted fair housing legislation, or that those interested in busing to eliminate racial isolation in Seattle conceivably might use the State's referendum process, does not justify the discriminatory reallocation of governmental decisionmaking. 56 In this case, the reallocation of power occurs in the judicial process—the major arena minorities have used to ensure the protection of rights "in their interest." Hunter v. Erickson, supra, at 395, 89 S.Ct., at 563 (Harlan, J., concurring). Certainly, Hunter and Seattle cannot be distinguished on the ground that they concerned the reallocation of legislative power, whereas Proposition I redistributes the inherent power of a court to tailor the remedy to the violation. As we have long recognized, courts too often have been "the sole practicable avenue open to a minority to petition for redress of grievances." NAACP v. Button, 371 U.S. 415, 430, 83 S.Ct. 328, 336, 9 L.Ed.2d 405 (1963). See Reitman v. Mulkey, 387 U.S., at 377, 87 S.Ct., at 1632 (invalidating state constitutional amendment because "[t]he right to discriminate, including the right to discriminate on racial grounds, was now embodied in the State's basic charter, immune from legislative, executive, or judicial regulation at any level of the state government") (emphasis added). It is no wonder, as the present case amply illustrates, that whatever progress has been made towards the elimination of de facto segregation has come from the California courts. Indeed, Proposition I, by denying full access to the only branch of government that has been willing to address this issue meaningfully, is far worse for those seeking to vindicate the plainly unpopular cause of racial integration in the public schools than a simple reallocation of an often unavailable and unresponsive legislative process. To paraphrase, "[i]t surely is an excessively formal exercise . . . to argue that the procedural revisions at issue in Hunter [and Seattle ] imposed special burdens on minorities, but that the selective allocation of decisionmaking authority worked by [Proposition I] does not erect comparable political obstacles." Seattle, 458 U.S., at 475, n. 17, 102 S.Ct., at 3197, n. 17. III 57 Even if the effects of Proposition I somehow can be distinguished from the enactments at issue in Hunter and Seattle, the result reached by the majority today is still plainly inconsistent with our precedents. Because it found that the segregation of the California public schools violated the Fourteenth Amendment, the state trial court never considered whether Proposition I was itself unconstitutional because it was the product of discriminatory intent. Despite the absence of any factual record on this issue, the Court of Appeal rejected petitioners' argument that the law was motivated by a discriminatory intent on the ground that the recitation of several potentially legitimate purposes in the legislation's preamble rendered any claim that it had been enacted for an invidious purpose "pure speculation." 113 Cal.App.3d 633, 655, 170 Cal.Rptr. 495, 509 (1981). 58 In Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 266, 97 S.Ct. 555, 564, 50 L.Ed.2d 450 (1977), we declared that "[d]etermining whether invidious discriminatory purpose was a motivating factor demands a sensitive inquiry into such circumstantial and direct evidence of intent as may be available." Petitioners assert that the disproportionate impact of Proposition I, combined with the circumstances surrounding its adoption and the history of opposition to integration cited supra, at 548-551, clearly indicates the presence of discriminatory intent. See Brief for Petitioners 64-96. Yet despite the fact that no inquiry has been conducted into these allegations by either the trial or the appellate court, this Court, in its haste to uphold the banner of "neighborhood schools," affirms a factual determination that was never made. Such blind allegiance to the conclusory statements of a lower court is plainly forbidden by our prior decisions.7 IV 59 Proposition I is in some sense "better" than the Washington initiative struck down in Seattle.8 In their generosity, California voters have allowed those seeking racial balance to petition the very school officials who have steadfastly maintained the color line at the schoolhouse door to comply voluntarily with their continuing state constitutional duty to desegregate. At the same time, the voters have deprived minorities of the only method of redress that has proved effective—the full remedial powers of the state judiciary. In the name of the State's "ability to experiment," ante, at 535, the Court today allows this placement of yet another burden in the path of those seeking to counter the effects of nearly three centuries of racial prejudice. Because this decision is neither justified by our prior decisions nor consistent with our duty to guarantee all citizens the equal protection of the laws, I must dissent. 1 In 1980 the District included 562 schools with 650,000 students in an area of 711 square miles. In 1968 when the case went to trial, the District was 53.6% white, 22.6% black, 20% Hispanic, and 3.8% Asian and other. By October 1980 the demographic composition had altered radically: 23.7% white, 23.3% black, 45.3% Hispanic, and 7.7% Asian and other. See 113 Cal.App.3d 633, 642, 170 Cal.Rptr. 495, 501 (1981). 2 "The findings in this case adequately support the trial court's conclusion that the segregation in the defendant school district is de jure in nature. We shall explain, however, that we do not rest our decision on this characterization because we continue to adhere to our conclusion in [Jackson v. Pasadena City School Dist., 59 Cal.2d 876, 31 Cal.Rptr. 606, 382 P.2d 878 (1963) ] that school boards in California bear a constitutional obligation to take reasonably feasible steps to alleviate school segregation 'regardless of its cause.' " Crawford v. Board of Education, 17 Cal.3d, at 285, 130 Cal.Rptr., at 726, 551 P.2d, at 30. The court explained that federal cases were not controlling: "In focusing primarily on . . . federal decisions . . . defendant ignores a significant line of California decisions, decisions which authoritatively establish that in this state school boards do bear a constitutional obligation to take reasonable steps to alleviate segregation in the public schools, whether the segregation be de facto or de jure in origin." Id., at 290, 130 Cal.Rptr., at 729-730, 551 P.2d, at 33-34. 3 In stating general principles to guide the trial court on remand, the State Supreme Court discussed the 'busing' question: "While critics have sometimes attempted to obscure the issue, court decisions time and time again emphasized that "busing" is not a constitutional end in itself but is simply one potential tool which may be utilized to satisfy a school district's constitutional obligation in this field. . . . [I]n some circumstances busing will be an appropriate and useful element in a desegregation plan, while in other instances its 'costs,' both in financial and educational terms, will render its use inadvisable." Id., at 309, 130 Cal.Rptr., at 743, 551 P.2d, at 47. It noted as well that a state court should not intervene to speed the desegregation process so long as the school board takes "reasonably feasible steps to alleviate school segregation," id., at 305, 130 Cal.Rptr., at 741, 551 P.2d, at 45, and that "a court cannot properly issue a 'busing' order so long as a school district continues to meet its constitutional obligations." Id., at 310, 130 Cal.Rptr., at 744, 551 P.2d, at 48. 4 The plan provided for the mandatory reassignment of approximately 40,000 students in the fourth through eighth grades. Some of these children were bused over long distances requiring daily round-trip bus rides of as long as two to four hours. In addition, the plan provided for the voluntary transfer of some 30,000 students. Respondent Bustop, Inc., unsuccessfully sought to stay implementation of the plan. See Bustop, Inc. v. Board of Education, 439 U.S. 1380, 99 S.Ct. 40, 58 L.Ed.2d 88 (1978) (REHNQUIST, J., in chambers); Bustop, Inc. v. Board of Education, 439 U.S. 1384, 99 S.Ct. 44, 58 L.Ed.2d 92 (1978) (POWELL, J., in chambers). 5 Proposition I was placed before the voters following a two-thirds vote of each house of the state legislature. Cal.Const., Art. 18, § 1. The State Senate approved the Proposition by a vote of 28 to 6, the State Assembly by a vote of 62 to 17. The voters favored the Proposition by a vote of 2,433,312 (68.6%) to 1,112,923 (31.4%). The Proposition received a majority of the vote in each of the State's 58 counties and in 79 of the State's 80 assembly districts. California Secretary of State, Statement of the Vote, November 6, 1979, Election 3-4, 43-49. 6 Proposition I added a lengthy proviso to Art. 1, § 7(a), of the California Constitution. Following passage of Proposition I, § 7 now provides, in relevant part: "(a) A person may not be deprived of life, liberty, or property without due process of law or denied equal protection of the laws; provided, that nothing contained herein or elsewhere in this Constitution imposes upon the State of California or any public entity, board, or official any obligations or responsibilities which exceed those imposed by the Equal Protection Clause of the 14th Amendment to the United States Constitution with respect to the use of pupil school assignment or pupil transportation. In enforcing this subdivision or any other provision of this Constitution, no court of this state may impose upon the State of California or any public entity, board, or official any obligation or responsibility with respect to the use of pupil school assignment or pupil transportation, (1) except to remedy a specific violation by such party that would also constitute a violation of the Equal Protection Clause of the 14th Amendment to the United States Constitution, and (2) unless a federal court would be permitted under federal decisional law to impose that obligation or responsibility upon such party to remedy the specific violation of the Equal Protection Clause of the 14th Amendment of the United States Constitution. * * * * * "Nothing herein shall prohibit the governing board of a school district from voluntarily continuing or commencing a school integration plan after the effective date of this subdivision as amended. "In amending this subdivision, the Legislature and people of the State of California find and declare that this amendment is necessary to serve compelling public interests, including those of making the most effective use of the limited financial resources now and prospectively available to support public education, maximizing the educational opportunities and protecting the health and safety of all public school pupils, enhancing the ability of parents to participate in the educational process, preserving harmony and tranquility in this state and its public schools, preventing the waste of scarce fuel resources, and protecting the environment." 7 The Superior Court ordered the immediate implementation of the revised plan. The District was unsuccessful in its effort to gain a stay of the plan pending appeal. See Board of Education v. Superior Court, 448 U.S. 1343, 101 S.Ct. 21, 65 L.Ed.2d 1166 (1980) (REHNQUIST, J., in chambers). 8 "When the 1970 findings of the trial court are reviewed in the light of the correct applicable federal law, it is apparent that no specific segregative intent with discriminatory purpose was found. The thrust of the findings of the trial court was that passive maintenance by the Board of a neighborhood school system in the face of widespread residential racial imbalance amounted to de jure segregation in violation of the Fourteenth Amendment. . . . But a school board has no duty under the Fourteenth Amendment to meet and overcome the effect of population movements." 113 Cal.App.3d, at 645-646, 170 Cal.Rptr., at 503. 9 The Court of Appeal also rejected the claim that Proposition I deprived minority children of a "vested right" to desegregated education in violation of due process. See id., at 655-656, 170 Cal.Rptr., at 509-510. Petitioners no longer advance this claim. 10 On March 16, 1981, the District directed that mandatory pupil reassignment under the Superior Court's revised plan be terminated on April 20, 1981. On that date, parents of children who had been reassigned were given the option of returning their children to neighborhood schools. According to respondent Board of Education, approximately 7,000 pupils took this option of whom 4,300 were minority students. Brief for Respondent Board of Education 10. The state courts refused to enjoin termination of the plan. On April 17, 1981, however, the United States District Court for the Central District of California issued a temporary restraining order preventing termination of the plan. Los Angeles NAACP v. Los Angeles Unified School District, 513 F.Supp. 717. The District Court found that there was a "fair chance" that intentional segregation by the District could be demonstrated. Id., at 720. The District Court's order was vacated on the following day by the United States Court of Appeals for the Ninth Circuit. Los Angeles Unified School District v. District Court, 650 F.2d 1004 (1981). On remand the District Court denied the District's motion to dismiss. This ruling has been certified for interlocutory appeal. See Brief for Respondent Board of Education 10, n. 4. On September 10, 1981, the Superior Court approved a new, voluntary desegregation plan. 11 Respondent Bustop, Inc., argues that far from doing "more" than the Fourteenth Amendment requires, the State actually violated the Amendment by assigning students on the basis of race when such assignments were not necessary to remedy a federal constitutional violation. See Brief for Respondent Bustop, Inc., 10-18. We do not reach this contention. 12 In this respect this case differs from the situation presented in Washington v. Seattle School District No. 1, 458 U.S. 457, 102 S.Ct. 3187, 73 L.Ed.2d 896. In an opinion delivered after Proposition I was enacted, the California Supreme Court stated that "the amendment neither releases school districts from their State Constitutional obligation to take reasonably feasible steps to alleviate segregation regardless of its cause, nor divests California courts of authority to order desegregation measures other than pupil school assignment or pupil transportation." McKinny v. Oxnard Union High School District Board of Trustees, 31 Cal.3d 79, 92-93, 181 Cal.Rptr. 549, 566, 642 P.2d 460, 467 (1982). Moreover, the Proposition only limits state courts when enforcing the State Constitution. Thus, the Proposition would not bar state-court enforcement of state statutes requiring busing for desegregation or for any other purpose. Cf. Brown v. Califano, 201 U.S.App.D.C. 235, 244, 627 F.2d 1221, 1230 (1980) (legislation limiting power of federal agency to require busing by local school boards held constitutional in view of the "effective avenues for desegregation" left open by the legislation). 13 "[I]t is racial discrimination in the judicial apparatus of the state, not racial discrimination in the state's schools, that petitioners challenge under the Fourteenth Amendment in this case." Brief for Petitioners 48. 14 In Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616 (1969), the Court invalidated a city charter amendment which placed a special burden on racial minorities in the political process. The Court considered that although the law was neutral on its face, "the reality is that the law's impact falls on the minority." Id., at 391, 89 S.Ct., at 560. In light of this reality and the distortion of the political process worked by the charter amendment, the Court considered that the amendment employed a racial classification despite its facial neutrality. In this case the elements underlying the holding in Hunter are missing. See infra. 15 A neighborhood school policy in itself does not offend the Fourteenth Amendment. See Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U.S. 1, 28, 91 S.Ct. 1267, 1282, 28 L.Ed.2d 554 (1971) ("Absent a constitutional violation there would be no basis for judicially ordering assignment of students on a racial basis. All things being equal, with no history of discrimination, it might well be desirable to assign pupils to schools nearest their homes"). Cf. 20 U.S.C. § 1701: "(a) The Congress declares it to be the policy of the United States that—(1) all children enrolled in public schools are entitled to equal educational opportunity without regard to race, color, sex, or national origin; and (2) the neighborhood is the appropriate basis for determining public school assignments." 16 In the Los Angeles School District, white students are now the racial minority, see n. 1, supra. Similarly, in Los Angeles County, racial minorities, including those of Spanish origin, constitute the majority of the population. See U.S. Dept. of Commerce, 1980 Census of Population and Housing, California, Advance Reports 6 (Mar.1981). 17 See Washington v. Davis, 426 U.S. 229, 238-248, 96 S.Ct. 2040, 2046-2052, 48 L.Ed.2d 597 (1976); Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 265, 97 S.Ct. 555, 563, 50 L.Ed.2d 450 (1977); James v. Valtierra, 402 U.S. 137, 141, 91 S.Ct. 1331, 1333, 28 L.Ed.2d 678 (1971). 18 Proposition I is not limited to busing for the purpose of racial desegregation. It applies neutrally to "pupil school assignment or pupil transportation" in general. Even so, it is clear that court-ordered busing in excess of that required by the Fourteenth Amendment, as one means of desegregating schools, prompted the initiation and probably the adoption of Proposition I. 19 See Dayton Bd. of Ed. v. Brinkman, 443 U.S., at 531, n. 5, 99 S.Ct., at 2976, n. 5 ("Racial imbalance, we noted in Dayton I, is not per se a constitutional violation, and rescission of prior resolutions proposing desegregation is unconstitutional only if the resolutions were required in the first place by the Fourteenth Amendment"). 20 In Hunter we noted that "we do not hold that mere repeal of an existing [antidiscrimination] ordinance violates the Fourteenth Amendment." 393 U.S., at 390, n. 5, 89 S.Ct., at 560, n. 5. In Reitman the Court held that California Proposition 14 was unconstitutional under the Fourteenth Amendment not because it repealed two pieces of antidiscrimination legislation, but because the Proposition involved the State in private racial discrimination: "Here we are dealing with a provision which does not just repeal an existing law forbidding private racial discriminations. Section 26 was intended to authorize, and does authorize, racial discrimination in the housing market." 387 U.S., at 380-381, 87 S.Ct., at 1634. 21 Of course, if the purpose of repealing legislation is to disadvantage a racial minority, the repeal is unconstitutional for this reason. See Reitman v. Mulkey, 387 U.S. 369, 87 S.Ct. 1627, 18 L.Ed.2d 830 (1967). 22 See Palmer v. Thompson, 403 U.S. 217, 228, 91 S.Ct. 1940, 1946, 29 L.Ed.2d 438 (1971) ("To hold . . . that every public facility or service, once opened, constitutionally 'locks in' the public sponsor so that it may not be dropped . . . would plainly discourage the expansion and enlargement of needed services in the long run") (BURGER, C. J., concurring); Reitman v. Mulkey, supra, 387 U.S., at 395, 87 S.Ct., at 1641 ("Opponents of state antidiscrimination statutes are now in a position to argue that such legislation should be defeated because, if enacted, it may be unrepealable") (Harlan, J., dissenting). 23 In his dissenting opinion in Reitman v. Mulkey, supra, at 395, 87 S.Ct., at 1641, Justice Harlan remarked upon the need for legislative flexibility when dealing with the "delicate and troublesome problems of race relations." He noted: "The lines that have been and must be drawn in this area, fraught as it is with human sensibilities and frailties of whatever race or creed, are difficult ones. The drawing of them requires understanding, patience, and compromise, and is best done by legislatures rather than by courts. When legislation in this field is unsuccessful there should be wide opportunities for legislative amendment, as well as for change through such processes as the popular initiative and referendum." 387 U.S., at 395-396, 87 S.Ct., at 1641. 24 Tr. of Oral Arg. 6. See id., at 7-8 ("The fact that a state may be free to remove a right or remove a duty, does not mean that it has the same freedom to leave the right in place but simply, in a discriminatory way we argue, provide less than full judicial remedy"). 25 "In the case before us . . . the city of Akron has not attempted to allocate governmental power on the basis of any general principle. Here, we have a provision that has the clear purpose of making it more difficult for certain racial and religious minorities to achieve legislation that is in their interest." 393 U.S., at 395, 89 S.Ct., at 563 (Harlan, J., concurring). 26 The Hunter Court noted that although "the law on its face treats Negro and white, Jew and gentile in an identical manner," id., at 391, 89 S.Ct., at 560, a charter amendment making it more difficult to pass antidiscrimination legislation could only disadvantage racial minorities in the governmental process. 27 Petitioners contend that Proposition I only restricts busing for the purpose of racial discrimination. The Proposition is neutral on its face, however, and respondents—as well as the State in its amicus brief—take issue with petitioners' interpretation of the provision. 28 Similarly, a "dual constitution" is not established when the State chooses to go beyond the requirements of the Federal Constitution in some areas but not others. Nor is a "dual executive branch" created when an agency is given enforcement powers in one area but not in another. Cf. Brown v. Califano, 201 U.S.App.D.C. 235, 627 F.2d 1221 (1980) (upholding federal legislation prohibiting a federal executive agency, but not local school officials or federal courts, from requiring busing). 29 The Proposition contains its own statement of purpose: "[T]he Legislature and people of the State of California find and declare that this amendment is necessary to serve compelling public interests, including those of making the most effective use of the limited financial resources now and prospectively available to support public education, maximizing the educational opportunities and protecting the health and safety of all public school pupils, enhancing the ability of parents to participate in the educational process, preserving harmony and tranquility in this state and its public schools, preventing the waste of scarce fuel, resources, and protecting the environment." 30 Cf. Washington v. Davis, 426 U.S., at 253, 96 S.Ct., at 2054 ("The extent of deference that one pays to the trial court's determination of the factual issue, and indeed, the extent to which one characterizes the intent issue as a question of fact or a question of law, will vary in different contexts.") (STEVENS, J., concurring). 31 In Brown v. Califano, supra, the Court of Appeals found that a federal statute preventing the Department of Health, Education, and Welfare (HEW) from requiring busing "to a school other than the school which is nearest the student's home," 42 U.S.C. § 2000d, was not unconstitutional. HEW retained authority to encourage school districts to desegregate through other means, and the enforcement powers of the Department of Justice were left untouched. The court therefore concluded that the limits on HEW's ability to order mandatory busing did not have a discriminatory effect. And, having done so, it refused to inquire into legislative motivation: "Absent discriminatory effect, judicial inquiry into legislative motivation is unnecessary, as well as undesirable." 201 U.S.App.D.C., at 248, 627 F.2d, at 1234 (footnote omitted). 32 Cf. Washington v. Davis, supra, at 253, 96 S.Ct., at 2054 (STEVENS, J., concurring) ("It is unrealistic . . . to invalidate otherwise legitimate action simply because an improper motive affected the deliberation of a participant in the decisional process. A law conscripting clerics should not be invalidated because an atheist voted for it"). 33 Proposition I received support from 73.9% of the voters in Los Angeles County which has a "minority" population—including persons of Spanish origin—of over 50%. California Secretary of State, Statement of the Vote, November 6, 1979, Election 3. See n. 16, supra. By contrast, the Proposition received its smallest percentage of the vote in Humboldt and Marin Counties which are nearly all-white in composition. 1 Just as in Seattle, the fact that other types of student transfers conceivably might be prohibited does not alter this conclusion: "Neither the initiative's sponsors, nor the District Court, nor the Court of Appeals had any difficulty perceiving the racial nature of the issue settled by" Proposition I. Seattle, at 471, 102 S.Ct., at 3195. Indeed in their response to the petition for certiorari, respondents characterized Proposition I as addressing but "one narrow area: the power of a state court to order mandatory student assignment or transportation as a desegregation remedy." Brief in Opposition 9. 2 It is therefore irrelevant whether the "benefits of neighborhood schooling are racially neutral," as the majority asserts. Ante, at 554; see ante, at 537. In Seattle, 458 U.S., at 472, 102 S.Ct., at 3196, we specifically rejected the argument that because some minorities as well as whites supported the initiative, it could not be considered a racial classification. 3 There can be no question that the practical effect of Proposition I will be to deprive state courts of "the sole and exclusive means of eliminating racial segregation in the schools." San Francisco Unified School District v. Johnson, 3 Cal.3d 937, 943, 92 Cal.Rptr. 309, 311, 479 P.2d 669, 671 (1971). As we have often noted, "bus transportation has long been an integral part of all public educational systems, and it is unlikely that a truly effective remedy could be devised without continued reliance upon it." North Carolina Board of Ed. v. Swann, 402 U.S. 43, 46, 91 S.Ct. 1284, 1286, 28 L.Ed.2d 586 (1971). Moreover, Proposition I prevents a state court from ordering school officials to take any action respecting pupil school assignment, as well as pupil transportation. Presumably, state courts could not design a remedy involving the "pairing" or "clustering" of schools, even if such a remedy did not involve any "busing." In the present case, the state trial court found that the voluntary programs proposed by the Los Angeles School Board were "constitutionally suspect" because they "place[d] the burden of relieving the racial isolation of the minority student upon the minority student." App. 160. Consequently, since "a voluntary program would not serve to integrate the community's schools," Seattle, 458 U.S., at 473, n. 16, 102 S.Ct., at 3197, n. 16, Proposition I, like the measures at issue in Lee v. Nyquist, 318 F.Supp. 710 (W.D.N.Y.1970) (three-judge court), summarily aff'd, 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105 (1971), and Seattle, precludes the effective enjoyment by California's minority children of their right to eliminate racially isolated schools. 4 In Reitman v. Mulkey, this Court struck down another California ballot measure, granting every resident the absolute constitutional right to sell or rent his property to whomever he or she chooses. We held that the provision amounted to an unconstitutional authorization of private discrimination. 5 Indeed Proposition I by its express terms allows for the modification of existing plans upon the application of any interested person. Art. 1, § 7(a). 6 As the majority notes, Proposition I states that the "people of the State of California find and declare that this amendment is necessary to serve compelling public interests," including, inter alia, "making the most efficient use of . . . limited financial resources," protecting the "health and safety" of all students, preserving "harmony and tranquility," and "protecting the environment." Ante, at 533, n. 6. These purported justifications, while undoubtedly meritorious, are clearly insufficient to sustain the racial classification established by Proposition I. As we have often noted, racial classifications may only be upheld where "necessary, and not merely rationally related, to the accomplishment of a permissible state policy." McLaughlin v. Florida, 379 U.S. 184, 196, 85 S.Ct. 283, 290, 13 L.Ed.2d 222 (1964). It goes without saying that a self-serving conclusory statement of necessity will not suffice to fulfill this burden. See Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 28, 29-31, 91 S.Ct. 1267, 1282, 1283, 28 L.Ed.2d 554 (1971) (rejecting a similar list of justifications for establishing a racial classification). "In any event, [respondents] have failed to show that the purpose[s] they impute to the [Proposition] could not be accomplished by alternative methods, not involving racial distinctions." Lee v. Nyquist, 318 F.Supp., at 720. Parenthetically, it is interesting to note that the allegedly compelling interest in establishing "neighborhood schools" so often referred to by the majority appears nowhere in the official list of justifications. The absence of any mention of this supposed justification is not surprising in light of the fact that the Proposition's ban on student "assignment" effectively prevents desegregation remedies that would not require a student to leave his "neighborhood." See n. 3, supra. 7 The majority's reliance on Reitman v. Mulkey, 387 U.S. 369, 87 S.Ct. 1627, 18 L.Ed.2d 830 (1967), is therefore misplaced. How can any deference be given to the state court's "knowledge of the facts and circumstances concerning the passage and potential impact" of Proposition I, id., at 378, 87 S.Ct., at 1633, when no such findings were ever made. 8 Initiative 350, however, at least did "not hinder [the] State from enforcing [the State] Constitution." Seattle, 458 U.S., at 490, n. 3, 102 S.Ct., at 3205, n. 3 (POWELL, J., dissenting).

12

458 U.S. 564 102 S.Ct. 3245 73 L.Ed.2d 973 Danny L. GRIFFIN, Petitionerv.OCEANIC CONTRACTORS, INC. No. 81-614. Argued April 26, 1982. Decided June 30, 1982. Syllabus Title 46 U.S.C. § 596, after obligating the master or owner of a vessel making coasting or foreign voyages to pay a seaman's unpaid wages within specified periods after his discharge, provides that a master or owner who fails to make such payment "without sufficient cause shall pay to the seaman a sum equal to two days' pay for each and every day during which payment is delayed beyond the respective periods." Petitioner, who was injured while working aboard respondent's vessel in foreign waters, brought suit under the Jones Act and general maritime law in Federal District Court after respondent refused to pay his medical expenses and to furnish transportation back to the United States. In addition to damages, petitioner sought to recover penalty wages under § 596 for respondent's failure to pay $412.50 in earned wages allegedly due upon discharge. The court found, inter alia, that petitioner had been discharged from respondent's employ on the day of the injury, and that respondent's failure to pay petitioner the $412.50 was "without sufficient cause." In assessing the penalty wages at $6,881.60, the court held that "[t]he period during which the penalty runs is to be determined by the sound discretion of the district court and depends on the equities of the case." It determined that the appropriate penalty period was the 34-day period from the date of discharge through the date when petitioner began work for another company. Petitioner appealed the award of damages as inadequate, but the Court of Appeals affirmed. Held : The district courts have no discretion to limit the period during which the wage penalty is assessed. Imposition of the penalty is mandatory for each day that payment is withheld in violation of § 596. Pp. 569-577. (a) The words chosen by Congress, given their plain meaning, leave no room for the exercise of discretion either in deciding whether to exact payment or in choosing the period of days by which the payment is to be calculated. After the District Court found that respondent had refused to pay petitioner the balance of his earned wages promptly after discharge and that its refusal was "without sufficient cause," nothing in § 596's language vested the court with discretion to limit the penalty assessment to the period of petitioner's unemployment. Pp. 569-571. (b) This is not the type of case where literal application of a statute would thwart its obvious purpose. Section 596's "evident purpose" is "to secure prompt payment of seamen's wages . . . and thus to protect them from the harsh consequences of arbitrary and unscrupulous action of their employers, to which, as a class, they are peculiarly exposed." Collie v. Fergusson, 281 U.S. 52, 55, 50 S.Ct. 189, 191, 74 L.Ed. 696. Although the statute's purpose is remedial, Congress has chosen to secure that purpose through the use of potentially punitive sanctions designed to deter negligent or arbitrary delays in payment. The legislative history confirms that Congress intended the statute to mean exactly what its plain language says. Pp. 571-574. (c) Nor is literal application of § 596 in this case precluded on the asserted ground that it would produce an absurd and unjust result which Congress could not have intended. Even though the penalty for respondent's failure to promptly pay the $412.50 in wages—if computed on the basis of the period from petitioner's discharge until the date respondent actually paid the wages by satisfying the District Court's judgment—would be over $300,000, awards made under § 596 were not intended to be merely compensatory. Since the District Court found that respondent's refusal to pay petitioner following his discharge was without sufficient cause, and since it made no finding that respondent's continuing delay in payment beyond the period petitioner was unable to work was for sufficient cause, its decision to limit the penalty was error. Pacific Mail S.S. Co. v. Schmidt, 241 U.S. 245, 36 S.Ct. 581, 60 L.Ed. 982. Pp. 574-577. 5th Cir., 664 F.2d 36, reversed and remanded. Robert A. Chaffin, Houston, Tex., for petitioner. Theodore Goller, Houston, Tex., for respondent. Justice REHNQUIST delivered the opinion of the Court. 1 This case concerns the application of 46 U.S.C. § 596, which requires certain masters and vessel owners to pay seamen promptly after their discharge and authorizes seamen to recover double wages for each day that payment is delayed without sufficient cause. The question is whether the district courts, in the exercise of discretion, may limit the period during which this wage penalty is assessed, or whether imposition of the penalty is mandatory for each day that payment is withheld in violation of the statute. 2 * On February 18, 1976, petitioner signed an employment contract with respondent in New Orleans, agreeing to work as a senior pipeline welder on board vessels operated by respondent in the North Sea. The contract specified that petitioner's employment would extend "until December 15, 1976 or until Oceanic's 1976 pipeline committal in the North Sea is fulfilled, whichever shall occur first." App. 41. The contract also provided that respondent would pay for transportation to and from the worksite, but that if petitioner quit the job prior to its termination date, or if his services were terminated for cause, he would be charged with the cost of transportation back to the United States. Respondent reserved the right to withhold $137.50 from each of petitioner's first four paychecks "as a cash deposit for the payment of your return transportation in the event you should become obligated for its payment." Id., at 47. On March 6, 1976, petitioner flew from the United States to Antwerp, Belgium, where he reported to work at respondent's vessel, the "Lay Barge 27," berthed in the Antwerp harbor for repairs. 3 On April 1, 1976, petitioner suffered an injury while working on the deck of the vessel readying it for sea. Two days later he underwent emergency surgery in Antwerp. On April 5, petitioner was discharged from the hospital and went to respondent's Antwerp office, where he spoke with Jesse Williams, the welding superintendent, and provided a physician's statement that he was not fit for duty. Williams refused to acknowledge that petitioner's injury was workrelated and denied that respondent was liable for medical and hospital expenses, maintenance, or unearned wages. Williams also refused to furnish transportation back to the United States, and continued to retain $412.50 in earned wages that had been deducted from petitioner's first three paychecks for that purpose. Petitioner returned to his home in Houston, Tex., the next day at his own expense. He was examined there by a physician who determined that he would be able to resume work on May 3, 1976. On May 5, petitioner began working as a welder for another company operating in the North Sea. 4 In 1978 he brought suit against respondent under the Jones Act, § 20, 38 Stat. 1185, as amended, 46 U.S.C. § 688, and under general maritime law, seeking damages for respondent's failure to pay maintenance, cure, unearned wages, repatriation expenses, and the value of certain personal effects lost on board respondent's vessel. Petitioner also sought penalty wages under Rev.Stat. § 4529, as amended, 46 U.S.C. § 596, for respondent's failure to pay over the $412.50 in earned wages allegedly due upon discharge. The District Court found for petitioner and awarded damages totalling $23,670.40. 5 Several findings made by that court are particularly relevant to this appeal. First, the court found that petitioner's injury was proximately caused by an unseaworthy condition of respondent's vessel. App. 17, ¶ 10; 23, ¶ 6. Second, the court found that petitioner was discharged from respondent's employ on the day of the injury, and that the termination of his employment was caused solely by that injury. Id., at 18, ¶ 16; 23, ¶ 7.1 Third, it found that respondent's failure to pay petitioner the $412.50 in earned wages was "without sufficient cause." Id., at 20, ¶ 20; 25, ¶ 11.2 Finally, the court found that petitioner had exercised due diligence in attempting to collect those wages. Id., at 20, ¶ 21. 6 In assessing penalty wages under 46 U.S.C. § 596, the court held that "[t]he period during which the penalty runs is to be determined by the sound discretion of the district court and depends on the equities of the case." App. 25, ¶ 11. It determined that the appropriate period for imposition of the penalty was from the date of discharge, April 1, 1976, through the date of petitioner's reemployment, May 5, 1976, a period of 34 days. Applying the statute, it computed a penalty of $6,881.60.3 Petitioner appealed the award of damages as inadequate. 7 The Court of Appeals for the Fifth Circuit affirmed. 664 F.2d 36 (1981). That court concluded, inter alia, that the District Court had not erred in limiting assessment of the penalty provided by 46 U.S.C. § 596 to the period beginning April 1 and ending May 5. The court recognized that the statute required payment of a penalty for each day during which wages were withheld until the date they were actually paid, which in this case did not occur until September 17, 1980, when respondent satisfied the judgment of the District Court. Id., at 40; see App. 30. Nevertheless, the court believed itself bound by prior decisions within the Circuit, which left calculation of the penalty period to the sound discretion of the district courts. 664 F.2d, at 40. It concluded that the District Court in this case had not abused its discretion by assessing a penalty only for the period during which petitioner was unemployed. 8 We granted certiorari to resolve a conflict among the Circuits regarding the proper application of the wage penalty statute.4 454 U.S. 1052, 102 S.Ct. 595, 70 L.Ed.2d 587 (1981). We reverse the judgment of the Court of Appeals as to that issue.5 II A. 9 The language of the statute first obligates the master or owner of any vessel making coasting or foreign voyages to pay every seaman the balance of his unpaid wages within specified periods after his discharge.6 It then provides: "Every master or owner who refuses or neglects to make payment in the manner hereinbefore mentioned without sufficient cause shall pay to the seaman a sum equal to two days' pay for each and every day during which payment is delayed beyond the respective periods. . . ." 10 The statute in straightforward terms provides for the payment of double wages, depending upon the satisfaction of two conditions. First, the master or owner must have refused or failed to pay the seaman his wages within the periods specified. Second, this failure or refusal must be "without sufficient cause." Once these conditions are satisfied, however, the unadorned language of the statute dictates that the master or owner "shall pay to the seaman" the sums specified "for each and every day during which payment is delayed." The words chosen by Congress, given their plain meaning, leave no room for the exercise of discretion either in deciding whether to exact payment or in choosing the period of days by which the payment is to be calculated. As this Court described the statute many years ago, it "affords a definite and reasonable procedure by which the seaman may establish his right to recover double pay where his wages are unreasonably withheld." McCrea v. United States, 294 U.S. 23, 32, 55 S.Ct. 291, 295, 79 L.Ed. 735 (1935). Our task is to give effect to the will of Congress, and where its will has been expressed in reasonably plain terms, "that language must ordinarily be regarded as conclusive." Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 108, 100 S.Ct. 2051, 2056, 64 L.Ed.2d 766 (1980). 11 The District Court found that respondent had refused to pay petitioner the balance of his earned wages promptly after discharge, and that its refusal was "without sufficient cause." Respondent challenges neither of these findings. Although the two statutory conditions were satisfied, however, the District Court obviously did not assess double wages "for each and every day" during which payment was delayed, but instead limited the assessment to the period of petitioner's unemployment. Nothing in the language of the statute vests the courts with the discretion to set such a limitation. B 12 Nevertheless, respondent urges that the legislative purpose of the statute is best served by construing it to permit some choice in determining the length of the penalty period. In respondent's view, the purpose of the statute is essentially remedial and compensatory, and thus it should not be interpreted literally to produce a monetary award that is so far in excess of any equitable remedy as to be punitive. 13 Respondent, however, is unable to support this view of legislative purpose by reference to the terms of the statute. "There is, of course, no more persuasive evidence of the purpose of a statute than the words by which the legislature undertook to give expression to its wishes." United States v. American Trucking Assns., Inc., 310 U.S. 534, 543, 60 S.Ct. 1059, 1063, 84 L.Ed. 1345 (1940). See Caminetti v. United States, 242 U.S. 470, 490, 37 S.Ct. 192, 196, 61 L.Ed. 442 (1917). Nevertheless, in rare cases the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters, and those intentions must be controlling. We have reserved "some 'scope for adopting a restricted rather than a literal or usual meaning of its words where acceptance of that meaning . . . would thwart the obvious purpose of the statute.' " Commissioner v. Brown, 380 U.S. 563, 571, 85 S.Ct. 1162, 1166, 14 L.Ed.2d 75 (1965) (quoting Helvering v. Hammel, 311 U.S. 504, 510-511, 61 S.Ct. 368, 371, 85 L.Ed. 303 (1941)). This, however, is not the exceptional case. 14 As the Court recognized in Collie v. Fergusson, 281 U.S. 52, 50 S.Ct. 189, 74 L.Ed. 696 (1930), the "evident purpose" of the statute is "to secure prompt payment of seamen's wages . . . and thus to protect them from the harsh consequences of arbitrary and unscrupulous action of their employers, to which, as a class, they are peculiarly exposed." Id., at 55, 50 S.Ct., at 191. This was to be accomplished "by the imposition of a liability which is not exclusively compensatory, but designed to prevent, by its coercive effect, arbitrary refusals to pay wages, and to induce prompt payment when payment is possible." Id., at 55-56, 50 S.Ct., at 191. Thus, although the sure purpose of the statute is remedial, Congress has chosen to secure that purpose through the use of potentially punitive sanctions designed to deter negligent or arbitrary delays in payment. 15 The legislative history of the statute leaves little if any doubt that this understanding is correct. The law owes its origins to the Act of July 20, 1790, ch. 29, § 6, 1 Stat. 133, passed by the First Congress. Although the statute as originally enacted gave every seaman the right to collect the wages due under his contract "as soon as the voyage is ended," it did not provide for the recovery of additional sums to encourage compliance. Such a provision was added by the Shipping Commissioners Act of 1872, ch. 322, § 35, 17 Stat. 269, which provided for the payment of "a sum not exceeding the amount of two days' pay for each of the days, not exceeding ten days, during which payment is delayed." The Act of 1872 obviously established a ceiling of 10 days on the period during which the penalty could be assessed and, by use of the words "not exceeding," left the courts with discretion to choose an appropriate penalty within that period.7 16 Congress amended the law again in 1898. As amended, it read in relevant part: 17 "Every master or owner who refuses or neglects to make payment in manner hereinbefore mentioned without sufficient cause shall pay to the seaman a sum equal to one day's pay for each and every day during which payment is delayed beyond the respective periods." Act of Dec. 21, 1898, ch. 28, § 4, 30 Stat. 756. 18 The amending legislation thus effected two changes: first, it removed the discretion theretofore existing by which courts might award less than an amount calculated on the basis of each day during which payment was delayed, and, second, it removed the 10-day ceiling which theretofore limited the number of days upon which an award might be calculated. The accompanying Committee Reports identify the purpose of the legislation as "the amelioration of the condition of the American seamen," and characterize the amended wage penalty in particular as "designed to secure the promptest possible payment of wages." H.R.Rep.No.1657, 55th Cong., 2d Sess., 2, 3 (1898). See also S.Rep.No.832, 54th Cong., 1st Sess., 2 (1896).8 Nothing in the legislative history of the 1898 Act suggests that Congress intended to do anything other than what the Act's enacted language plainly demonstrates: to strengthen the deterrent effect of the statute by removing the courts' latitude in assessing the wage penalty. 19 The statute was amended for the last time in 1915 to increase further the severity of the penalty by doubling the wages due for each day during which payment of earned wages was delayed. Seamen's Act of 1915, ch. 153, § 3, 38 Stat. 1164. There is no suggestion in the Committee Reports or in the floor debates that, in so doing, Congress intended to reinvest the courts with the discretion it had removed in the Act of 1898. Resort to the legislative history, therefore, merely confirms that Congress intended the statute to mean exactly what its plain language says. III 20 Respondent argues, however, that a literal construction of the statute in this case would produce an absurd and unjust result which Congress could not have intended. The District Court found that the daily wage to be used in computing the penalty was $101.20. If the statute is applied literally, petitioner would receive twice this amount for each day after his discharge until September 17, 1980, when respondent satisfied the District Court's judgment.9 Petitioner would receive over $300,000 simply because respondent improperly withheld $412.50 in wages. In respondent's view, Congress could not have intended seamen to receive windfalls of this nature without regard to the equities of the case. 21 It is true that interpretations of a statute which would produce absurd results are to be avoided if alternative interpretations consistent with the legislative purpose are available. See United States v. American Trucking Assns., Inc., 310 U.S., at 542-543, 60 S.Ct., at 1063; Haggar Co. v. Helvering, 308 U.S. 389, 394, 60 S.Ct. 337, 339, 84 L.Ed. 340 (1940). In refusing to nullify statutes, however hard or unexpected the particular effect, this Court has said: 22 "Laws enacted with good intention, when put to the test, frequently, and to the surprise of the law maker himself, turn out to be mischievous, absurd or otherwise objectionable. But in such case the remedy lies with the law making authority, and not with the courts." Crooks v. Harrelson, 282 U.S. 55, 60, 51 S.Ct. 49, 50, 75 L.Ed. 156 (1930). 23 It is highly probable that respondent is correct in its contention that a recovery in excess of $300,000 in this case greatly exceeds any actual injury suffered by petitioner as a result of respondent's delay in paying his wages. But this Court has previously recognized that awards made under this statute were not intended to be merely compensatory: 24 "We think the use of this language indicates a purpose to protect seamen from delayed payments of wages by the imposition of a liability which is not exclusively compensatory, but designed to prevent, by its coercive effect, arbitrary refusals to pay wages, and to induce prompt payment when payment is possible." Collie v. Fergusson, 281 U.S., at 55-56, 50 S.Ct., at 191. 25 It is in the nature of punitive remedies to authorize awards that may be out of proportion to actual injury; such remedies typically are established to deter particular conduct, and the legislature not infrequently finds that harsh consequences must be visited upon those whose conduct it would deter. It is probably true that Congress did not precisely envision the grossness of the difference in this case between the actual wages withheld and the amount of the award required by the statute. But it might equally well be said that Congress did not precisely envision the trebled amount of some damages awards in private antitrust actions, see Reiter v. Sonotone Corp., 442 U.S. 330, 344-345, 99 S.Ct. 2326, 2333-2334, 60 L.Ed.2d 931 (1979), or that, because it enacted the Endangered Species Act, "the survival of a relatively small number of three-inch fish . . . would require the permanent halting of a virtually completed dam for which Congress ha[d] expended more than $1 million," TVA v. Hill, 437 U.S. 153, 172, 98 S.Ct. 2279, 2290, 57 L.Ed.2d 117 (1978). It is enough that Congress intended that the language it enacted would be applied as we have applied it. The remedy for any dissatisfaction with the results in particular cases lies with Congress and not with this Court. Congress may amend the statute; we may not. See Consumer Product Safety Comm'n v. GTE Sylvania, Inc., 447 U.S., at 123-124, 100 S.Ct., at 2063-2064; Reiter v. Sonotone, supra, at 344-345, 99 S.Ct., at 2333-2334. 26 Finally, we note that our holding is consistent with Pacific Mail S. S. Co. v. Schmidt, 241 U.S. 245, 36 S.Ct. 581, 60 L.Ed. 982 (1916). The employer in that case challenged a decision by the Court of Appeals to apply the wage penalty to the delay after the District Court's judgment occasioned by the employer's appeal. The Court held that on the facts of that case, application of the penalty beyond the date of the District Court's judgment was error. Contrary to respondent's assertion, however, the holding does not reflect the discretionary tailoring of the penalty to the equities of the case. Instead, the Court held that the delay pending appeal was not "without sufficient cause," as required by the statute before the penalty can attach. Id., at 250, 36 S.Ct., at 582.10 As we explained earlier, a condition to the imposition of the wage penalty is a finding that the delay in payment is "without sufficient cause." To the extent that the equities of the situation are to be considered, see Collie v. Fergusson, supra, they bear on that finding, and not on the calculation of the penalty period once that finding has been made. IV 27 The District Court found that respondent's refusal to pay petitioner earned wages following his discharge was without sufficient cause. It applied the wage penalty only for the period of nonpayment during which petitioner was unable to work. It made no finding, however, that respondent's continuing delay in payment beyond that period was for sufficient cause. Under the plain language of the statute, therefore, its decision to limit the penalty period was error. The judgment of the Court of Appeals affirming that decision accordingly is reversed, and the case is remanded for proceedings consistent with this opinion. 28 It is so ordered. 29 Justice STEVENS, with whom Justice BLACKMUN joins, dissenting. 30 In final analysis, any question of statutory construction requires the judge to decide how the legislature intended its enactment to apply to the case at hand. The language of the statute is usually sufficient to answer that question, but "the reports are full of cases" in which the will of the legislature is not reflected in a literal reading of the words it has chosen.1 In my opinion this is such a case. 31 Qualifying language in 46 U.S.C. § 596 supports a much narrower construction than the Court adopts. For over 50 years after the statute's most recent amendment in 1915, federal judges consistently construed it to avoid the absurd result the Court sanctions today. Their reading of the statute was consistent with the specific purposes achieved by the amendments in 1898 and 1915, as well as with the meaning of the statute when an award for unearned wages was first authorized. 32 * On April 1, 1976, petitioner, a welder, suffered a temporarily disabling injury aboard respondent's vessel. On April 5, 1976, petitioner met with respondent's welding superintendent, who refused to acknowledge that respondent was responsible for the injury and who also refused to pay petitioner $412.50 in earned wages. Petitioner fully recovered from the injury by May 3, 1976, and two days later obtained comparable work with another employer. He filed this action on February 3, 1978. It is now settled that respondent was responsible for petitioner's injury and that respondent wrongfully refused to pay him $412.50 on April 5, 1976. 33 The question of statutory construction that is before us is what "sum shall be recoverable as wages" to compensate petitioner for respondent's refusal to pay him $412.50 on April 5, 1976. 46 U.S.C. § 596.2 The District Court computed that sum by doubling his daily wage of $101.20 and multiplying that amount by 34—the number of days between the injury on April 1 and petitioner's reemployment on May 5, 1976. The District Court's award thus amounted to $6,881.60.3 This Court holds that the sum recoverable as wages amounts to at least $302,790.40.4 II 34 In pertinent part, § 596 provides as follows: 35 "Every master or owner who refuses or neglects to make payment [of a seaman's earned wages within four days after the seaman's discharge] without sufficient cause shall pay to the seaman a sum equal to two days' pay for each and every day during which payment is delayed beyond the [4-day period], which sum shall be recoverable as wages in any claim made before the court. . . ."5 (Emphasis added.) 36 The text of the statute admittedly supports the construction given it by the Court—if there was not sufficient cause for the refusal to make payment within four days of the discharge, then the seaman is entitled to double wages for the entire period between the fourth day and the date the payment is finally made. The statute, however, is susceptible of another interpretation. Indeed, for a half century following its latest amendment the federal courts, including this Court, consistently exercised some discretion in determining the sum recoverable as wages under this section. A. 37 In fixing the amount of the award of double wages, the District Court in this case may have reasoned that respondent had sufficient cause for its delay in paying the earned wages after petitioner obtained employment with another shipmaster, but that there was not sufficient cause for its failure to make payment before that time. Although this reasoning conflicts with a literal reading of § 596, it is perfectly consistent with this Court's contemporary construction of the statute in Pacific Mail S. S. Co. v. Schmidt, 241 U.S. 245, 36 S.Ct. 581, 60 L.Ed. 982 (1916). The teaching of Justice Holmes' opinion for the Court in that case is that the wrongful character of the initial refusal to pay does not mean that all subsequent delay in payment is also "without sufficient cause" within the meaning of the statute. 38 The controversy in Pacific Mail arose in 1913, when the statute provided that the sum recoverable as wages was measured by one day's pay, rather than double that amount, for each day that the wages were withheld without sufficient cause; the statute was otherwise exactly as it is today. The seaman was discharged on October 1, 1913, but $30.33 was withheld from his wages because he was believed responsible for the loss of some silverware. He filed an action on October 20, 1913, and on November 5, 1913, obtained a judgment for his wages and an additional sum of $151.59, representing the sum recoverable as wages for the period between October 1 and November 5, 1913. The District Court's decree established the proposition that the vessel owner's defenses did not constitute sufficient cause for refusing to pay the wages and requiring the seaman to sue to recover them. 39 The vessel owner prosecuted an unsuccessful appeal. The Court of Appeals not only affirmed the decision of the District Court, but also added an additional recovery of daily wages for the period between the entry of the original judgment on November 5 and the actual payment of the disputed wages. The Court of Appeals thus read the statute literally and ordered the result that the District Court's finding seemed to dictate. This Court, however, set aside the additional recovery, reaching a conclusion that cannot be reconciled with a wooden, literal reading of the statute. Concurrent findings of the District Court and the Court of Appeals established that the refusal to make the wage payment when due was without sufficient cause. Justice Holmes and his Brethren accepted that finding for purposes of decision, but reasoned that there was sufficient cause for the owner's decision to appeal and his refusal to pay while the appeal was pending. 40 The curious character of this Court's conclusion that reasons insufficient to justify the refusal to pay before the trial court's decision somehow became sufficient to justify a subsequent refusal to pay is not the most significant point to Justice Holmes' opinion. The case is primarily significant because its holding cannot be squared with a literal reading of the statute.6 Even though the initial refusal is without sufficient cause, statutory wages are not necessarily recoverable for the entire period until payment is made either to the seaman or to a stakeholder.7 A subsequent event—even though not expressly mentioned in the statute itself—may foreshorten the recovery period. 41 In Pacific Mail the subsequent event was the vessel owner's decision to appeal. The finding that that event provided sufficient cause for the delay after November 5, 1913, was made sua sponte by this Court. In this case the subsequent event was the reemployment of petitioner in a comparable job on May 5, 1976. The finding that that event—coupled with the failure to make any additional demand for almost two years thereafter—was sufficient cause for the delay after May 5, 1976, was made by the District Court. It is true that the judge did not expressly frame his decision in these terms, but his actual decision fits precisely the mold established by Pacific Mail. Both cases give a flexible reading to the "sufficient cause" language in the statute. They differ with respect to the nature of the subsequent event but not with respect to their departure from the statutory text.8 B 42 The second case in which this Court construed § 596, Collie v. Ferguson, 281 U.S. 52, 50 S.Ct. 189, 74 L.Ed. 696 (1930), also focused on the meaning of the phrase "without sufficient cause." In that case the unpaid seamen claimed that the financial necessities of the owner could not constitute sufficient cause for delay in wage payments; that contention was surely consistent with the plain language of the statute. This Court nevertheless denied recovery, construing the statute as implicitly containing a requirement that the refusal be "in some sense arbitrary or wilful, or at least a failure not attributable to impossibility of payment." Id., at 55, 50 S.Ct., at 191. The Court adopted this nonliteral construction of the statute because it recognized the significance of the provision that a seaman's double-wage claim "shall be recoverable as wages in any claim made before the Court." See id., at 54, 50 S.Ct., at 190. In any proceeding arising out of the insolvency of the vessel owner, this provision accords this type of claim priority over general creditors and various lienors who have stronger equitable claims on limited assets. The construction of the words "without sufficient cause" to narrow the protection of the statute was consistent with the intent of Congress even though it involved a rather flexible reading of the text of the statute itself.9 43 This Court's third occasion to interpret § 596 was McCrea v. United States, 294 U.S. 23, 55 S.Ct. 291, 79 L.Ed. 735 (1935), and, once again, the Court construed the statute narrowly, this time by taking a literal approach. In that case the seaman, citing specific sections of federal legislation, demanded from the shipmaster his discharge, his earned wages, and other benefits. The master was unfamiliar with the cited sections and asked the seaman to meet with him at noon the next day for an informed discussion of the demands. The seaman missed the appointment and left the country without contacting the master. After his return to the United States, the seaman filed an action in which he claimed entitlement to, inter alia, his earned wages and double wages for the delay in payment. The District Court, affirmed by the Court of Appeals, held that the owner of the ship, the United States, was immune from the double-wage provision of § 596 because the double wages constituted a penalty. This Court granted the seaman's petition for certiorari, but avoided decision of the sovereign immunity question by holding that there was sufficient cause for the failure of the shipmaster to make the wage payment within four days of the seaman's discharge. The Court then rejected the seaman's rather reasonable argument that even if the shipmaster had cause to withhold payment during those four days, there was not sufficient cause for the continued refusal once the seaman filed his action and formally made his claim to earned wages. The Court, without citing Pacific Mail, held that if the master's failure to pay earned wages at the time specified in the statute was justified by sufficient cause, the fact that he later refused to pay pursuant to a proper demand could not give rise to statutory liability even though there was no sufficient cause for the subsequent refusal. 44 These early interpretations of § 596 dispel any notion that the statute means exactly what it says. The Court has construed the statute "to effect its purpose," Isbrandtsen Co. v. Johnson, 343 U.S. 779, 783, 72 S.Ct. 1011, 1014, 96 L.Ed. 1294 (1952), and, as the early cases demonstrate, the purpose of the statute does not always require the award of double wages in the amount that the statute literally specifies. C 45 Flexibility also has characterized the applications of the statute rendered by the lower federal courts. For decades those courts consistently concluded that Congress intended to allow judicial discretion to play a part in determining the amount of the double-wage recovery.10 Whether those decisions were entirely consistent with the meaning a grammarian might have placed on the statute is less significant than the fact that they were entirely consistent with this Court's decisions and with one another,11 and the fact that their holdings must have come to the attention of Congress. 46 It was not until 1966 that a contrary reading of the statute was adopted by the Third Circuit in Swain v. Isthmian Lines, Inc., 360 F.2d 81,12 and another eight years before that case was followed in another Circuit.13 I cannot deny that there is wisdom in the rule of construction that mandates close adherence to literal statutory text,14 but it is also true that a consistent course of judicial construction can become as much a part of a statute as words inserted by the legislature itself. The construction consistently followed by the federal judiciary between 1898 and 1966 was presumably acceptable to Congress, and I find this more persuasive than the literal reading on which the Court places its entire reliance.15 Moreover, since the result that construction produces in this case is both absurd and palpably unjust, this is one of the cases in which the exercise of judgment dictates a departure from the literal text in order to be faithful to the legislative will.16 III 47 The construction permitting the district court to exercise some discretion in tailoring the double-wage award to the particular equities of the case is just as consistent with the legislative history of § 596 as the Court's new literal approach to this statute. In 1872, when Congress authorized the recovery of additional wages by seamen who were not paid within five days of their discharge, it used the word "shall" to make it clear that such a recovery must be awarded, but it allowed the district courts a limited discretion in setting the amount of such recovery.17 The judge's discretion as to amount was limited in two ways: (1) the statutory wage rate could not be more than double the amount of the seaman's daily wage; and (2) the period for which the statutory wage could be awarded could not exceed 10 days. 48 Subsequent amendments to the statute did not remove the requirement that some recovery "shall" be awarded, but did modify both of the limits on the judge's discretion. With respect to the wage rate, Congress first specified that it should equal the daily rate—rather than double the daily rate—and later specified that the rate should be the double rate.18 With respect to the period for which the statutory wage was payable, the 1898 amendment simply removed the 10-day limit. This amendment is subject to two different interpretations, one that would represent a rather unremarkable change and the other that would be both drastic and dramatic. 49 The unremarkable change would amount to nothing more than a removal of the narrow 10-day limit on the scope of the judge's discretion. The word "shall" would continue to do nothing more than require some recovery in an amount to be fixed by the judge, but in recognition of the reality that seamen might be stranded for more than 10 days, the recovery period could extend beyond 10 days. This sort of unremarkable change is consistent with the purpose of the statute,19 as well as with a legislative history that fails to make any comment on its significance. As Justice REHNQUIST has perceptively observed in another context, the fact that the dog did not bark can itself be significant.20 50 The Court's construction of the amendment is, however, both drastic and dramatic. Instead of effecting a modest enlargement of the judge's discretion to do justice in these cases, the Court's construction effects a complete prohibition of judicial discretion. Instead of permitting recoveries for a period somewhat longer than 10 days, the amendment is construed as a command that even when the unresolved dispute persists for two or three years without any special hardship to the seaman, an automatic recovery must be ordered for the entire period regardless of the equitable considerations that may arise after the shipmaster's initial mistake has been made. Such a major change in both the potential amount of the statutory recovery and the character of the judge's authority would normally be explained in the commitee reports or the debates if it had been intended.21 IV 51 It is ironic that the same seven Justices—who today are transfixed by a literal reading of § 596—only a few days ago blithely ignored the text of the Tax Injunction Act in order to reach the conclusion that a federal court has no jurisdiction to entertain a suit for a declaratory judgment against the United States Secretary of Labor to determine whether a federal statute violates the Federal Constitution. California v. Grace Brethren Church, 457 U.S. 393, 102 S.Ct. 2498, 73 L.Ed.2d 93 (1982). The inconsistency in the Court's approach to the task of statutory construction in these two cases is less troublesome, however, than its failure in each case to consider whether its conclusion could reasonably be thought to represent the will of Congress. I am not persuaded that the 1898 amendment, removing the 10-day limit on the scope of the trial judge's discretion, was intended to be read as a command to award $302,790.40 to a seaman who was not paid $412.50 in wages when due. 52 I respectfully dissent. 1 According to respondent, petitioner was not formally discharged until June 1, 1976, but his termination was made retroactive to April 1. Brief for Respondent 5. 2 The court also found: "Defendant did not begin a thorough investigation of plaintiff's claim until September 30, 1976. The investigation was not made with reasonable diligence. Defendant's failure to pay maintenance and cure, repatriation expenses, the cost of his personal effects, and earned and unearned wages to plaintiff constituted arbitrary, unreasonable, callous, and willful disregard of plaintiff's rights." App. 20, ¶ 21. 3 The court found that the daily wage rate to be used in calculating the penalty was $101.20. In accordance with the statute, the court assessed a penalty of twice this rate ($202.40) for each of the 34 days of the penalty period. 4 The Courts of Appeals for the Third and Ninth Circuits have interpreted the statute to mandate imposition of the penalty for each day until the wages are paid and to leave no room for the district court's exercise of discretion. Swain v. Isthmian Lines, Inc., 360 F.2d 81 (CA3 1966); Larkins v. Hudson Waterways Corp., 640 F.2d 997 (CA9 1981); Thomas v. §§ Santa Mercedes, 572 F.2d 1331 (CA9 1978). The Courts of Appeals for the First, Second, and Fourth Circuits have adopted the interpretation followed by the Fifth Circuit. Mavromatis v. United Greek Shipowners Corp., 179 F.2d 310 (CA1 1950); Forster v. Oro Navigation Co., 228 F.2d 319 (CA2 1955), aff'g 128 F.Supp. 113 (SDNY 1954); Southern Cross S.S. Co. v. Firipis, 285 F.2d 651 (CA4 1960), cert. denied, 365 U.S. 869, 81 S.Ct. 903, 5 L.Ed.2d 859 (1961). We noted this conflict in American Foreign S.S. Co. v. Matise, 423 U.S. 150, 152, n. 1, 96 S.Ct. 410, 412 n. 1, 46 L.Ed.2d 354 (1975). 5 Petitioner has not questioned the other holdings of the Court of Appeals in his case. Respondent did not appeal from the judgment of the District Court and has not cross-petitioned for certiorari here. 6 The statute reads in full: "The master or owner of any vessel making coasting voyages shall pay to every seaman his wages within two days after the termination of the agreement under which he was shipped, or at the time such seaman is discharged, whichever first happens; and in case of vessels making foreign voyages, or from a port on the Atlantic to a port on the Pacific, or vice versa, within twenty-four hours after the cargo has been discharged, or within four days after the seaman has been discharged, whichever first happens; and in all cases the seaman shall be entitled to be paid at the time of his discharge on account of wages a sum equal to one-third part of the balance due him. Every master or owner who refuses or neglects to make payment in the manner hereinbefore mentioned without sufficient cause shall pay to the seaman a sum equal to two days' pay for each and every day during which payment is delayed beyond the respective periods, which sum shall be recoverable as wages in any claim made before the court; but this section shall not apply to masters or owners of any vessel the seamen of which are entitled to share in the profits of the cruise or voyage. This section shall not apply to fishing or whaling vessels or yachts." 7 The Act of 1790 and the Act of 1872 provided the basis for § 4529 of the Revised Statutes, codified in 1878. Section 4529 read as follows: "The master or owner of every vessel making voyages from a port on the Atlantic to a port on the Pacific, or vice versa, shall pay to every seaman his wages, within two days after the termination of the agreement, or at the time such seaman is discharged, whichever first happens; and, in the case of vessels making foreign voyages, within three days after the cargo has been delivered, or within five days after the seaman's discharge, whichever first happens; and in all cases the seaman shall, at the time of his discharge, be entitled to be paid, on account, a sum equal to one-fourth part of the balance due him. Every master or owner who neglects or refuses to make payment in manner hereinbefore mentioned, without sufficient cause, shall pay to the seaman a sum not exceeding the amount of two days' pay for each of the days, not exceeding ten days, during which payment is delayed beyond the respective periods; which sum shall be recoverable as wages in any claim made before the court. But this section shall not apply to the masters or owners of any vessel the seamen on which are entitled to share in the profits of the cruise or voyage." 8 The 1898 Act was substantially identical to legislation that had passed the House in the previous Congress, and had been favorably reported in the Senate, but had failed to come to a vote before the end of the session. Thus, the House Report of the legislation enacted in 1898 contained little more than a reproduction of the House Report of the previous Congress, and the relevant Senate Report also dates from that Congress. 9 Respondent assumes that the penalty would run until September 17, 1980, since that was the date on which it finally paid petitioner the $412.50. Brief for Respondent 17. Petitioner, on the other hand, apparently assumes that the penalty period expired on May 6, 1980, the date of the District Court's judgment. Brief for Petitioner 19. Under our construction of the statute, the District Court's entry of judgment will not toll the running of the penalty period unless delays beyond that date are explained by sufficient cause. See Pacific Mail S.S. Co. v. Schmidt, 241 U.S. 245, 250-251, 36 S.Ct. 581, 582-583, 60 L.Ed. 982 (1916) (holding that when an appeal is taken on reasonable grounds, the penalty should not apply to delays in payment beyond the date on which the district court's decree is entered, since those delays are supported by sufficient cause). The Court of Appeals for the Fourth Circuit, in Southern Cross S.S. Co. v. Firipis, 285 F.2d, at 660, and the Court of Appeals for the Third Circuit in Swain v. Isthmian Lines, Inc., 360 F.2d, at 88, n. 26, have interpreted this Court's decision in Pacific Mail to permit the employer to toll the running of the penalty period by placing in the hands of the court the allegedly unlawfully withheld wages. 10 The Court found that the employer "had strong and reasonable ground for believing that the statute ought not to be held to apply," 241 U.S., at 250, 36 S.Ct., at 582, because the work for which the seaman claimed unpaid wages did not occur during a voyage and was the result of an oral contract. 1 "It is a familiar rule, that a thing may be within the letter of the statute and yet not within the statute, because not within its spirit, nor within the intention of its makers. This has been often asserted, and the reports are full of cases illustrating its application. This is not the substitution of the will of the judge for that of the legislator, for frequently words of general meaning are used in a statute, words broad enough to include an act in question, and yet a consideration of the whole legislation, or of the circumstances surrounding its enactment, or of the absurd results which follow from giving such broad meaning to the words, makes it unreasonable to believe that the legislator intended to include the particular act." Holy Trinity Church v. United States, 143 U.S. 457, 459, 12 S.Ct. 511, 512, 36 L.Ed. 226 (1892). 2 The statute is quoted in full, ante, at 569-570, n. 6. 3 The Court of Appeals held that the award did not constitute an abuse of discretion. 664 F.2d 36, 40 (CA5 1981). 4 This figure is computed by reference to the period between April 1, 1976 (date of discharge), and May 6, 1980 (date of judgment). But see ante, at 574-575, n. 9. 5 The Court omits the italicized clause of the statute when it quotes the statute in the text of its opinion. Ante, at 570 and 573. Because wage claims, unlike penalties, have consistently been accorded a high priority in insolvency proceedings against employers, it seems to me the clause is pertinent to our task of discerning the intent of Congress in authorizing an award of double wages. See also infra, at 582-583. 6 In a petition for rehearing, after pointing out that the Court had adopted an interpretation of the statute that had not been urged in any of the briefs or in any of the opinions of the lower courts, the seaman argued: "According to the grammatical, natural and unambiguous meaning conveyed by the words of section 4529, R.S., Congress has limited the running of penalties only by a 'sufficient cause' for original non-payment of earned wages—a sufficient cause operating to prevent penalties from ever beginning to accrue; once these penalties begin to run, nothing short of actual payment or tender can suffice to prevent the continuous accrual of the per diem penalties." Pet. for Rehearing, O.T. 1915, No. 323, p. 5. 7 Petitioner argues that literal application of § 596 will not yield harsh results because the shipmaster may toll the period of delay by tendering the disputed wage claim into the registry of the court. Brief for Petitioner 27. In petitioner's words, the master may make a "constructive payment of wages." Id., at 28. The Court seems to accept this argument. See ante, at 574-575, n.9. Neither petitioner nor the Court seems to recognize that acceptance of this minimal tolling rule conflicts with a literal reading of the statute. A tender of wages to the court is a "constructive payment of wages" only because a court has added that reasonable gloss to the statute. 8 If the Court today were to read Pacific Mail as requiring that the subsequent-event finding be made expressly, I would either follow Justice Holmes' lead by making a comparable finding in this Court or I would remand to the District Court for additional findings on this issue. 9 The District Court's finding of arbitrariness in this case which, in view of the holding in Collie, was necessary if any penalty wage were to be recovered—must be read in the context of its actual award. It was not clearly erroneous to find that the refusal to pay petitioner $412.50 was arbitrary while he was unemployed; it surely was not equally arbitrary during the ensuing 2-year period when he was employed by a competitor and did not renew his demand. 10 See Mystic S.S. Co. v. Stromland, 20 F.2d 342, 344 (CA4) ("The District Court, by limiting the right of recovery to 10 days after the libel was filed, in effect placed a limitation on the amount of the recovery under the statute. Was there objection to this? We think not"), rehearing denied, 21 F.2d 607 (1927), cert. denied, 276 U.S. 618, 48 S.Ct. 213, 72 L.Ed. 734 (1928); Mavromatis v. United Greek Shipowners Corp., 179 F.2d 310, 316 (CA1 1950) ("The language of § 596 has been given a somewhat free reading so as to accord to the courts a considerable margin of discretion in adjusting the duration of the penalty to the equities of the particular case"); Prindes v. S.S. African Pilgrim, 266 F.2d 125, 128 (CA4 1959) ("The period during which the penalty accumulates is to be determined by the equities of the particular case"); Southern Cross §§ Co. v. Firipis, 285 F.2d 651, 658 (CA4 1960) ("With regard to liability for double pay, a doctrine had developed before the McCrea case, and has continued to the present time, that the District Court has a measure of discretion through the application of equitable principles in determining the number of days for which double wages should be assessed"), cert. denied, 365 U.S. 869, 81 S.Ct. 903, 5 L.Ed.2d 859 (1961); Caribbean Federation Lines v. Dahl, 315 F.2d 370, 374 (CA5) ("The time for which the penalty provision runs rests within the sound discretion of the court and depends upon the equities of the case"), cert. denied, 375 U.S. 831, 84 S.Ct. 76, 11 L.Ed.2d 62 (1963); McConville v. Florida Towing Corp., 321 F.2d 162, 168, n. 11 (CA5 1963) ("The Court has wide equitable discretion in fixing the time for which the penalty provision runs"); The Chester, 25 F.2d 908, 911 (Md.1928) ("[I]n spite of the seeming rigidity of the statute, there is still left to the courts certain discretionary power to limit the penalties"); The Victoria, 76 F.Supp. 54, 56 (SDNY 1947) ("In spite of the seeming rigidity of the statute, the court still has discretionary power to limit the penalties"), rev'd on other grounds, 172 F.2d 434 (CA2 1949); Forster v. Oro Navigation Co., 128 F.Supp. 113, 116 (SDNY 1954) ("The number of days for which the defendant must pay double wages rests in the discretion of the court and depends on the equities of the particular case"), aff'd, 228 F.2d 319 (CA2 1955); Samad v. The Etivebank, 134 F.Supp. 530, 542 (ED Va.1955) ("The number of days for which respondents must pay double wages rests in the discretion of the Court and depends upon the equities of the particular case"); Spero v. Steamship The Argodon, 150 F.Supp. 1, 6 (ED Va.1957) ("It is well settled that the number of days for which respondents must pay double wages rests in the discretion of the Court and depends upon the equities of the particular case"); Kontos v. §§ Sophie C., 236 F.Supp. 664, 674 (ED Pa.1964) ("Although the penalty is applicable, its duration seems to be committed to the discretion of the trial judge to tailor to the equities of the particular case"); Ventiadis v. C. J. Thibodeaux & Co., 295 F.Supp. 135, 138 (SD Tex.1968) (" 'The time for which the penalty provision runs rests within the sound discretion of the court and depends upon the equities of the case' "); see also Swanson v. Torry, 25 F.2d 835 (CA4 1928); The Lake Galewood, 21 F.2d 987 (Md.1927), aff'd, 25 F.2d 1020 (CA4), cert. denied, 278 U.S. 637, 49 S.Ct. 33, 73 L.Ed. 553 (1928). 11 See Southern Cross §§ Co. v. Firipis, supra, at 655-658. 12 See 360 F.2d, at 85 ("Despite this rather precise statutory directive, of those cases which we have uncovered, where the question—whether Section 596 of the statute may still be read with a measure of judicial discretion when supposed equitable considerations present themselves—was considered, all have found proper the balancing of the statutory language with a judicial sense of the equities of each case."). 13 See Escobar v. §§ Washington Trader, 503 F.2d 271, 274 (CA9 1974), vacated and remanded on other grounds sub nom. American Trading Transportation Co. v. Escobar, 423 U.S. 1070, 96 S.Ct. 852, 47 L.Ed.2d 80 (1976). 14 See, most recently, Weinberger v. Romero-Barcelo, 456 U.S. 305, 322-335, 102 S.Ct. 1798, 1808-1815, 72 L.Ed.2d 91 (1982) (STEVENS, J., dissenting). 15 This Court based its interpretation of "sufficient cause" in Collie v. Ferguson, 281 U.S. 52, 50 S.Ct. 189, 74 L.Ed. 696 (1930), in part upon "the conclusion reached with practical unanimity by the lower federal courts." Id., at 56, 50 S.Ct., at 191. 16 "The Court has had several occasions within the last few years to construe statutes in which conflicts between reasonable intention and literal meaning occurred. We have refused to nullify statutes, however hard or unexpected the particular effect, where unambiguous language called for a logical and sensible result. Any other course would be properly condemned as judicial legislation. However, to construe statutes so as to avoid results glaringly absurd, has long been a judicial function. Where, as here, the language is susceptible of a construction which preserves the usefulness of the section, the judicial duty rests upon this Court to give expression to the intendment of the law." Armstrong Paint & Varnish Works v. Nu-Enamel Corp., 305 U.S. 315, 332-333, 59 S.Ct. 191, 199-200, 83 L.Ed. 195 (1938) (footnotes omitted). 17 The 1872 version of § 596 provided in pertinent part: "[E]very master or owner who neglects or refuses to make payment [of a seaman's earned wages within five days after the seaman's discharge] without sufficient cause shall pay to the seaman a sum not exceeding the amount of two days' pay for each of the days, not exceeding ten days, during which payment is delayed beyond the [five-day period]; and such sum shall be recoverable as wages in any claim made before the court . . . ." Act of June 7, 1872, ch. 322, § 35, 17 Stat. 269. 18 The 1898 version of § 596 provided in pertinent part: "Every master or owner who refuses or neglects to make payment [of a seaman's earned wages within four days of the seaman's discharge] without sufficient cause shall pay to the seaman a sum equal to one day's pay for each and every day during which payment is delayed beyond the [four-day period], which sum shall be recoverable as wages in any claim made before the court . . . ." Act of Dec. 21, 1898, § 4, 30 Stat. 756. The 1915 amendment substituted "two days' pay" for "one day's pay." See Act of Mar. 4, 1915, § 3, 38 Stat. 1164-1165. 19 Justice Cardozo, while a member of the New York Court of Appeals, explained the purpose of the statute: "The purpose, or at least the predominant one, was, not punishment of the master or owner, but compensation to the seaman. Delay means loss of opportunity to ship upon another vessel. It means hardship during the term of waiting, the sufferer often improvident, and stranded far from home. 'In all fairness he should recover more than the amount due him for wages earned' (Calvin v. Huntley, 178 Mass. 29, 32, 59 N.E. 435)." Cox v. Lykes Brothers, 237 N.Y. 376, 379, 143 N.E. 226, 227 (1924). The District Court's award in this case, tolling the period for computing the double wages on the date petitioner obtained employment with another seagoing vessel, was also perfectly consistent with the statutory purpose. 20 Harrison v. PPG Industries, Inc., 446 U.S. 578, 602, 100 S.Ct. 1889, 1902, 64 L.Ed.2d 525 (1980) (dissenting opinion); cf. A. Conan Doyle, Silver Blaze, in The Complete Sherlock Holmes 383 (1938). 21 The House Report's description of the 1898 amendment was that it "increases from one-fourth to one-third the amount of balance of wages due the seaman, to which he is entitled immediately upon discharge, and, in general, provides for prompter payment of wages of seamen." H.R.Rep.No.1657, 55th Cong., 2d Sess., 3 (1898). It is noteworthy that the first change described, respecting the part of the seaman's wages to which he is entitled at the time of his discharge in every case, is rather trivial. See the full text of the statute, ante, at 569-570, n. 6. And the description of the provision for prompter payment of wages most likely refers to the amendment of the time period during which the shipmaster had to pay the seaman his full wages in order to avoid the double-wage provision of § 596. In the cases of seamen of vessels making foreign voyages, the time period was changed from five days in 1872, see n. 17, supra, to four days in 1898, see n. 18, supra.

78

458 U.S. 457 102 S.Ct. 3187 73 L.Ed.2d 896 WASHINGTON, et al., Appellantsv.SEATTLE SCHOOL DISTRICT NO. 1, et al. No. 81-9. Argued March 22, 1982. Decided June 30, 1982. Syllabus In 1978, appellee Seattle School District No. 1 (District) enacted the so-called Seattle Plan for desegregation of its schools. The plan makes extensive use of mandatory busing. Subsequently, a statewide initiative (Initiative 350) was drafted to terminate the use of mandatory busing for purposes of racial integration in the public schools of the State of Washington. The initiative prohibits school boards from requiring any student to attend a school other than the one geographically nearest or next nearest to his home. It sets out a number of broad exceptions to this requirement, however: a student may be assigned beyond his neighborhood school if he requires special educational programs, or if the nearest or next nearest school is overcrowded or unsafe, or if it lacks necessary physical facilities. These exceptions permit school boards to assign students away from their neighborhood schools for virtually all of the nonintegrative purposes required by their educational policies. After the initiative was passed at the November 1978 general election, the District, together with two other districts, brought suit against appellant State in Federal District Court, challenging the constitutionality of Initiative 350 under the Equal Protection Clause of the Fourteenth Amendment. The District Court held the initiative unconstitutional on the ground, inter alia, that it established an impermissible racial classification in violation of Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616, and Lee v. Nyquist, 318 F.Supp. 710 (WDNY), summarily aff'd, 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105, "because it permits busing for non-racial reasons but forbids it for racial reasons." The court permanently enjoined implementation of the initiative's restrictions. The Court of Appeals affirmed. Held : Initiative 350 violates the Equal Protection Clause. Pp. 467-487. (a) When a State allocates governmental power nonneutrally, by explicitly using the racial nature of a decision to determine the decisionmaking process, its action "places special burdens on racial minorities within the governmental process," Hunter v. Erickson, 393 U.S., at 391, 89 S.Ct., at 560, thereby "making it more difficult for certain racial and religious minorities [than for other members of the community] to achieve legislation that is in their interest." Id., at 395, 89 S.Ct., at 563. Such a structuring of the political process is "no more permissible than [is] denying [members of a racial minority] the vote, on an equal basis with others." Id., at 391, 89 S.Ct., at 560. Pp. 467-470. (b) Initiative 350 must fall because it does "not attemp[t] to allocate governmental power on the basis of any general principle," Hunter v. Erickson, 393 U.S., at 395, 89 S.Ct., at 563, but instead uses the racial nature of an issue to define the governmental decisionmaking structure, thus imposing substantial and unique burdens on racial minorities. The initiative worked a major reordering of the State's educational decisionmaking process. Before adoption of the initiative, the power to determine what programs would most appropriately fill a school district's educational needs—including programs involving student assignment and desegregation—was committed to the local board's discretion. After passage of Initiative 350, authority over all but one of these areas remained in the local board's hands. By placing power over desegregative busing at the state level, the initiative thus "differentiates between the treatment of problems involving racial matters and that afforded other problems in the same area." Lee v. Nyquist, 318 F.Supp., at 718. And Initiative 350 works something more than the "mere repeal" of a desegregation law by the political entity that created it. It burdens all future attempts to integrate Washington schools by lodging decisionmaking authority over the question at a new and remote level of government. This makes the enactment of racially beneficial legislation uniquely difficult, and therefore imposes direct and undeniable burdens on minority interests. Pp. 470-484. (c) Contrary to appellants' suggestion, Hunter v. Erickson was not effectively overruled by Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597, and Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 97 S.Ct. 555, 50 L.Ed.2d 450. While Washington v. Davis and Arlington Heights considered classifications facially unrelated to race, Hunter like this case—involved an attempt to use explicitly racial criteria to define the community's decisionmaking structure. In so doing, the legislation at issue there directly and invidiously curtailed "the operation of those political processes ordinarily to be relied upon to protect minorities." United States v. Carolene Products Co., 304 U.S. 144, 152-153, n. 4, 58 S.Ct. 778, 783-784, n. 4, 82 L.Ed. 1234. Hunter § principle—that meaningful and unjustified distinctions based on race are impermissible—is still vital. Pp. 484-487. 9th Cir., 633 F.2d 1338, affirmed. Kenneth O. Eikenberry, Olympia, Wash., for appellants. Michael W. Hoge, Seattle, Wash., for appellees. Justice BLACKMUN delivered the opinion of the Court. 1 We are presented here with an extraordinary question: whether an elected local school board may use the Fourteenth Amendment to defend its program of busing for integration from attack by the State. 2 * A. 3 Seattle School District No. 1 (District), which is largely coterminous with the city of Seattle, Wash., is charged by state law with administering 112 schools and educating approximately 54,000 public school students. About 37% of these children are of Negro, Asian, American Indian, or Hispanic ancestry. Because segregated housing patterns in Seattle have created racially imbalanced schools, the District historically has taken steps to alleviate the isolation of minority students; since 1963, it has permitted students to transfer from their neighborhood schools to help cure the District's racial imbalance.1 4 Despite these efforts, the District in 1977 came under increasing pressure to accelerate its program of desegregation.2 In response, the District's Board of Directors (School Board) enacted a resolution defining "racial imbalance" as "the situation that exists when the combined minority student enrollment in a school exceeds the districtwide combined average by 20 percentage points, provided that the single minority enrollment . . . of no school will exceed 50 percent of the student body." 473 F.Supp. 996, 1006 (WD Wash.1979). The District resolved to eliminate all such imbalance from the Seattle public schools by the beginning of the 1979-1980 academic year.3 5 In September 1977, the District implemented a "magnet" program, designed to alleviate racial isolation by enhancing educational offerings at certain schools, thereby encouraging voluntary student transfers. A "disproportionate amount of the overall movement" inspired by the program was undertaken by Negro students, however, ibid., and racial imbalance in the Seattle schools was found to have actually increased between the 1970-1971 and 1977-1978 academic years. The District therefore concluded that mandatory reassignment of students was necessary if racial isolation in its schools was to be eliminated. Accordingly, in March 1978, the School Board enacted the so-called "Seattle Plan" for desegregation. The plan, which makes extensive use of busing and mandatory reassignments, desegregates elementary schools by "pairing" and "triading" predominantly minority with predominantly white attendance areas, and by basing student assignments on attendance zones rather than on race. The racial makeup of secondary schools is moderated by "feeding" them from the desegregated elementary schools. App. 142-143. The District represents that the plan results in the reassignment of roughly equal numbers of white and minority students, and allows most students to spend roughly half of their academic careers attending a school near their homes. Brief for Appellee Seattle School District No. 1, p. 5. 6 The desegregation program, implemented in the 1978-1979 academic year, apparently was effective: the District Court found that the Seattle Plan "has substantially reduced the number of racially imbalanced schools in the district and has substantially reduced the percentage of minority students in those schools which remain racially imbalanced." 473 F.Supp., at 1007. B 7 In late 1977, shortly before the Seattle Plan was formally adopted by the District, a number of Seattle residents who opposed the desegregation strategies being discussed by the School Board formed an organization called the Citizens for Voluntary Integration Committee (CiVIC). This organization, which the District Court found "was formed because of its founders' opposition to The Seattle Plan," ibid., attempted to enjoin implementation of the Board's mandatory desegregation program through litigation in state court; when these efforts failed, CiVIC drafted a statewide initiative designed to terminate the use of mandatory busing for purposes of racial integration.4 This proposal, known as Initiative 350, provided that "no school board . . . shall directly or indirectly require any student to attend a school other than the school which is geographically nearest or next nearest the student's place of residence . . . and which offers the course of study pursued by such student. . . ." See Wash.Rev.Code § 28A.26.010 (1981).5 The initiative then set out, however, a number of broad exceptions to this requirement: a student may be assigned beyond his neighborhood school if he "requires special education, care or guidance," or if "there are health or safety hazards, either natural or man made, or physical barriers or obstacles . . . between the student's place of residence and the nearest or next nearest school," or if "the school nearest or next nearest to his place of residence is unfit or inadequate because of overcrowding, unsafe conditions or lack of physical facilities." See ibid. Initiative 350 also specifically proscribed use of seven enumerated methods of "indirec[t]" student assignment—among them the redefinition of attendance zones, the pairing of schools, and the use of "feeder" schools—that are a part of the Seattle Plan. See § 28A.26.030. The initiative envisioned busing for racial purposes in only one circumstance: it did not purport to "prevent any court of competent jurisdiction from adjudicating constitutional issues relating to the public schools." See § 28A.26.060. 8 Its proponents placed Initiative 350 on the Washington ballot for the November 1978 general election. During the ensuing campaign, the District Court concluded, the leadership of CiVIC "acted legally and responsibly," and did not address "its appeals to the racial biases of the voters." 473 F.Supp., at 1009. At the same time, however, the court's findings demonstrate that the initiative was directed solely at desegregative busing in general, and at the Seattle Plan in particular. Thus, "[e]xcept for the assignment of students to effect racial balancing, the drafters of Initiative 350 attempted to preserve to school districts the maximum flexibility in the assignment of students," id., at 1008, and "[e]xcept for racially-balancing purposes" the initiative "permits local school districts to assign students other than to their nearest or next nearest schools for most, if not all, of the major reasons for which students are at present assigned to schools other than their nearest or next nearest schools." Id., at 1010.6 In campaigning for the measure, CiVIC officials accurately represented that its passage would result in "no loss of school district flexibility other than in busing for desegregation purposes," id., at 1008, and it is evident that the campaign focused almost exclusively on the wisdom of "forced busing" for integration. See id., at 1009. 9 On November 8, 1978, two months after the Seattle Plan went into effect, Initiative 350 passed by a substantial margin, drawing almost 66% of the vote statewide. The initiative failed to attract majority support in two state legislative districts, both in Seattle. In the city as a whole, however, the initiative passed with some 61% of the vote. Within the month, the District, together with the Tacoma and Pasco School Districts,7 initiated this suit against the State in the United States District Court for the Western District of Washington, challenging the constitutionality of Initiative 350 under the Equal Protection Clause of the Fourteenth Amendment. The United States and several community organizations intervened in support of the District;8 CiVIC intervened on behalf of the defendants. 10 After a 9-day trial, the District Court made extensive and detailed findings of fact. The court determined that "[t]hose Seattle schools which are most crowded are located in those areas of the city where the preponderance of minority families live." Id., at 1001. Yet the court found that Initiative 350, if implemented, "will prevent the racial balancing of a significant number of Seattle schools and will cause the school system to become more racially imbalanced than it presently is," "will make it impossible for Tacoma schools to maintain their present racial balance," and will make "doubtful" the prospects for integration of the Pasco schools. Id., at 1010; see id., at 1001, 1011. Except for desegregative busing, however, the court found that "almost all of the busing of students currently taking place in [Washington] is permitted by Initiative 350." Id., at 1010. And while the court found that "racial bias . . . is a factor in the opposition to the 'busing' of students to obtain racial balance," id., at 1001, it also found that voters were moved to support Initiative 350 for "a number of reasons," so that "[i]t is impossible to ascertain all of those reasons [o]r to determine the relative impact of those reasons upon the electorate." Id., at 1010. 11 The District Court then held Initiative 350 unconstitutional for three independent reasons. The court first concluded that the initiative established an impermissible racial classification in violation of Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616 (1969), and Lee v. Nyquist, 318 F.Supp. 710 (WDNY 1970) (three-judge court), summarily aff'd, 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105 (1971), "because it permits busing for non-racial reasons but forbids it for racial reasons." 473 F.Supp., at 1012. The court next held Initiative 350 invalid because "a racially discriminatory purpose was one of the factors which motivated the conception and adoption of the initiative." Id., at 1013.9 Finally, the District Court reasoned that Initiative 350 was unconstitutionally overbroad, because in the absence of a court order it barred even school boards that had engaged in de jure segregation from taking steps to foster integration.10 Id., at 1016. The court permanently enjoined implementation of the initiative's restrictions. 12 On the merits, a divided panel of the United States Court of Appeals for the Ninth Circuit affirmed, relying entirely on the District Court's first rationale. 633 F.2d 1338 (1980).11 By subjecting desegregative student assignments to unique treatment, the Court of Appeals concluded, Initiative 350 "both creates a constitutionally-suspect racial classification and radically restructures the political process of Washington by allowing a state-wide majority to usurp traditional local authority over local school board educational policies." Id., at 1344. In doing so, the court continued, the initiative "remove[s] from local school boards their existing authority, and in large part their capability, to enact programs designed to desegregate the schools." Id., at 1346 (emphasis in original and footnote omitted). The court found such a result contrary to the principles of Hunter v. Erickson, supra, and Lee v. Nyquist, supra. The court acknowledged that the issue would be a different one had a successor school board attempted to rescind the Seattle Plan. Here, however, "a different governmental body—the state-wide electorate—rescinded a policy voluntarily enacted by locally elected school boards already subject to local political control." 633 F.2d, at 1346.12 13 The State and various state officers appealed to this Court. We noted probable jurisdiction to address an issue of significance to our Nation's system of education. 454 U.S. 890, 102 S.Ct. 384, 70 L.Ed.2d 204 (1981). II 14 The Equal Protection Clause of the Fourteenth Amendment guarantees racial minorities the right to full participation in the political life of the community. It is beyond dispute, of course, that given racial or ethnic groups may not be denied the franchise, or precluded from entering into the political process in a reliable and meaningful manner. See White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973); Nixon v. Herndon, 273 U.S. 536, 47 S.Ct. 446, 71 L.Ed. 759 (1927). But the Fourteenth Amendment also reaches "a political structure that treats all individuals as equals," Mobile v. Bolden, 446 U.S. 55, 84, 100 S.Ct. 1490, 1509, 64 L.Ed.2d 47 (1980) (STEVENS, J., concurring in judgment), yet more subtly distorts governmental processes in such a way as to place special burdens on the ability of minority groups to achieve beneficial legislation. 15 This principle received its clearest expression in Hunter v. Erickson, supra, a case that involved attempts to overturn antidiscrimination legislation in Akron, Ohio. The Akron City Council, pursuant to its ordinary legislative processes, had enacted a fair housing ordinance. In response, the local citizenry, using an established referendum procedure, see 393 U.S., at 390, and n. 6; at id., 393-394, and n., 89 S.Ct., at 560, and n. 6; 562, and n. (Harlan, J., concurring), amended the city charter to provide that ordinances regulating real estate transactions " 'on the basis of race, color, religion, national origin or ancestry must first be approved by a majority of the electors voting on the question at a regular or general election before said ordinance shall be effective.' " Id., at 387, 89 S.Ct., at 558. This action "not only suspended the operation of the existing ordinance forbidding housing discrimination, but also required the approval of the electors before any future [fair housing] ordinance could take effect." Id., at 389-390, 89 S.Ct., at 559-560. In essence, the amendment changed the requirements for the adoption of one type of local legislation: to enact an ordinance barring housing discrimination on the basis of race or religion, proponents had to obtain the approval of the City Council and of a majority of the voters citywide. To enact an ordinance preventing housing discrimination on other grounds, or to enact any other type of housing ordinance, proponents needed the support of only the City Council. 16 In striking down the charter amendment, the Hunter Court recognized that, on its face, the provision "draws no distinctions among racial and religious groups." Id., at 390, 89 S.Ct., at 560. But it did differentiate "between those groups who sought the law's protection against racial . . . discriminatio[n] in the sale and rental of real estate and those who sought to regulate real property transactions in the pursuit of other ends," ibid., thus "disadvantag[ing] those who would benefit from laws barring racial . . . discriminatio[n] as against those who would bar other discriminations or who would otherwise regulate the real estate market in their favor." Id., at 391, 89 S.Ct., at 560. In "reality," the burden imposed by such an arrangement necessarily "falls on the minority. The majority needs no protection against discrimination and if it did, a referendum might be bothersome but no more than that." Ibid. In effect, then, the charter amendment served as an "explicitly racial classification treating racial housing matters differently from other racial and housing matters." Id., at 389, 89 S.Ct., at 559. This made the amendment constitutionally suspect: "the State may no more disadvantage any particular group by making it more difficult to enact legislation in its behalf than it may dilute any person's vote or give any group a smaller representation than another of comparable size." Id., at 393, 89 S.Ct., at 562 (emphasis added). 17 Lee v. Nyquist, 318 F.Supp. 710 (WDNY 1970) (three-judge court), offers an application of the Hunter doctrine in a setting strikingly similar to the one now before us. That case involved the New York education system, which made use of both elected and appointed school boards and which conferred extensive authority on state education officials. In an effort to eliminate de facto segregation in New York's schools, those officials had directed the city of Buffalo—a municipality with an appointed school board to implement an integration plan. While these developments were proceeding, however, the New York Legislature enacted a statute barring state education officials and appointed—though not elected school boards from "assign[ing] or compell[ing] [students] to attend any school on account of race . . . or for the purpose of achieving [racial] equality in attendance . . . at any school." Id., at 712.13 18 Applying Hunter, the three-judge District Court invalidated the statute, noting that under the provision "[t]he Commissioner [of Education] and local appointed officials are prohibited from acting in [student assignment] matters only where racial criteria are involved." Id., at 719. In the court's view, the statute therefore "place[d] burdens on the implementation of educational policies designed to deal with race on the local level" by "treating educational matters involving racial criteria differently from other educational matters and making it more difficult to deal with racial imbalance in the public schools." Ibid. (emphasis in original). This drew an impermissible distinction "between the treatment of problems involving racial matters and that afforded other problems in the same area." Id., at 718. This Court affirmed the District Court's judgment without opinion. 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105 (1971). 19 These cases yield a simple but central principle. As Justice Harlan noted while concurring in the Court's opinion in Hunter, laws structuring political institutions or allocating political power according to "neutral principles"—such as the executive veto, or the typically burdensome requirements for amending state constitutions—are not subject to equal protection attack, though they may "make it more difficult for minorities to achieve favorable legislation." 393 U.S., at 394, 89 S.Ct., at 562. Because such laws make it more difficult for every group in the community to enact comparable laws, they "provid[e] a just framework within which the diverse political groups in our society may fairly compete." Id., at 393, 89 S.Ct., at 562. Thus, the political majority may generally restructure the political process to place obstacles in the path of everyone seeking to secure the benefits of governmental action. But a different analysis is required when the State allocates governmental power nonneutrally, by explicitly using the racial nature of a decision to determine the decisionmaking process. State action of this kind, the Court said, "places special burdens on racial minorities within the governmental process," id., at 391, 89 S.Ct., at 560 (emphasis added), thereby "making it more difficult for certain racial and religious minorities [than for other members of the community] to achieve legislation that is in their interest." Id., at 395, 89 S.Ct., at 563 (emphasis added) (Harlan, J., concurring). Such a structuring of the political process, the Court said, was "no more permissible than [is] denying [members of a racial minority] the vote, on an equal basis with others." Id., at 391, 89 S.Ct., at 560. III 20 We believe that the Court of Appeals properly focused on Hunter and Lee, for we find the principle of those cases dispositive of the issue here. In our view, Initiative 350 must fall because it does "not attemp[t] to allocate governmental power on the basis of any general principle." Hunter v. Erickson, 393 U.S., at 395, 89 S.Ct., at 563 (Harlan, J., concurring). Instead, it uses the racial nature of an issue to define the governmental decisionmaking structure, and thus imposes substantial and unique burdens on racial minorities. 21 * Noting that Initiative 350 nowhere mentions "race" or "integration," appellants suggest that the legislation has no racial overtones; they maintain that Hunter is inapposite because the initiative simply permits busing for certain enumerated purposes while neutrally forbidding it for all other reasons. We find it difficult to believe that appellants' analysis is seriously advanced, however, for despite its facial neutrality there is little doubt that the initiative was effectively drawn for racial purposes. Neither the initiative's sponsors, nor the District Court, nor the Court of Appeals had any difficulty perceiving the racial nature of the issue settled by Initiative 350. Thus, the District Court found that the text of the initiative was carefully tailored to interfere only with desegregative busing.14 Proponents of the initiative candidly "represented that there would be no loss of school district flexibility other than in busing for desegregation purposes." 473 F.Supp., at 1008. And, as we have noted, Initiative 350 in fact allows school districts to bus their students "for most, if not all," of the nonintegrative purposes required by their educational policies. Id., at 1010. The Washington electorate surely was aware of this, for it was "assured" by CiVIC officials that " '99% of the school districts in the state' "—those that lacked mandatory integration programs—"would not be affected by the passage of 350." Id., at 1008-1009. It is beyond reasonable dispute, then, that the initiative was enacted " 'because of,' not merely 'in spite of,' its adverse effects upon" busing for integration. Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 279, 99 S.Ct. 2282, 2296, 60 L.Ed.2d 870 (1979). 22 Even accepting the view that Initiative 350 was enacted for such a purpose, the United States—which has changed its position during the course of this litigation, and now supports the State maintains that busing for integration, unlike the fair housing ordinance involved in Hunter, is not a peculiarly "racial" issue at all. Brief for United States 17, n. 18. Again, we are not persuaded. It undoubtedly is true, as the United States suggests, that the proponents of mandatory integration cannot be classified by race: Negroes and whites may be counted among both the supporters and the opponents of Initiative 350. And it should be equally clear that white as well as Negro children benefit from exposure to "ethnic and racial diversity in the classroom." Columbus Board of Education v. Penick, 443 U.S. 449, 486, 99 S.Ct. 2941, 2991, 61 L.Ed.2d 666 (1979) (POWELL, J., dissenting). See Milliken v. Bradley, 418 U.S. 717, 783, 94 S.Ct. 3112, 3146, 41 L.Ed.2d 1069 (1974) (MARSHALL, J., dissenting).15 But neither of these factors serves to distinguish Hunter, for we may fairly assume that members of the racial majority both favored and benefited from Akron's fair housing ordinance. Cf. Havens Realty Corp. v. Coleman, 455 U.S. 363, 376-377, and n. 17, 102 S.Ct. 1114, 1122-1123, and n. 17, 71 L.Ed.2d 214 (1982); Gladstone, Realtors v. Village of Bellwood, 441 U.S. 91, 111, 115, 99 S.Ct. 1601, 1614, 1615, 60 L.Ed. 66 (1979). 23 In any event, our cases suggest that desegregation of the public schools, like the Akron open housing ordinance, at bottom inures primarily to the benefit of the minority, and is designed for that purpose. Education has come to be "a principal instrument in awakening the child to cultural values, in preparing him for later professional training, and in helping him to adjust normally to his environment." Brown v. Board of Education, 347 U.S. 483, 493, 74 S.Ct. 686, 691, 98 L.Ed. 873 (1954). When that environment is largely shaped by members of different racial and cultural groups, minority children can achieve their full measure of success only if they learn to function in—and are fully accepted by—the larger community. Attending an ethnically diverse school may help accomplish this goal by preparing minority children "for citizenship in our pluralistic society," Estes v. Metropolitan Branches of Dallas NAACP, 444 U.S. 437, 451, 100 S.Ct. 716, 723, 62 L.Ed.2d 626 (1980) (POWELL, J., dissenting), while, we may hope, teaching members of the racial majority "to live in harmony and mutual respect" with children of minority heritage. Columbus Board of Education v. Penick, 443 U.S., at 485, n. 5, 99 S.Ct., at 2946, n. 5 (POWELL, J., dissenting). Lee v. Nyquist settles this point, for the Court there accepted the proposition that mandatory desegregation strategies present the type of racial issue implicated by the Hunter doctrine.16 24 It is undeniable that busing for integration—particularly when ordered by a federal court—now engenders considerably more controversy than does the sort of fair housing ordinance debated in Hunter. See Estes v. Metropolitan Branches of Dallas NAACP, 444 U.S., at 448-451, 100 S.Ct., at 722-723 (POWELL, J., dissenting). But in the absence of a constitutional violation, the desirability and efficacy of school desegregation are matters to be resolved through the political process. For present purposes, it is enough that minorities may consider busing for integration to be "legislation that is in their interest." Hunter v. Erickson, 393 U.S., at 395, 89 S.Ct., at 563 (Harlan, J., concurring). Given the racial focus of Initiative 350, this suffices to trigger application of the Hunter doctrine. B 25 We are also satisfied that the practical effect of Initiative 350 is to work a reallocation of power of the kind condemned in Hunter. The initiative removes the authority to address a racial problem—and only a racial problem—from the existing decisionmaking body, in such a way as to burden minority interests. Those favoring the elimination of de facto school segregation now must seek relief from the state legislature, or from the statewide electorate. Yet authority over all other student assignment decisions, as well as over most other areas of educational policy, remains vested in the local school board. Indeed, by specifically exempting from Initiative 350's proscriptions most nonracial reasons for assigning students away from their neighborhood schools, the initiative expressly requires those championing school integration to surmount a considerably higher hurdle than persons seeking comparable legislative action. As in Hunter, then, the community's political mechanisms are modified to place effective decisionmaking authority over a racial issue at a different level of government.17 In a very obvious sense, the initiative thus "disadvantages those who would benefit from laws barring" de facto desegregation "as against those who . . . would otherwise regulate" student assignment decisions; "the reality is that the law's impact falls on the minority." Hunter v. Erickson, 393 U.S., at 391, 89 S.Ct., at 560. 26 The state appellants and the United States, in response to this line of analysis, argue that Initiative 350 has not worked any reallocation of power. They note that the State necessarily retains plenary authority over Washington's system of education, and therefore they suggest that the initiative amounts to nothing more than an unexceptional example of a State's intervention in its own school system. In effect, they maintain that the State functions as a "super school board," Tr. of Oral Arg. 5, 17, which typically involves itself in all areas of educational policy. And, the argument continues, if the State is the body that usually makes decisions in this area, Initiative 350 worked a simple change in policy rather than a forbidden reallocation of power. Cf. Crawford v. Los Angeles Board of Education, 458 U.S. 527, 102 S.Ct. 3211, 73 L.Ed.2d 948. 27 This at first glance would seem to be a potent argument, for States traditionally have been accorded the widest latitude in ordering their internal governmental processes, see Holt Civic Club v. Tuscaloosa, 439 U.S. 60, 71, 99 S.Ct. 383, 390, 58 L.Ed.2d 292 (1978), and school boards, as creatures of the State, obviously must give effect to policies announced by the state legislature. But "insisting that a State may distribute legislative power as it desires . . . furnish[es] no justification for a legislative structure which otherwise would violate the Fourteenth Amendment. Nor does the implementation of this change through popular referendum immunize it." Hunter v. Erickson, 393 U.S., at 392, 89 S.Ct., at 561. The issue here, after all, is not whether Washington has the authority to intervene in the affairs of local school boards; it is, rather, whether the State has exercised that authority in a manner consistent with the Equal Protection Clause. As the Court noted in Hunter: "[T]hough Akron might have proceeded by majority vote . . . on all its municipal legislation, it has instead chosen a more complex system. Having done so, the State may no more disadvantage any particular group by making it more difficult to enact legislation in its behalf than it may dilute any person's vote." Id., at 392-393, 89 S.Ct., at 561-562.18 Washington also has chosen to make use of a more complex governmental structure, and a close examination both of the Washington statutes and of the Court's decisions in related areas convinces us that Hunter is fully applicable here. 28 At the outset, it is irrelevant that the State might have vested all decisionmaking authority in itself, so long as the political structure it in fact erected imposes comparative burdens on minority interests; that much is settled by Hunter and by Lee.19 And until the passage of Initiative 350, Washington law in fact had established the local school board, rather than the State, as the entity charged with making decisions of the type at issue here. Like all 50 States, see Brief for National School Boards Assn. as Amicus Curiae 11, 14-16, Washington of course is ultimately responsible for providing education within its borders, see Wash.Const., Art. IX; Wash.Rev.Code § 28A.02.010 (1981); ch. 28A.41 (establishing a uniform school financing system); Seattle School District No. 1 v. State, 90 Wash.2d 476, 585 P.2d 71 (1978), and it therefore has set certain procedural requirements and minimum educational standards to be met by each school. See, e.g., §§ 28A.01.010, 28A.01.020 (length of school day and year); ch. 28A.27 (mandatory attendance); ch. 28A.67 (teacher qualifications); ch. 28A.05 and §§ 28A.58.750-28A.58.754 (curriculum). But Washington has chosen to meet its educational responsibilities primarily through "state and local officials, boards, and committees," § 28A.02.020, and the responsibility to devise and tailor educational programs to suit local needs has emphatically been vested in the local school boards. 29 Thus "each common school district board of directors" is made "accountable for the proper operation of [its] district to the local community and its electorate." § 28A.58.758(1). To this end, each school board is "vested with the final responsibility for the setting of policies ensuring quality in the content and extent of its educational program" (emphasis added). Ibid. School boards are given responsibility for, among many other things, "[e]stablish[ing] performance criteria" for personnel and programs, for assigning staff "according to board enumerated classroom and program needs," for setting requirements concerning hours of instruction, for establishing curriculum standards "relevant to the particular needs of district students or the unusual characteristics of the district," and for evaluating teaching materials. § 28A.58.758(2). School boards are generally directed to "develop a program identifying student learning objectives for their district[s]," § 28A.58.090; see also § 28A.58.092, to select instructional materials, § 28A.58.103, to stock libraries as they deem necessary, § 28A.58.104, and to initiate a variety of optional programs. See, e.g., §§ 28A.34.010, 28A.35.010, 28A.58.105. School boards, of course, are given broad corporate powers. §§ 28A.58.010, 28A.58.075, 28A.59.180. Significantly for present purposes, school boards are directed to determine which students should be bused to school and to provide those students with transportation. § 28A.24.055. 30 Indeed, the notion of school board responsibility for local educational programs is so firmly rooted that local boards are subject to disclosure and reporting provisions specifically designed to ensure the board's "accountability" to the people of the community for "the educational programs in the school distric[t]." § 28A.58.758(3). And, perhaps most relevant here, before the adoption of Initiative 350 the Washington Supreme Court had found it within the general discretion of local school authorities to settle problems related to the denial of "equal educational opportunity."20 Citizens Against Mandatory Bussing v. Palmason, 80 Wash.2d 445, 453, 495 P.2d 657, 663 (1972). It therefore had squarely held that a program of desegregative busing was a proper means of furthering the school board's responsibility to "administe[r] the schools in such a way as to provide a sound education for all children." Id., at 456, 495 P.2d, at 664.21 See State ex rel. Citizens Against Mandatory Bussing v. Brooks, 80 Wash.2d 121, 492 P.2d 536 (1972); State ex rel. Lukens v. Spokane School District No. 81, 147 Wash. 467, 474, 266 P. 189, 191 (1928).22 31 Given this statutory structure, we have little difficulty concluding that Initiative 350 worked a major reordering of the State's educational decisionmaking process. Before adoption of the initiative, the power to determine what programs would most appropriately fill a school district's educational needs—including programs involving student assignment and desegregation—was firmly committed to the local board's discretion. The question whether to provide an integrated learning environment rather than a system of neighborhood schools surely involved a decision of that sort. See Citizens Against Mandatory Bussing v. Palmason, 80 Wash.2d, at 459-460, 495 P.2d, at 666-667. After passage of Initiative 350, authority over all but one of those areas remained in the hands of the local board. By placing power over desegregative busing at the state level, then, Initiative 350 plainly "differentiates between the treatment of problems involving racial matters and that afforded other problems in the same area." Lee v. Nyquist, 318 F.Supp., at 718.23 The District Court and the Court of Appeals similarly concluded that the initiative restructured the Washington political process, and we see no reason to challenge the determinations of courts familiar with local law. Cf. Milliken v. Bradley, 418 U.S., at 769, 94 S.Ct., at 3139 (WHITE, J., dissenting). 32 That we reach this conclusion should come as no surprise, for when faced with a similar educational scheme in Milliken v. Bradley, supra,24 the Court concluded that the actions of a local school board could not be attributed to the State that had created it. We there addressed the Michigan education system, which vests in the State constitutional responsibility for providing education: " 'The policy of [Michigan] has been to retain control of its school system, to be administered throughout the State under State laws by local State agencies . . . to carry out the delegated functions given [them] by the legislature.' " Milliken v. Bradley, 418 U.S., at 794, 94 S.Ct., at 3151 (MARSHALL, J., dissenting), quoting School District of City of Lansing v. State Board of Education, 367 Mich. 591, 595, 116 N.W.2d 866, 868 (1962). See Milliken v. Bradley, 418 U.S., at 726, n. 5, 94 S.Ct., at 3118, n. 5. To fulfill this responsibility, the State of Michigan provided a substantial measure of school district funding, established standards for teacher certification, determined part of the curriculum, set a minimum school term, approved bus routes and textbooks, established disciplinary procedures, and under certain circumstances had the power even to remove local school board members. See id., at 795-796, 94 S.Ct., at 3151-3152 (MARSHALL, J., dissenting). See also id., at 726, n. 5, 727, 94 S.Ct., at 3118, n. 5, 3118 (describing state controls over education); id., at 768, and n. 4, 94 S.Ct., at 3139, and n. 4 (WHITE, J., dissenting) (same); id., at 794, 94 S.Ct., at 3151 (MARSHALL, J., dissenting) (same). 33 Yet the Court, noting that "[n]o single tradition in public education is more deeply rooted than local control over the operation of schools," concluded that the "Michigan educational structure . . . in common with most States, provides for a large measure of local control." Id., at 741-742, 94 S.Ct., at 3125-3126. Relying on this analysis, the Court determined that a Michigan school board's assignment policies could not be attributed to the State, and therefore declined to permit interdistrict busing as a remedy for one school district's acts of unconstitutional segregation. If local school boards operating under a similar statutory structure are considered separate entities for purposes of constitutional adjudication when they make segregative assignment decisions, it is difficult to see why a different analysis should apply when a local board's desegregative policy is at issue. 34 In any event, we believe that the question here is again settled by Lee. There, state control of the educational system was fully as complete as it now is in Washington. See generally N.Y.Educ.Law §§ 305, 306, 308-310 (McKinney 1969 and Supp.1981). The state statute under attack reallocated power over mandatory desegregation in two ways: it transferred authority from the State Commissioner of Education to local elected school boards, and it shifted authority from local appointed school boards to the state legislature.25 When presented with this restructuring of the political process, the District Court declared that it could "conceive of no more compelling case for the application of the Hunter principle." 318 F.Supp., at 719. This Court of course affirmed the District Court's judgment. We see no relevant distinction between this case and Lee ; indeed, it is difficult to imagine a more precise parallel.26 C 35 To be sure, "the simple repeal or modification of desegregation or antidiscrimination laws, without more, never has been viewed as embodying a presumptively invalid racial classification." Crawford v. Los Angeles Board of Education, 458 U.S. 527, 539, 102 S.Ct. 3211, 3218, 73 L.Ed.2d 948. See Dayton Board of Education v. Brinkman, 443 U.S. 526, 531, n. 5, 99 S.Ct. 2971, 2976, n. 5, 61 L.Ed.2d 720 (1979); Hunter v. Erickson, 393 U.S., at 390, n. 5, 89 S.Ct., at 560, n. 5. As Justice Harlan noted in Hunter, the voters of the polity may express their displeasure through an established legislative or referendum procedure when particular legislation "arouses passionate opposition." Id., at 395, 89 S.Ct., at 563 (concurring opinion). Had Akron's fair housing ordinance been defeated at a referendum, for example, "Negroes would undoubtedly [have lost] an important political battle, but they would not thereby [have been] denied equal protection." Id., at 394, 89 S.Ct., at 562. 36 Initiative 350, however, works something more than the "mere repeal" of a desegregation law by the political entity that created it. It burdens all future attempts to integrate Washington schools in districts throughout the State, by lodging decisionmaking authority over the question at a new and remote level of government. Indeed, the initiative, like the charter amendment at issue in Hunter, has its most pernicious effect on integration programs that do "not arouse extraordinary controversy." Id., at 396, 89 S.Ct., at 563 (emphasis in original). In such situations the initiative makes the enactment of racially beneficial legislation difficult, though the particular program involved might not have inspired opposition had it been promulgated through the usual legislative processes used for comparable legislation.27 This imposes direct and undeniable burdens on minority interests. "If a governmental institution is to be fair, one group cannot always be expected to win," id., at 394, 89 S.Ct., at 562; by the same token, one group cannot be subjected to a debilitating and often insurmountable disadvantage. IV 37 In the end, appellants are reduced to suggesting that Hunter has been effectively overruled by more recent decisions of this Court. As they read it, Hunter applied a simple "disparate impact" analysis: it invalidated a facially neutral ordinance because of the law's adverse effects upon racial minorities. Appellants therefore contend that Hunter was swept away, along with the disparate-impact approach to equal protection, in Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976), and Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977). Cf. James v. Valtierra, 402 U.S. 137, 91 S.Ct. 1331, 28 L.Ed.2d 678 (1971). 38 Appellants unquestionably are correct when they suggest that "purposeful discrimination is 'the condition that offends the Constitution,' " Personnel Administrator of Massachusetts v. Feeney, 442 U.S., at 274, 99 S.Ct., at 2293, quoting Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 16, 91 S.Ct. 1267, 1276, 28 L.Ed.2d 554 (1971), for the "central purpose of the Equal Protection Clause . . . is the prevention of official conduct discriminating on the basis of race." Washington v. Davis, 426 U.S., at 239, 96 S.Ct., at 2047. Thus, when facially neutral legislation is subjected to equal protection attack, an inquiry into intent is necessary to determine whether the legislation in some sense was designed to accord disparate treatment on the basis of racial considerations. Appellants' suggestion that this analysis somehow conflicts with Hunter, however, misapprehends the basis of the Hunter doctrine. We have not insisted on a particularized inquiry into motivation in all equal protection cases: "A racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification." Personnel Administrator of Massachusetts v. Feeney, 442 U.S., at 272, 99 S.Ct., at 2292. And legislation of the kind challenged in Hunter similarly falls into an inherently suspect category.28 39 There is one immediate and crucial difference between Hunter and the cases cited by appellants. While decisions such as Washington v. Davis and Arlington Heights considered classifications facially unrelated to race, the charter amendment at issue in Hunter dealt in explicitly racial terms with legislation designed to benefit minorities "as minorities," not legislation intended to benefit some larger group of underprivileged citizens among whom minorities were disproportionately represented. This does not mean, of course, that every attempt to address a racial issue gives rise to an impermissible racial classification. See Crawford v. Los Angeles Board of Education, 458 U.S. 527, 102 S.Ct. 3211, 73 L.Ed.2d 948. But when the political process or the decisionmaking mechanism used to address racially conscious legislation—and only such legislation—is singled out for peculiar and disadvantageous treatment, the governmental action plainly "rests on 'distinctions based on race.' "29 James v. Valtierra, 402 U.S., at 141, 91 S.Ct., at 1333, quoting Hunter v. Erickson, 393 U.S., at 391, 89 S.Ct., at 560. And when the State's allocation of power places unusual burdens on the ability of racial groups to enact legislation specifically designed to overcome the "special condition" of prejudice, the governmental action seriously "curtail[s] the operation of those political processes ordinarily to be relied upon to protect minorities." United States v. Carolene Products Co., 304 U.S. 144, 153, n. 4, 58 S.Ct. 778, 783-784, n. 4, 82 L.Ed. 1234 (1938). In a most direct sense, this implicates the judiciary's special role in safeguarding the interests of those groups that are "relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process." San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 28, 93 S.Ct. 1278, 1296, 36 L.Ed.2d 16 (1973).30 40 Hunter recognized the considerations addressed above, and it therefore rested on a principle that has been vital for over a century—that "the core of the Fourteenth Amendment is the prevention of meaningful and unjustified official distinctions based on race." 393 U.S., at 391, 89 S.Ct., at 560. Just such distinctions infected the reallocation of decisionmaking authority considered in Hunter, for minorities are no less powerless with the vote than without it when a racial criterion is used to assign governmental power in such a way as to exclude particular racial groups "from effective participation in the political proces[s]." Mobile v. Bolden, 446 U.S., at 94, 100 S.Ct., at 1514 (WHITE, J., dissenting). Certainly, a state requirement that "desegregation or antidiscrimination laws," Crawford v. Los Angeles Board of Education, 458 U.S., at 539, 102 S.Ct., at 3218, and only such laws, be passed by unanimous vote of the legislature would be constitutionally suspect. It would be equally questionable for a community to require that laws or ordinances "designed to ameliorate race relations or to protect racial minorities," ibid., be confirmed by popular vote of the electorate as a whole, while comparable legislation is exempted from a similar procedure. The amendment addressed in Hunter —and, as we have explained, the legislation at issue here—was less obviously pernicious than are these examples, but was no different in principle. V 41 In reaching this conclusion, we do not undervalue the magnitude of the State's interest in its system of education. Washington could have reserved to state officials the right to make all decisions in the areas of education and student assignment. It has chosen, however, to use a more elaborate system; having done so, the State is obligated to operate that system within the confines of the Fourteenth Amendment. That, we believe, it has failed to do.31 42 Accordingly, the judgment of the Court of Appeals is 43 Affirmed. Justice POWELL, with whom THE CHIEF JUSTICE, Justice REHNQUIST, and Justice O'CONNOR join, dissenting. 44 The people of the State of Washington, by a two-to-one vote, have adopted a neighborhood school policy. The policy is binding on local school districts but in no way affects the authority of state or federal courts to order school transportation to remedy violations of the Fourteenth Amendment. Nor does the policy affect the power of local school districts to establish voluntary transfer programs for racial integration or for any other purpose. 45 In the absence of a constitutional violation, no decision of this Court compels a school district to adopt or maintain a mandatory busing program for racial integration.1 Accordingly, the Court does not hold that the adoption of a neighborhood school policy by local school districts would be unconstitutional. Rather, it holds that the adoption of such a policy at the state level—rather than at the local level—violates the Equal Protection Clause of the Fourteenth Amendment. 46 I dissent from the Court's unprecedented intrusion into the structure of a state government. The School Districts in this case were under no federal constitutional obligation to adopt mandatory busing programs. The State of Washington, the governmental body ultimately responsible for the provision of public education, has determined that certain mandatory busing programs are detrimental to the education of its children. "[T]he Fourteenth Amendment leaves the States free to distribute the powers of government as they will between their legislative and judicial branches." Hughes v. Superior Court, 339 U.S. 460, 467, 70 S.Ct. 718, 722, 94 L.Ed. 985 (1950). In my view, that Amendment leaves the States equally free to decide matters of concern to the State at the state, rather than local, level of government. 47 * At the November 1978 general election, the voters of the State adopted Initiative 350 by a two-to-one majority.2 The Initiative sets forth a neighborhood school policy binding on local school districts. It establishes a general rule prohibiting school districts from "directly or indirectly requir[ing] any student to attend a school other than the school which is geographically nearest or next nearest the student's place of residence." Wash.Rev.Code § 28A.26.010 (1981). The rule may be avoided in individual instances only if the student requires special education; if there are health or safety hazards between the student's residence and the nearest or next nearest school; or if the nearby schools are overcrowded, unsafe, or lacking in physical facilities. Ibid. 48 The Initiative includes two significant limitations upon the scope of its neighborhood school policy. It expressly provides that nothing in the Initiative shall "preclude the establishment of schools offering specialized or enriched educational programs which students may voluntarily choose to attend, or of any other voluntary option offered to students." § 28A.26.050. Moreover, and critical to this case, the authority of state and federal courts to order mandatory school assignments to remedy constitutional violations is left untouched by the Initiative: "This chapter shall not prevent any court of competent jurisdiction from adjudicating constitutional issues relating to the public schools." § 28A.26.060.3 49 This suit was filed in United States District Court shortly after the Initiative was enacted. The Seattle School District, joined by the Tacoma and Pasco School Districts4 and certain individual plaintiffs, argued that the Initiative violated the Equal Protection Clause of the Fourteenth Amendment. The District Court agreed, and, in a split decision, the Court of Appeals affirmed. Relying on Hunter v. Erickson, 393 U.S. 385, 89 S.Ct. 557, 21 L.Ed.2d 616 (1969), the Court of Appeals concluded that Initiative 350 "both creates a constitutionally-suspect racial classification and radically restructures the political process of Washington by allowing a state-wide majority to usurp traditional local authority over local school board educational policies." 633 F.2d 1338, 1344 (CA9 1980).5 II 50 The principles that should guide us in reviewing the constitutionality of Initiative 350 are well established. To begin with, we have never held, or even intimated, that absent a federal constitutional violation, a State must choose to treat persons differently on the basis of race. In the absence of a federal constitutional violation requiring race-specific remedies, a policy of strict racial neutrality by a State would violate no federal constitutional principle. Cf. University of California Regents v. Bakke, 438 U.S. 265, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978). 51 In particular, a neighborhood school policy and a decision not to assign students on the basis of their race, does not offend the Fourteenth Amendment.6 The Court has never held that there is an affirmative duty to integrate the schools in the absence of a finding of unconstitutional segregation. See Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 24, 91 S.Ct. 1267, 1280, 28 L.Ed.2d 554 (1971); Dayton Board of Education v. Brinkman, 433 U.S. 406, 417, 97 S.Ct. 2766, 2774, 53 L.Ed.2d 851 (1977). Certainly there is no constitutional duty to adopt mandatory busing in the absence of such a violation. Indeed, even where desegregation is ordered because of a constitutional violation, the Court has never held that racial balance itself is a constitutional requirement. Ibid. And even where there have been segregated schools, once desegregation has been accomplished no further constitutional duty exists upon school boards or States to maintain integration. See Pasadena City Board of Education v. Spangler, 427 U.S. 424, 96 S.Ct. 2697, 49 L.Ed.2d 599 (1976). 52 Moreover, it is a well-established principle that the States have "extraordinarily wide latitude . . . in creating various types of political subdivisions and conferring authority upon them." Holt Civic Club v. Tuscaloosa, 439 U.S. 60, 71, 99 S.Ct. 383, 390, 58 L.Ed.2d 292 (1978).7 The Constitution does not dictate to the States a particular division of authority between legislature and judiciary or between state and local governing bodies. It does not define institutions of local government. 53 Thus, a State may choose to run its schools from the state legislature or through local school boards just as it may choose to address the matter of race relations at the state or local level. There is no constitutional requirement that the State establish or maintain local institutions of government or that it delegate particular powers to these bodies. The only relevant constitutional limitation on a State's freedom to order its political institutions is that it may not do so in a fashion designed to "plac[e] special burdens on racial minorities within the governmental process." Hunter v. Erickson, 393 U.S., at 391, 89 S.Ct., at 560-61 (emphasis added). 54 In sum, in the absence of a prior constitutional violation, the States are under no constitutional duty to adopt integration programs in their schools, and certainly they are under no duty to establish a regime of mandatory busing. Nor does the Federal Constitution require that particular decisions concerning the schools or any other matter be made on the local as opposed to the state level. It does not require the States to establish local governmental bodies or to delegate unreviewable authority to them. III 55 Application of these settled principles demonstrates the serious error of today's decision—an error that cuts deeply into the heretofore unquestioned right of a State to structure the decisionmaking authority of its government. In this case, by Initiative 350, the State has adopted a policy of racial neutrality in student assignments. The policy in no way interferes with the power of state or federal courts to remedy constitutional violations. And if such a policy had been adopted by any of the School Districts in this litigation there could have been no question that the policy was constitutional.8 56 The issue here arises only because the Seattle School District—in the absence of a then-established state policy—chose to adopt race specific school assignments with extensive busing. It is not questioned that the District itself, at any time thereafter, could have changed its mind and canceled its integration program without violating the Federal Constitution. Yet this Court holds that neither the legislature nor the people of the State of Washington could alter what the District had decided. 57 The Court argues that the people of Washington by Initiative 350 created a racial classification, and yet must agree that identical action by the Seattle School District itself would have created no such classification. This is not an easy argument to answer because it seems to make no sense. School boards are the creation of supreme state authority, whether in a State Constitution or by legislative enactment. Until today's decision no one would have questioned the authority of a State to abolish school boards altogether, or to require that they conform to any lawful state policy. And in the State of Washington, a neighborhood school policy would have been lawful. 58 Under today's decision this heretofore undoubted supreme authority of a State's electorate is to be curtailed whenever a school board—or indeed any other state board or local instrumentality—adopts a race-specific program that arguably benefits racial minorities. Once such a program is adopted, only the local or subordinate entity that approved it will have authority to change it. The Court offers no authority or relevant explanation for this extraordinary subordination of the ultimate sovereign power of a State to act with respect to racial matters by subordinate bodies. It is a strange notion—alien to our system that local governmental bodies can forever preempt the ability of a State—the sovereign power—to address a matter of compelling concern to the State. The Constitution of the United States does not require such a bizarre result. 59 This is certainly not a case where a State—in moving to change a locally adopted policy—has established a racially discriminatory requirement. Initiative 350 does not impede enforcement of the Fourteenth Amendment. If a Washington school district should be found to have established a segregated school system, Initiative 350 will place no barrier in the way of a remedial busing order. Nor does Initiative 350 authorize or approve segregation in any form or degree. It is neutral on its face, and racially neutral as public policy. Children of all races benefit from neighborhood schooling, just as children of all races benefit from exposure to " 'ethnic and racial diversity in the classroom.' " Ante, at 472, quoting Columbus Board of Education v. Penick, 443 U.S. 449, 486, 99 S.Ct. 2941, 2991, 61 L.Ed.2d 666 (1979) (POWELL, J., dissenting).9 60 Finally, Initiative 350 places no "special burdens on racial minorities within the governmental process," Hunter v. Erickson, supra, 393 U.S., at 391, 89 S.Ct., at 560-61, such that interference with the State's distribution of authority is justified. Initiative 350 is simply a reflection of the State's political process at work. It does not alter that process in any respect. It does not require, for example, that all matters dealing with race—or with integration in the schools—must henceforth be submitted to a referendum of the people. Cf. Hunter v. Erickson, supra. The State has done no more than precisely what the Court has said that it should do: It has "resolved through the political process" the "desirability and efficacy of [mandatory] school desegregation" where there has been no unlawful segregation. Ante, at 474. 61 The political process in Washington, as in other States, permits persons who are dissatisfied at a local level to appeal to the state legislature or the people of the State for redress. It permits the people of a State to pre-empt local policies, and to formulate new programs and regulations. Such a process is inherent in the continued sovereignty of the States. This is our system. Any time a State chooses to address a major issue some persons or groups may be disadvantaged. In a democratic system there are winners and losers. But there is no inherent unfairness in this and certainly no constitutional violation.10 IV 62 Nonetheless, the Court holds that Initiative 350 "imposes substantial and unique burdens on racial minorities" in the governmental process. See ante, at 470. Its authority for this holding is said to be Hunter v. Erickson, supra.11 In Hunter the people of Akron passed a charter amendment that "not only suspended the operation of the existing ordinance forbidding housing discrimination, but also required the approval of the electors before any future [antidiscrimination] ordinance could take effect." 393 U.S., at 389-390, 89 S.Ct., at 559-60. Although the charter amendment was facially neutral, the Court found that it could be said to embody a racial classification: "[T]he reality is that the law's impact falls on the minority. The majority needs no protection against discrimination." Id., at 391, 89 S.Ct., at 560. By making it more difficult to pass legislation in favor of racial minorities, the amendment placed "special burdens on racial minorities within the governmental process." Ibid. 63 Nothing in Hunter supports the Court's extraordinary invasion into the State's distribution of authority. Even could it be assumed that Initiative 350 imposed a burden on racial minorities,12 it simply does not place unique political obstacles in the way of racial minorities. In this case, unlike in Hunter, the political system has not been redrawn or altered. The authority of the State over the public school system, acting through initiative or the legislature, is plenary. Thus, the State's political system is not altered when it adopts for the first time a policy, concededly within the area of its authority, for the regulation of local school districts. And certainly racial minorities are not uniquely or comparatively burdened by the State's adoption of a policy that would be lawful if adopted by any school district in the State.13 64 Hunter, therefore, is simply irrelevant. It is the Court that by its decision today disrupts the normal course of State government.14 Under its unprecedented theory of a vested constitutional right to local decisionmaking, the State apparently is now forever barred from addressing the perplexing problems of how best to educate fairly all children in a multi-racial society where, as in this case, the local school board has acted first.15 V 65 We are not asked to decide the wisdom of a state policy that limits the ability of local school districts to adopt—on their own volition—mandatory reassignments for racial balance. We must decide only whether the Federal Constitution permits the State to adopt such a policy. The School Districts in this case were under no federal constitutional obligation to adopt mandatory busing. Absent such an obligation, the State—exercising its sovereign authority over all subordinate agencies—should be free to reject this debatable restriction on liberty. But today's decision denies this right to a State. In this case, it deprives the State of Washington of all opportunity to address the unresolved questions resulting from extensive mandatory busing.16 The Constitution does not dictate to the States at what level of government decisions affecting the public schools must be taken. It certainly does not strip the States of their sovereignty. It therefore does not authorize today's intrusion into the State's internal structure.17 1 In 1971, the District implemented a program of mandatory reassignments to integrate certain of its middle schools. This prompted an attempt to recall four School Board members who had voted for the program. That attempt narrowly failed. See 473 F.Supp. 996, 1006 (WD Wash.1979). 2 Several community organizations threatened legal action if the District did not initiate a more effective integration effort, while the Mayor of Seattle and a number of community leaders, by letter dated May 20, 1977, urged the District to adopt "a definition of racial isolation and measurable goals leading to the elimination of racial isolation in the Seattle Public Schools prior to a Court ordered and mandated desegregation remedy." App. 139. 3 The District Court found that the actions of the School Board were prompted by its members' "desire to ward off threatened litigation, their desire to prevent the threatened loss of federal funds, their desire to relieve the black students of the disproportionate burden which they had borne in the voluntary efforts to balance the schools racially and their perception that racial balance in the schools promotes the attainment of equal educational opportunity and is beneficial in the preparation of all students for democratic citizenship regardless of their race." 473 F.Supp., at 1007. 4 Washington's Constitution reserves to the people of the State "the power to propose bills, laws, and to enact or reject the same at the polls, independent of the legislature." Wash.Const., Art. II, § 1. Such initiatives are placed on the ballot upon the petition of 8% of the State's voters registered and voting for governor at the last preceding regular gubernatorial election. § 1a. If passed by the electorate, an initiative may not be repealed by the state legislature for two years, although it may be amended within two years by a vote of two-thirds of each house of the legislature. § 41. See generally Comment, Judicial Review of Laws Enacted by Popular Vote, 55 Wash.L.Rev. 175 (1979). 5 The text of Initiative 350 is now codified as Wash.Rev.Code §§ 28A.26.010-28A.26.900 (1981). 6 At the beginning of the 1978-1979 academic year, approximately 300,000 of the 769,040 students enrolled in Washington's public schools were bused to school. Ninety-five percent of these students were transported for reasons unrelated to race. 473 F.Supp., at 1002. 7 Along with Seattle, Tacoma School District No. 10 and Pasco School District No. 1 are the only districts in the State of Washington with comprehensive integration programs, and therefore the three are the only districts affected by Initiative 350. See id., at 1009. Since 1965, Pasco has made use of school closures and a mandatory busing program to overcome the racial isolation caused by segregated housing patterns; if students attended the schools nearest their homes, three of Pasco's seven elementary schools would have a primarily white and three a primarily minority student body. Id., at 1002-1003. The Tacoma School District has made use of school closures, racially controlled enrollment at magnet schools, and voluntary transfers—though not mandatory busing—to enhance racial balance in its schools. Id., at 1003-1004. 8 Several of the intervenor plaintiffs also alleged that the District had engaged in de jure segregation, and therefore was operating an unconstitutional dual school system. The District Court therefore bifurcated the litigation, first addressing the constitutionality of Initiative 350. Because of the court's conclusions on that question, the allegations of de jure segregation did not go to trial and have not been addressed by the District Court or by the Court of Appeals. 9 The District Court acknowledged that it was impossible to determine whether the supporters of Initiative 350 "subjectively [had] a racially discriminatory intent or purpose," because "[a]s to that subjective intent the secret ballot raises an impenetrable barrier." Id., at 1014. The court looked instead to objective factors, noting that it "marked [a] departure from the norm . . . for the autonomy of school boards to be restricted relative to the assignment of students," and that it marked a similar "departure from the procedural norm" for "an administrative decision of a subordinate local unit of government . . . [to be] overridden in a statewide initiative." Id., at 1016. These factors, when coupled with the "racially disproportionate impact of the initiative," its "historical background," and "the sequence of events leading to its adoption," were found to demonstrate that a "racially discriminatory intent or purpose was at least one motivating factor in the adoption of the initiative." Ibid. 10 The District Court noted that school boards that had practiced de jure segregation are under an affirmative obligation to eliminate the effects of that practice. Ibid. See Columbus Board of Education v. Penick, 443 U.S. 449, 458-459, 99 S.Ct. 2941, 2947, 61 L.Ed.2d 666 (1979). 11 The Court of Appeals therefore did not address the District Court's alternative finding that Initiative 350 had been adopted for discriminatory reasons, or its conclusion that the initiative was overbroad. 633 F.2d, at 1342. 12 After the decision on the merits, the District Court had declined to award attorney's fees to the plaintiff School Districts because the Districts are state-funded entities. App. to Juris. Statement C-1. The Court of Appeals reversed on this issue, concluding that the District Court had abused its discretion in denying fees. The Court of Appeals determined that the School Districts fell within the language of the attorney's fees statutes, 42 U.S.C. § 1988 and 20 U.S.C. § 3205 (1976 ed., Supp.IV), see n. 31, infra, and it reasoned that "[a]s long as a publicly-funded organization advances important constitutional values, it is eligible for fees under the statutes." 633 F.2d, at 1348. 13 As does Initiative 350, the New York statute apparently permitted voluntary student transfers to achieve integration. See n. 16, infra. 14 The Court of Appeals accepted the District Court's characterization of the initiative, and even the dissenting judge in the Court of Appeals agreed that Initiative 350 addresses a "racial" problem. 633 F.2d, at 1353. 15 Appellants and the United States do not challenge the propriety of race-conscious student assignments for the purpose of achieving integration, even absent a finding of prior de jure segregation. We therefore do not specifically pass on that issue. See generally Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 16, 91 S.Ct. 1267, 1276, 28 L.Ed.2d 586 (1971); North Carolina State Board of Education v. Swann, 402 U.S. 43, 45, 91 S.Ct. 1284, 1286, 28 L.Ed.2d 554 (1971). Cf. University of California Regents v. Bakke, 438 U.S. 265, 300, n. 39, 312-314, 98 S.Ct. 2733, 2753, n. 39, 2759-2760, 57 L.Ed.2d 750 (1978) (opinion of POWELL, J.). 16 The United States seeks to distinguish Lee by suggesting that the statute there at issue "clearly prohibited" all attempts to ameliorate racial imbalance in the schools, while Initiative 350 permits voluntary desegregation efforts. Brief for United States 25. Even assuming that this distinction would otherwise be of constitutional significance, its premise is not accurate. The legislation challenged in Lee did permit voluntary integration efforts, for it expressly exempted from its restrictions "the assignment of a pupil in the manner requested or authorized by his parents or guardian." 318 F.Supp., at 712. Thus, as the District Court in Lee noted, the statute "denie[d] appointed officials the power to implement non-voluntary programs for the improvement of racial balance." Id., at 715 (emphasis added). The difficulty in Lee —as in this case—stemmed from the Lee District Court's conclusion that a voluntary program would not serve to integrate the community's schools: "Voluntary plans for achieving racial balance . . . have not had a significant impact on the problems of racial segregation in the Buffalo public schools; indeed it would appear that racial isolation is actually increasing." Ibid. Thus the statute challenged in Lee and Initiative 350 operated in precisely the same way to "deny . . . student[s] the right to attend a fully integrated school." Brief for United States 25. 17 Justice POWELL finds Hunter completely irrelevant, dismissing it with the conclusory statement that "the political system [of Washington] has not been redrawn or altered." Post, at 498 (emphasis in original). But the dissent entirely fails to address the relevance of Hunter to the reallocation of decisionmaking authority worked by Initiative 350. The evil condemned by the Hunter Court was not the particular political obstacle of mandatory referenda imposed by the Akron charter amendment; it was, rather, the comparative structural burden placed on the political achievement of minority interests. Thus, in Hunter, the procedures for enacting racial legislation were modified in such a way as to place effective control in the hands of the citywide electorate. Similarly here, the power to enact racial legislation has been reallocated. In each case, the effect of the challenged action was to redraw decisionmaking authority over racial matters and only over racial matters—in such a way as to place comparative burdens on minorities. While Justice POWELL and the United States find it crucial that the proponents of integrated schools remain free to use Washington's initiative system to further their ends, that was true in Hunter as well: proponents of open housing were not barred from invoking Akron's initiative procedures to repeal the charter amendment, or to enact fair housing legislation of their own. It surely is an excessively formal exercise, then, to argue that the procedural revisions at issue in Hunter imposed special burdens on minorities, but that the selective allocation of decisionmaking authority worked by Initiative 350 does not erect comparable political obstacles. Indeed, Hunter would have been virtually identical to this case had the Akron charter amendment simply barred the City Council from passing any fair housing ordinance, as Initiative 350 forbids the use of virtually all mandatory desegregation strategies. Surely, however, Hunter would not have come out the other way had the charter amendment made no provision for the passage of fair housing legislation, instead of subjecting such legislation to ratification by referendum. The United States also would note that Initiative 350's "modification of state policy [was] not the result of any unusual political procedure," Brief for United States 30, for initiatives and referenda are often used by the Washington electorate. But that observation hardly serves to distinguish this case from Hunter, since the fair housing charter amendment was added through the unexceptional use of Akron's initiative procedure. See 393 U.S., at 387, 89 S.Ct., at 558. 18 Despite the force with which it is written, then, Justice POWELL's essay on "the heretofore unquestioned right of a State to structure the decisionmaking authority of its government," post, at 493 —as well as his observations on a State's right to repeal programs designed to eliminate de facto segregation—is largely beside the point. The State's power has not been questioned at any point during this litigation. The single narrow question before us is whether the State has exercised its power in such a way as to place special, and therefore impermissible, burdens on minority interests. 19 The Court noted in Hunter that Akron "might have proceeded by majority vote . . . on all its municipal legislation," 393 U.S., at 392, 89 S.Ct., at 561; the charter amendment was invalidated because the citizens of Akron did not reserve all power to themselves, but rather distributed it in a nonneutral manner. In Lee, of course, the State had unquestioned authority to vest all power over education in state officials. 20 Indeed, even the State's efforts to help ensure equal opportunity in education and to encourage desegregation are cast in cooperative terms, and are designed to assist school districts in implementing programs of their choosing. See, e.g., Wash.Rev.Code §§ 28A.21.010(3), 28A.21.136(1) and (3) (1981); cf. § 28A.58.245(3). 21 The Washington Supreme Court noted: "[A]s long as the school board authorized or required students to attend schools geographically situated close to their homes, they had such a right. But the right existed only because it was given to them by the school authorities." 80 Wash.2d, at 452, 495 P.2d, at 662. 22 We also note that the State has not attempted to reserve to itself exclusive power to deal with racial issues generally. Municipalities in Washington have been given broad powers of self-government, see generally Wash.Const., Amdt. 40; Wash.Rev.Code §§ 35.22.020, 35.23.440, 35.27.370, 35.30.010 (1981); Wash.Rev.Code, Tit. 35A (Optional Municipal Code), and Washington courts specifically have held that municipalities have the power to enact antidiscrimination ordinances. See, e.g., Seattle Newspaper-Web Pressmen's Union Local No. 26 v. Seattle, 24 Wash.App. 462, 604 P.2d 170 (1979). Cf. 5 E. McQuillin, Law of Municipal Corporations § 19.23, p. 425 (3d rev. ed. 1981). 23 Throughout his dissent, Justice POWELL insists that the Court has created a "vested constitutional right to local decisionmaking," post, at 498-499, that under our holding "the people of the State of Washington apparently are forever barred from developing a different policy on mandatory busing where a school district previously has adopted one of its own," post, at 498, n. 14, and that today's decision somehow raises doubts about "the authority of a State to abolish school boards altogether." Post, at 494. See also post, at 495, and 498-499, n. 14. These statements evidence a basic misunderstanding of our decision. Our analysis vests no rights, and has nothing to do with whether school board action predates that taken by the State. Instead, what we find objectionable about Initiative 350 is the comparative burden it imposes on minority participation in the political process—that is, the racial nature of the way in which it structures the process of decisionmaking. It is evident, then, that the horribles paraded by the dissent, post, at 498-499, n. 14 which have nothing to do with the ability of minorities to participate in the process of self-government—are entirely unrelated to this case. It is equally clear, as we have noted at several points in our opinion, that the State remains free to vest all decisionmaking power in state officials, or to remove authority from local school boards in a race-neutral manner. 24 One amicus observes that many States employ a similar educational structure. See Brief for National School Boards Assn. as Amicus Curiae 11, 14-16, App. 1a-10a. 25 When authority to initiate desegregation programs was removed from appointed school boards and from state education officials, the only body capable of exercising power over such programs was the state legislature. 26 The United States makes only one attempt to distinguish Lee in this regard: Lee is inapposite, the United States maintains, because the statute at issue there "blocked desegregation efforts even by 'a school district subject to a pre-existing order to eliminate segregation in its schools,' " and therefore—purportedly in contrast to Initiative 350—"interfere[d] with the efforts of individual school districts to eliminate de jure segregation." Brief for United States 25, quoting Lee v. Nyquist, 318 F.Supp., at 715. If by this statement the United States seeks to place the District Court's holding and this Court's affirmance in Lee on the ground that the New York statute interfered with Buffalo's attempts to eliminate de jure segregation, its submission is simply inaccurate. At the time of the Lee litigation, Buffalo had not been found guilty of practicing intentional segregation. See Arthur v. Nyquist, 573 F.2d 134, 137 (CA2 1978). As the United States notes, Buffalo was under a "pre-existing order to eliminate segregation in its schools"—but that order was issued by the New York Commissioner of Education, because he had found Buffalo's schools de facto segregated. Appeal of Dixon, 4 N.Y.Educ. Dept. Reports 115 (1965). See Lee v. Nyquist, 318 F.Supp., at 714-715. Lee did not concern de jure segregation; it is to be explained only as a straightforward application of the Hunter doctrine. 27 That phenomenon is graphically demonstrated by the circumstances of this litigation. The longstanding desegregation programs in Pasco and Tacoma, as well as the Seattle middle school integration plan, have functioned for years without creating undue controversy. Yet they have been swept away, along with the Seattle Plan, by Initiative 350. As a practical matter, it seems most unlikely that proponents of desegregative busing in smaller communities such as Tacoma or Pasco will be able to obtain the statewide support now needed to permit them to desegregate the schools in their communities. 28 The State does not suggest that Initiative 350 furthers the kind of compelling interest necessary to overcome the strict scrutiny applied to explicit racial classifications. 29 Thus we do not hold, as the dissent implies, post, at 494, that the State's attempt to repeal a desegregation program creates a racial classification, while "identical action" by the Seattle School Board does not. It is the State's race-conscious restructuring of its decisionmaking process that is impermissible, not the simple repeal of the Seattle Plan. 30 We also note that singling out the political processes affecting racial issues for uniquely disadvantageous treatment inevitably raises dangers of impermissible motivation. When political institutions are more generally restructured, as JUSTICE BRENNAN has noted in another context, "[t]he very breadth of [the] scheme . . . negates any suggestion" of improper purpose. Walz v. Tax Comm'n, 397 U.S. 664, 689, 90 S.Ct. 1409, 1422, 25 L.Ed.2d 697 (1970) (concurring opinion). 31 Appellants also challenge the Court of Appeals' award of attorney's fees to the School District plaintiffs, see n. 12, supra, arguing that state-funded entities are not eligible to receive such awards from the State. In our view, this contention is without merit. The Districts are plainly parties covered by the language of the fees statutes. See 42 U.S.C. § 1988 (1976 ed., Supp. IV) ("In any action . . . to enforce a provision of sections 1981, 1982, 1983, 1985, and 1986 of this title . . . the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney's fee as part of its costs") (emphasis added); 20 U.S.C. § 3205 (1976 ed., Supp. IV) ("Upon the entry of a final order by a court of the United States against a . . . State . . . for failure to comply with . . . the fourteenth amendment to the Constitution of the United States as [it] pertain[s] to elementary and secondary education, the court, in its discretion . . . may allow the prevailing party, other than the United States, a reasonable attorney's fee as part of its costs") (emphasis added). Nothing in the history of the statutes suggests that this language was meant to exclude state-funded entities. To the contrary, the Courts of Appeals have held with substantial unanimity that publicly funded legal services organizations may be awarded fees. See, e.g., Dennis v. Chang, 611 F.2d 1302 (CA9 1980); Holley v. Lavine, 605 F.2d 638 (CA2 1979), cert. denied sub nom. Blum v. Holley, 446 U.S. 913, 100 S.Ct. 1843, 64 L.Ed.2d 266 (1980); Lund v. Affleck, 587 F.2d 75 (CA1 1978). And when it enacted § 1988, Congress cited with approval a decision awarding fees to a state-funded organization. See H.R.Rep.No.94-1558, p. 8, n. 16 (1976) (citing Incarcerated Men of Allen County Jail v. Fair, 507 F.2d 281 (CA6 1974). In any event, the underlying congressional policies are served by awarding fees in cases such as the one before us: no matter what the source of their funds, school boards have limited budgets, and allowing them fees "encourage[s] compliance with and enforcement of the civil rights laws." Dennis v. Chang, 611 F.2d, at 1306. See id., at 1306-1307. While appellants suggest that it is incongruous for a State to pay attorney's fees to one of its school boards, it seems no less incongruous that a local board would feel the need to sue the State for a violation of the Fourteenth Amendment. We see no reason to disturb the judgment of the Court of Appeals on this point. 1 Throughout this dissent, I use the term "mandatory busing" to refer to busing—or mandatory student reassignments—for the purpose of achieving racial integration. 2 The Initiative passed by almost 66% of the statewide vote. In Seattle the Initiative passed by over 61% of the vote. It failed in only two of Seattle's legislative districts—one predominantly black and one predominantly white. 3 Unlike the constitutional amendment at issue in Crawford v. Los Angeles Board of Education of the City of Los Angeles, 458 U.S. 527, 102 S.Ct. 3211, 73 L.Ed.2d 948, Initiative 350 places no limits on the state courts in their interpretation of the State Constitution. Thus, if mandatory school assignments were required by the State Constitution—although not by the Fourteenth Amendment of the Federal Constitution—Initiative 350 would not hinder a State from enforcing its Constitution. 4 Tacoma School District No. 10 and Pasco School District No. 1 are the only other school districts in Washington with extensive integration programs. Pasco has relied upon school closings and mandatory busing to achieve racial integration in its schools. Only minority children are bused under the Pasco plan. 473 F.Supp. 996, 1002 (WDWash.1979). In addition to school closings, the Tacoma integration plan relies upon voluntary techniques—magnet schools and voluntary transfers. 5 Judge Wright dissented. In his view Initiative 350 could not be said to embody a racial classification. The Initiative does not classify individuals on the basis of their race. It simply deals with a matter bearing on race relations. Moreover, no racial classification is created because the citizens of a State favor mandatory school reassignments for some purposes but not for reasons of race. The benefits and problems associated with busing for one reason—e.g., for racial integration—are not the same as for another—e.g., to avoid safety hazards. Finally, Judge Wright could not understand how the exercise of authority by the State could create a racial classification. The State had not intervened by altering the legislative process in a way that burdened racial minorities. Charged by the State Constitution with the responsibility for the provision of public education, the State had simply exercised its authority to run its own school system. Judge Wright also addressed the District Court's alternative holdings that Initiative 350 is overbroad or that it was motivated by discriminatory intent. He found no basis for either conclusion. These alternative holdings were not addressed by the Court of Appeals majority. Nor are they relied upon by the Court today. Accordingly, they are not discussed in this dissent. 6 See Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 28, 91 S.Ct. 1267, 1282, 28 L.Ed.2d 554 (1971) ("Absent a constitutional violation there would be no basis for judicially ordering assignment of students on a racial basis. All things being equal, with no history of discrimination, it might well be desirable to assign pupils to schools nearest their homes"). Indeed, in the absence of a finding of segregation by the School District, mandatory busing on the basis of race raises constitutional difficulties of its own. Extensive pupil transportation may threaten liberty or privacy interests. See University of California Regents v. Bakke, 438 U.S. 265, 300, n. 39, 98 S.Ct. 2733, 2753, n. 39, 57 L.Ed.2d 750 (1978) (opinion of POWELL, J.); Keyes v. School District No. 1, Denver, Colo., 413 U.S. 189, 240-250, 93 S.Ct. 2686, 2713-18, 37 L.Ed.2d 548 (1973) (POWELL, J., concurring in part and dissenting in part). Moreover, when a State or school board assigns students on the basis of their race, it acts on the basis of a racial classification, and we have consistently held that "[a] racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification." Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 272, 99 S.Ct. 2282, 2292, 60 L.Ed.2d 870 (1979). 7 "[A]ccording to the institutions of this country, the sovereignty in every State resides in the people of the State, and . . . they may alter and change their form of government at their own pleasure." Luther v. Borden, 7 How. 1, 47, 12 L.Ed. 581 (1849). See Community Communications Co. v. Boulder, 455 U.S. 40, 53-54, 102 S.Ct. 835, 842, 70 L.Ed.2d 810 (1982); Sailors v. Board of Education, 387 U.S. 105, 109, 87 S.Ct. 1549, 1552, 18 L.Ed.2d 650 (1967) ("Save and unless the state, county, or municipal government runs afoul of a federally protected right, it has vast leeway in the management of its internal affairs"); United States v. Kagama, 118 U.S. 375, 379, 6 S.Ct. 1109, 1111, 30 L.Ed. 228 (1886) (under the Constitution, sovereign authority resides either with the States or the Federal Government, and "[t]here exist . . . but these two"). 8 The Court consistently has held that "the Equal Protection Clause is not violated by the mere repeal of race-related legislation or policies that were not required by the Federal Constitution in the first place." Crawford v. Los Angeles Board of Education, 458 U.S., at 538, 102 S.Ct., at 3218. 9 The policies in support of neighborhood schooling are various but all of them are racially neutral. The people of the State legitimately could decide that unlimited mandatory busing places too great a burden on the liberty and privacy interests of families and students of all races. It might decide that the reassignment of students to distant schools, on the basis of race, was too great a departure from the ideal of racial neutrality in state action. And, in light of the experience with mandatory busing in other cities, the State might conclude that such a program ultimately would lead to greater racial imbalance in the schools. See Estes v. Metropolitan Branches of Dallas NAACP, 444 U.S. 437, 451, 100 S.Ct. 716, 723, 62 L.Ed.2d 626 (1980) (POWELL, J., dissenting). 10 Cf. James v. Valtierra, 402 U.S. 137, 142, 91 S.Ct. 1331, 1334, 28 L.Ed.2d 678 (1971) ("[O]f course a lawmaking procedure that 'disadvantages' a particular group does not always deny equal protection. Under any such holding, presumably a State would not be able to require referendums on any subject unless referendums were required on all, because they would always disadvantage some group. And this Court would be required to analyze governmental structures to determine whether a gubernatorial veto provision or a filibuster rule is likely to 'disadvantage' any of the diverse and shifting groups that make up the American people"). 11 The Court also relies at certain critical points in its discussion on the summary affirmance in Lee v. Nyquist, 318 F.Supp. 710 (WDNY 1970), summarily aff'd, 402 U.S. 935, 91 S.Ct. 1618, 29 L.Ed.2d 105 (1971). As we have often noted, however, summary affirmances by this Court are of little precedential force. See Metromedia, Inc. v. San Diego, 453 U.S. 490, 500, 101 S.Ct. 2882, 2888, 69 L.Ed.2d 800 (1981). A summary affirmance "is not to be read as an adoption of the reasoning supporting the judgment under review." Zobel v. Williams, 457 U.S. 55, 64, n. 13, 102 S.Ct. 2309, 2315, n. 13, 72 L.Ed.2d 672 (1982). 12 It is far from clear that in the absence of a constitutional violation, mandatory busing necessarily benefits racial minorities or that it is even viewed with favor by racial minorities. See Crawford v. Los Angeles Board of Education, 458 U.S. 527, 545, n. 32, 102 S.Ct. 3211, 3222, n. 32, 73 L.Ed.2d 948. As the Court indicates, the busing question is complex and is best resolved by the political process. Ante, at 474. Moreover, it is significant that Initiative 350 places no limits on voluntary programs or on court-ordered reassignments. It permits school districts to order school closings for purposes of racial balance. § 28A.26.030. And it permits school districts to order a student to attend the "next nearest"—rather than nearest—school to promote racial integration. 13 The Court repeatedly states that the effect of Initiative 350 is "to redraw decisionmaking authority over racial matters—and only over racial matters —in such a way as to place comparative burdens on minorities." Ante, at 475, n. 17 (emphasis added). But the decision by the State to exercise its authority over the schools and over racial matters in the schools does not place a comparative burden on racial minorities. In Hunter, as we have understood it, "fair housing legislation alone was subject to an automatic referendum requirement." Gordon v. Lance, 403 U.S. 1, 5, 91 S.Ct. 1889, 1891, 29 L.Ed.2d 273 (1971) (emphasis added). By contrast, Initiative 350 merely places mandatory busing among the much larger group of matters—covering race relations, administration of the schools, and a variety of other matters addressed at the state level. See n. 15, infra. Racial minorities, if indeed they are burdened by Initiative 350, are not comparatively burdened. In this respect, they are in the same position as any other group of persons who are disadvantaged by regulations drawn at the state level. 14 The Court's decision intrudes deeply into normal state decisionmaking. Under its holding the people of the State of Washington apparently are forever barred from developing a different policy on mandatory busing where a school district previously has adopted one of its own. This principle would not seem limited to the question of mandatory busing. Thus, if the admissions committee of a state law school developed an affirmative-action plan that came under fire, the Court apparently would find it unconstitutional for any higher authority to intervene unless that authority traditionally dictated admissions policies. As a constitutional matter, the dean of the law school, the faculty of the university as a whole, the university president, the chancellor of the university system, and the board of regents might be powerless to intervene despite their greater authority under state law. After today's decision it is unclear whether the State may set policy in any area of race relations where a local governmental body arguably has done "more" than the Fourteenth Amendment requires. If local employment or benefits are distributed on a racial basis to the benefit of racial minorities, the State apparently may not thereafter ever intervene. Indeed, under the Court's theory one must wonder whether—under the equal protection component of the Fifth Amendment—even the Federal Government could assert its superior authority to regulate in these areas. 15 Even accepting the dubious notion that a State must demonstrate some past control over public schooling or race relations before now intervening in these matters, ante, at 477, the Court's attempt to demonstrate that Initiative 350 represents a unique thrust by the State into these areas is unpersuasive. The Court's own discussion indicates the comprehensive character of the State's activity. The Common School Provisions of the State's Code of Laws are nearly 200 pages long, governing a broad variety of school matters. The State has taken seriously its constitutional obligation to provide public education. See Wash.Const., Art. IX, § 2; Seattle School District No. 1 v. State, 90 Wash.2d 476, 518, 585 P.2d 71, 95 (1978). In light of the wide range of regulation of the public schools by the State, it is wholly unclear what degree of prior concern or control by the State would satisfy the Court's new doctrine. In addition to public school affairs generally, the State has taken a direct interest in ending racial discrimination in the schools and elsewhere. See Wash.Rev.Code § 49.60.010 et seq. (1981). Article IX, § 1, of the State Constitution specifically prohibits discrimination in public schools: "It is the paramount duty of the state to make ample provision for the education of all children residing within its borders without distinction or preference on account of race, color, caste, or sex." The State Supreme Court has not interpreted this section of the State Constitution to prohibit race-conscious school assignments in the absence of a violation of the Fourteenth Amendment. Cf. Citizens Against Mandatory Bussing v. Palmason, 80 Wash.2d 445, 495 P.2d 657 (1972). But until today's decision one would have thought that the state court could have rendered such a decision without violating the Federal Constitution. 16 Responding to this dissent, the Court denies that its opinion limits the authority of the people of the State of Washington and the legislature to control or regulate school boards. It further states that "the State remains free to vest all decisionmaking power in state officials, or to remove authority from local school boards in a race-neutral manner." Ante, at 480, n. 23. These are puzzling statements that seem entirely at odds with much of the text of the Court's opinion. It will be surprising if officials of the State of Washington—with the one exception mentioned below—will have any clear idea as to what the State now lawfully may do. The Court does say that "[i]t is the State's race-conscious restructuring of its decisionmaking process that is impermissible, not the simple repeal of the Seattle Plan." Ante, at 485-486, n. 29. Apparently the Court is saying that, despite what else may be said in its opinion, the people of the State—or the state legislature—may repeal the Seattle Plan, even though neither the people nor the legislature validly may prescribe statewide standards. I perceive no logic in—and certainly no constitutional basis for—a distinction between repealing the Seattle Plan of mandatory busing and establishing a statewide policy to the same effect. The people of a State have far greater interest in the general problems associated with compelled busing for the purpose of integration than in the plan of a single school board. 17 As a former school board member for many years, I accept the privilege of a dissenting Justice to add a personal note. In my view, the local school board—responsible to the people of the district it serves—is the best qualified agency of a state government to make decisions affecting education within its district. As a policy matter, I would not favor reversal of the Seattle Board's decision to experiment with a reasonable mandatory busing program, despite my own doubts as to the educational or social merit of such a program. See Estes v. Metropolitan Branches of Dallas NAACP, 444 U.S., at 438-448, 100 S.Ct., at 716-722 (POWELL, J., dissenting). But this case does not present a question of educational policy or even the merits of busing for racial integration. The question is one of a State's sovereign authority to structure and regulate its own subordinate bodies.

12

458 U.S. 502 102 S.Ct. 3231 73 L.Ed.2d 928 Marjorie LEHMAN, etc., Petitionerv.LYCOMING COUNTY CHILDREN'S SERVICES AGENCY. No. 80-2177. Argued March 30, 1982. Decided June 30, 1982. Syllabus Petitioner voluntarily placed her three sons in the legal custody of respondent county agency, which in turn placed them in foster homes. Thereafter, a Pennsylvania state court terminated petitioner's parental rights with respect to her sons because of parental incapacity, and the Pennsylvania Supreme Court affirmed. Petitioner then filed an action in Federal District Court, seeking a writ of habeas corpus under 28 U.S.C. § 2254(a), which requires a district court to entertain an application for such a writ in behalf "of a person in custody" pursuant to a state-court judgment in alleged violation of the Federal Constitution. She requested a declaration of the invalidity of the Pennsylvania statute under which her parental rights were terminated and an order releasing her sons to her custody. The District Court dismissed the petition on the ground that respondent's custody over petitioner's sons was not the type of custody to which § 2254(a) may be addressed. The Court of Appeals affirmed. Held: Section 2254(a) does not confer jurisdiction on federal courts to consider collateral challenges to state-court judgments involuntarily terminating parental rights. Pp. 508-516. (a) Although the scope of the federal writ of habeas corpus has been extended beyond that which the most literal reading of the statute might require, the writ has not been considered a generally available federal remedy for every violation of federal rights. The writ's availability has been limited to challenges to state-court judgments in situations where, as a result of a state-court criminal conviction, a petitioner has suffered substantial restraints not shared by the public generally, and the petitioner has been found to be "in custody" within the meaning of § 2254(a). Here, petitioner's children are not in the "custody" of the State in the way in which this term has been used in determining the availability of the writ of habeas corpus. They are in the "custody" of their foster parents in essentially the same way, and to the same extent, other children are in the custody of their natural or adoptive parents. They suffer no restraint on liberty not shared by the public generally, cf. Jones v. Cunningham, 371 U.S. 236, 83 S.Ct. 373, 9 L.Ed.2d 285; Hensley v. Municipal Court, 411 U.S. 345, 93 S.Ct. 1571, 36 L.Ed.2d 294, nor do they suffer "collateral consequences" sufficient to outweigh the need for finality, cf. Carafas v. LaVallee, 391 U.S. 234, 88 S.Ct. 1556, 20 L.Ed.2d 554. To extend the federal writ to challenges to state child-custody decisions based on alleged constitutional defects collateral to the actual custody decision would be an unprecedented expansion of the jurisdiction of the federal courts. Pp. 508-512. (b) Federalism and the exceptional need for finality in child-custody disputes also argue strongly against the grant of the writ here. Extended uncertainty for the children would be inevitable in many cases if federal courts had jurisdiction to relitigate state custody decisions. Pp. 512-514. (c) Habeas corpus has been used in child-custody cases in many States and in England, and 28 U.S.C. § 2255, authorizing federal-court collateral review of federal decisions, could be construed to include the type of custody to which petitioner's children are subject. But reliance on what may be appropriate within the federal system or within a state system is of little force where, as in this case, a state judgment is attacked collaterally in a federal court. Petitioner would have the federal judicial system entertain a writ that is not time-barred to challenge collaterally a final judgment entered in a state judicial system. Pp. 514-515. 3rd Cir., 648 F.2d 135, affirmed. Martin Guggenheim, Brooklyn, N. Y., for petitioner. Charles F. Greevy, III, Williamsport, Pa., for respondent. Justice POWELL delivered the opinion of the Court. 1 The question presented is whether the habeas corpus statute, 28 U.S.C. § 2254, confers jurisdiction on the federal courts to consider collateral challenges to state-court judgments involuntarily terminating parental rights. 2 * The facts of this case are described in detail in In re William L., 477 Pa. 322, 383 A.2d 1228, cert. denied, sub nom. Lehman v. Lycoming County Children's Services, 439 U.S. 880, 99 S.Ct. 216, 58 L.Ed.2d 192 (1978), the Pennsylvania Supreme Court decision terminating the parental rights of petitioner Marjorie Lehman with respect to three sons born in 1963, 1965, and 1969.1 In 1971, Ms. Lehman discovered that she was pregnant again. Because of housing and other problems related to the care of her sons, Ms. Lehman voluntarily placed them in the legal custody of the Lycoming County Children's Services Agency, and it placed them in foster homes. 3 Although Ms. Lehman visited her sons monthly, she did not request their return until 1974. At that point, the Lycoming County Children's Services Agency initiated parental termination proceedings. In those proceedings, the Orphan's Court Division of the Lycoming County Court of Common Pleas heard testimony from Agency caseworkers, a psychologist, nutrition aides, petitioner, and the three sons.2 The judge concluded: "[I]t is absolutely clear to the court that, by reason of her very limited social and intellectual development combined with her five-year separation from the children, the mother is incapable of providing minimal care, control and supervision for the three children. Her incapacity cannot and will not be remedied."3 In re Lehman, Nos. 2986, 2987 and 2988, p. 4 (Ct.Common Pleas, Lycoming County, Pa., June 3, 1976).4 The court therefore declared that petitioner's parental rights respecting the three sons were terminated. 4 The Pennsylvania Supreme Court affirmed the termination order based on "parental incapacity, which does not involve parental misconduct." In re William L., supra, at 331, 383 A.2d, at 1232. It held that the legislature's power to protect the physical and emotional needs of children authorized termination in the absence of serious harm or risk of serious harm to the children and in the absence of parental misconduct. The court stressed that, "[i]n the instant cases, the basis for termination is several years of demonstrated parental incapacity. . . ." Ibid. It also held that the statute was not unconstitutionally vague either on its face or as applied. 5 Petitioner sought this Court's review in a petition for certiorari rather than by appeal.5 We denied the petition. Lehman v. Lycoming County Children's Services, 439 U.S. 880, 99 S.Ct. 216, 58 L.Ed.2d 192 (1978). Petitioner then filed the instant proceeding on January 16, 1979, in the United States District Court for the Middle District of Pennsylvania, seeking a writ of habeas corpus pursuant to 28 U.S.C. §§ 2241 and 2254. Petitioner requested (i) a declaration of the invalidity of the Pennsylvania statute under which her parental rights were terminated; (ii) a declaration that petitioner was the legal parent of the children; and (iii) an order releasing the children to her custody unless within 60 days an appropriate state court judicially determined that the best interest of the children required that temporary custody remain with the State. 6 The District Court dismissed the petition without a hearing. Relying primarily on Sylvander v. New England Home for Little Wanderers, 584 F.2d 1103 (CA 1 1978), the court concluded that "the custody maintained by the Respondent over the three Lehman children is not that type of custody to which the federal habeas corpus remedy may be addressed." Lehman v. Lycoming County Children's Services Agency, Civ. No. 79-65 (MD Pa.1979), reprinted in App. to Pet. for Cert. 135a, 147a. 7 Sitting en banc, the Court of Appeals for the Third Circuit affirmed the District Court's order of dismissal by a divided vote of six to four. 648 F.2d 135 (1981). No majority opinion was written. A plurality of four, in an opinion written by Judge Garth, concluded that "disputes of the nature addressed here and which essentially involve no more than the question of who shall raise a child to maturity, do not implicate the federal interest in personal liberty sufficiently to warrant the extension of federal habeas corpus." Id., at 146. In support of this conclusion, Judge Garth reasoned that "[i]t is not the liberty interest of the children that is sought to be protected in such a case, but only the right of the particular parent to raise them." Id., at 140 (footnote omitted). 8 A second plurality of four, in an opinion written by Judge Adams wrote that it "would appear to be both unwise and impolitic for the federal courts to uncover a whole new font of jurisdiction. . . ." Id., at 151. He would have disposed of the case on the ground that Ms. Lehman did not have standing to assert a habeas corpus action on behalf of her children. See id., at 151-155. This view was based on the conclusion that once a parent's rights have been terminated in a state proceeding, a parent is no longer presumed to represent the interest of the child. See id., at 153-154.6 9 The question presented to this Court can be stated more fully as whether federal habeas corpus jurisdiction, under § 2254, may be invoked to challenge the constitutionality of a state statute under which a State has obtained custody of children and has terminated involuntarily the parental rights of their natural parent. As this is a question of importance not heretofore considered by this Court, and one over which the Circuits are divided,7 we granted certiorari. 454 U.S. 813, 102 S.Ct. 89, 70 L.Ed.2d 82 (1981). We now affirm. II A. 10 Petitioner seeks habeas corpus collateral review by a federal court of the Pennsylvania decision. Her application was filed under 28 U.S.C. § 2254(a): 11 "The Supreme Court, a Justice thereof, a circuit judge, or a district court shall entertain an application for a writ of habeas corpus in behalf of a person in custody pursuant to the judgment of a State court only on the ground that he is in custody in violation of the Constitution or laws or treaties of the United States." 12 Although the language of § 2254(a), especially in light of § 2241, suggests that habeas corpus is available only to challenge the convictions of prisoners actually in the physical custody of the State,8 three modern cases have extended it to other situations involving challenges to state-court decisions.9 The first of these cases is Jones v. Cunningham, 371 U.S. 236, 83 S.Ct. 373, 9 L.Ed.2d 285 (1963), in which the Court allowed a parolee to challenge his conviction by a habeas petition. The Court considered the parolee in "custody" for purposes of § 2254(b) because "the custody and control of the Parole Board involve significant restraints on petitioner's liberty . . . which are in addition to those imposed by the State upon the public generally." 371 U.S., at 242, 83 S.Ct., at 377. And in Carafas v. LaVallee, 391 U.S. 234, 88 S.Ct. 1556, 20 L.Ed.2d 554 (1968), the Court allowed the writ in a challenge to a state-court judgment even though the prisoner, incarcerated at the time the writ was filed, had finished serving his sentence during the proceedings. The custody requirement had, of course, been met at the time the writ was filed, and the case was not moot because Carafas was subject to " 'collateral consequences' " as a result of his conviction, id., at 237, 88 S.Ct., at 1559, and "is suffering, and will continue to suffer, serious disabilities . . . ." Id., at 239, 88 S.Ct., at 1560. Most recently, in Hensley v. Municipal Court, 411 U.S. 345, 93 S.Ct. 1571, 36 L.Ed.2d 294 (1973), the Court allowed the writ to be used to challenge a state-court conviction even though the defendant had been released on his own recognizance after sentencing but prior to the commencement of his incarceration. The Court held that the defendant was in the custody of the State for purposes of § 2254(b) because he was "subject to restraints 'not shared by the public generally,' " 411 U.S., at 351, 93 S.Ct., at 1575 (citation omitted)—indeed, his arrest was imminent.10 13 Thus, although the scope of the writ of habeas corpus has been extended beyond that which the most literal reading of the statute might require, the Court has never considered it a generally available federal remedy for every violation of federal rights. Instead, past decisions have limited the writ's availability to challenges to state-court judgments in situations where—as a result of a state-court criminal conviction—a petitioner has suffered substantial restraints not shared by the public generally. In addition, in each of these cases the Court considered whether the habeas petitioner was "in custody" within the meaning of § 2254.11 14 Ms. Lehman argues that her sons are involuntarily in the custody of the State for purposes of § 2254 because they are in foster homes pursuant to an order issued by a state court. Her sons, of course, are not prisoners. Nor do they suffer any restrictions imposed by a state criminal justice system. These factors alone distinguish this case from all other cases in which this Court has sustained habeas challenges to state-court judgments. Moreover, although the children have been placed in foster homes pursuant to an order of a Pennsylvania court, they are not in the "custody" of the State in the sense in which that term has been used by this Court in determining the availability of the writ of habeas corpus. They are in the "custody" of their foster parents in essentially the same way, and to the same extent, other children are in the custody of their natural or adoptive parents. Their situation in this respect differs little from the situation of other children in the public generally; they suffer no unusual restraints not imposed on other children. They certainly suffer no restraint on liberty as that term is used in Hensley and Jones, and they suffer no "collateral consequences"—like those in Carafas—sufficient to outweigh the need for finality. The "custody" of foster or adoptive parents over a child is not the type of custody that traditionally has been challenged through federal habeas.12 Ms. Lehman simply seeks to relitigate, through federal habeas, not any liberty interest of her sons, but the interest in her own parental rights.13 15 Although a federal habeas corpus statute has existed ever since 1867, federal habeas has never been available to challenge parental rights or child custody.14 Indeed, in two cases, the Court refused to allow the writ in such instances. Matters v. Ryan, 249 U.S. 375, 39 S.Ct. 315, 63 L.Ed. 654 (1919); In re Burrus, 136 U.S. 586, 10 S.Ct. 850, 34 L.Ed. 1500 (1890). These decisions rest on the absence of a federal question, but the opinions suggest that federal habeas corpus is not available to challenge child custody. Moreover, federal courts consistently have shown special solicitude for state interests "in the field of family and family-property arrangements." United States v. Yazell, 382 U.S. 341, 352, 86 S.Ct. 500, 507, 16 L.Ed.2d 404 (1966). Under these circumstances, extending the federal writ to challenges to state child-custody decisions—challenges based on alleged constitutional defects collateral to the actual custody decision—would be an unprecedented expansion of the jurisdiction of the lower federal courts.15 B 16 Federalism concerns and the exceptional need for finality in child-custody disputes argue strongly against the grant of Ms. Lehman's petition.16 The writ of habeas corpus is a major exception to the doctrine of res judicata, as it allows relitigation of a final state-court judgment disposing of precisely the same claims. Because of this tension between the State's interest in finality and the asserted federal interest, federal courts properly have been reluctant to extend the writ beyond its historic purpose. As Judge Campbell noted in Sylvander v. New England Home for Little Wanderers: 17 "Federal habeas involves a substantial thrust by the federal system into the sphere normally reserved to the states and hence a change in the federal-state balance. This is so because the federal habeas remedy, as recently fashioned, offers a federal forum regardless of what state proceedings have already taken place and in effect allows a single federal district judge to overrule the judgment of the highest state court, unfettered by the constraints of collateral estoppel and res judicata." 584 F.2d, at 1111-1112.17 18 The State's interest in finality is unusually strong in child-custody disputes. The grant of federal habeas would prolong uncertainty for children such as the Lehman sons, possibly lessening their chances of adoption. It is undisputed that children require secure, stable, long-term, continuous relationships with their parents or foster parents. There is little that can be as detrimental to a child's sound development as uncertainty over whether he is to remain in his current "home," under the care of his parents or foster parents, especially when such uncertainty is prolonged. Extended uncertainty would be inevitable in many cases if federal courts had jurisdiction to relitigate state custody decisions.18 III 19 Petitioner argues that habeas corpus should be available to her because it has been used as a procedure in child-custody cases in various States and in England. She notes that, in Jones v. Cunningham, 371 U.S., at 238-240, 83 S.Ct., at 374-375, the Court indicated that in construing the habeas statute, reference may be made to the common law and to practices in the States and in England. It is true that habeas has been used in child-custody cases in England and in many of the States. See id., at 239-240, and nn. 8, 12, and 13, 83 S.Ct., at 375-376, and nn. 8, 12, and 13, citing Ford v. Ford, 371 U.S. 187, 83 S.Ct. 273, 9 L.Ed.2d 240 (1962); Boardman v. Boardman, 135 Conn. 124, 138, 62 A.2d 521, 528 (1948); Ex parte Swall, 36 Nev. 171, 174, 134 P. 96, 97 (1913); Ex parte M'Clellan, 1 Dowl. 81 (K.B.1831); Earl of Westmeath v. Countess of Westmeath, as set out in reporter's footnote in Lyons v. Blenkin, 1 Jac. 245, 264, 37 Eng.Rep. 842, 848 (Ch. 1821). As these cases illustrate, the term "custody" in 28 U.S.C. § 2255—authorizing federal-court collateral review of federal decisions—could be construed to include the type of custody the Lehman children are subject to, since they are in foster homes pursuant to court orders. But reliance on what may be appropriate within the federal system or within a state system is of little force where—as in this case—a state judgment is attacked collaterally in a federal court. It is one thing to use a proceeding called "habeas corpus" in resolving child-custody disputes within a single system obligated to resolve such disputes. The question in such a case may be which procedure is most appropriate. The system is free to set time limits on the bringing of such actions as well as to impose other requirements to ensure finality and a speedy resolution of disputes in cases involving child custody or termination of parental rights. In this case, however, petitioner would have the federal judicial system entertain a writ that is not time-barred to challenge collaterally a final judgment entered in a state judicial system. In Sylvander v. New England Home for Little Wanderers, the Court of Appeals for the First Circuit gave a compelling answer to this argument: 20 "Federal habeas when applied to persons under state control is a procedure of unique potency within the federal-state framework, having far different and more far-reaching consequences than a state's utilization of habeas within its own system. State utilization of habeas to test the legal custody of a child is part of the fabric of its reserved jurisdiction over child custody matters. If a habeas remedy were not provided, some other procedure would be needed to effectuate the state's substantive interest in these relationships. It is purely a matter of procedural detail whether the remedy is called 'habeas' or something else. The federal government, however, has no parallel substantive interest in child custody matters that federal habeas would serve. The sole federal interest is in the constitutional issues collateral to such disputes. At bottom, the question is whether these constitutional issues can be adequately raised through the usual channels—appeal, certiorari and the civil rights statutes—or whether the vehicle of federal habeas, with its unique features, is required." 584 F.2d, at 1111. IV 21 The considerations in a child-custody case are quite different from those present in any prior case in which this Court has sustained federal-court jurisdiction under § 2254. The federal writ of habeas corpus, representing as it does a profound interference with state judicial systems and the finality of state decisions, should be reserved for those instances in which the federal interest in individual liberty is so strong that it outweighs federalism and finality concerns.19 Congress has indicated no intention that the reach of § 2254 encompass a claim like that of petitioner. We therefore hold that § 2254 does not confer federal-court jurisdiction. The decision below, affirming the denial of a writ of habeas corpus, therefore is affirmed. 22 It is so ordered. 23 Justice BLACKMUN, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 24 Although I can sympathize with what the Court seeks to accomplish in this case today, I cannot reconcile myself to its holding that "§ 2254 does not confer federal-court jurisdiction," ante, this page, to consider collateral challenges to state-court judgments involuntarily terminating parental rights. In my view, the literal statutory requisites for the exercise of § 2254 federal habeas corpus jurisdiction are satisfied here—in particular, the requirement that petitioner's children must be "in custody." Because I believe the Court could have achieved much the same practical result in this area without decreeing a complete withdrawal of federal jurisdiction, I respectfully dissent. 25 * Justice Black, speaking for a unanimous Court in Jones v. Cunningham, 371 U.S. 236, 243, 83 S.Ct. 373, 377, 9 L.Ed.2d 285 (1963), observed that the federal writ of habeas corpus "is not now and never has been a static, narrow, formalistic remedy." 26 "While limiting its availability to those 'in custody,' the statute does not attempt to mark the boundaries of 'custody' nor in any way other than by use of that word attempt to limit the situations in which the writ can be used. To determine whether habeas corpus could be used to test the legality of a given restraint on liberty, this Court has generally looked to common-law usages and the history of habeas corpus both in England and in this country." Id., at 238, 83 S.Ct., at 374. 27 Even a brief historical examination of common-law usages teaches two lessons: first, for centuries, the English and American common-law courts have had the undisputed power to issue writs of habeas corpus ordering the release of children from unlawful custody; and, second, those courts have exercised broad discretion in deciding whether or not to invoke that power in a given case. English common-law courts traditionally were authorized to order the release of minor children from unlawful custody.1 Relying on the English tradition, American state courts very early asserted their own power to issue common-law habeas writs in child-custody matters. See generally Oaks, Habeas Corpus in the States—1776-1865, 32 U.Chi.L.Rev. 243, 270-274 (1965). 28 While acknowledging that "habeas has been used in child-custody cases in England and in many of the States," ante, at 514, the Court suggests that a state court derives its authority to issue a writ of habeas corpus in such disputes not from the common law, but from " 'the fabric of its reserved jurisdiction over child custody matters.' " Ante, at 515, quoting Sylvander v. New England Home for Little Wanderers, 584 F.2d 1103, 1111 (CA1 1978). While such a conclusion is not illogical, it is surely ahistorical. Contrary to the Court's suggestion, it is not " 'purely a matter of procedural detail whether the [state] remedy is called "habeas" or something else.' " Ibid. A state court's traditional power to issue a writ of habeas corpus to free a confined child always has been derived directly from the nature of the writ, not from any reserved jurisdiction over child-custody matters.2 29 The codification of the writ into federal law indicates no congressional intent to contract its common-law scope. The sparse legislative history of the predecessor statute to 28 U.S.C. § 2254, the Habeas Corpus Act of February 5, 1867, ch. 28, § 1, 14 Stat. 385, gave "no indication whatever that the bill intended to change the general nature of the classical habeas jurisdiction." Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 Harv.L.Rev. 441, 476-477 (1963) (emphasis in original).3 Nor, since that time, has this Court ever held that the congressional purpose originally underlying the statute barred use of the federal writ to free children from unlawful state custody.4 The Court's more recent precedents have firmly established § 2254's "in custody" requirement as its most flexible element, stressing that the test of "custody" is not present physical restraint, but whether "there are other restraints on a man's liberty, restraints not shared by the public generally, which have been thought sufficient in the English-speaking world to support the issuance of habeas corpus." Jones v. Cunningham, 371 U.S., at 240, 83 S.Ct., at 376. 30 Today the Court bows in the direction of this historical precedent only by leaving open the possible availability of federal habeas if a child is actually confined in a state institution, rather than in the custody of a foster parent pursuant to a court order.5 Ante, at 511, n. 12. At the same time, however, the Court presents three reasons why federal courts lack "jurisdiction" to issue writs of federal habeas corpus to release children from the latter form of state custody. Not one of these reasons is sufficient to erect a jurisdictional, as opposed to a prudential, bar to federal habeas relief.6 31 First, the Court restrictively reads Jones v. Cunningham, supra; Carafas v. LaVallee, 391 U.S. 234, 88 S.Ct. 1556, 20 L.Ed.2d 554 (1968); and Hensley v. Municipal Court, 411 U.S. 345, 93 S.Ct. 1571, 36 L.Ed.2d 294 (1973), and deems those three cases to involve only substantial and unusual restraints suffered by individuals "as a result of a state-court criminal conviction." Ante, at 511 (emphasis added). Yet those decisions plainly drew no distinction between criminal and civil detention. To the contrary, they declared in unusually broad and expansive language that the habeas writ must be widely available "as a remedy for severe restraints on individual liberty." Hensley v. Municipal Court, 411 U.S., at 351, 93 S.Ct., at 1575.7 Indeed, for its interpretation of the statutory "custody" requirement, Jones itself expressly relied on the fact that at common law, English courts had "permitted a parent to use habeas corpus to obtain his children from the other parent, even though the children were 'not under imprisonment, restraint, or duress of any kind.' " 371 U.S., at 239, 83 S.Ct., at 375, citing Earl of Westmeath v. Countess of Westmeath, as set out in a reporter's footnote in Lyons v. Blenkin, 1 Jac. 245, 264, 37 Eng.Rep. 842, 848 (Ch. 1821). 32 Second, the Court argues that children living with foster parents somehow are not in the State's "custody" because "they suffer no unusual restraints not imposed on other children." Ante, at 511. Yet because unadopted children whose ties with their natural parents have been severed are wards of the State, the State decides where they will live, reserves the right to move them to new physical settings at will, and consents to their marriage, their enlistment in the Armed Forces, as well as all major decisions regarding medical, psychiatric, and surgical treatment. See Tr. of Oral Arg. 7 and 18, citing 23 Pa.Cons.Stat. § 2521(c) (1980). 33 This Court has found the statutory concept of "custody" broad enough to confer jurisdiction on federal courts to hear and determine habeas applications from petitioners who have freely traveled across state borders while released on their own recognizance, Hensley v. Municipal Court, supra, and who are on unattached, inactive Army Reserve duty, Strait v. Laird, 406 U.S. 341, 92 S.Ct. 1693, 32 L.Ed.2d 141 (1972). Under these precedents, I have difficulty finding that minor children, who as state wards are fully subject to state-court custody orders, are not sufficiently and peculiarly restrained to be deemed "in custody" for the purposes of the habeas corpus statute. Cf. Braden v. 30th Judicial Circuit Court of Ky., 410 U.S. 484, 501, 93 S.Ct. 1123, 1133, 35 L.Ed.2d 443 (1973) (opinion concurring in result); Hensley v. Municipal Court, 411 U.S., at 353, 93 S.Ct., at 1576 (opinion concurring in result). Equally important, "[w]ith respect to the argument, that some force or improper restraint must be used, in order to authorize the Court in removing an infant from the custody of any one," historical authorities show that "it is not necessary that any force or restraint should exist on the part of the person having the custody of the infant towards it." Ex parte M'Clellan, 1 Dowl. 81, 84 (K.B.1831) (Patteson, J.). Accord: R. Hurd, A Treatise on the Right of Personal Liberty and on the Writ of Habeas Corpus 455 (1858); W. Church, A Treatise of the Writ of Habeas Corpus 555 (1886). 34 Third, the Court asserts that "[f]ederalism concerns and the exceptional need for finality in child-custody disputes argue strongly against the grant of Ms. Lehman's petition." Ante, at 512. While I am fully sensitive to these concerns, once again I cannot understand how they deprive federal courts of statutory jurisdiction to entertain habeas petitions. Although the Court's decisions involving collateral attack by state prisoners against state criminal convictions have recognized similar federalism and finality concerns, they have never held that those interests erect jurisdictional bars to relief. To the contrary, the Court has carefully separated the question whether federal courts have the power to issue a writ of habeas corpus from the question whether "in some circumstances considerations of comity and concerns for the orderly administration of criminal justice require a federal court to forgo the exercise of its habeas corpus power." Francis v. Henderson, 425 U.S. 536, 539, 96 S.Ct. 1708, 1710, 48 L.Ed.2d 149 (1976). See also Stone v. Powell, 428 U.S. 465, 478, n. 11, and 495, n. 37, 96 S.Ct. 3037, 3044, n. 11, and 3052, n. 37, 49 L.Ed.2d 1067 (1976) ("Our decision does not mean that the federal court lacks jurisdiction over such a claim . . ."); Fay v. Noia, 372 U.S. 391, 425-426, 83 S.Ct. 822, 842, 9 L.Ed.2d 827 (1963). II 35 As a matter of history and precedent, then, "[t]here can be no question of a federal district court's power to entertain an application for a writ of habeas corpus in a case such as this. . . . The issue . . . goes rather to the appropriate exercise of that power." Francis v. Henderson, 425 U.S., at 538-539, 96 S.Ct., at 1709-1710. Cf. 648 F.2d 135, 155 (CA3 1981) (en banc) (Seitz, C. J., concurring). In my view, the difficult discretionary question in this case is whether, 11 years after petitioner voluntarily relinquished her sons to state custody and 4 years after the involuntary termination of her parental rights was affirmed on direct appeal, she remains a proper "next friend" to apply for the federal habeas writ on behalf of her natural children. 36 As amended in 1948, the federal habeas statute permits a third-party application for habeas relief only if it is "signed and verified by the person for whose relief it is intended or by someone acting in his behalf." 28 U.S.C. § 2242 (emphasis added). "But one who so signs and verifies does not thereby become the applicant"; the person under detention remains the real party in interest. Nash ex rel. Hashimoto v. MacArthur, 87 U.S.App.D.C. 268, 270, 184 F.2d 606, 608 (1950), cert. denied, 342 U.S. 838, 72 S.Ct. 64, 96 L.Ed. 634 (1951). For that reason, the "next friend" application has been uncommonly granted, see Weber v. Garza, 570 F.2d 511, 513-514 (CA5 1978); United States ex rel. Bryant v. Houston, 273 F. 915, 916 (CA2 1921); United States ex rel. Funaro v. Watchorn, 164 F. 152, 153 (CC SDNY 1908), and has not been made available automatically even to the natural parents of a habeas petitioner. See, e.g., Evans v. Bennett, 467 F.Supp. 1108, 1110 (SD Ala.1979). Cf. Gilmore v. Utah, 429 U.S. 1012, 1013-1014, 97 S.Ct. 436, 50 L.Ed.2d 632 (1976) (BURGER, C. J., concurring). 37 Historically, the English common-law courts permitted parents to use the habeas writ to obtain custody of a child as a way of vindicating their own rights. American common-law courts, however, soon relied on Lord Mansfield's language in King v. Delaval, see n. 1, supra, to resolve custody disputes initiated by way of a habeas writ in a manner best adapted to serve the welfare of the child. See Oaks, Habeas Corpus in the States—1776-1865, 32 U.Chi.L.Rev., at 270 and 274. Thus, the American common-law rule came to be that "the parent stands in court as the real party in interest, upon his natural right of parent; but he is liable to be defeated by his own wrongdoing or unfitness and by the demands and requirements of society that the well-being of the child shall be deemed paramount to the natural rights of an unworthy parent." Hand, Habeas Corpus Proceedings for the Release of Infants, 56 Cent.L.J. 385, 389 (1903). 38 Similarly, the federal courts have interpreted the writ as being available only to serve the best interest of the child. " 'When a party comes here, using the privilege of acting on the behalf and as the next friend of infants, it is his bounden duty to show that he really acts for the benefit of the infants, and not to promote purposes of his own.' " King v. McLean Asylum of Massachusetts General Hospital, 64 F. 331, 356 (CA1 1894), quoting Sale v. Sale, 1 Beav. 586, 587, 48 Eng.Rep. 1068, 1069 (1839). "[I]n such cases the court exercises a discretion in the interest of the child to determine what care and custody are best for it in view of its age and requirements." New York Foundling Hospital v. Gatti, 203 U.S. 429, 439, 27 S.Ct. 53, 55, 51 L.Ed. 254 (1906).8 39 Against this historical background, then, I find most telling the Court's observation that "Ms. Lehman simply seeks to relitigate, through federal habeas, not any liberty interest of her sons, but the interest in her own parental rights." Ante, at 511. As the Court notes, the record reveals no evidence that any of the sons wanted to return to their natural mother. Seeante, at 504, n. 2. Moreover, in filing her federal habeas petition, petitioner expressly did not seek to disturb the state trial court's factual findings. See Brief for Petitioner 6. Those findings made "absolutely clear . . . that, by reason of her very limited social and intellectual development combined with her five-year separation from the children, [petitioner] is incapable of providing minimal care, control and supervision for the three children. Her incapacity cannot and will not be remedied." In re William L., 477 Pa. 322, 345, 383 A.2d 1228, 1239-1240, cert. denied sub nom. Lehman v. Lycoming County Children's Services, 439 U.S. 880, 99 S.Ct. 216, 58 L.Ed. 192 (1978). 40 On such a record, I believe that the District Court could have found, as a discretionary matter, that petitioner had not made a sufficient showing that she acted in the interests of the children to warrant issuing her the writ as their "next friend."9 Indeed, I believe that the common-law habeas corpus tradition would have supported recognition of broad district court discretion to withhold the writ in all but the most extraordinary cases, where the district court had strong reason to believe both that the conditions of the child's confinement unconstitutionally constrained that child's liberty, and that release of the child to his natural parent very likely would serve the child's best interest. 41 Such a ruling would not have been inconsistent with the Court's decision today, which expressly bases denial of habeas relief on a need to reserve the federal writ "for those instances in which the federal interest in individual liberty is so strong that it outweighs federalism and finality concerns." Ante, at 516. Indeed, I cannot understand why the Court's explicit balancing approach yields a strict jurisdictional bar. A discretionary limit would have allowed the writ to issue only in those very rare cases that demanded its unique "capacity to . . . cut through barriers of form and procedural mazes." Harris v. Nelson, 394 U.S. 286, 291, 89 S.Ct. 1082, 1086, 22 L.Ed.2d 281 (1969). Because the Court overrides contrary history and precedent to find that habeas jurisdiction does not lie, I dissent. 1 Petitioner has never been married. The fathers to these sons voluntarily have relinquished their parental rights in state-court proceedings. 2 There was no evidence that any of the sons wanted to return to their mother. See Tr. 82, 117-118, 122-125, 127-129. 3 It has now been over a decade since the sons were removed from the custody of their mother. Frank, the oldest, is now 18, and the case is moot with respect to him since he is free to seek adoption by anyone, including his natural mother. See Tr. of Oral Arg. 25-26. The other two sons, Bill and Mark, are now 12 and 16 respectively. 4 The judge relied on the Pennsylvania statute which provides, in relevant part: "The rights of a parent in regard to a child may be terminated after a petition filed on any of the following grounds: * * * * * "(2) The repeated and continued incapacity, abuse, neglect or refusal of the parent has caused the child to be without essential parental care, control or subsistence necessary for his physical or mental well-being and the conditions and causes of the incapacity, abuse, neglect or refusal cannot or will not be remedied by the parent." 23 Pa.Cons.Stat. § 2511(a) (1980). 5 This decision appeared to have been a strategic one, making possible, in the event this Court did not grant plenary review, the filing of a habeas corpus petition in federal district court without any problem of res judicata on the federal issues as a result of this Court's summary affirmance or dismissal of the appeal for want of a substantial federal question. At oral argument, however, petitioner's lawyer also explained that he was confused as to whether he could appeal both the facial attack on the statute and the challenge to the statute as applied, and had therefore chosen the more conservative route of seeking a petition for a writ of certiorari on both issues. See Tr.Oral Arg. 21-22. 6 Chief Judge Seitz filed a separate concurring opinion. He found the case "most difficult," noting that "the literal statutory requirements for exercise of section 2254 federal habeas corpus jurisdiction can be said to be satisfied." 648 F.2d, at 155. But he nevertheless concurred in the result because habeas corpus has never been used to challenge state child-custody decisions, and "such a major departure from traditional uses of federal habeas corpus to challenge state-court judgments" should "await a congressional directive on the matter." Id., at 156. Judge Rosenn, joined by two other judges, dissented. He stressed that "[t]he total extinction of a familial relationship between children and their biological parents is the most drastic measure that a state can impose, short of criminal sanctions." Id., at 163. Judge Gibbons also filed a dissenting opinion, arguing that there was federal subject-matter jurisdiction and that habeas corpus should be an available remedy because a decision terminating parental rights has on-going effects. Id., at 177. 7 The federal courts have split on this issue. Only one court other than the Court of Appeals for the Third Circuit has addressed the question in a full opinion; in Sylvander v. New England Home for Little Wanderers, 584 F.2d 1103 (1978), the Court of Appeals for the First Circuit held that habeas corpus could not be used to avoid the finality of prior state-court child-custody proceedings, with a rationale much like Judge Garth's in the instant case. Other federal courts have assumed—without full analysis—that habeas jurisdiction lies. See Davis v. Page, 640 F.2d 599, 602 (CA5 1981) (en banc); Rowell v. Oesterle, 626 F.2d 437 (CA5 1980). 8 See 28 U.S.C. § 2241 (empowering federal judges to grant such writs; subsection (c) provides that "[t]he writ of habeas corpus shall not extend to a prisoner unless. . .") (emphasis added); see also 28 U.S.C. § 2254(b) ("An application for a writ of habeas corpus in behalf of a person in custody pursuant to the judgment of a State court shall . . . be granted unless [state remedies have not been exhausted or are not available, or there are] circumstances rendering such process ineffective to protect the rights of the prisoner") (emphasis added). 9 When habeas corpus is made available by a federal court to challenge custody by federal entities, federalism concerns are not implicated. The only relevant question then is what federal remedy may be available. The grant of habeas relief in such instances—e.g., Strait v. Laird, 406 U.S. 341, 92 S.Ct. 1693, 32 L.Ed.2d 141 (1972) (inactive Army Reserve member allowed to bring habeas petition to challenge his military obligation); Brownell v. Tom We Shung, 352 U.S. 180, 182-184, 77 S.Ct. 252, 254-255, 1 L.Ed.2d 225 (1956) (alien allowed to use habeas to challenge his exclusion from the United States)—is not precedent for the use of federal habeas to challenge judgments of state courts. As Judge Garth noted in his decision below: "[T]he writ assumes even more profound implications when its operation cuts across the federal and state judicial systems. In this latter context, the writ empowers a single federal judge to overrule determinations of federal issues which have been adjudicated by the highest court of a state." 648 F.2d, at 139. Jurisdiction to challenge both state and federal judgments is conferred by § 2241. But § 2254, conferring general jurisdiction to consider collateral attacks on state judgments, has no relevance to federal habeas proceedings challenging federal custody of nonprisoners. Thus, federal decisions made pursuant to § 2241 constitute no authority for the claim of jurisdiction under § 2254 in this case. 10 In Hensley, the State would have placed the petitioner behind bars, but was prevented by a stay entered by the state trial court that subsequently was extended by two Justices of this Court. 411 U.S., at 351, 93 S.Ct., at 1575. Thus, although Hensley held the writ to be available in a case in which there was no actual custody in a state penal institution at the time the writ was filed, the extension was in the context of a person who had a strong claim to be treated as a prisoner for jurisdictional purposes. 11 See Hensley, 411 U.S., at 345, 93 S.Ct., at 1571 ("This case requires us to determine whether a person released on his own recognizance is 'in custody' within the meaning of the federal habeas corpus statute. . ."); Carafas v. LaVallee, 391 U.S., at 238, 88 S.Ct., at 1559 (similar); Jones v. Cunningham, 371 U.S., at 236, 83 S.Ct., at 373 (similar). 12 We express no view as to the availability of federal habeas when a child is actually confined in a state institution rather than being at liberty in the custody of a foster parent pursuant to a court order. 13 At the hearing before the Pennsylvania trial court, petitioner's lawyer actually stated "[t]his is not a custody proceeding . . . ." Tr. 67. 14 The Court has considered constitutional challenges to custody or parental-rights proceedings, but these cases have reached the Court on direct review of the final state-court decision, not on federal habeas. See, e.g., Santosky v. Kramer, 455 U.S. 745, 102 S.Ct. 1388, 71 L.Ed.2d 599 (1982). Justice BLACKMUN's dissenting opinion states that the legislative history, though admittedly sparse, supports its interpretation of the scope of § 2254 because "[t]he codification of the writ into federal law indicates no congressional intent to contract its common-law scope." See post, at 518. But the dissenting opinion cites no legislative history relevant to state-court custody decisions. Moreover, for at least 100 years after passage of the statute in 1867, the writ was not used in child-custody cases. This history strongly suggests that the extension of federal habeas corpus to state custody cases was never contemplated by Congress, nor understood by the Bar to have been an available remedy. 15 Petitioner maintains that the approval of habeas jurisdiction in this case may be limited. She suggests that it could be available only when the State takes the child away from its natural parents, but not when the State simply determines custody in a routine intrafamily dispute. It is not apparent that such distinctions are possible, either in legal theory or as a practical matter. The circumstances of custody vary widely, though in each disputed case the child is in the custody of one person—over the objections of someone else—by order of a state court. We see no principled basis for distinguishing between the claim of a natural parent and the claim of grandparents or even the claim of an unrelated person who has been given legal custody that is challenged by a third party. Moreover, the arguments of res judicata and federalism apply with equal force in every collateral attack on a state custody decision in a federal court. 16 The dissent suggests that comity and federalism concerns cannot inform a court's construction of a statute in determining a question of jurisdiction over certain kinds of cases. Post, at 522-523. But in Fair Assessment in Real Estate Assn. v. McNary, 454 U.S. 100, 102 S.Ct. 177, 70 L.Ed.2d 271 (1981), precisely those concerns lead this Court to conclude that 42 U.S.C. § 1983 does not confer jurisdiction on the federal courts to hear suits for tax refunds when state law provides an adequate remedy. 17 In his decision below, Judge Garth expressed similar views: "While the ability to avoid res judicata is an extraordinary characteristic of habeas when the relitigation takes place within the same judicial system—that is, when a state court entertains the writ on behalf of a person in custody pursuant to the judgment of a court of that same state—the writ assumes even more profound implications when its operation cuts across the federal and state judicial systems. [T]he assumption of habeas jurisdiction by a federal court on behalf of a party complaining of a judgment rendered against him by a state court, represents an unparalleled assertion of federal authority over the state judicial system. Such an intrusion upon state judicial authority deeply implicates the principles of comity and may impair the smooth workings of our federal system. "The awesome power of the writ to avoid res judicata, and its implications for our federalism, demand that its use be confined to its proper role: the preservation of individual liberty and the relief from unlawful custody." 648 F.2d, at 139. 18 There is also the danger that "if litigation expenses mount, social workers and charitable organizations . . . may well become less willing to seek placements for children over their parents' objections, whether rational or irrational, even though in their honest judgment the child's best interests demand it." Sylvander v. New England Home for Little Wanderers, 584 F.2d, at 1112. 19 In Hensley, this Court observed: "The custody requirement of the habeas corpus statute is designed to preserve the writ of habeas corpus as a remedy for severe restraints on individual liberty. Since habeas corpus is an extraordinary remedy whose operation is to a large extent uninhibited by traditional rules of finality and federalism, its use has been limited to cases of special urgency, leaving more conventional remedies for cases in which the restraints on liberty are neither severe nor immediate." 411 U.S., at 351, 93 S.Ct., at 1575. 1 In King v. Delaval, 3 Burr. 1434, 1436-1437, 97 Eng.Rep. 913, 914 (K.B.1763), Lord Mansfield declared: "In cases of writs of habeas corpus directed to private persons 'to bring up infants,' the Court is bound, ex debito justitiae, to set the infant free from an improper restraint: but they are not bound to deliver them over to any body nor to give them any privilege. This must be left to their discretion, according to the circumstances that shall appear before them. * * * * * "The true rule is, 'that the Court[s] are to judge upon the circumstances of the particular case; and to give their directions accordingly.' " 2 See, e.g., R. Hurd, A Treatise on the Right of Personal Liberty and on the Writ of Habeas Corpus 454-521 (1858); W. Church, A Treatise of the Writ of Habeas Corpus 555-557 (1886); L. Hochheimer, A Treatise on the Law Relating to the Custody of Infants 156-162 (1887); H. Clark, The Law of Domestic Relations in the United States 578-580 (1968); Bantz, Habeas Corpus—Custody of Infant, 15 Cent.L.J. 281, 281-282 (1882) (footnote omitted) (The writ "is granted on the application of the parent, guardian or master to inquire into the legality of the restraint of the child, ward, etc.; and its object is, not to enforce a right of custody, but to remove unlawful restraint"); Hand, Habeas Corpus Proceedings for the Release of Infants, 56 Cent.L.J. 385, 388 (1903) ("Whenever the parent seeks to recover a child from any third person, the approved remedy is habeas campus."); Oaks, Habeas Corpus in the States—1776-1865, 32 U.Chi.L.Rev. 243, 273 (1965). 3 To the contrary, the legislators plainly intended to enact "a bill of the largest liberty" that would not "restrain the writ of habeas corpus at all" and would "enable the courts of the United States to enforce the freedom of the wife and children of soldiers of the United States, and also to enforce the liberty of all persons." Cong.Globe, 39th Cong., 1st Sess., 4151 (1866) (remarks of Rep. Lawrence) (emphasis added). 4 In Wales v. Whitney, 114 U.S. 564, 5 S.Ct. 1050, 29 L.Ed. 277 (1885), which early delineated the forms of "custody" subject to the writ, the Court stated: "There is no very satisfactory definition to be found in the adjudged cases of the character of the restraint or imprisonment suffered by a party applying for the writ of habeas corpus, which is necessary to sustain the writ. . . . Wives restrained by husbands, children withheld from the proper parent or guardian, persons held under arbitrary custody by private individuals, as in a madhouse, as well as those under military control, may all become proper subjects of relief by the writ of habeas corpus." Id., at 571, 5 S.Ct., at 1053 (emphasis added). In In re Burrus, 136 U.S. 586, 10 S.Ct. 850, 34 L.Ed. 1500 (1890), and Matters v. Ryan, 249 U.S. 375, 39 S.Ct. 315, 63 L.Ed. 654 (1919), this Court refused to permit the federal writ to be used in private child-custody disputes, stating in dictum that matters of family law are reserved for the States. As the Court correctly notes, however, ante, at 511-512, those cases dismissed habeas petitions for want of federal-question jurisdiction, and thus did not generally deny the federal courts power to issue writs of habeas corpus in child-custody cases. 5 Notwithstanding their conclusions that federal habeas "jurisdiction" does not lie in child-custody cases, neither plurality opinion in the Court of Appeals was willing to foreclose a federal court's power to issue the writ to secure a child's release from state custody under extreme circumstances. See 648 F.2d 135, 144 (CA3 1981) (en banc) (Garth, J., announcing the judgment of the court) (" 'Were [the Lehman boys] incarcerated in a state home, or were there other issues making this truly a struggle for liberty by one imprisoned under the aegis of the state,' the writ might well be available") (citation omitted; emphasis in original); id., at 152, n. 35 (Adams, J., concurring) ("Should the children be in state custody against their will, it is even possible that habeas would be an appropriate vehicle for the legal attack"). See also Sylvander v. New England Home for Little Wanderers, 584 F.2d 1103, 1113 (CA1 1978) (leaving open the possibility that federal habeas corpus might be available to free a child from state custody). 6 I disagree with the Court's announcement that "no principled basis" would exist for limiting the approval of federal habeas jurisdiction in child-custody disputes. Ante, at 512, n. 15. When, as in this case, the State both initiates the challenged judicial proceedings and remains the ongoing legal custodian of the child, subject to state-court order, the state action is plainly sufficient to create "custody in violation of the Constitution . . . of the United States" for § 2254 purposes. Intrafamily disputes, however, are ordinarily privately initiated and result in private custody. If a child's natural parents disputed custody, and a state court awarded custody to one of them, a legitimate question would arise whether that person "may fairly be said to be a state actor." Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S.Ct. 2744, 2754, 73 L.Ed.2d 482 (1982). See also Dennis v. Sparks, 449 U.S. 24, 28, 101 S.Ct. 183, 186, 66 L.Ed.2d 185 (1980) ("Of course, merely resorting to the courts and being on the winning side of a lawsuit does not make a party a . . . joint actor with the judge"). 7 See Jones v. Cunningham, 371 U.S., at 243, 83 S.Ct., at 377, (the "grand purpose" of the writ is "the protection of individuals against erosion of their right to be free from wrongful restraints upon their liberty"); Carafas v. LaVallee, 391 U.S., at 238, 88 S.Ct., at 1559 (the "province" of the writ "is to provide an effective and speedy instrument by which judicial inquiry may be had into the legality of the detention of a person"); Hensley v. Municipal Court, 411 U.S., at 350, 93 S.Ct., at 1574 ("[W]e have consistently rejected interpretations of the habeas corpus statute that would suffocate the writ in stifling formalisms or hobble its effectiveness with the manacles of arcane and scholastic procedural requirements. . . . That same theme has indelibly marked our construction of the statute's custody requirement"). 8 Presiding over United States v. Green, 26 F.Cas. 30 (No. 15,256) (CC RI 1824), Justice Story concluded: "[T]he right of the father to have the custody of his infant child . . . is not on account of any absolute right of the father, but for the benefit of the infant, the law presuming it to be for his interest to be under the nurture and care of his natural protector, both for maintenance and education. When, therefore, the court is asked to lend its aid to put the infant in the custody of the father, and to withdraw him from other persons, it will look into all the circumstances, and ascertain whether it will be for the real permanent interests of the infant; and if the infant be of sufficient discretion it will also consult its personal wishes. . . . It is an entire mistake to suppose the court is at all events bound to deliver over the infant to his father, or that the latter has an absolute vested right in the custody." Id., at 31-32. 9 Petitioner's colorable claim that her own constitutional rights were infringed would not have entitled her automatically to serve as a "next friend." As Judge Adams' concurring opinion in the Court of Appeals observed: "Even if we assume that the statute under which the termination occurred, and which survived attack in the state courts, is unconstitutional, it is highly possible that Mrs. Lehman, in challenging the statute ostensibly on behalf of the children, may actually be asserting an interest that derogates from the child's interest. That is, the child's interest in a sound family environment that the state statute was intended to protect may not be properly represented by the parent's demand for family unity. . . . [W]hat is questionable here is her right to resort to a habeas petition, which can be framed only on behalf of her children." 648 F.2d, at 154 (footnote omitted). I disagree, however, with Judge Adams' conclusion that petitioner lacks "standing to assert [a federal habeas] action on behalf of the three children." Id., at 155. As Judge Rosenn correctly responded in dissent, petitioner plainly has standing in a constitutional sense to challenge the violation of her own rights. The question here, however, is whether "Ms. Lehman may not be the best—or even a proper—relator in this action." Id., at 156, n. 2. Cf. id., at 154, n. 47 (Adams, J., concurring).

01

458 U.S. 419 102 S.Ct. 3164 73 L.Ed.2d 868 Jean LORETTO, on behalf of Herself and all Others Similarly Situated, Appellantv.TELEPROMPTER MANHATTAN CATV CORP. et al. No. 81-244. Argued March 30, 1982. Decided June 30, 1982. Syllabus A New York statute provides that a landlord must permit a cable television (CATV) company to install its CATV facilities upon his property and may not demand payment from the company in excess of the amount determined by a State Commission to be reasonable. Pursuant to the statute, the Commission ruled that a one-time $1 payment was a reasonable fee. After purchasing a five-story apartment building in New York City, appellant landlord discovered that appellee CATV companies had installed cables on the building, both "crossovers" for serving other buildings and "noncrossovers" for serving appellant's tenants. Appellant then brought a class action for damages and injunctive relief in a New York state court, alleging, inter alia, that installation of the cables insofar as appellee companies relied on the New York statute constituted a taking without just compensation. Appellee New York City, which had granted the companies an exclusive franchise to provide CATV within certain areas of the city, intervened. Upholding the New York statute, the trial court granted summary judgment to appellees. The Appellate Division of the New York Supreme Court affirmed, and on further appeal the New York Court of Appeals also upheld the statute, holding that it serves the legitimate police power purpose of eliminating landlord fees and conditions that inhibit the development of CATV, which has important educational and community benefits. Rejecting appellant's argument that a physical occupation authorized by government is necessarily a taking, the court further held that the statute did not have an excessive economic impact upon appellant when measured against her aggregate property rights, did not interfere with any reasonable investment-backed expectations, and accordingly did not work a taking of appellant's property. Held : The New York statute works a taking of a portion of appellant's property for which she is entitled to just compensation under the Fifth Amendment, as made applicable to the States by the Fourteenth Amendment. Pp. 425-441. (a) When the "character of the governmental action," Penn Central Transportation Co. v. New York City, 438 U.S. 104, 124, 98 S.Ct. 2646, 2659, 57 L.Ed.2d 631, is a permanent physical occupation of real property, there is a taking to the extent of the occupation without regard to whether the action achieves an important public benefit or has only minimal economic impact on the owner. Pp. 426-435. (b) To the extent that the government permanently occupies physical property, it effectively destroys the owner's rights to possess, use, and dispose of the property. Moreover, the owner suffers a special kind of injury when a stranger invades and occupies the owner's property. Such an invasion is qualitatively more severe than a regulation of the use of property, since the owner may have no control over the timing, extent, or nature of the invasion. And constitutional protection for the rights of private property cannot be made to depend on the size of the area permanently occupied. Pp. 435-438. (c) Here, the cable installation on appellant's building constituted a taking under the traditional physical occupation test, since it involved a direct physical attachment of plates, boxes, wires, bolts, and screws to the building, completely occupying space immediately above and upon the roof and along the building's exterior wall. There is no constitutional difference between a crossover and noncrossover installation, since portions of the installation necessary for both types of installation permanently appropriated appellant's property. The fact that the New York statute applies only to buildings used as rental property does not make it simply a regulation of the use of real property. Physical occupation of one type of property but not another is no less a physical occupation. The New York statute does not purport to give the tenant any enforceable property rights with respect to CATV installation, and thus cannot be construed as merely granting a tenant a property right as an appurtenance to his leasehold. Application of the physical occupation rule in this case will not have dire consequences for the government's power to adjust landlord-tenant relationships, since it in no way alters the usual analysis governing a State's power to require landlords to comply with building codes. Pp. 438-440. 53 N.Y.2d 124, 440 N.Y.S.2d 843, 423 N.E.2d 320, reversed and remanded. Michael S. Gruen, New York City, for appellant. Erwin N. Griswold, Washington, D. C., for appellees. Justice MARSHALL delivered the opinion of the Court. 1 This case presents the question whether a minor but permanent physical occupation of an owner's property authorized by government constitutes a "taking" of property for which just compensation is due under the Fifth and Fourteenth Amendments of the Constitution. New York law provides that a landlord must permit a cable television company to install its cable facilities upon his property. N.Y.Exec.Law § 828(1) (McKinney Supp. 1981-1982). In this case, the cable installation occupied portions of appellant's roof and the side of her building. The New York Court of Appeals ruled that this appropriation does not amount to a taking. 53 N.Y.2d 124, 440 N.Y.S.2d 843, 423 N.E.2d 320 (1981). Because we conclude that such a physical occupation of property is a taking, we reverse. 2 * Appellant Jean Loretto purchased a five-story apartment building located at 303 West 105th Street, New York City, in 1971. The previous owner had granted appellees Teleprompter Corp. and Teleprompter Manhattan CATV (collectively Teleprompter)1 permission to install a cable on the building and the exclusive privilege of furnishing cable television (CATV) services to the tenants. The New York Court of Appeals described the installation as follows: 3 "On June 1, 1970 TelePrompter installed a cable slightly 4 less than one-half inch in diameter and of approximately 30 feet in length along the length of the building about 18 inches above the roof top, and directional taps, approximately 4 inches by 4 inches by 4 inches, on the front and rear of the roof. By June 8, 1970 the cable had been extended another 4 to 6 feet and cable had been run from the directional taps to the adjoining building at 305 West 105th Street." Id., at 135, 440 N.Y.S.2d, at 847, 423 N.E.2d, at 324. 5 Teleprompter also installed two large silver boxes along the roof cables. The cables are attached by screws or nails penetrating the masonry at approximately two-foot intervals, and other equipment is installed by bolts. 6 Initially, Teleprompter's roof cables did not service appellant's building. They were part of what could be described as a cable "highway" circumnavigating the city block, with service cables periodically dropped over the front or back of a building in which a tenant desired service. Crucial to such a network is the use of so-called "crossovers"—cable lines extending from one building to another in order to reach a new group of tenants.2 Two years after appellant purchased the building, Teleprompter connected a "noncrossover" line—i.e., one that provided CATV service to appellant's own tenants—by dropping a line to the first floor down the front of appellant's building. 7 Prior to 1973, Teleprompter routinely obtained authorization for its installations from property owners along the cable's route, compensating the owners at the standard rate of 5% of the gross revenues that Teleprompter realized from the particular property. To facilitate tenant access to CATV, the State of New York enacted § 828 of the Executive Law, effective January 1, 1973. Section 828 provides that a landlord may not "interfere with the installation of cable television facilities upon his property or premises," and may not demand payment from any tenant for permitting CATV, or demand payment from any CATV company "in excess of any amount which the [State Commission on Cable Television] shall, by regulation, determine to be reasonable."3 The landlord may, however, require the CATV company or the tenant to bear the cost of installation and to indemnify for any damage caused by the installation. Pursuant to § 828(1)(b), the State Commission has ruled that a one-time $1 payment is the normal fee to which a landlord is entitled. In the Matter of Implementation of Section 828 of the Executive Law, No. 90004, Statement of General Policy (New York State Commission on Cable Television, Jan. 15, 1976) (Statement of General Policy), App. 51-52; Clarification of General Policy (Aug. 27, 1976), App. 68-69. The Commission ruled that this nominal fee, which the Commission concluded was equivalent to what the landlord would receive if the property were condemned pursuant to New York's Transportation Corporations Law, satisfied constitutional requirements "in the absence of a special showing of greater damages attributable to the taking." Statement of General Policy, App. 52. 8 Appellant did not discover the existence of the cable until after she had purchased the building. She brought a class action against Teleprompter in 1976 on behalf of all owners of real property in the State on which Teleprompter has placed CATV components, alleging that Teleprompter's installation was a trespass and, insofar as it relied on § 828, a taking without just compensation. She requested damages and injunctive relief.4 Appellee City of New York, which has granted Teleprompter an exclusive franchise to provide CATV within certain areas of Manhattan, intervened. The Supreme Court, Special Term, granted summary judgment to Teleprompter and the city, upholding the constitutionality of § 828 in both crossover and noncrossover situations. 98 Misc.2d 944, 415 N.Y.S.2d 180 (1979). The Appellate Division affirmed without opinion. 73 A.D.2d 849, 422 N.Y.S.2d 550 (1979). 9 On appeal, the Court of Appeals, over dissent, upheld the statute. 53 N.Y.2d 124, 440 N.Y.S.2d 843, 423 N.E.2d 320 (1981). The court concluded that the law requires the landlord to allow both crossover and noncrossover installations but permits him to request payment from the CATV company under § 828(1)(b), at a level determined by the State Cable Commission, only for noncrossovers. The court then ruled that the law serves a legitimate police power purpose—eliminating landlord fees and conditions that inhibit the development of CATV, which has important educational and community benefits. Rejecting the argument that a physical occupation authorized by government is necessarily a taking, the court stated that the regulation does not have an excessive economic impact upon appellant when measured against her aggregate property rights, and that it does not interfere with any reasonable investment-backed expectations. Accordingly, the court held that § 828 does not work a taking of appellant's property. Chief Judge Cooke dissented, reasoning that the physical appropriation of a portion of appellant's property is a taking without regard to the balancing analysis courts ordinarily employ in evaluating whether a regulation is a taking. 10 In light of its holding, the Court of Appeals had no occasion to determine whether the $1 fee ordinarily awarded for a noncrossover installation was adequate compensation for the taking. Judge Gabrielli, concurring, agreed with the dissent that the law works a taking but concluded that the $1 presumptive award, together with the procedures permitting a landlord to demonstrate a greater entitlement, affords just compensation. We noted probable jurisdiction. 454 U.S. 938, 102 S.Ct. 472, 70 L.Ed.2d 246 (1981). II 11 The Court of Appeals determined that § 828 serves the legitimate public purpose of "rapid development of and maximum penetration by a means of communication which has important educational and community aspects," 53 N.Y.2d, at 143-144, 440 N.Y.S.2d, at 852, 423 N.E.2d, at 329, and thus is within the State's police power. We have no reason to question that determination. It is a separate question, however, whether an otherwise valid regulation so frustrates property rights that compensation must be paid. See Penn Central Transporta- tion Co. v. New York City, 438 U.S. 104, 127-128, 98 S.Ct. 2646, 2660-2661, 57 L.Ed.2d 631 (1978); Delaware, L. & W. R. Co. v. Morristown, 276 U.S. 182, 193, 48 S.Ct. 276, 278, 72 L.Ed. 523 (1928). We conclude that a permanent physical occupation authorized by government is a taking without regard to the public interests that it may serve. Our constitutional history confirms the rule, recent cases do not question it, and the purposes of the Takings Clause compel its retention. A. 12 In Penn Central Transportation Co. v. New York City, supra, the Court surveyed some of the general principles governing the Takings Clause. The Court noted that no "set formula" existed to determine, in all cases, whether compensation is constitutionally due for a government restriction of property. Ordinarily, the Court must engage in "essentially ad hoc, factual inquiries." Id., at 124, 98 S.Ct., at 2659. But the inquiry is not standardless. The economic impact of the regulation, especially the degree of interference with investment-backed expectations, is of particular significance. "So, too, is the character of the governmental action. A 'taking' may more readily be found when the interference with property can be characterized as a physical invasion by government, than when interference arises from some public program adjusting the benefits and burdens of economic life to promote the common good." Ibid. (citation omitted). 13 As Penn Central affirms, the Court has often upheld substantial regulation of an owner's use of his own property where deemed necessary to promote the public interest. At the same time, we have long considered a physical intrusion by government to be a property restriction of an unusually serious character for purposes of the Takings Clause. Our cases further establish that when the physical intrusion reaches the extreme form of a permanent physical occupation, a taking has occurred. In such a case, "the character of the government action" not only is an important factor in resolving whether the action works a taking but also is determinative. 14 When faced with a constitutional challenge to a permanent physical occupation of real property, this Court has invariably found a taking.5 As early as 1872, in Pumpelly v. Green Bay Co., 13 Wall. (80 U.S.) 166, 20 L.Ed. 557, this Court held that the defendant's construction, pursuant to state authority, of a dam which permanently flooded plaintiff's property constituted a taking. A unanimous Court stated, without qualification, that "where real estate is actually invaded by superinduced additions of water, earth, sand, or other material, or by having any artificial structure placed on it, so as to effectually destroy or impair its usefulness, it is a taking, within the meaning of the Constitution." Id., 13 Wall. (80 U.S.) at 181. Seven years later, the Court reemphasized the importance of a physical occupation by distinguishing a regulation that merely restricted the use of private property. In Northern Transportation Co. v. Chicago, 99 U.S. 635, 25 L.Ed. 336 (1879), the Court held that the city's construction of a temporary dam in a river to permit construction of a tunnel was not a taking, even though the plaintiffs were thereby denied access to their premises, because the obstruction only impaired the use of plaintiffs' property. The Court distinguished earlier cases in which permanent flooding of private property was regarded as a taking, e.g., Pumpelly, supra, as involving "a physical invasion of the real estate of the private owner, and a practical ouster of his possession." In this case, by contrast, "[n]o entry was made upon the plaintiffs' lot." 99 U.S., at 642. 15 Since these early cases, this Court has consistently distinguished between flooding cases involving a permanent physical occupation, on the one hand, and cases involving a more temporary invasion, or government action outside the owner's property that causes consequential damages within, on the other. A taking has always been found only in the former situation. See United States v. Lynah, 188 U.S. 445, 468-470, 23 S.Ct. 349, 356-357, 47 L.Ed. 539 (1903); Bedford v. United States, 192 U.S. 217, 225, 24 S.Ct. 238, 240, 48 L.Ed. 414 (1904); United States v. Cress, 243 U.S. 316, 327-328, 37 S.Ct. 380, 384-385, 61 L.Ed. 746 (1917); Sanguinetti v. United States, 264 U.S. 146, 149, 44 S.Ct. 264, 265, 68 L.Ed. 608 (1924) (to be a taking, flooding must "constitute an actual, permanent invasion of the land, amounting to an appropriation of, and not merely an injury to, the property"); United States v. Kansas City Life Ins. Co., 339 U.S. 799, 809-810, 70 S.Ct. 885, 890-891, 94 L.Ed. 1277 (1950). 16 In St. Louis v. Western Union Telegraph Co., 148 U.S. 92, 13 S.Ct. 485, 37 L.Ed. 380 (1893), the Court applied the principles enunciated in Pumpelly to a situation closely analogous to the one presented today. In that case, the Court held that the city of St. Louis could exact reasonable compensation for a telegraph company's placement of telegraph poles on the city's public streets. The Court reasoned: 17 "The use which the [company] makes of the streets is an exclusive and permanent one, and not one temporary, shifting and in common with the general public. The ordinary traveler, whether on foot or in a vehicle, passes to and fro along the streets, and his use and occupation thereof are temporary and shifting. The space he occupies one moment he abandons the next to be occupied by any other traveller. . . . But the use made by the telegraph company is, in respect to so much of the space as it occupies with its poles, permanent and exclusive. It as effectually and permanently dispossesses the general public as if it had destroyed that amount of ground. Whatever benefit the public may receive in the way of transportation of messages, that space is, so far as respects its actual use for purposes of highway and personal travel, wholly lost to the public. . . . 18 * * * * * 19 ". . . It matters not for what that exclusive appropriation is taken, whether for steam railroads or street railroads, telegraphs or telephones, the state may if it chooses exact from the party or corporation given such exclusive use pecuniary compensation to the general public for being deprived of the common use of the portion thus appropriated." Id., at 98-99, 101-102, 13 S.Ct., at 488-489 (emphasis added).6 20 Similarly, in Western Union Telegraph Co. v. Pennsylvania R. Co., 195 U.S. 540, 25 S.Ct. 133, 49 L.Ed. 312 (1904), a telegraph company constructed and operated telegraph lines over a railroad's right of way. In holding that federal law did not grant the company the right of eminent domain or the right to operate the lines absent the railroad's consent, the Court assumed that the invasion of the telephone lines would be a compensable taking. Id., at 570, 25 S.Ct., at 141 (the right-of-way "cannot be appropriated in whole or in part except upon the payment of compensation"). Later cases, relying on the character of a physical occupation, clearly establish that permanent occupations of land by such installations as telegraph and telephone lines, rails, and underground pipes or wires are takings even if they occupy only relatively insubstantial amounts of space and do not seriously interfere with the landowner's use of the rest of his land. See, e.g., Lovett v. West Va. Central Gas Co., 65 W.Va. 739, 65 S.E. 196 (1909); Southwestern Bell Telephone Co. v. Webb, 393 S.W.2d 117, 121 (Mo.App.1965). Cf. Portsmouth Harbor Land & Hotel Co. v. United States, 260 U.S. 327, 43 S.Ct. 135, 67 L.Ed. 287 (1922). See generally 2 J. Sackman, Nichols' Law of Eminent Domain § 6.21 (rev. 3d ed. 1980).7 21 More recent cases confirm the distinction between a permanent physical occupation, a physical invasion short of an occupation, and a regulation that merely restricts the use of property. In United States v. Causby, 328 U.S. 256, 66 S.Ct. 1062, 90 L.Ed. 1206 (1946), the Court ruled that frequent flights immediately above a landowner's property constituted a taking, comparing such overflights to the quintessential form of a taking: 22 "If, by reason of the frequency and altitude of the flights, respondents could not use this land for any purpose, their loss would be complete. It would be as complete as if the United States had entered upon the surface of the land and taken exclusive possession of it." Id., at 261, 66 S.Ct., at 1065 (footnote omitted). 23 As the Court further explained, 24 "We would not doubt that, if the United States erected an elevated railway over respondents' land at the precise altitude where its planes now fly, there would be a partial taking, even though none of the supports of the structure rested on the land. The reason is that there would be an intrusion so immediate and direct as to subtract from the owner's full enjoyment of the property and to limit his exploitation of it." Id., at 264-265, 66 S.Ct., at 1067. 25 The Court concluded that the damages to the respondents "were not merely consequential. They were the product of a direct invasion of respondents' domain." Id., at 265-266, 66 S.Ct., at 1067-1068. See also Griggs v. Allegheny County, 369 U.S. 84, 82 S.Ct. 531, 7 L.Ed.2d 585 (1962). 26 Two wartime takings cases are also instructive. In United States v. Pewee Coal Co., 341 U.S. 114, 71 S.Ct. 670, 95 L.Ed. 809 (1951), the Court unanimously held that the Government's seizure and direction of operation of a coal mine to prevent a national strike of coal miners constituted a taking, though members of the Court differed over which losses suffered during the period of Government control were compensable. The plurality had little difficulty concluding that because there had been an "actual taking of possession and control," the taking was as clear as if the Government held full title and ownership. Id., at 116, 71 S.Ct., at 671 (plurality opinion of Black, J., with whom Frankfurter, Douglas, and Jackson, JJ., joined; no other Justice challenged this portion of the opinion). In United States v. Central Eureka Mining Co., 357 U.S. 155, 78 S.Ct. 1097, 2 L.Ed.2d 1228 (1958), by contrast, the Court found no taking where the Government had issued a wartime order requiring nonessential gold mines to cease operations for the purpose of conserving equipment and manpower for use in mines more essential to the war effort. Over dissenting Justice Harlan's complaint that "as a practical matter the Order led to consequences no different from those that would have followed the temporary acquisition of physical possession of these mines by the United States," id., at 181, 78 S.Ct., at 1110; the Court reasoned that "the Government did not occupy, use, or in any manner take physical possession of the gold mines or of the equipment connected with them." Id., at 165-166, 78 S.Ct., at 1102-1103. The Court concluded that the temporary though severe restriction on use of the mines was justified by the exigency of war.8 Cf. YMCA v. United States, 395 U.S. 85, 92, 89 S.Ct. 1511, 1515, 23 L.Ed.2d 117 (1969) ("Ordinarily, of course, government occupation of private property deprives the private owner of his use of the property, and it is this deprivation for which the Constitution requires compensation"). 27 Although this Court's most recent cases have not addressed the precise issue before us, they have emphasized that physical invasion cases are special and have not repudiated the rule that any permanent physical occupation is a taking. The cases state or imply that a physical invasion is subject to a balancing process, but they do not suggest that a permanent physical occupation would ever be exempt from the Takings Clause. 28 Penn Central Transportation Co. v. New York City, as noted above, contains one of the most complete discussions of the Takings Clause. The Court explained that resolving whether public action works a taking is ordinarily an ad hoc inquiry in which several factors are particularly significant—the economic impact of the regulation, the extent to which it interferes with investment-backed expectations, and the character of the governmental action. 438 U.S., at 124, 98 S.Ct., at 2659. The opinion does not repudiate the rule that a permanent physical occupation is a government action of such a unique character that it is a taking without regard to other factors that a court might ordinarily examine.9 29 In Kaiser Aetna v. United States, 444 U.S. 164, 100 S.Ct. 383, 62 L.Ed.2d 332 (1979), the Court held that the Government's imposition of a navigational servitude requiring public access to a pond was a taking where the landowner had reasonably relied on Government consent in connecting the pond to navigable water. The Court emphasized that the servitude took the landowner's right to exclude, "one of the most essential sticks in the bundle of rights that are commonly characterized as property." Id., at 176, 100 S.Ct., at 391. The Court explained: 30 "This is not a case in which the Government is exercising its regulatory power in a manner that will cause an insubstantial devaluation of petitioner's private property; rather, the imposition of the navigational servitude in this context will result in an actual physical invasion of the privately owned marina. . . . And even if the Government physically invades only an easement in property, it must nonetheless pay compensation. See United States v. Causby, 328 U.S. 256, 265 [66 S.Ct. 1062, 1067, 90 L.Ed. 1206] (1946); Portsmouth Co. v. United States, 260 U.S. 327 [43 S.Ct. 135, 67 L.Ed. 287] (1922)." Id., at 180, 100 S.Ct., at 393 (emphasis added). 31 Although the easement of passage, not being a permanent occupation of land, was not considered a taking per se, Kaiser Aetna reemphasizes that a physical invasion is a government intrusion of an unusually serious character.10 32 Another recent case underscores the constitutional distinction between a permanent occupation and a temporary physical invasion. In PruneYard Shopping Center v. Robins, 447 U.S. 74, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980), the Court upheld a state constitutional requirement that shopping center owners permit individuals to exercise free speech and petition rights on their property, to which they had already invited the general public. The Court emphasized that the State Constitution does not prevent the owner from restricting expressive activities by imposing reasonable time, place, and manner restrictions to minimize interference with the owner's commercial functions. Since the invasion was temporary and limited in nature, and since the owner had not exhibited an interest in excluding all persons from his property, "the fact that [the solicitors] may have 'physically invaded' [the owners'] property cannot be viewed as determinative." Id., at 84, 100 S.Ct., at 338.11 33 In short, when the "character of the governmental action," Penn Central, 438 U.S., at 124, 98 S.Ct., at 2659, is a permanent physical occupation of property, our cases uniformly have found a taking to the extent of the occupation, without regard to whether the action achieves an important public benefit or has only minimal economic impact on the owner. B 34 The historical rule that a permanent physical occupation of another's property is a taking has more than tradition to commend it. Such an appropriation is perhaps the most serious form of invasion of an owner's property interests. To borrow a metaphor, cf. Andrus v. Allard, 444 U.S. 51, 65-66, 100 S.Ct. 318, 326-327, 62 L.Ed.2d 210 (1979), the government does not simply take a single "strand" from the "bundle" of property rights: it chops through the bundle, taking a slice of every strand. 35 Property rights in a physical thing have been described as the rights "to possess, use and dispose of it." United States v. General Motors Corp., 323 U.S. 373, 378, 65 S.Ct. 357, 359, 89 L.Ed. 311 (1945). To the extent that the government permanently occupies physical property, it effectively destroys each of these rights. First, the owner has no right to possess the occupied space himself, and also has no power to exclude the occupier from possession and use of the space. The power to exclude has traditionally been considered one of the most treasured strands in an owner's bundle of property rights.12 See Kaiser Aetna, 444 U.S., at 179-180, 100 S.Ct., at 392-393; see also Restatement of Property § 7 (1936). Second, the permanent physical occupation of property forever denies the owner any power to control the use of the property; he not only cannot exclude others, but can make no nonpossessory use of the property. Although deprivation of the right to use and obtain a profit from property is not, in every case, independently sufficient to establish a taking, see Andrus v. Allard, supra, at 66, 100 S.Ct., at 327, it is clearly relevant. Finally, even though the owner may retain the bare legal right to dispose of the occupied space by transfer or sale, the permanent occupation of that space by a stranger will ordinarily empty the right of any value, since the purchaser will also be unable to make any use of the property. 36 Moreover, an owner suffers a special kind of injury when a stranger directly invades and occupies the owner's property. As Part II-A, supra, indicates, property law has long protected an owner's expectation that he will be relatively undisturbed at least in the possession of his property. To require, as well, that the owner permit another to exercise complete dominion literally adds insult to injury. See Michelman, Property, Utility, and Fairness: Comments on the Ethical Foundations of "Just Compensation" Law, 80 Harv.L.Rev. 1165, 1228, and n. 110 (1967). Furthermore, such an occupation is qualitatively more severe than a regulation of the use of property, even a regulation that imposes affirmative duties on the owner, since the owner may have no control over the timing, extent, or nature of the invasion. See n. 19, infra. 37 The traditional rule also avoids otherwise difficult line-drawing problems. Few would disagree that if the State required landlords to permit third parties to install swimming pools on the landlords' rooftops for the convenience of the tenants, the requirement would be a taking. If the cable installation here occupied as much space, again, few would disagree that the occupation would be a taking. But constitutional protection for the rights of private property cannot be made to depend on the size of the area permanently occupied.13 Indeed, it is possible that in the future, additional cable installations that more significantly restrict a landlord's use of the roof of his building will be made. Section 828 requires a landlord to permit such multiple installations.14 38 Finally, whether a permanent physical occupation has occurred presents relatively few problems of proof. The placement of a fixed structure on land or real property is an obvious fact that will rarely be subject to dispute. Once the fact of occupation is shown, of course, a court should consider the extent of the occupation as one relevant factor in determining the compensation due.15 For that reason, moreover, there is less need to consider the extent of the occupation in determining whether there is a taking in the first instance. C 39 Teleprompter's cable installation on appellant's building constitutes a taking under the traditional test. The installation involved a direct physical attachment of plates, boxes, wires, bolts, and screws to the building, completely occupying space immediately above and upon the roof and along the building's exterior wall.16 40 In light of our analysis, we find no constitutional difference between a crossover and a noncrossover installation. The portions of the installation necessary for both crossovers and noncrossovers permanently appropriate appellant's property. Accordingly, each type of installation is a taking. 41 Appellees raise a series of objections to application of the traditional rule here. Teleprompter notes that the law applies only to buildings used as rental property, and draws the conclusion that the law is simply a permissible regulation of the use of real property. We fail to see, however, why a physical occupation of one type of property but not another type is any less a physical occupation. Insofar as Teleprompter means to suggest that this is not a permanent physical invasion, we must differ. So long as the property remains residential and a CATV company wishes to retain the installation, the landlord must permit it.17 42 Teleprompter also asserts the related argument that the State has effectively granted a tenant the property right to have a CATV installation placed on the roof of his building, as an appurtenance to the tenant's leasehold. The short answer is that § 828(1)(a) does not purport to give the tenant any enforceable property rights with respect to CATV installation, and the lower courts did not rest their decisions on this ground.18 Of course, Teleprompter, not appellant's tenants, actually owns the installation. Moreover, the government does not have unlimited power to redefine property rights. See Webb's Fabulous Pharmacies, Inc. v. Beckwith, 449 U.S. 155, 164, 101 S.Ct. 446, 452, 66 L.Ed.2d 358 (1980) ("a State, by ipse dixit, may not transform private property into public property without compensation"). 43 Finally, we do not agree with appellees that application of the physical occupation rule will have dire consequences for the government's power to adjust landlord-tenant relationships. This Court has consistently affirmed that States have broad power to regulate housing conditions in general and the landlord-tenant relationship in particular without paying compensation for all economic injuries that such regulation entails. See, e.g., Heart of Atlanta Motel, Inc. v. United States, 379 U.S. 241, 85 S.Ct. 348, 13 L.Ed.2d 258 (1964) (discrimination in places of public accommodation); Queenside Hills Realty Co. v. Saxl, 328 U.S. 80, 66 S.Ct. 850, 90 L.Ed. 1096 (1946) (fire regulation); Bowles v. Willingham, 321 U.S. 503, 64 S.Ct. 641, 88 L.Ed. 892 (1944) (rent control); Home Building & Loan Assn. v. Blaisdell, 290 U.S. 398, 54 S.Ct. 231, 78 L.Ed. 413 (1934) (mortgage moratorium); Edgar A. Levy Leasing Co. v. Siegel, 258 U.S. 242, 42 S.Ct. 289, 66 L.Ed. 595 (1922) (emergency housing law); Block v. Hirsh, 256 U.S. 135, 41 S.Ct. 458, 65 L.Ed. 865 (1921) (rent control). In none of these cases, however, did the government authorize the permanent occupation of the landlord's property by a third party. Consequently, our holding today in no way alters the analysis governing the State's power to require landlords to comply with building codes and provide utility connections, mailboxes, smoke detectors, fire extinguishers, and the like in the common area of a building. So long as these regulations do not require the landlord to suffer the physical occupation of a portion of his building by a third party, they will be analyzed under the multifactor inquiry generally applicable to nonpossessory governmental activity. See Penn Central Transportation Co. v. New York City, 438 U.S. 104, 98 S.Ct. 2646, 57 L.Ed.2d 631 (1978).19 III 44 Our holding today is very narrow. We affirm the traditional rule that a permanent physical occupation of property is a taking. In such a case, the property owner entertains a historically rooted expectation of compensation, and the character of the invasion is qualitatively more intrusive than perhaps any other category of property regulation. We do not, however, question the equally substantial authority upholding a State's broad power to impose appropriate restrictions upon an owner's use of his property. 45 Furthermore, our conclusion that § 828 works a taking of a portion of appellant's property does not presuppose that the fee which many landlords had obtained from Teleprompter prior to the law's enactment is a proper measure of the value of the property taken. The issue of the amount of compensation that is due, on which we express no opinion, is a matter for the state courts to consider on remand.20 46 The judgment of the New York Court of Appeals is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. 47 It is so ordered. 48 Justice BLACKMUN, with whom Justice BRENNAN and Justice WHITE join, dissenting. 49 If the Court's decisions construing the Takings Clause state anything clearly, it is that "[t]here is no set formula to determine where regulation ends and taking begins." Goldblatt v. Town of Hempstead, 369 U.S. 590, 594, 82 S.Ct. 987, 990, 8 L.Ed.2d 130 (1962).1 50 In a curiously anachronistic decision, the Court today acknowledges its historical disavowal of set formulae in almost the same breath as it constructs a rigid per se takings rule: "a permanent physical occupation authorized by government is a taking without regard to the public interests that it may serve." Ante, at 426. To sustain its rule against our recent precedents, the Court erects a strained and untenable distinction between "temporary physical invasions," whose constitutionality concededly "is subject to a balancing process," and "permanent physical occupations," which are "taking[s] without regard to other factors that a court might ordinarily examine." Ante, at 432. 51 In my view, the Court's approach "reduces the constitutional issue to a formalistic quibble" over whether property has been "permanently occupied" or "temporarily invaded." Sax, Takings and the Police Power, 74 Yale L.J. 36, 37 (1964). The Court's application of its formula to the facts of this case vividly illustrates that its approach is potentially dangerous as well as misguided. Despite its concession that "States have broad power to regulate . . . the landlord-tenant relationship . . . without paying compensation for all economic injuries that such regulation entails," ante, at 440, the Court uses its rule to undercut a carefully considered legislative judgment concerning landlord-tenant relationships. I therefore respectfully dissent. 52 * Before examining the Court's new takings rule, it is worth reviewing what was "taken" in this case. At issue are about 36 feet of cable one-half inch in diameter and two 4" x 4" x 4" metal boxes. Jointly, the cable and boxes occupy only about one-eighth of a cubic foot of space on the roof of appellant's Manhattan apartment building. When appellant purchased that building in 1971, the "physical invasion" she now challenges had already occurred.2 Appellant did not bring this action until about five years later, demanding 5% of appellee Teleprompter's gross revenues from her building, and claiming that the operation of N.Y.Exec.Law § 828 (McKinney Supp. 1981-1982) "took" her property. The New York Supreme Court, the Appellate Division, and the New York Court of Appeals all rejected that claim, upholding § 828 as a valid exercise of the State's police power. The Court of Appeals held that 53 "the State may proscribe a trespass action by landlords generally against a cable TV company which places a cable and other fixtures on the roof of any landlord's building, in order to protect the right of the tenants of rental property, who will ultimately have to pay any charge a landlord is permitted to collect from the cable TV company, to obtain TV service in their respective apartments." 53 N.Y.2d 124, 153, 440 N.Y.S.2d 843, 858, 423 N.E.2d 320, 335 (1981). 54 In so ruling, the court applied the multifactor balancing test prescribed by this Court's recent Takings Clause decisions. Those decisions teach that takings questions should be resolved through "essentially ad hoc, factual inquiries," Kaiser Aetna v. United States, 444 U.S. 164, 175, 100 S.Ct. 383, 390, 62 L.Ed.2d 332 (1979), into "such factors as the character of the governmental action, its economic impact, and its interference with reasonable investment-backed expectations." PruneYard Shopping Center v. Robins, 447 U.S. 74, 83, 100 S.Ct. 2035, 2042, 64 L.Ed.2d 741 (1980). See 53 N.Y.2d, at 144-151, 440 N.Y.S.2d, at 853-858, 423 N.E.2d, at 330-334. 55 The Court of Appeals found, first, that § 828 represented a reasoned legislative effort to arbitrate between the interests of tenants and landlords and to encourage development of an important educational and communications medium.3 Id., at 143-145, 440 N.Y.S.2d, at 852-853, 423 N.E.2d, at 329-330. Moreover, under PruneYard Shopping Center v. Robins, 447 U.S., at 83-84, 100 S.Ct., at 2042, the fact that § 828 authorized Teleprompter to make a minor physical intrusion upon appellant's property was in no way determinative of the takings question. 53 N.Y.2d, at 146-147, 440 N.Y.S.2d, at 854, 423 N.E.2d, at 331.4 56 Second, the court concluded that the statute's economic impact on appellant was de minimis because § 828 did not affect the fair return on her property. 53 N.Y.2d, at 148-150, 440 N.Y.S.2d, at 855-856, 423 N.E.2d, at 332-333. Third, the statute did not interfere with appellant's reasonable investment-backed expectations. Id., at 150-151, 440 N.Y.S.2d, at 856-857, 423 N.E.2d, at 333-334. When appellant purchased the building, she was unaware of the existence of the cable. See n. 2, supra. Thus, she could not have invested in the building with any reasonable expectation that the one-eighth cubic foot of space occupied by the cable television installment would become income-productive. 53 N.Y.2d, at 155, 440 N.Y.S.2d, at 859, 423 N.E.2d, at 336. II 57 Given that the New York Court of Appeals' straightforward application of this Court's balancing test yielded a finding of no taking, it becomes clear why the Court now constructs a per se rule to reverse. The Court can escape the result dictated by our recent takings cases only by resorting to bygone precedents and arguing that "permanent physical occupations" somehow differ qualitatively from all other forms of government regulation. 58 The Court argues that a per se rule based on "permanent physical occupation" is both historically rooted, see ante, at 426-435, and jurisprudentially sound, see ante, at 435-438. I disagree in both respects. The 19th-century precedents relied on by the Court lack any vitality outside the agrarian context in which they were decided.5 But if, by chance, they have any lingering vitality, then, in my view, those cases stand for a constitutional rule that is uniquely unsuited to the modern urban age. Furthermore, I find logically untenable the Court's assertion that § 828 must be analyzed under a per se rule because it "effectively destroys" three of "the most treasured strands in an owner's bundle of property rights," ante, at 435. 59 The Court's recent Takings Clause decisions teach that nonphysical government intrusions on private property, such as zoning ordinances and other land-use restrictions, have become the rule rather than the exception. Modern government regulation exudes intangible "externalities" that may diminish the value of private property far more than minor physical touchings. Nevertheless, as the Court recognizes, it has "often upheld substantial regulation of an owner's use of his own property where deemed necessary to promote the public interest." Ante, at 426. See, e.g., Agins v. City of Tiburon, 447 U.S. 255, 100 S.Ct. 2138, 65 L.Ed.2d 106 (1980); Penn Central Transportation Co. v. New York City, 438 U.S. 104, 124-125, 98 S.Ct. 2646, 2659, 57 L.Ed.2d 631 (1978); Village of Euclid v. Ambler Realty Co., 272 U.S. 365, 47 S.Ct. 114, 71 L.Ed. 303 (1926). 60 Precisely because the extent to which the government may injure private interests now depends so little on whether or not it has authorized a "physical contact," the Court has avoided per se takings rules resting on outmoded distinctions between physical and nonphysical intrusions. As one commentator has observed, a takings rule based on such a distinction is inherently suspect because "its capacity to distinguish, even crudely, between significant and insignificant losses is too puny to be taken seriously." Michelman, Property, Utility, and Fairness: Comments on the Ethical Foundations of "Just Compensation" Law, 80 Harv.L.Rev. 1165, 1227 (1967). 61 Surprisingly, the Court draws an even finer distinction today between "temporary physical invasions" and "permanent physical occupations." When the government authorizes the latter type of intrusion, the Court would find "a taking without regard to the public interests" the regulation may serve. Ante, at 426. Yet an examination of each of the three words in the Court's "permanent physical occupation" formula illustrates that the newly-created distinction is even less substantial than the distinction between physical and nonphysical intrusions that the Court already has rejected. 62 First, what does the Court mean by "permanent"? Since all "temporary limitations on the right to exclude" remain "subject to a more complex balancing process to determine whether they are a taking," ante, at 435, n. 12, the Court presumably describes a government intrusion that lasts forever. But as the Court itself concedes, § 828 does not require appellant to permit the cable installation forever, but only "[s]o long as the property remains residential and a CATV company wishes to retain the installation." Ante, at 439. This is far from "permanent." 63 The Court reaffirms that "States have broad power to regulate housing conditions in general and the landlord-tenant relationship in particular without paying compensation for all economic injuries that such regulation entails." Ante, at 440. Thus, § 828 merely defines one of the many statutory responsibilities that a New Yorker accepts when she enters the rental business. If appellant occupies her own building, or converts it into a commercial property, she becomes perfectly free to exclude Teleprompter from her one-eighth cubic foot of roof space. But once appellant chooses to use her property for rental purposes, she must comply with all reasonable government statutes regulating the landlord-tenant relationship.6 If § 828 authorizes a "permanent" occupation, and thus works a taking "without regard to the public interests that it may serve," then all other New York statutes that require a landlord to make physical attachments to his rental property also must constitute takings, even if they serve indisputably valid public interests in tenant protection and safety.7 64 The Court denies that its theory invalidates these statutes, because they "do not require the landlord to suffer the physical occupation of a portion of his building by a third party." Ante, at 440. But surely this factor cannot be determinative, since the Court simultaneously recognizes that temporary invasions by third parties are not subject to a per se rule. Nor can the qualitative difference arise from the incidental fact that, under § 828, Teleprompter, rather than appellant or her tenants, owns the cable installation. Cf. ante, at 440, and n. 19. If anything, § 828 leaves appellant better off than do other housing statutes, since it ensures that her property will not be damaged esthetically or physically, see n. 4, supra, without burdening her with the cost of buying or maintaining the cable. 65 In any event, under the Court's test, the "third party" problem would remain even if appellant herself owned the cable. So long as Teleprompter continuously passed its electronic signal through the cable, a litigant could argue that the second element of the Court's formula—a "physical touching" by a stranger—was satisfied and that § 828 therefore worked a taking.8 Literally read, the Court's test opens the door to endless metaphysical struggles over whether or not an individual's property has been "physically" touched. It was precisely to avoid "permit[ting] technicalities of form to dictate consequences of substance," United States v. Central Eureka Mining Co., 357 U.S. 155, 181, 78 S.Ct. 1097, 1110, 2 L.Ed.2d 1228 (1958) (Harlan, J., dissenting), that the Court abandoned a "physical contacts" test in the first place. 66 Third, the Court's talismanic distinction between a continuous "occupation" and a transient "invasion" finds no basis in either economic logic or Takings Clause precedent. In the landlord-tenant context, the Court has upheld against takings challenges rent control statutes permitting "temporary" physical invasions of considerable economic magnitude. See, e.g., Block v. Hirsh, 256 U.S. 135, 41 S.Ct. 458, 65 L.Ed. 865 (1921) (statute permitting tenants to remain in physical possession of their apartments for two years after the termination of their leases). Moreover, precedents record numerous other "temporary" officially authorized invasions by third parties that have intruded into an owner's enjoyment of property far more deeply than did Teleprompter's long-unnoticed cable. See, e.g., PruneYard Shopping Center v. Robins, 447 U.S. 74, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980) (leafletting and demonstrating in busy shopping center); Kaiser Aetna v. United States, 444 U.S. 164, 100 S.Ct. 383, 62 L.Ed.2d 332 (1979) (public easement of passage to private pond); United States v. Causby, 328 U.S. 256, 66 S.Ct. 1062, 90 L.Ed. 1206 (1946) (noisy airplane flights over private land). While, under the Court's balancing test, some of these "temporary invasions" have been found to be takings, the Court has subjected none of them to the inflexible per se rule now adapted to analyze the far less obtrusive "occupation" at issue in the present case. Cf. ante, at 430-431, 432-435. 67 In sum, history teaches that takings claims are properly evaluated under a multifactor balancing test. By directing that all "permanent physical occupations" automatically are compensable, "without regard to whether the action achieves an important public benefit or has only minimal economic impact on the owner," ante, at 434-435, the Court does not further equity so much as it encourages litigants to manipulate their factual allegations to gain the benefit of its per se rule. Cf. n. 8, supra. I do not relish the prospect of distinguishing the inevitable flow of certiorari petitions attempting to shoehorn insubstantial takings claims into today's "set formula." B 68 Setting aside history, the Court also states that the permanent physical occupation authorized by § 828 is a per se taking because it uniquely impairs appellant's powers to dispose of, use, and exclude others from, her property. See ante, at 435-438. In fact, the Court's discussion nowhere demonstrates how § 828 impairs these private rights in a manner qualitatively different from other garden-variety landlord-tenant legislation. 69 The Court first contends that the statute impairs appellant's legal right to dispose of cable-occupied space by transfer and sale. But that claim dissolves after a moment's reflection. If someone buys appellant's apartment building, but does not use it for rental purposes, that person can have the cable removed, and use the space as he wishes. In such a case, appellant's right to dispose of the space is worth just as much as if § 828 did not exist. 70 Even if another landlord buys appellant's building for rental purposes, § 828 does not render the cable-occupied space valueless. As a practical matter, the regulation ensures that tenants living in the building will have access to cable television for as long as that building is used for rental purposes, and thereby likely increases both the building's resale value and its attractiveness on the rental market.9 71 In any event, § 828 differs little from the numerous other New York statutory provisions that require landlords to install physical facilities "permanently occupying" common spaces in or on their buildings. As the Court acknowledges, the States traditionally—and constitutionally—have exercised their police power "to require landlords to . . . provide utility connections, mailboxes, smoke detectors, fire extinguishers, and the like in the common area of a building." Ante, at 440. Like § 828, these provisions merely ensure tenants access to services the legislature deems important, such as water, electricity, natural light, telephones, intercommunication systems, and mail service. See n. 7, supra. A landlord's dispositional rights are affected no more adversely when he sells a building to another landlord subject to § 828, than when he sells that building subject only to these other New York statutory provisions. 72 The Court also suggests that § 828 unconstitutionally alters appellant's right to control the use of her one-eighth cubic foot of roof space. But other New York multiple dwelling statutes not only oblige landlords to surrender significantly larger portions of common space for their tenants' use, but also compel the landlord—rather than the tenants or the private installers—to pay for and to maintain the equipment. For example, New York landlords are required by law to provide and pay for mailboxes that occupy more than five times the volume that Teleprompter's cable occupies on appellant's building. See Tr. of Oral Arg. 42-43, citing N.Y.Mult.Dwell.Law § 57 (McKinney 1974). If the State constitutionally can insist that appellant make this sacrifice so that her tenants may receive mail, it is hard to understand why the State may not require her to surrender less space, filled at another's expense, so that those same tenants can receive television signals. 73 For constitutional purposes, the relevant question cannot be solely whether the State has interfered in some minimal way with an owner's use of space on her building. Any intelligible takings inquiry must also ask whether the extent of the State's interference is so severe as to constitute a compensable taking in light of the owner's alternative uses for the property.10 Appellant freely admitted that she would have had no other use for the cable-occupied space, were Teleprompter's equipment not on her building. See App. 97 (Deposition of Jean A. Loretto). 74 The Court's third and final argument is that § 828 has deprived appellant of her "power to exclude the occupier from possession and use of the space" occupied by the cable. Ante, at 435. This argument has two flaws. First, it unjustifiably assumes that appellant's tenants have no countervailing property interest in permitting Teleprompter to use that space.11 Second, it suggests that the New York Legislature may not exercise its police power to affect appellant's common-law right to exclude Teleprompter even from one-eighth cubic foot of roof space. But this Court long ago recognized that new social circumstances can justify legislative modification of a property owner's common-law rights, without compensation, if the legislative action serves sufficiently important public interests. See Munn v. Illinois, 94 U.S. 113, 134, 24 L.Ed. 77 (1877) ("A person has no property, no vested interest, in any rule of the common law. . . . Indeed, the great office of statutes is to remedy defects in the common law as they are developed, and to adapt it to the changes of time and circumstance"); United States v. Causby, 328 U.S., at 260-261, 66 S.Ct., at 1065 (In the modern world, "[c]ommon sense revolts at the idea" that legislatures cannot alter common-law ownership rights). 75 As the Court of Appeals recognized, § 828 merely deprives appellant of a common-law trespass action against Teleprompter, but only for as long as she uses her building for rental purposes, and as long as Teleprompter maintains its equipment in compliance with the statute. Justice MARSHALL recently and most aptly observed: 76 "[Appellant's] claim in this case amounts to no less than a suggestion that the common law of trespass is not subject to revision by the State . . . . If accepted, that claim would represent a return to the era of Lochner v. New York, 198 U.S. 45 [25 S.Ct. 539, 49 L.Ed. 937] (1905), when common-law rights were also found immune from revision by State or Federal Government. Such an approach would freeze the common law as it has been constructed by the courts, perhaps at its 19th-century state of development. It would allow no room for change in response to changes in circumstance. The Due Process Clause does not require such a result." PruneYard Shopping Center v. Robins, 447 U.S., at 93, 100 S.Ct., at 2047 (concurring opinion). III 77 In the end, what troubles me most about today's decision is that it represents an archaic judicial response to a modern social problem. Cable television is a new and growing, but somewhat controversial, communications medium. See Brief for New York State Cable Television Association as Amicus Curiae 6-7 (about 25% of American homes with televisions—approximately 20 million families—currently subscribe to cable television, with the penetration rate expected to double by 1990). The New York Legislature not only recognized, but also responded to, this technological advance by enacting a statute that sought carefully to balance the interests of all private parties. See nn. 3 and 4, supra. New York's courts in this litigation, with only one jurist in dissent, unanimously upheld the constitutionality of that considered legislative judgment. 78 This Court now reaches back in time for a per se rule that disrupts that legislative determination.12 Like Justice Black, I believe that "the solution of the problems precipitated by . . . technological advances and new ways of living cannot come about through the application of rigid constitutional restraints formulated and enforced by the courts." United States v. Causby, 328 U.S., at 274, 66 S.Ct., at 1072 (dissenting opinion). I would affirm the judgment and uphold the reasoning of the New York Court of Appeals. 1 Teleprompter Manhattan CATV was formerly a subsidiary, and is now a division, of Teleprompter Corp. 2 The Court of Appeals defined a "crossover" more comprehensively as occurring: "[W]hen (1) the line servicing the tenants in a particular building is extended to adjacent or adjoining buildings, (2) an amplifier which is placed on a building is used to amplify signals to tenants in that building and in a neighboring building or buildings, and (3) a line is placed on a building, none of the tenants of which are provided CATV service, for the purpose of providing service to an adjoining or adjacent building." 53 N.Y.2d, at 133, n. 6, 440 N.Y.S.2d, at 846, n. 6, 423 N.E.2d, at 323, n. 6. 3 New York Exec.Law § 828 (McKinney Supp. 1981-1982) provides in part: "1. No landlord shall "a. interfere with the installation of cable television facilities upon his property or premises, except that a landlord may require: "i. that the installation of cable television facilities conform to such reasonable conditions as are necessary to protect the safety, functioning and appearance of the premises, and the convenience and well-being of other tenants; "ii. that the cable television company or the tenant or a combination thereof bear the entire cost of the installation, operation or removal of such facilities; and "iii. that the cable television company agree to indemnify the landlord for any damage caused by the installation, operation or removal of such facilities. "b. demand or accept payment from any tenant, in any form, in exchange for permitting cable television service on or within his property or premises, or from any cable television company in exchange therefor in excess of any amount which the commission shall, by regulation, determine to be reasonable; or "c. discriminate in rental charges, or otherwise, between tenants who receive cable television service and those who do not." 4 Class-action status was granted in accordance with appellant's request, except that owners of single-family dwellings on which a CATV component had been placed were excluded. Notice to the class has been postponed, however, by stipulation. 5 Professor Michelman has accurately summarized the case law concerning the role of the concept of physical invasions in the development of takings jurisprudence: "At one time it was commonly held that, in the absence of explicit expropriation, a compensable 'taking' could occur only through physical encroachment and occupation. The modern significance of physical occupation is that courts, while they sometimes do hold nontrespassory injuries compensable, never deny compensation for a physical takeover. The one incontestable case for compensation (short of formal expropriation) seems to occur when the government deliberately brings it about that its agents, or the public at large, 'regularly' use, or 'permanently' occupy, space or a thing which theretofore was understood to be under private ownership." Michelman, Property, Utility, and Fairness: Comments on the Ethical Foundations of "Just Compensation" Law, 80 Harv.L.Rev. 1165, 1184 (1967) (emphasis in original; footnotes omitted). See also 2 J. Sackman, Nichols' Law of Eminent Domain 6-50, 6-51 (rev. 3d ed. 1980); L. Tribe, American Constitutional Law 460 (1978). For historical discussions, see 53 N.Y.2d, at 157-158, 440 N.Y.S.2d, at 860-861, 423 N.E.2d, at 337-338 (Cooke, C.J., dissenting); F. Bosselman, D. Callies, & J. Banta, The Taking Issue 51 (1973); Stoebuck, A General Theory of Eminent Domain, 47 Wash.L.Rev. 553, 600-601 (1972); Dunham, Griggs v. Allegheny County in Perspective: Thirty Years of Supreme Court Expropriation Law, 1962 S.Ct.Rev. 63, 82; Cormack, Legal Concepts in Cases of Eminent Domain, 41 Yale L.J. 221, 225 (1931). 6 The City of New York objects that this case only involved a city's right to charge for use of its streets, and not the power of eminent domain; the city could have excluded the company from any use of its streets. But the physical occupation principle upon which the right to compensation was based has often been cited as authority in eminent domain cases. See, e.g., Western Union Telegraph Co. v. Pennsylvania R. Co., 195 U.S. 540, 566-567, 25 S.Ct. 133, 139-140, 49 L.Ed. 312 (1904); California v. United States, 395 F.2d 261, 263, n. 4 (CA9 1968). Also, the Court squarely held that insofar as the company relied on a federal statute authorizing its use of post roads, an appropriation of state property would require compensation. St. Louis v. Western Union Telegraph Co., 148 U.S., at 101, 13 S.Ct., at 489. 7 Early commentators viewed a physical occupation of real property as the quintessential deprivation of property. See, e.g., 1 W. Blackstone, Commentaries *139; J. Lewis, Law of Eminent Domain in the United States 197 (1888) ("Any invasion of property, except in case of necessity . . ., either upon, above or below the surface, and whether temporary or permanent, is a taking : as by constructing a ditch through it, passing under it by a tunnel, laying gas, water or sewer pipes in the soil, or extending structures over it, as a bridge or telephone wire" (footnote omitted; emphasis in original)); 1 P. Nichols, Law of Eminent Domain 282 (2d ed. 1917). 8 Indeed, although dissenting Justice Harlan would have treated the restriction as if it were a physical occupation, it is significant that he relied on physical appropriation as the paradigm of a taking. See United States v. Central Eureka Mining Co., 357 U.S., at 181, 183-184, 78 S.Ct., at 1110, 1111-1112. 9 The City of New York and the opinion of the Court of Appeals place great emphasis on Penn Central's reference to a physical invasion "by government," 438 U.S., at 124, 98 S.Ct., at 2659, and argue that a similar invasion by a private party should be treated differently. We disagree. A permanent physical occupation authorized by state law is a taking without regard to whether the State, or instead a party authorized by the State, is the occupant. See, e.g., Pumpelly v. Green Bay Co., 13 Wall. (80 U.S.) 166, 20 L.Ed. 557 (1872). Penn Central simply holds that in cases of physical invasion short of permanent appropriation, the fact that the government itself commits an invasion from which it directly benefits is one relevant factor in determining whether a taking has occurred. 438 U.S., at 124, 128, 98 S.Ct., at 2659, 2661. 10 See also Andrus v. Allard, 444 U.S. 51, 100 S.Ct. 318, 62 L.Ed.2d 210 (1979). That case held that the prohibition of the sale of eagle feathers was not a taking as applied to traders of bird artifacts. "The regulations challenged here do not compel the surrender of the artifacts, and there is no physical invasion or restraint upon them. . . . In this case, it is crucial that appellees retain the rights to possess and transport their property, and to donate or devise the protected birds. . . . [L]oss of future profits—unaccompanied by any physical property restriction—provides a slender reed upon which to rest a takings claim." Id., at 65-66, 100 S.Ct., at 326-327. 11 Teleprompter's reliance on labor cases requiring companies to permit access to union organizers, see, e.g., Hudgens v. NLRB, 424 U.S. 507, 96 S.Ct. 1029, 47 L.Ed.2d 196 (1976); Central Hardware Co. v. NLRB, 407 U.S. 539, 92 S.Ct. 2238, 33 L.Ed.2d 122 (1972); NLRB v. Babcock & Wilcox Co., 351 U.S. 105, 76 S.Ct. 679, 100 L.Ed. 975 (1956), is similarly misplaced. As we recently explained: "[T]he allowed intrusion on property rights is limited to that necessary to facilitate the exercise of employees' § 7 rights [to organize under the National Labor Relations Act]. After the requisite need for access to the employer's property has been shown, the access is limited to (i) union organizers; (ii) prescribed non-working areas of the employer's premises; and (iii) the duration of the organization activity. In short, the principle of accommodation announced in Babcock is limited to labor organization campaigns, and the 'yielding' of property rights it may require is both temporary and limited." Central Hardware Co., supra, at 545, 92 S.Ct., at 2242. 12 The permanence and absolute exclusivity of a physical occupation distinguish it from temporary limitations on the right to exclude. Not every physical invasion is a taking. As PruneYard Shopping Center v. Robins, 447 U.S. 74, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980); Kaiser Aetna v. United States, 444 U.S. 164, 100 S.Ct. 383, 62 L.Ed.2d 332 (1979), and the intermittent flooding cases reveal, such temporary limitations are subject to a more complex balancing process to determine whether they are a taking. The rationale is evident: they do not absolutely dispossess the owner of his rights to use, and exclude others from, his property. The dissent objects that the distinction between a permanent physical occupation and a temporary invasion will not always be clear. Post, at 448. This objection is overstated, and in any event is irrelevant to the critical point that a permanent physical occupation is unquestionably a taking. In the antitrust area, similarly, this Court has not declined to apply a per se rule simply because a court must, at the boundary of the rule, apply the rule of reason and engage in a more complex balancing analysis. 13 In United States v. Causby, 328 U.S. 256, 66 S.Ct. 1062, 90 L.Ed. 1206 (1946), the Court approvingly cited Butler v. Frontier Telephone Co., 186 N.Y. 486, 79 N.E. 716 (1906), holding that ejectment would lie where a telephone wire was strung across the plaintiff's property without touching the soil. The Court quoted the following language: " '[A]n owner is entitled to the absolute and undisturbed possession of every part of his premises, including the space above, as much as a mine beneath. If the wire had been a huge cable, several inches thick and but a foot above the ground, there would have been a difference in degree, but not in principle. Expand the wire into a beam supported by posts standing upon abutting lots without touching the surface of plaintiff's land, and the difference would still be one of degree only. Enlarge the beam into a bridge, and yet space only would be occupied. Erect a house upon the bridge, and the air above the surface of the land would alone be disturbed.' " 328 U.S., at 265, n. 10, 66 S.Ct., at 1067, n. 10, quoting Butler Frontier Telephone Co., supra, at 491-492, 79 N.E., at 718. 14 Although the City of New York has granted an exclusive franchise to Teleprompter, it is not required to do so under state law, see N.Y.Exec.Law § 811 et seq. (McKinney Supp. 1981-1982), and future changes in technology may cause the city to reconsider its decision. Indeed, at present some communities apparently grant nonexclusive franchises. Brief for National Satellite Cable Association et al. as Amici Curiae 21. 15 In this case, the Court of Appeals noted testimony preceding the enactment of § 828 that the landlord's interest in excluding cable installation "consists entirely of insisting that some negligible unoccupied space remain unoccupied." 53 N.Y.2d, at 141, 440 N.Y.S.2d, at 851, 423 N.E.2d, at 328 (emphasis omitted). The State Cable Commission referred to the same testimony in establishing a $1 presumptive award. Statement of General Policy, App. 48. A number of the dissent's arguments—that § 828 "likely increases both the building's resale value and its attractiveness on the rental market," post, at 452, and that appellant might have no alternative use for the cable-occupied space, post, at 453-454 may also be relevant to the amount of compensation due. It should be noted, however, that the first argument is speculative and is contradicted by appellant's testimony that she and "the whole block" would be able to sell their buildings for a higher price absent the installation. App. 100. 16 It is constitutionally irrelevant whether appellant (or her predecessor in title) had previously occupied this space, since a "landowner owns at least as much of the space above the ground as he can occupy or use in connection with the land." United States v. Causby, supra, at 264, 66 S.Ct., at 1067. The dissent asserts that a taking of about one-eighth of a cubic foot of space is not of constitutional significance. Post, at 443. The assertion appears to be factually incorrect, since it ignores the two large silver boxes that appellant identified as part of the installation. App. 90; Loretto Affidavit in Support of Motion for Summary Judgment (Apr. 21, 1978), Appellants' Appendix in No. 8300/76 (N.Y.App.), p. 77. Although the record does not reveal their size, appellant states that they are approximately 18" x 12" x 6", Brief for Appellant 6 n.*, and appellees do not dispute this statement. The displaced volume, then, is in excess of 11/2 cubic feet. In any event, these facts are not critical: whether the installation is a taking does not depend on whether the volume of space it occupies is bigger than a breadbox. 17 It is true that the landlord could avoid the requirements of § 828 by ceasing to rent the building to tenants. But a landlord's ability to rent his property may not be conditioned on his forfeiting the right to compensation for a physical occupation. Teleprompter's broad "use-dependency" argument proves too much. For example, it would allow the government to require a landlord to devote a substantial portion of his building to vending and washing machines, with all profits to be retained by the owners of these services and with no compensation for the deprivation of space. It would even allow the government to requisition a certain number of apartments as permanent government offices. The right of a property owner to exclude a stranger's physical occupation of his land cannot be so easily manipulated. 18 We also decline to hazard an opinion as to the respective rights of the landlord and tenant under state law prior to enactment of § 828 to use the space occupied by the cable installation, an issue over which the parties sharply disagree. 19 If § 828 required landlords to provide cable installation if a tenant so desires, the statute might present a different question from the question before us, since the landlord would own the installation. Ownership would give the landlord rights to the placement, manner, use, and possibly the disposition of the installation. The fact of ownership is, contrary to the dissent, not simply "incidental," post, at 450; it would give a landlord (rather than a CATV company) full authority over the installation except only as government specifically limited that authority. The landlord would decide how to comply with applicable government regulations concerning CATV and therefore could minimize the physical, esthetic, and other effects of the installation. Moreover, if the landlord wished to repair, demolish, or construct in the area of the building where the installation is located, he need not incur the burden of obtaining the CATV company's cooperation in moving the cable. In this case, by contrast, appellant suffered injury that might have been obviated if she had owned the cable and could exercise control over its installation. The drilling and stapling that accompanied installation apparently caused physical damage to appellant's building. App. 83, 95-96, 104. Appellant, who resides in her building, further testified that the cable installation is "ugly." Id., at 99. Although § 828 provides that a landlord may require "reasonable" conditions that are "necessary" to protect the appearance of the premises and may seek indemnity for damage, these provisions are somewhat limited. Even if the provisions are effective, the inconvenience to the landlord of initiating the repairs remains a cognizable burden. 20 In light of our disposition of appellant's takings claim, we do not address her contention that § 828 deprives her of property without due process of law. 1 See Kaiser Aetna v. United States, 444 U.S. 164, 175, 100 S.Ct. 383, 390, 62 L.Ed.2d 332 (1979); Andrus v. Allard, 444 U.S. 51, 65, 100 S.Ct. 318, 326, 62 L.Ed.2d 210 (1979) ("There is no abstract or fixed point at which judicial intervention under the Takings Clause becomes appropriate"); Penn Central Transportation Co. v. New York City, 438 U.S. 104, 124, 98 S.Ct. 2646, 2659, 57 L.Ed.2d 631 (1978); United States v. Caltex, Inc., 344 U.S. 149, 156, 73 S.Ct. 200, 203, 97 L.Ed. 157 (1952) ("No rigid rules can be laid down to distinguish compensable losses from noncompensable losses"); Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 416, 43 S.Ct. 158, 160, 67 L.Ed. 322 (1922) (a takings question "is a question of degree—and therefore cannot be disposed of by general propositions"). 2 In January 1968, appellee Teleprompter signed a 5-year installation agreement with the building's previous owner in exchange for a flat fee of $50. Appellee installed both the 30-foot main cable and its 4- to 6-foot "crossover" extension in June 1970. For two years after taking possession of the building and the appurtenant equipment, appellant did not object to the cable's presence. Indeed, despite numerous inspections, appellant had never even noticed the equipment until Teleprompter first began to provide cable television service to one of her tenants. 53 N.Y.2d 124, 134-135, 440 N.Y.S.2d 843, 847, 423 N.E.2d 320, 324 (1981). Nor did appellant thereafter ever specifically ask Teleprompter to remove the components from her building. App. 107, 108, 110. Although the Court alludes to the presence of "two large silver boxes" on appellant's roof, ante, at 438, n. 16, the New York Court of Appeals' opinion nowhere mentions them, nor are their dimensions stated anywhere in the record. 3 The court found that the state legislature had enacted § 828 to "prohibit gouging and arbitrary action" by "landlords [who] in many instances have imposed extremely onerous fees and conditions on cable access to their buildings." 53 N.Y.2d, at 141, 440 N.Y.S.2d, at 851, 423 N.E.2d, at 328, citing testimony of Joseph C. Swidler, Chairman of the Public Service Commission, before the Joint Legislative Committee considering the CATV bill. Given the growing importance of cable television, the legislature decided that urban tenants' need for access to that medium justified a minor intrusion upon the landlord's interest, which "consists entirely of insisting that some negligible unoccupied space remain unoccupied. The tenant's interest clearly is more substantial, consisting of a right to receive (and perhaps send) communications from and to the outside world. In the electronic age, the landlord should not be able to preclude a tenant from obtaining CATV service (or to exact a surcharge for allowing the service) any more than he could preclude a tenant from receiving mail or telegrams directed to him." Ibid., citing Regulation of Cable Television by the State of New York, Report to the New York Public Service Commission by Commissioner William K. Jones 207 (1970). 4 Section 828 carefully regulates the cable television company's physical intrusion onto the landlord's property. If the landlord requests, the company must conform its installations "to such reasonable conditions as are necessary to protect the safety, functioning and appearance of the premises, and the convenience and well-being of other tenants." N.Y.Exec.Law § 828(1)(a)(i) (McKinney Supp.1981-1982). Furthermore, the company must "agree to indemnify the landlord for any damage caused by the installation, operation or removal of such facilities." § 828(1)(a)(iii). Finally, the statute authorizes the landlord to require either "the cable television company or the tenant or a combination thereof [to] bear the entire cost of the installation, operation or removal" of any equipment. § 828(1)(a)(ii). 5 The Court properly acknowledges that none of our recent takings decisions have adopted a per se test for either temporary physical invasions or permanent physical occupations. See ante, at 432-435, and 435 n. 12. While the Court relies on historical dicta to support its per se rule, the only holdings it cites fall into two categories: a number of cases involving flooding, ante, at 427-428, and St. Louis v. Western Union Telegraph Co., 148 U.S. 92, 13 S.Ct. 485, 37 L.Ed. 380 (1893), cited ante, at 428. In 1950, the Court noted that the first line of cases stands for "the principle that the destruction of privately owned land by flooding is 'a taking' to the extent of the destruction caused," and that those rulings had already "been limited by later decisions in some respects." United States v. Kansas City Life Ins. Co., 339 U.S. 799, 809-810, 70 S.Ct. 885, 890-891, 94 L.Ed. 1277. Even at the time of its decision, St. Louis v. Western Union Telegraph Co. addressed only the question "[w]hether the city has power to collect rental for the use of streets and public places" when a private company seeks exclusive use of land whose "use is common to all members of the public, and . . . [is] open equally to citizens of other States with those of the State in which the street is situate." 148 U.S., at 98-99, 13 S.Ct., at 487-488. On its face, that issue is distinct from the question here: whether appellant may extract from Teleprompter a fee for the continuing use of her roof space above and beyond the fee set by statute, namely, "any amount which the commission shall, by regulation, determine to be reasonable." N.Y.Exec.Law § 828(1)(b) (McKinney Supp.1982). 6 In my view, the fact that § 828 incidentally protects so-called "crossover" wires that do not currently serve tenants, see ante, at 422, n. 2, does not affect § 828's fundamental character as a piece of landlord-tenant legislation. As the Court recognizes, ante, at 422, crossovers are crucial links in the cable "highway," and represent the simplest and most economical way to provide service to tenants in a group of buildings in close proximity. Like the Court, I find "no constitutional difference between a crossover and a noncrossover installation," ante, at 438. Even assuming, arguendo, that the crossover extension in this case works a taking, I would be prepared to hold that the incremental governmental intrusion caused by that 4- to 6-foot wire, which occupies the cubic volume of a child's building block, is a de minimis deprivation entitled to no compensation. 7 See, e.g., N.Y.Mult.Dwell.Law § 35 (McKinney 1974) (requiring entrance doors and lights); § 36 (windows and skylights for public halls and stairs); § 50-a (Supp.1982) (locks and intercommunication systems); § 50-c (lobby attendants); § 51-a (peepholes); § 51-b (elevator mirrors); § 53 (fire escapes); § 57 (bells and mail receptacles); § 67(3) (fire sprinklers). See also Queenside Hills Realty Co. v. Saxl, 328 U.S. 80, 66 S.Ct. 850, 90 L.Ed. 1096 (1946) (upholding constitutionality of New York fire sprinkler provision). These statutes specify in far greater detail than § 828 what types of physical facilities a New York landlord must provide his tenants and where he must provide them. See, e.g., N.Y.Mult.Dwell.Law § 75 (McKinney 1974) (owners of multiple dwellings must provide "proper appliances to receive and distribute an adequate supply of water," including "a proper sink with running water and with a two-inch waste and trap"); § 35 (owners of multiple dwellings with frontage exceeding 22 feet must provide "at least two lights, one at each side of the entrance way, with an aggregate illumination of one hundred fifty watts or equivalent illumination"); § 50-a(2) (Supp.1981-1982) (owners of Class A multiple dwellings must provide intercommunication system "located at an automatic self-locking door giving public access to the main entrance hall or lobby"). Apartment building rooftops are not exempted. See § 62 (landlords must place parapet walls and guardrails on their roofs "three feet six inches or more in height above the level of such area"). 8 Indeed, appellant's counsel made precisely this claim at oral argument. Urging the rule which the Court now adopts, appellant's counsel suggested that a taking would result even if appellant owned the cable. "[T]he precise location of the easement [taken by Teleprompter changes] from the surface of the roof to inside the wire.. . . [T]he wire itself is owned by the landlord, but the cable company has the right to pass its signal through the wire without compensation to the landlord, for its commercial benefit." Tr. of Oral Arg. 15. 9 In her pretrial deposition, appellant conceded not only that owners of other apartment buildings thought that the cable's presence had enhanced the market value of their buildings, App. 102-103, but also that her own tenants would have been upset if the cable connection had been removed. Id., at 107, 108, 110. 10 For this reason, the Court provides no support for its per se rule by asserting that the State could not require landlords, without compensation, "to permit third parties to install swimming pools," ante, at 436, or vending and washing machines, ante, at 439, n. 17, for the convenience of tenants. Presumably, these more intrusive government regulations would create difficult takings problems even under our traditional balancing approach. Depending on the character of the governmental action, its economic impact, and the degree to which it interfered with an owner's reasonable investment-backed expectations, among other things, the Court's hypothetical examples might or might not constitute takings. These examples hardly prove, however, that a permanent physical occupation that works a de minimis interference with a private property interest is a taking per se. 11 It is far from clear that, under New York law, appellant's tenants would lack all property interests in the few square inches on the exterior of the building to which Teleprompter's cable and hardware attach. Under modern landlord-tenant law, a residential tenancy is not merely a possessory interest in specified space, but also a contract for the provision of a package of services and facilities necessary and appurtenant to that space. See R. Schoshinski, American Law of Landlord and Tenant § 3:14 (1980). A modern urban tenant's leasehold often includes not only contractual, but also statutory rights, including the rights to an implied warranty of habitability, rent control, and such services as the landlord is obliged by statute to provide. Cf. n. 7, supra. 12 Happily, the Court leaves open the question whether § 828 provides landlords like appellant sufficient compensation for their actual losses. See ante, at 441. Since the State Cable Television Commission's regulations permit higher than nominal awards if a landlord makes "a special showing of greater damages," App. 52, the concurring opinion in the New York Court of Appeals found that the statute awards just compensation. See 53 N.Y.2d, at 155, 440 N.Y.S.2d, at 859, 423 N.E.2d, at 336 ("[I]t is obvious that a landlord who actually incurs damage to his property or is restricted in the use to which he might put that property will receive compensation commensurate with the greater injury"). If, after the remand following today's decision, this minor physical invasion is declared to be a taking deserving little or no compensation, the net result will have been a large expenditure of judicial resources on a constitutional claim of little moment.

34

458 U.S. 591 102 S.Ct. 3503 73 L.Ed.2d 994 Richard W. VELDE, et al., petitioners,v.NATIONAL BLACK POLICE ASSOCIATION, INC., et al No. 80-1074 Supreme Court of the United States June 30, 1982 On writ of certiorari to the United States Court of Appeals for the District of Columbia Circuit. PER CURIAM. 1 The judgment is vacated, and the case is remanded to the United States Court of Appeals for the District of Columbia Circuit for further consideration in light of Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396 (1982). 2 Opinion on remand, 229 U.S.App.D.C. 255, 712 F.2d 569. 3 Justice POWELL and Justice STEVENS took no part in the consideration or decision of this case.

78

458 U.S. 613 102 S.Ct. 3272 73 L.Ed.2d 1012 Quentin ROGERS, et al., Appellants,v.Herman LODGE et al. No. 80-2100. Argued Feb. 23, 1982. Decided July 1, 1982. Rehearing Denied Oct. 4, 1982. See 459 U.S. 899, 103 S.Ct. 198. Syllabus Burke County, Ga., a large, predominately rural county, has an at-large system for electing members of its governing Board of Commissioners. No Negro has ever been elected to the Board. Appellee black citizens of the county filed a class action in Federal District Court, alleging that the at-large system of elections violated, inter alia, appellees' Fourteenth and Fifteenth Amendment rights by diluting the voting power of black citizens. Finding that blacks have always made up a substantial majority of the county's population but that they are a minority of the registered voters, that there had been bloc voting along racial lines, and that past discrimination had restricted the present opportunity of blacks to participate effectively in the political process, the District Court held that although the state policy behind the at-large electoral system was "neutral in origin," the policy was being maintained for invidious purposes in violation of appellees' Fourteenth and Fifteenth Amendment rights. The court then ordered the county to be divided into districts for purposes of electing County Commissioners. The Court of Appeals affirmed, holding that the District Court properly required appellees to prove that the at-large system was maintained for a discriminatory purpose, that the District Court's findings were not clearly erroneous, and that its conclusion that the at-large system was maintained for invidious purposes was "virtually mandated by the overwhelming proof." Held: 1. The Court of Appeals did not err in concluding that the District Court applied the proper legal standard, where it appears that the District Court demonstrated its understanding of the controlling standard by observing that a determination of discriminatory intent was "a requisite to a finding of unconstitutional vote dilution" under the Fourteenth and Fifteenth Amendments. Pp. 616-622. 2. Where neither the District Court's ultimate findings of intentional discrimination nor its subsidiary findings of fact appear to be clearly erroneous and such findings were agreed to by the Court of Appeals, this Court will not disturb the findings. Pp. 622-627. 3. Nor is there any reason to overturn the relief ordered by the District Court, where neither that court nor the Court of Appeals discerned any special circumstances that would militate against utilizing single-member districts. Pp.627-628. 5th Cir., 639 F.2d 1358, affirmed. E. Freeman Leverett, Elberton, Ga., for appellants. David F. Walbert, Atlanta, Ga., for appellees. Justice WHITE delivered the opinion of the Court. 1 The issue in this case is whether the at-large system of elections in Burke County, Ga., violates the Fourteenth Amendment rights of Burke County's black citizens. 2 * Burke County is a large, predominately rural county located in eastern Georgia. Eight hundred and thirty-one square miles in area,1 it is approximately two-thirds the size of the State of Rhode Island. According to the 1980 census, Burke County had a total population of 19,349, of whom 10,385, or 53.6%, were black.2 The average age of blacks living there is lower than the average age of whites and therefore whites constitute a slight majority of the voting age population. As of 1978, 6,373 persons were registered to vote in Burke County, of whom 38% were black.3 3 The Burke County Board of Commissioners governs the county. It was created in 1911, see 1911 Ga.Laws 310-311, and consists of five members elected at large to concurrent 4-year terms by all qualified voters in the county. The county has never been divided into districts, either for the purpose of imposing a residency requirement on candidates or for the purpose of requiring candidates to be elected by voters residing in a district. In order to be nominated or elected, a candidate must receive a majority of the votes cast in the primary or general election, and a runoff must be held if no candidate receives a majority in the first primary or general election. Ga.Code § 34-1513 (Supp.1980). Each candidate must run for a specific seat on the Board, Ga.Code § 34-1015 (1978), and a voter may vote only once for any candidate. No Negro has ever been elected to the Burke County Board of Commissioners. 4 Appellees, eight black citizens of Burke County filed this suit in 1976 in the United States District Court for the Southern District of Georgia. The suit was brought on behalf of all black citizens in Burke County. The class was certified in 1977. The complaint alleged that the county's system of at-large elections violates appellees' First, Thirteenth, Fourteenth, and Fifteenth Amendment rights, as well as their rights under 42 U.S.C. §§ 1971, 1973, and 1983, by diluting the voting power of black citizens. Following a bench trial at which both sides introduced extensive evidence, the court issued an order on September 29, 1978, stating that appellees were entitled to prevail and ordering that Burke County be divided into five districts for purposes of electing County Commissioners. App. to Juris. Statement 62a. The court later issued detailed findings of fact and conclusions of law in which it stated that while the present method of electing County Commissioners was "racially neutral when adopted, [it] is being maintained for invidious purposes" in violation of appellees' Fourteenth and Fifteenth Amendment rights. Id., at 71a, 96a. 5 The Court of Appeals affirmed. Lodge v. Buxton, 639 F.2d 1358 (CA5 1981). It stated that while the proceedings in the District Court took place prior to the decision in Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), the District Court correctly anticipated Mobile and required appellees to prove that the at-large voting system was maintained for a discriminatory purpose. 639 F.2d, at 1375-1376. The Court of Appeals also held that the District Court's findings were not clearly erroneous, and that its conclusion that the at-large system was maintained for invidious purposes was "virtually mandated by the overwhelming proof." Id., at 1380. We noted probable jurisdiction, 454 U.S. 811, 102 S.Ct. 86, 70 L.Ed.2d 80 (1981), and now affirm.4 II 6 At-large voting schemes and multimember districts tend to minimize the voting strength of minority groups by permitting the political majority to elect all representatives of the district. A distinct minority, whether it be a racial, ethnic, economic, or political group, may be unable to elect any representatives in an at-large election, yet may be able to elect several representatives if the political unit is divided into single-member districts. The minority's voting power in a multimember district is particularly diluted when bloc voting occurs and ballots are cast along strict majority-minority lines. While multimember districts have been challenged for "their winner-take-all aspects, their tendency to submerge minorities and to overrepresent the winning party," Whitcomb v. Chavis, 403 U.S. 124, 158-159, 91 S.Ct. 1858, 1877, 29 L.Ed.2d 363 (1971), this Court has repeatedly held that they are not unconstitutional per se. Mobile v. Bolden, supra, 446 U.S., at 66, 100 S.Ct., at 1499; White v. Regester, 412 U.S. 755, 765, 93 S.Ct. 2332, 2339, 37 L.Ed.2d 314 (1973); Whitcomb v. Chavis, supra, 403 U.S., at 142, 91 S.Ct., at 1868. The Court has recognized, however, that multimember districts violate the Fourteenth Amendment if "conceived or operated as purposeful devices to further racial discrimination" by minimizing, cancelling out or diluting the voting strength of racial elements in the voting population. Whitcomb v. Chavis, supra, 403 U.S., at 149, 91 S.Ct., at 1872. See also White v. Regester, supra, 412 U.S., at 765, 93 S.Ct., at 2339. Cases charging that multimember districts unconstitutionally dilute the voting strength of racial minorities are thus subject to the standard of proof generally applicable to Equal Protection Clause cases. Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597 (1976), and Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977), made it clear that in order for the Equal Protection Clause to be violated, "the invidious quality of a law claimed to be racially discriminatory must ultimately be traced to a racially discriminatory purpose." Washington v. Davis, supra, 426 U.S., at 240, 96 S.Ct., at 2048. Neither case involved voting dilution, but in both cases the Court observed that the requirement that racially discriminatory purpose or intent be proved applies to voting cases by relying upon, among others, Wright v. Rockefeller, 376 U.S. 52, 84 S.Ct. 603, 11 L.Ed.2d 512 (1964), a districting case, to illustrate that a showing of discriminatory intent has long been required in all types of equal protection cases charging racial discrimination. Arlington Heights, supra, 429 U.S., at 265, 97 S.Ct., at 563; Washington v. Davis, supra, 426 U.S., at 240, 96 S.Ct., at 2048.5 7 Arlington Heights and Washington v. Davis both rejected the notion that a law is invalid under the Equal Protection Clause simply because it may affect a greater proportion of one race than another. Arlington Heights, supra, 429 U.S., at 265, 97 S.Ct., at 563; Washington v. Davis, 426 U.S., at 242, 96 S.Ct., at 2049. However, both cases recognized that discriminatory intent need not be proved by direct evidence. "Necessarily, an invidious discriminatory purpose may often be inferred from the totality of the relevant facts, including the fact, if it is true, that the law bears more heavily on one race than another." Ibid. Thus determining the existence of a discriminatory purpose "demands a sensitive inquiry into such circumstantial and direct evidence of intent as may be available." Arlington Heights, supra, 429 U.S., at 266, 97 S.Ct., at 564. 8 In Mobile v. Bolden, supra, the Court was called upon to apply these principles to the at-large election system in Mobile, Ala. Mobile is governed by three commissioners who exercise all legislative, executive, and administrative power in the municipality. 446 U.S., at 59, 100 S.Ct., at 1495. Each candidate for the City Commission runs for one of three numbered posts in an at-large election and can only be elected by a majority vote. Id., at 59-60, 100 S.Ct., at 1495-96. Plaintiffs brought a class action on behalf of all Negro citizens of Mobile alleging that the at-large scheme diluted their voting strength in violation of several statutory and constitutional provisions. The District Court concluded that the at-large system "violates the constitutional rights of the plaintiffs by improperly restricting their access to the political process," Bolden v. Mobile, 423 F.Supp. 384, 399 (SD Ala.1976), and ordered that the commission form of government be replaced by a mayor and a nine-member City Council elected from single-member districts. Id., at 404. The Court of Appeals affirmed. 571 F.2d 238 (CA5 1978). This Court reversed. 9 Justice Stewart, writing for himself and three other Justices, noted that to prevail in their contention that the at-large voting system violates the Equal Protection Clause of the Fourteenth Amendment, plaintiffs had to prove the system was " 'conceived or operated as [a] purposeful devic[e] to further racial . . . discrimination.' " 446 U.S., at 66, 100 S.Ct., at 1499, quoting Whitcomb v. Chavis, supra, 403 U.S., at 149, 91 S.Ct., at 1872.6 Such a requirement "is simply one aspect of the basic principle that only if there is purposeful discrimination can there be a violation of the Equal Protection Clause of the Fourteenth Amendment," 446 U.S., at 66, 100 S.Ct., at 1499, and White v. Regester is consistent with that principle. 446 U.S., at 69, 100 S.Ct., at 1501. Another Justice agreed with the standard of proof recognized by the plurality. Id., at 101, 100 S.Ct., at 1517 (WHITE, J., dissenting). 10 The plurality went on to conclude that the District Court had failed to comply with this standard. The District Court had analyzed plaintiffs' claims in light of the standard which had been set forth in Zimmer v. McKeithen, 485 F.2d 1297 (CA5 1973), aff'd on other grounds sub nom. East Carroll Parish School Bd. v. Marshall, 424 U.S. 636, 96 S.Ct. 1083, 47 L.Ed. 296 (1975) (per curiam ).7 Zimmer set out a list of factors8 gleaned from Whitcomb v. Chavis, supra, and White v. Regester, supra, that a court should consider in assessing the constitutionality of at-large and multimember district voting schemes. Under Zimmer, voting dilution is established "upon proof of the existence of an aggregate of these factors." 485 F.2d, at 1305. 11 The plurality in Mobile was of the view that Zimmer was "decided upon the misunderstanding that it is not necessary to show a discriminatory purpose in order to prove a violation of the Equal Protection Clause—that proof of a discriminatory effect is sufficient." 446 U.S., at 71, 100 S.Ct., at 1502. The plurality observed that while "the presence of the indicia relied on in Zimmer may afford some evidence of a discriminatory purpose," the mere existence of those criteria is not a substitute for a finding of discriminatory purpose. Id., at 73, 100 S.Ct., at 1503. The District Court's standard in Mobile was likewise flawed. Finally, the plurality concluded that the evidence upon which the lower courts had relied was "insufficient to prove an unconstitutionally discriminatory purpose in the present case." Ibid. Justice STEVENS rejected the intentional discrimination standard but concluded that the proof failed to satisfy the legal standard that in his view was the applicable rule. He therefore concurred in the judgment of reversal. Four other Justices, however, thought the evidence sufficient to satisfy the purposeful discrimination standard. One of them, Justice BLACKMUN, nevertheless concurred in the Court's judgment because he believed an erroneous remedy had been imposed. 12 Because the District Court in the present case employed the evidentiary factors outlined in Zimmer, it is urged that its judgment is infirm for the same reasons that led to the reversal in Mobile. We do not agree. First, and fundamentally, we are unconvinced that the District Court in this case applied the wrong legal standard. Not only was the District Court's decision rendered a considerable time after Washington v. Davis and Arlington Heights, but the trial judge also had the benefit of Nevett v. Sides, 571 F.2d 209 (1978), where the Court of Appeals for the Fifth Circuit assessed the impact of Washington v. Davis and Arlington Heights and held that "a showing of racially motivated discrimination is a necessary element in an equal protection voting dilution claim . . . ." 571 F.2d, at 219. The court stated that "[t]he ultimate issue in a case alleging unconstitutional dilution of the votes of a racial group is whether the districting plan under attack exists because it was intended to diminish or dilute the political efficacy of that group." Id., at 226. The Court of Appeals also explained that although the evidentiary factors outlined in Zimmer were important considerations in arriving at the ultimate conclusion of discriminatory intent, the plaintiff is not limited to those factors. "The task before the fact finder is to determine, under all the relevant facts, in whose favor the 'aggregate' of the evidence preponderates. This determination is peculiarly dependent upon the facts of each case." 571 F.2d, at 224 (footnote omitted). 13 The District Court referred to Nevett v. Sides and demonstrated its understanding of the controlling standard by observing that a determination of discriminatory intent is "a requisite to a finding of unconstitutional vote dilution" under the Fourteenth and Fifteenth Amendments. App. to Juris. Statement 68a. Furthermore, while recognizing that the evidentiary factors identified in Zimmer were to be considered, the District Court was aware that it was "not limited in its determination only to the Zimmer factors" but could consider other relevant factors as well. App. to Juris. Statement, at 70a. The District Court then proceeded to deal with what it considered to be the relevant proof and concluded that the at-large scheme of electing commissioners, "although racially neutral when adopted, is being maintained for invidious purposes." Id, at 71a. That system "while neutral in origin . . . has been subverted to invidious purposes." Id., at 90a. For the most part, the District Court dealt with the evidence in terms of the factors set out in Zimmer and its progeny, but as the Court of Appeals stated: 14 "Judge Alaimo employed the constitutionally required standard . . . . [and] did not treat the Zimmer criteria as absolute, but rather considered them only to the extent they were relevant to the question of discriminatory intent." 639 F.2d, at 1376. 15 Although a tenable argument can be made to the contrary, we are not inclined to disagree with the Court of Appeals' conclusion that the District Court applied the proper legal standard. III A. 16 We are also unconvinced that we should disturb the District Court's finding that the at-large system in Burke County was being maintained for the invidious purpose of diluting the voting strength of the black population. In White v. Regester, 412 U.S., at 769-770, 93 S.Ct., at 2341, we stated that we were not inclined to overturn the District Court's factual findings, "representing as they do a blend of history and an intensely local appraisal of the design and impact of the Bexar County multimember district in the light of past and present reality, political and otherwise." See also Columbus Board of Education v. Penick, 443 U.S. 449, 468, 99 S.Ct. 2941, 2952, 61 L.Ed.2d 666 (1979) (BURGER, C.J., concurring in judgment). Our recent decision in Pullman-Standard v. Swint, 456 U.S. 273, 102 S.Ct. 1781, 72 L.Ed.2d 66 (1982), emphasizes the deference Federal Rule of Civil Procedure 52 requires reviewing courts to give a trial court's findings of fact. "Rule 52a broadly requires that findings of fact not be set aside unless clearly erroneous. It does not make exceptions or purport to exclude certain categories of factual findings. . . ." 456 U.S. at 287, 102 S.Ct., at 1789. The Court held that the issue of whether the differential impact of a seniority system resulted from an intent to discriminate on racial grounds "is a pure question of fact, subject to Rule 52a's clearly-erroneous standard." Id., at 287-288, 102 S.Ct., at 1789. The Swint Court also noted that issues of intent are commonly treated as factual matters. Id., at 288, 102 S.Ct., at 1790. We are of the view that the same clearly-erroneous standard applies to the trial court's finding in this case that the at-large system in Burke County is being maintained for discriminatory purposes, as well as to the court's subsidiary findings of fact. The Court of Appeals did not hold any of the District Court's findings of fact to be clearly erroneous, and this Court has frequently noted its reluctance to disturb findings of fact concurred in by two lower courts. See, e.g., Berenyi v. Information Director, 385 U.S. 630, 635, 87 S.Ct. 666, 670, 17 L.Ed.2d 656 (1967); Blau v. Lehman, 368 U.S. 403, 408-409, 82 S.Ct. 451, 454, 7 L.Ed.2d 403 (1962); Graver Tank & Mfg. Co. v. Linde Co., 336 U.S. 271, 275, 69 S.Ct. 535, 537, 93 L.Ed. 672 (1949). We agree with the Court of Appeals that on the record before us, none of the factual findings are clearly erroneous. B 17 The District Court found that blacks have always made up a substantial majority of the population in Burke County, App. to Juris. Statement 66a, n. 3, but that they are a distinct minority of the registered voters. Id., at 71a-72a. There was also overwhelming evidence of bloc voting along racial lines. Id., at 72a-73a. Hence, although there had been black candidates, no black had ever been elected to the Burke County Commission. These facts bear heavily on the issue of purposeful discrimination. Voting along racial lines allows those elected to ignore black interests without fear of political consequences, and without bloc voting the minority candidates would not lose elections solely because of their race. Because it is sensible to expect that at least some blacks would have been elected in Burke County, the fact that none have ever been elected is important evidence of purposeful exclusion. See White v. Regester, supra, at 766, 93 S.Ct., at 2339. 18 Under our cases, however, such facts are insufficient in themselves to prove purposeful discrimination absent other evidence such as proof that blacks have less opportunity to participate in the political processes and to elect candidates of their choice. United Jewish Organizations v. Carey, 430 U.S. 144, 167, 97 S.Ct. 996, 1010, 51 L.Ed.2d 229 (1977); White v. Regester, supra, at 765-766, 93 S.Ct., at 2339; Whitcomb v. Chavis, 403 U.S., at 149-150, 91 S.Ct., at 1872. See also Mobile v. Bolden, 446 U.S., at 66, 100 S.Ct., at 1499 (plurality opinion). Both the District Court and the Court of Appeals thought the supporting proof in this case was sufficient to support an inference of intentional discrimination. The supporting evidence was organized primarily around the factors which Nevett v. Sides, 571 F.2d 209 (CA5 1978), had deemed relevant to the issue of intentional discrimination. These factors were primarily those suggested in Zimmer v. McKeithen, 485 F.2d 1297 (CA5 1973). 19 The District Court began by determining the impact of past discrimination on the ability of blacks to participate effectively in the political process. Past discrimination was found to contribute to low black voter registration, because prior to the Voting Rights Act of 1965, blacks had been denied access to the political process by means such as literacy tests, poll taxes, and white primaries. The result was that "Black suffrage in Burke County was virtually non-existent." App. to Juris. Statement 71a. Black voter registration in Burke County has increased following the Voting Rights Act to the point that some 38% of blacks eligible to vote are registered to do so. Id., at 72a. On that basis the District Court inferred that "past discrimination has had an adverse effect on black voter registration which lingers to this date." Ibid. Past discrimination against blacks in education also had the same effect. Not only did Burke County schools discriminate against blacks as recently as 1969, but also some schools still remain essentially segregated and blacks as a group have completed less formal education than whites. Id., at 74a. 20 The District Court found further evidence of exclusion from the political process. Past discrimination had prevented blacks from effectively participating in Democratic Party affairs and in primary elections. Until this lawsuit was filed, there had never been a black member of the County Executive Committee of the Democratic Party. There were also property ownership requirements that made it difficult for blacks to serve as chief registrar in the county. There had been discrimination in the selection of grand jurors, the hiring of county employees, and in the appointments to boards and committees which oversee the county government. Id., at 74a-76a. The District Court thus concluded that historical discrimination had restricted the present opportunity of blacks effectively to participate in the political process. Evidence of historical discrimination is relevant to drawing an inference of purposeful discrimination, particularly in cases such as this one where the evidence shows that discriminatory practices were commonly utilized, that they were abandoned when enjoined by courts or made illegal by civil rights legislation, and that they were replaced by laws and practices which, though neutral on their face, serve to maintain the status quo. 21 Extensive evidence was cited by the District Court to support its finding that elected officials of Burke County have been unresponsive and insensitive to the needs of the black community,9 which increases the likelihood that the political process was not equally open to blacks. This evidence ranged from the effects of past discrimination which still haunt the county courthouse to the infrequent appointment of blacks to county boards and committees; the overtly discriminatory pattern of paving county roads; the reluctance of the county to remedy black complaints, which forced blacks to take legal action to obtain school and grand jury desegregation; and the role played by the County Commissioners in the incorporation of an all-white private school to which they donated public funds for the purchase of band uniforms. Id., at 77a-82a. 22 The District Court also considered the depressed socio-economic status of Burke County blacks. It found that proportionately more blacks than whites have incomes below the poverty level. Id., at 83a. Nearly 53% of all black families living in Burke County had incomes equal to or less than three-fourths of a poverty-level income. Ibid. Not only have blacks completed less formal education than whites, but also the education they have received "was qualitatively inferior to a marked degree." Id., at 84a. Blacks tend to receive less pay than whites, even for similar work, and they tend to be employed in menial jobs more often than whites. Id., at 85a. Seventy-three percent of houses occupied by blacks lacked all or some plumbing facilities; only 16% of white-occupied houses suffered the same deficiency. Ibid. The District Court concluded that the depressed socio-economic status of blacks results in part from "the lingering effects of past discrimination." Ibid. 23 Although finding that the state policy behind the at-large electoral system in Burke County was "neutral in origin," the District Court concluded that the policy "has been subverted to invidious purposes." Id., at 90a. As a practical matter, maintenance of the state statute providing for at-large elections in Burke County is determined by Burke County's state representatives, for the legislature defers to their wishes on matters of purely local application. The court found that Burke County's state representatives "have retained a system which has minimized the ability of Burke County Blacks to participate in the political system." Ibid. The trial court considered, in addition, several factors which this Court has indicated enhance the tendency of multimember districts to minimize the voting strength of racial minorities. See Whitcomb v. Chavis, 403 U.S., at 143-144, 91 S.Ct., at 1869. It found that the sheer geographic size of the county, which is nearly two-thirds the size of Rhode Island, "has made it more difficult for Blacks to get to polling places or to campaign for office." App. to Juris. Statement 91a. The court concluded, as a matter of law, that the size of the county tends to impair the access of blacks to the political process. Id., at 92a. The majority vote requirement, Ga.Code § 34-1513 (Supp.1980), was found "to submerge the will of the minority" and thus "deny the minority's access to the system." App. to Juris. Statement 92a. The court also found the requirement that candidates run for specific seats, Ga.Code § 34-1015 (1978), enhances appellee's lack of access because it prevents a cohesive political group from concentrating on a single candidate. Because Burke County has no residency requirement, "[a]ll candidates could reside in Waynesboro, or in 'lilly-white' [sic] neighborhoods. To that extent, the denial of access becomes enhanced." App. to Juris. Statement 93a. 24 None of the District Court's findings underlying its ultimate finding of intentional discrimination appears to us to be clearly erroneous; and as we have said, we decline to overturn the essential finding of the District Court, agreed to by the Court of Appeals, that the at-large system in Burke County has been maintained for the purpose of denying blacks equal access to the political processes in the county. As in White v. Regester, 412 U.S., at 767, 93 S.Ct., at 2340, the District Court's findings were "sufficient to sustain [its] judgment . . . and, on this record, we have no reason to disturb them." IV 25 We also find no reason to overturn the relief ordered by the District Court. Neither the District Court nor the Court of Appeals discerned any special circumstances that would militate against utilizing single-member districts. Where "a constitutional violation has been found, the remedy does not 'exceed' the violation if the remedy is tailored to cure the 'condition that offends the Constitution.' " Milliken v. Bradley, 433 U.S. 267, 282, 97 S.Ct. 2749, 2758, 53 L.Ed.2d 745 (1977), (emphasis deleted), quoting Milliken v. Bradley, 418 U.S. 717, 738, 94 S.Ct. 3112, 3124, 41 L.Ed.2d 1069 (1974).10 The judgment of the Court of Appeals is 26 Affirmed. 27 Justice POWELL, with whom Justice REHNQUIST joins, dissenting. 28 * Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), establishes that an at-large voting system must be upheld against constitutional attack unless maintained for a discriminatory purpose. In Mobile we reversed a finding of unconstitutional vote dilution because the lower courts had relied on factors insufficient as a matter of law to establish discriminatory intent. See id., at 73, 100 S.Ct., at 1503 (plurality opinion of Stewart, J.). The District Court and Court of Appeals in this case based their findings of unconstitutional discrimination on the same factors held insufficient in Mobile. Yet the Court now finds their conclusion unexceptionable. The Mobile plurality also affirmed that the concept of "intent" was no mere fiction, and held that the District Court had erred in "its failure to identify the state officials whose intent it considered relevant." Id., at 74, n. 20, 100 S.Ct., at 1503, n. 20. Although the courts below did not answer that question in this case, the Court today affirms their decision. 29 Whatever the wisdom of Mobile, the Court's opinion cannot be reconciled persuasively with that case. There are some variances in the largely sociological evidence presented in the two cases. But Mobile held that this kind of evidence was not enough. Such evidence, we found in Mobile, did not merely fall short, but "fell far short[,] of showing that [an at-large electoral scheme was] 'conceived or operated [as a] purposeful devic[e] to further racial . . . discrimination.' " Id., at 70, 100 S.Ct., at 1501 (emphasis added), quoting Whitcomb v. Chavis, 403 U.S. 124, 149, 91 S.Ct., 1858, 1872, 29 L.Ed.2d 363 (1971). Because I believe that Mobile controls this case, I dissent. II 30 The Court's decision today relies heavily on the capacity of the federal district courts—essentially free from any standards propounded by this Court—to determine whether at-large voting systems are "being maintained for the invidious purpose of diluting the voting strength of the black population." Ante, at 622. Federal courts thus are invited to engage in deeply subjective inquiries into the motivations of local officials in structuring local governments. Inquiries of this kind not only can be "unseemly," see Karst, The Costs of Motive-Centered Inquiry, 15 San Diego L.Rev. 1163, 1164 (1978); they intrude the federal courts—with only the vaguest constitutional direction—into an area of intensely local and political concern. 31 Emphasizing these considerations, Justice STEVENS, post, at 642-650, argues forcefully that the Court's focus of inquiry is seriously mistaken. I agree with much of what he says. As I do not share his views entirely, however, I write separately. 32 * As I understand it, Justice STEVENS' critique of the Court's approach rests on three principles with which I am in fundamental agreement. 33 First, it is appropriate to distinguish between "state action that inhibits an individual's right to vote and state action that affects the political strength of various groups." Mobile v. Bolden, supra, at 83, 100 S.Ct., at 1508 (STEVENS, J., concurring in judgment); see post, at 632, 637-638, n. 16. Under this distinction, this case is fundamentally different from cases involving direct barriers to voting. There is no claim here that blacks may not register freely and vote for whom they choose. This case also differs from one-man, one-vote cases, in which districting practices make a person's vote less weighty in some districts than in others. 34 Second, I agree with Justice STEVENS that vote dilution cases of this kind are difficult if not impossible to distinguish especially in their remedial aspect—from other actions to redress gerrymanders. See post, at 650-653. 35 Finally, Justice STEVENS clearly is correct in arguing that the standard used to identify unlawful racial discrimination in this area should be defined in terms that are judicially manageable and reviewable. See post, at 633, 642-650. In the absence of compelling reasons of both law and fact, the federal judiciary is unwarranted in undertaking to restructure state political systems. This is inherently a political area, where the identification of a seeming violation does not necessarily suggest an enforceable judicial remedy—or at least none short of a system of quotas or group representation. Any such system, of course, would be antithetical to the principles of our democracy. B 36 Justice STEVENS would accommodate these principles by holding that subjective intent is irrelevant to the establishment of a case of racial vote dilution under the Fourteenth Amendment. See post, at 637. Despite sharing the concerns from which his position is developed, I would not accept this view. "The central purpose of the Equal Protection Clause of the Fourteenth Amendment is the prevention of official conduct discriminating on the basis of race." Washington v. Davis, 426 U.S. 229, 239, 96 S.Ct. 2040, 2047, 48 L.Ed.2d 597 (1976). Because I am unwilling to abandon this central principle in cases of this kind, I cannot join Justice STEVENS' opinion. 37 Nonetheless, I do agree with him that what he calls "objective" factors should be the focus of inquiry in vote-dilution cases. Unlike the considerations on which the lower courts relied in this case and in Mobile, the factors identified by Justice STEVENS as "objective" in fact are direct, reliable, and unambiguous indices of discriminatory intent. If we held, as I think we should, that the district courts must place primary reliance on these factors to establish discriminatory intent, we would prevent federal-court inquiries into the subjective thought processes of local officials—at least until enough objective evidence had been presented to warrant discovery into subjective motivations in this complex, politically charged area. By prescribing such a rule we would hold federal courts to a standard that was judicially manageable. And we would remain faithful to the central protective purpose of the Equal Protection Clause. 38 In the absence of proof of discrimination by reliance on the kind of objective factors identified by Justice STEVENS, I would hold that the factors cited by the Court of Appeals are too attenuated as a matter of law to support an inference of discriminatory intent. I would reverse its judgment on that basis. 39 Justice STEVENS, dissenting. 40 Our legacy of racial discrimination has left its scars on Burke County, Georgia.1 The record in this case amply supports the conclusion that the governing officials of Burke County have repeatedly denied black citizens rights guaranteed by the Fourteenth and Fifteenth Amendments to the Federal Constitution. No one could legitimately question the validity of remedial measures, whether legislative or judicial, designed to prohibit discriminatory conduct by public officials and to guarantee that black citizens are effectively afforded the rights to register and to vote. Public roads may not be paved only in areas in which white citizens live;2 black citizens may not be denied employment opportunities in county government;3 segregated schools may not be maintained.4 41 Nor, in my opinion, could there be any doubt about the constitutionality of an amendment to the Voting Rights Act that would require Burke County and other covered jurisdictions to abandon specific kinds of at-large voting schemes that perpetuate the effects of past discrimination. "As against the reserved powers of the States, Congress may use any rational means to effectuate the constitutional prohibition of racial discrimination in voting." South Carolina v. Katzenbach, 383 U.S. 301, 324, 86 S.Ct. 803, 816, 15 L.Ed.2d 769. It might indeed be wise policy to accelerate the transition of minority groups to a position of political power commensurate with their voting strength by amending the Act to prohibit the use of multimember districts in all covered jurisdictions. 42 The Court's decision today, however, is not based on either its own conception of sound policy or any statutory command. The decision rests entirely on the Court's interpretation of the requirements of the Federal Constitution. Despite my sympathetic appraisal of the Court's laudable goals, I am unable to agree with its approach to the constitutional issue that is presented. In my opinion, this case raises questions that encompass more than the immediate plight of disadvantaged black citizens. I believe the Court errs by holding the structure of the local governmental unit unconstitutional without identifying an acceptable, judicially manageable standard for adjudicating cases of this kind. 43 * The Court's entry into the business of electoral reapportionment in 1962 was preceded by a lengthy and scholarly debate over the role the judiciary legitimately could play in what Justice Frankfurter described in Colegrove v. Green, 328 U.S. 549, 66 S.Ct. 1198, 90 L.Ed. 1432, as a "political thicket."5 In that case, decided in 1946, the Court declined to entertain a challenge to single-member congressional districts in Illinois that had been created in 1901 and had become grossly unequal by reason of the great growth in urban population.6 In dissent, Justice Black advocated the use of a statewide, at-large election of representatives; he argued that an at-large election "has an element of virtue that the more convenient method does not have—namely, it does not discriminate against some groups to favor others, it gives all the people an equally effective voice in electing their representatives as is essential under a free government, and it is constitutional." Id., at 574, 66 S.Ct., at 1212. 44 In 1962, the Court changed course. In another challenge to the constitutionality of a 1901 districting statute, it held that the political question doctrine did not foreclose judicial review. Baker v. Carr, 369 U.S. 186, 82 S.Ct. 691, 7 L.Ed.2d 663. That decision represents one of the great landmarks in the history of this Court's jurisprudence. 45 Two aspects of the Court's opinion in Baker v. Carr are of special relevance to the case the Court decides today. First, the Court's scholarly review of the political question doctrine focused on the dominant importance of satisfactory standards for judicial determination.7 Second, the Court's articulation of the relevant constitutional standard made no reference to subjective intent.8 The host of cases that have arisen in the wake ofBaker v. Carr have shared these two characteristics. They have formulated, refined, and applied a judicially manageable standard that has become known as the one-person, one-vote rule; they have attached no significance to the subjective intent of the decisionmakers who adopted or maintained the official rule under attack. 46 In reviewing the constitutionality of the structure of a local government, two quite different methods of analysis could be employed. The Court might identify the specific features of the government that raise constitutional concerns and decide whether, singly or in combination, they are valid. This is the approach the Court has used in testing the constitutionality of rules conditioning the right to vote on payment of a poll tax,9 imposing burdens on independent candidates,10 denying new residents or members of the Armed Forces the right to vote,11 prohibiting crossovers in party primaries,12 requiring political candidates to pay filing fees,13 and disadvantaging minority parties in Presidential elections.14 In none of these cases did the validity of the electoral procedure turn on whether the legislators who enacted the rule subjectively intended to discriminate against minority voters. Under the approach employed by the Court in those cases, the objective circumstances that led to a declaration that an election procedure was unconstitutional would invalidate a similar law wherever it might be found. 47 Alternatively, the Court could employ a subjective approach under which the constitutionality of a challenged procedure depends entirely on federal judges' appraisals of the reasons why particular localities have chosen to govern themselves in a particular way. The Constitution would simply protect a right to have an electoral machinery established and maintained without the influence of impermissible factors. Constitutional challenges to identical procedures in neighboring communities could produce totally different results, for the subjective motivations of the legislators who enacted the procedures—or at least the admissible evidence that might be discovered concerning such motivation—could be quite different. 48 In deciding the question presented in this case, the Court abruptly rejects the former approach and considers only the latter. It starts from the premise that Burke County's atlarge method of electing its five county commissioners is, on its face, unobjectionable. The otherwise valid system is unconstitutional, however, because it makes it more difficult for the minority to elect commissioners and because the majority that is now in power has maintained the system for that very reason. Two factors are apparently of critical importance: (1) the intent of the majority to maintain control; and (2) the racial character of the minority.15 49 I am troubled by each aspect of the Court's analysis. In my opinion, the question whether Burke County's at-large system may survive scrutiny under a purely objective analysis is not nearly as easy to answer as the Court implies. Assuming, however, that the system is otherwise valid, I do not believe that the subjective intent of the persons who adopted the system in 1911, or the intent of those who have since declined to change it, can determine its constitutionality. Even if the intent of the political majority were the controlling constitutional consideration, I could not agree that the only political groups that are entitled to protection under the Court's rule are those defined by racial characteristics. II 50 At-large voting systems generally tend to maximize the political power of the majority. See ante, at 616.16 There are, however, many types of at-large electoral schemes. Three features of Burke County's electoral system are noteworthy, not in my opinion because they shed special light on the subjective intent of certain unidentified people, but rather because they make it especially difficult for a minority candidate to win an election. First, although the qualifications and the duties of the office are identical for all five commissioners, each runs for a separately designated position.17 Second, in order to be elected, each commissioner must receive a majority of all votes cast in the primary and in the general election; if the leading candidate receives only a plurality, a runoff election must be held. Third, there are no residency requirements; thus, all candidates could reside in a single, all-white neighborhood.18 51 Even if one assumes that a system of local government in which power is concentrated in the hands of a small group of persons elected from the community at large is an acceptable—or perhaps even a preferred—form of municipal government,19 it is not immediately apparent that these additional features that help to perpetuate the power of an entrenched majority are either desirable or legitimate.20 If the only purpose these features serve—particularly when viewed in combination—is to assist a dominant party to maintain its political power, they are no more legitimate than the Tennessee districts described in Baker v. Carr as "no policy, but simply arbitrary and capricious action." 369 U.S., at 226, 82 S.Ct., at 714 (emphasis in original). Unless these features are independently justified, they may be invalid simply because there is no legitimate justification for their impact on minority participation in elections.21 52 In this case, appellees have not argued—presumably because they assumed that this Court's many references to the requirement of proving an improper motive in equal protection cases are controlling in this new context—that the special features of Burke County's at-large system have such an adverse impact on the minority's opportunity to participate in the political process that this type of government deprives the minority of equal protection of the law. Nor have the appellants sought to identify legitimate local policies that might justify the use of such rules. As a result, this record does not provide an adequate basis for determining the validity of Burke County's governmental structure on the basis of traditional objective standards.22 53 If the governmental structure were itself found to lack a legitimate justification, inquiry into subjective intent would clearly be unnecessary. As Justice MARSHALL stated in his dissent in Mobile : "Whatever may be the merits of applying motivational analysis to the allocation of constitutionally gratuitous benefits, that approach is completely misplaced where, as here, it is applied to the distribution of a constitutionally protected interest." 446 U.S., at 121, 100 S.Ct., at 1529.23 Under the Court's analysis, however, the characteristics of the particular form of government under attack are virtually irrelevant. Not only would the Court's approach uphold an arbitrary—but not invidious—system that lacked independent justification, it would invalidate—if a discriminatory intent were proved—a local rule that would be perfectly acceptable absent a showing of invidious intent. The Court's standard applies not only to Burke County and to multimember districts, but to any other form of government as well. III 54 Ever since I joined the Court, I have been concerned about the Court's emphasis on subjective intent as a criterion for constitutional adjudication.24 Although that criterion is often regarded as a restraint on the exercise of judicial power, it may in fact provide judges with a tool for exercising power that otherwise would be confined to the legislature.25 My principal concern with the subjective-intent standard, however, is unrelated to the quantum of power it confers upon the judiciary. It is based on the quality of that power. For in the long run constitutional adjudication that is premised on a case-by-case appraisal of the subjective intent of local decisionmakers cannot possibly satisfy the requirement of impartial administration of the law that is embodied in the Equal Protection Clause of the Fourteenth Amendment. 55 The facts of this case illustrate the ephemeral character of a constitutional standard that focuses on subjective intent. When the suit was filed in 1976, approximately 58 percent of the population of Burke County was black and approximately 42 percent was white. Because black citizens had been denied access to the political process—through means that have since been outlawed by the Voting Rights Act of 1965—and because there had been insufficient time to enable the registration of black voters to overcome the history of past injustice, the majority of registered voters in the county were white. The at-large electoral system therefore served, as a result of the presence of bloc voting, to maintain white control of the local government. Whether it would have continued to do so would have depended on a mix of at least three different factors—the continuing increase in voter registration among blacks, the continuing exodus of black residents from the county, and the extent to which racial bloc voting continued to dominate local politics. 56 If those elected officials in control of the political machinery had formed the judgment that these factors created a likelihood that a bloc of black voters was about to achieve sufficient strength to elect an entirely new administration, they might have decided to abandon the at-large system and substitute five single-member districts with the boundary lines drawn to provide a white majority in three districts and a black majority in only two. Under the Court's intent standard, such a change presumably would violate the Fourteenth Amendment. It is ironic that the remedy ordered by the District Court fits that pattern precisely.26 57 If votes continue to be cast on a racial basis, the judicial remedy virtually guarantees that whites will continue to control a majority of seats on the County Board. It is at least possible that white control of the political machinery has been frozen by judicial decree at a time when increased black voter registration might have led to a complete change of administration. Since the federal judge's intent was unquestionably benign rather than invidious—and, unlike that of state officials, is presumably not subject in any event to the Court's standard—that result has been accomplished without violating the Federal Constitution. 58 In the future, it is not inconceivable that the white officials who are likely to remain in power under the District Court's plan will desire to perpetuate that system and to continue to control a majority of seats on the County Board. Under this Court's standard, if some of those officials harbor such an intent for an "invidious" reason, the District Court's plan will itself become unconstitutional. It is not clear whether the invidious intent would have to be shared by all three white commissioners, by merely a majority of two, or by simply one if he were influential. It is not clear whether the issue would be affected by the intent of the two black commissioners, who might fear that a return to an at-large system would undermine the certainty of two black seats.27 Of course, if the subjective intent of these officials were such as to mandate a change to a governmental structure that would permit black voters to elect an all-black commission—and if black voters did so—those black officials could not harbor an intent to maintain the system to keep whites from returning to power. In sum, as long as racial consciousness exists in Burke County, its governmental structure is subject to attack. Perhaps those more familiar than I with political maneuvering will be able to identify with greater accuracy and reliability those subjective intentions that are legitimate and those that are not. Because judges may not possess such expertise, however, I am afraid the Court is planting seeds that may produce an unexpected harvest. 59 The costs and the doubts associated with litigating questions of motive, which are often significant in routine trials, will be especially so in cases involving the "motives" of legislative bodies.28 Often there will be no evidence that the governmental system was adopted for a discriminatory reason.29 The reform movement in municipal government, see n. 19, supra, or an attempt to comply with the strictures of Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506, may account for the enactment of countless at-large systems. In such a case the question becomes whether the system was maintained for a discriminatory purpose. Whose intentions control? Obviously not the voters, although they may be most responsible for the attitudes and actions of local government.30 Assuming that it is the intentions of the "state actors" that is critical, how will their mental processes be discovered? Must a specific proposal for change be defeated? What if different motives are held by different legislators or, indeed, by a single official? Is a selfish desire to stay in office sufficient to justify a failure to change a governmental system? 60 The Court avoids these problems by failing to answer the very question that its standard asks. Presumably, according to the Court's analysis, the Burke County governmental structure is unconstitutional because it was maintained at some point for an invidious purpose. Yet the Court scarcely identifies the manner in which changes to a county governmental structure are made. There is no reference to any unsuccessful attempt to replace the at-large system with single-member districts. It is incongruous that subjective intent is identified as the constitutional standard and yet the persons who allegedly harbored an improper intent are never identified or mentioned. Undoubtedly, the evidence relied on by the Court proves that racial prejudice has played an important role in the history of Burke County and has motivated many wrongful acts by various community leaders. But unless that evidence is sufficient to prove that every governmental action was motivated by a racial animus—and may be remedied by a federal court—the Court has failed under its own test to demonstrate that the governmental structure of Burke County was maintained for a discriminatory purpose. 61 Certainly governmental action should not be influenced by irrelevant considerations. I am not convinced, however, that the Constitution affords a right—and this is the only right the Court finds applicable in this case—to have every official decision made without the influence of considerations that are in some way "discriminatory." Is the failure of a state legislature to ratify the Equal Rights Amendment invalid if a federal judge concludes that a majority of the legislators harbored stereotypical views of the proper role of women in society? Is the establishment of a memorial for Jews slaughtered in World War II unconstitutional if civic leaders believe that their cause is more meritorious than that of victimized Palestinian refugees? Is the failure to adopt a state holiday for Martin Luther King, Jr., invalid if it is proved that state legislators believed that he does not deserve to be commemorated? Is the refusal to provide Medicaid funding for abortions unconstitutional if officials intend to discriminate against women who would abort a fetus?31 62 A rule that would invalidate all governmental action motivated by racial, ethnic, or political considerations is too broad. Moreover, in my opinion the Court is incorrect in assuming that the intent of elected officials is invidious when they are motivated by a desire to retain control of the local political machinery. For such an intent is surely characteristic of politicians throughout the country. In implementing that sort of purpose, dominant majorities have used a wide variety of techniques to limit the political strength of aggressive minorities. In this case the minority is defined by racial characteristics, but minority groups seeking an effective political voice can, of course, be identified in many other ways. The Hasidic Jews in Kings County, N.Y.,32 the Puerto Ricans in Chicago,33 the Spanish-speaking citizens in Dallas,34 the Bohemians in Cedar Rapids,35 the Federalists in Massachusetts,36 the Democrats in Indiana,37 and the Republicans in California38 have all been disadvantaged by deliberate political maneuvers by the dominant majority. As I have stated, a device that serves no purpose other than to exclude minority groups from effective political participation is unlawful under objective standards. But if a political majority's intent to maintain control of a legitimate local government is sufficient to invalidate any electoral device that makes it more difficult for a minority group to elect candidates—regardless of the nature of the interest that gives the minority group cohesion—the Court is not just entering a "political thicket"; it is entering a vast wonderland of judicial review of political activity. 63 The obvious response to this suggestion is that this case involves a racial group and that governmental decisions that disadvantage such a group must be subject to special scrutiny under the Fourteenth Amendment. I therefore must consider whether the Court's holding can legitimately be confined to political groups that are identified by racial characteristics. IV 64 Governmental action that discriminates between individuals on the basis of their race is, at the very least, presumptively irrational.39 For an individual's race is virtually always irrelevant to his right to enjoy the benefits and to share the responsibilities of citizenship in a democratic society. Persons of different races, like persons of different religious faiths and different political beliefs, are equal in the eyes of the law. 65 Groups of every character may associate together to achieve legitimate common goals. If they voluntarily identify themselves by a common interest in a specific issue, by a common ethnic heritage, by a common religious belief, or by their race, that characteristic assumes significance as the bond that gives the group cohesion and political strength. When referring to different kinds of political groups, this Court has consistently indicated that, to borrow Justice BRENNAN's phrasing, the Equal Protection Clause does not make some groups of citizens more equal than others. See Zobel v. Williams, 457 U.S. 55, 71, 102 S.Ct. 2309, 2318, 72 L.Ed.2d 672 (BRENNAN, J., concurring). Thus, the Court has considered challenges to discrimination based on "differences of color, race, nativity, religious opinions [or] political affiliations,"American Sugar Refining Co. v. Louisiana, 179 U.S. 89, 92, 21 S.Ct. 43, 44, 45 L.Ed. 102, to redistricting plans that serve "to further racial or economic discrimination," Whitcomb v. Chavis, 403 U.S. 124, 149, 91 S.Ct. 1858, 1872, 29 L.Ed.2d 363; to biases "tending to favor particular political interests or geographic areas." Abate v. Mundt, 403 U.S. 182, 187, 91 S.Ct. 1904, 1907, 29 L.Ed.2d 399. Indeed, in its opinion today the Court recognizes that the practical impact of the electoral system at issue applies equally to any "distinct minority, whether it be a racial, ethnic, economic, or political group." Ante, at 616. 66 A constitutional standard that gave special protection to political groups identified by racial characteristics would be inconsistent with the basic tenet of the Equal Protection Clause. Those groups are no more or no less able to pursue their interests in the political arena than are groups defined by other characteristics. Nor can it be said that racial alliances are so unrelated to political action that any electoral decision that is influenced by racial consciousness—as opposed to other forms of political consciousness—is inherently irrational. For it is the very political power of a racial or ethnic group that creates a danger that an entrenched majority will take action contrary to the group's political interests. "The mere fact that a number of citizens share a common ethnic, racial, or religious background does not create the need for protection against gerrymandering. It is only when their common interests are strong enough to be manifested in political action that the need arises. Thus the characteristic of the group which creates the need for protection is its political character." Cousins v. City Council of Chicago, 466 F.2d 830, 852 (CA7 1972) (Stevens, J., dissenting), cert. denied, 409 U.S. 893, 93 S.Ct. 85, 34 L.Ed.2d 151. It would be unrealistic to distinguish racial groups from other political groups on the ground that race is an irrelevant factor in the political process. 67 Racial consciousness and racial association are not desirable features of our political system. We all look forward to the day when race is an irrelevant factor in the political process. In my opinion, however, that goal will best be achieved by eliminating the vestiges of discrimination that motivate disadvantaged racial and ethnic groups to vote as identifiable units. Whenever identifiable groups in our society are disadvantaged, they will share common political interests and tend to vote as a "bloc." In this respect, racial groups are like other political groups. A permanent constitutional rule that treated them differently would, in my opinion, itself tend to perpetuate race as a feature distinct from all others; a trait that makes persons different in the eyes of the law. Such a rule would delay—rather than advance—the goal advocated by Justice Douglas: 68 "When racial or religious lines are drawn by the State, the multiracial, multireligious communities that our Constitution seeks to weld together as one become separatist; antagonisms that relate to race or to religion rather than to political issues are generated; communities seek not the best representative but the best racial or religious partisan. Since that system is at war with the democratic ideal, it should find no footing here." Wright v. Rockefeller, 376 U.S. 52, 67, 84 S.Ct. 603, 611, 11 L.Ed.2d 512 (dissenting opinion). 69 My conviction that all minority groups are equally entitled to constitutional protection against the misuse of the majority's political power does not mean that I would abandon judicial review of such action. As I have written before, a gerrymander as grotesque as the boundaries condemned in Gomillion v. Lightfoot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110, is intolerable whether it fences out black voters, Republican voters, or Irish-Catholic voters. Mobile v. Bolden, 446 U.S., at 86, 100 S.Ct., at 1510 (opinion concurring in judgment). But if the standard the Court applies today extends to all types of minority groups, it is either so broad that virtually every political device is vulnerable or it is so undefined that federal judges can pick and choose almost at will among those that will be upheld and those that will be condemned. 70 There are valid reasons for concluding that certain minority groups—such as the black voters in Burke County, Georgia—should be given special protection from political oppression by the dominant majority. But those are reasons that justify the application of a legislative policy choice rather than a constitutional principle that cannot be confined to special circumstances or to a temporary period in our history. Any suggestion that political groups in which black leadership predominates are in need of a permanent constitutional shield against the tactics of their political opponents underestimates the resourcefulness, the wisdom, and the demonstrated capacity of such leaders. I cannot accept the Court's constitutional holding.40 71 I respectfully dissent. 1 U.S. Dept. of Commerce, Bureau of the Census, County and City Data Book 1977, p. 90 (1978). 2 U.S. Dept. of Commerce, Bureau of the Census, 1980 Census of Population and Housing, PHC80-V-12, p. 5 (Mar.1981). In 1930, Burke County had a total population of 29,224, of whom 22,698 or 78% were black. U.S. Dept. of Commerce, Bureau of the Census, II Characteristics of the Population, pt. 2, p. 229 (1943). The percentage of blacks in the total population of Burke County has steadily diminished over the last 50 years. 3 App. to Juris. Statement 72a. 4 The District Court's judgment was stayed pending appeal to the Court of Appeals. 439 U.S. 948, 99 S.Ct. 342, 58 L.Ed.2d 339 (1978). The Court of Appeals stayed its mandate on April 6, 1981, pending disposition of the case here. 5 Purposeful racial discrimination invokes the strictest scrutiny of adverse differential treatment. Absent such purpose, differential impact is subject only to the test of rationality. Washington v. Davis, 426 U.S., at 247-248, 96 S.Ct., at 2051. 6 With respect to the Fifteenth Amendment, the plurality held that the Amendment prohibits only direct, purposefully discriminatory interference with the freedom of Negroes to vote. "Having found that Negroes in Mobile 'register and vote without hindrance,' the District Court and Court of Appeals were in error in believing that the appellants invaded the protection of [the Fifteenth] Amendment in the present case." Mobile v. Bolden, 446 U.S., at 65, 100 S.Ct., at 1499. Three Justices disagreed with the plurality's basis for putting aside the Fifteenth Amendment. Id., at 84, n. 3, 100 S.Ct. at 1509 (STEVENS, J., concurring in judgment); id., at 102, 100 S.Ct., at 1518 (WHITE, J., dissenting); id., at 125-135, 100 S.Ct., at 1531-1537 (MARSHALL, J., dissenting). We express no view on the application of the Fifteenth Amendment to this case. The plurality noted that plaintiffs' claim under § 2 of the Voting Rights Act, 79 Stat. 437, as amended, 42 U.S.C. § 1973, added nothing to their Fifteenth Amendment claim because the "legislative history of § 2 makes clear that it was intended to have an effect no different from that of the Fifteenth Amendment itself." 446 U.S., at 60-61, 100 S.Ct., at 1496. 7 We specifically affirmed the judgment below "without approval of the constitutional views expressed by the Court of Appeals." 424 U.S., at 638, 96 S.Ct., at 1084. 8 The primary factors listed in Zimmer include a lack of minority access to the candidate selection process, unresponsiveness of elected officials to minority interests, a tenuous state policy underlying the preference for multimember or at-large districting, and the existence of past discrimination which precludes effective participation in the elector process. 485 F.2d, at 1305. Factors which enhance the proof of voting dilution are the existence of large districts, anti-single-shot voting provisions, and the absence of any provision for at-large candidates to run from geographic subdistricts. Ibid. 9 The Court of Appeals held that "proof of unresponsiveness by the public body in question to the group claiming injury" is an essential element of a claim of voting dilution under the Fourteenth Amendment. 639 F.2d, at 1375. Under our cases, however, unresponsiveness is an important element but only one of a number of circumstances a court should consider in determining whether discriminatory purpose may be inferred. 10 Appellants contend that the District Court should not have divided Burke County into five districts but should have allowed appellants to devise a plan for subdividing the county and to submit their plan for preclearance under § 5 of the Voting Rights Act, 79 Stat. 439, as amended, 42 U.S.C. § 1973c. This contention was not properly raised in the Court of Appeals and was not addressed by that court. We therefore do not address it. See Adickes v. S.H. Kress & Co., 398 U.S. 144, 147, n. 2, 90 S.Ct. 1598, 1602, n. 2, 26 L.Ed.2d 142 (1970). Appellants also contend that the doctrine of unconstitutional dilution of voting rights arising from an at-large election system does not apply to county governing bodies. We find no merit to this contention, having previously affirmed a judgment that at-large elections for the governing body of a parish (county) unconstitutionally diluted black voting strength. East Carroll Parish School Bd. v. Marshall, 424 U.S. 636, 96 S.Ct. 1083, 47 L.Ed. 296 (1976). 1 Certain vestiges of discrimination—although clearly not the most pressing problems facing black citizens today—are a haunting reminder of an all too recent period of our Nation's history. The District Court found that a segregated laundromat is operated within a few blocks of the county courthouse; at the courthouse itself, faded paint over restroom doors does not entirely conceal the words "colored" and "white." 2 See Dowdell v. City of Apopka, 511 F.Supp. 1375 (MD Fla.1981). 3 42 U.S.C. § 2000e-2. 4 Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873. 5 In his much criticized opinion announcing the judgment of the Court, Justice Frankfurter wrote: "Nothing is clearer than that this controversy concerns matters that bring courts into immediate and active relations with party contests. From the determination of such issues this Court has traditionally held aloof. It is hostile to a democratic system to involve the judiciary in the politics of the people. And it is not less pernicious if such judicial intervention in an essentially political contest be dressed up in the abstract phrases of the law. * * * * * "Courts ought not to enter this political thicket. The remedy for unfairness in districting is to secure State legislatures that will apportion properly, or to invoke the ample powers of Congress. The Constitution has many commands that are not enforceable by courts because they clearly fall outside the conditions and purposes that circumscribe judicial action. Thus, 'on Demand of the executive Authority,' Art. IV, § 2, of a State it is the duty of a sister State to deliver up a fugitive from justice. But the fulfilment of this duty cannot be judicially enforced. Kentucky v. Dennison, 24 How. 66, 16 L.Ed. 717. The duty to see to it that the laws are faithfully executed cannot be brought under legal compulsion, Mississippi v. Johnson, 4 Wall. 475. Violation of the great guaranty of a republican form of government in States cannot be challenged in the courts. Pacific Telephone Co. v. Oregon, 223 U.S. 118, 32 S.Ct. 224, 56 L.Ed. 377. The Constitution has left the performance of many duties in our governmental scheme to depend on the fidelity of the executive and legislative action and, ultimately, on the vigilance of the people in exercising their political rights." 328 U.S., at 553-554, 556, 66 S.Ct., at 1200, 1201. 6 The districts ranged in population from 112,000 to 914,000 persons. Id., at 557, 66 S.Ct., at 1202. 7 The Court stated that the "nonjusticiability of a political question is primarily a function of the separation of powers." 369 U.S., at 210, 82 S.Ct., at 706. It emphasized, however, that "the lack of satisfactory criteria for a judicial determination" was a dominant consideration in Coleman v. Miller, 307 U.S. 433, 454-455, 59 S.Ct. 972, 982, 83 L.Ed. 1385, that whether a foreign relations question is justiciable turns, in part, on "its susceptibility to judicial handling"; that in the presence of clearly definable criteria for decision "the political question barrier falls away"; and that "even in private litigation which directly implicates no feature of separation of powers, lack of judicially discoverable standards and the drive for even-handed application may impel reference to the political departments' determination of dates of hostilities' beginning and ending." 369 U.S., at 210, 211, 214, 82 S.Ct., at 706, 707, 708. Luther v. Borden, 7 How. 1, 12 L.Ed. 581, was distinguished, in part, because that case involved "the lack of criteria by which a court could determine which form of government was republican"; the Court stated that "the only significance that Luther could have for our immediate purposes is in its holding that the Guaranty Clause is not a repository of judicially manageable standards which a court could utilize independently in order to identify a State's lawful government." 369 U.S., at 222, 223, 82 S.Ct., at 712, 713. In concluding that the reapportionment question before it was justiciable, the Court emphasized that it would not be necessary "to enter upon policy determinations for which judicially manageable standards are lacking." Id., at 226, 82 S.Ct., at 714. 8 The Court simply stated: "Judicial standards under the Equal Protection Clause are well developed and familiar, and it has been open to courts since the enactment of the Fourteenth Amendment to determine, if on the particular facts they must, that a discrimination reflects no policy, but simply arbitrary and capricious action." Ibid. 9 Harper v. Virginia Board of Elections, 383 U.S. 663, 86 S.Ct. 1079, 16 L.Ed.2d 169. The Court concluded that "a State violates the Equal Protection Clause of the Fourteenth Amendment whenever it makes the affluence of the voter or payment of any fee an electoral standard. Voter qualifications have no relation to wealth nor to paying or not paying this or any other tax." Id., at 666, 86 S.Ct., at 1081. "To introduce wealth or payment of a fee as a measure of a voter's qualifications is to introduce a capricious or irrelevant factor." Id., at 668, 86 S.Ct., at 1082. In dissent, Justice Black noted: "It should be pointed out at once that the Court's decision is to no extent based on a finding that the Virginia law as written or as applied is being used as a device or mechanism to deny Negro citizens of Virginia the right to vote on account of their color." Id., at 672, 86 S.Ct., at 1084. 10 Storer v. Brown, 415 U.S. 724, 94 S.Ct. 1274, 39 L.Ed.2d 714. The Court stated that, in determining the constitutionality of eligibility requirements for independent candidates, the "inevitable question for judgment" is "could a reasonably diligent independent candidate be expected to satisfy the signature requirements, or will it be only rarely that the unaffiliated candidate will succeed in getting on the ballot?" Id., at 742, 94 S.Ct., at 1285. See Mandel v. Bradley, 432 U.S. 173, 177, 97 S.Ct. 2238, 2241, 53 L.Ed.2d 199, id., at 181, 97 S.Ct., at 2243 (STEVENS, J., dissenting). See also American Party of Texas v. White, 415 U.S. 767, 795, 94 S.Ct. 1296, 1313, 39 L.Ed.2d 744. 11 Dunn v. Blumstein, 405 U.S. 330, 92 S.Ct. 995, 31 L.Ed.2d 274; Carrington v. Rash, 380 U.S. 89, 85 S.Ct. 775, 13 L.Ed.2d 675. 12 Kusper v. Pontikes, 414 U.S. 51, 94 S.Ct. 303, 38 L.Ed.2d 260. 13 Lubin v. Panish, 415 U.S. 709, 94 S.Ct. 1315, 39 L.Ed.2d 702; Bullock v. Carter, 405 U.S. 134, 92 S.Ct. 849, 31 L.Ed.2d 92. 14 Williams v. Rhodes, 393 U.S. 23, 89 S.Ct. 5, 21 L.Ed.2d 24. 15 The Court's articulation of the applicable standard in this case is somewhat puzzling. It states that this case is subject to "the standard of proof generally applicable to Equal Protection Clause cases." Ante, at 617. But later in the same paragraph, the Court indicates that its special requirement of a showing of discriminatory intent merely applies to equal protection cases "charging racial discrimination." Ibid. The Court seems to imply that plaintiffs in equal protection cases charging racial discrimination must surmount a special hurdle in order to prevail. Yet the Court has unequivocally stated that a "racial classification, regardless of purported motivation, is presumptively invalid and can be upheld only upon an extraordinary justification." Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 272, 99 S.Ct. 2282, 2292, 60 L.Ed.2d 870. 16 In the words of Chancellor Kent, the requirement of districting "was recommended by the wisdom and justice of giving, as far as possible, to the local subdivisions of the people of each state, a due influence in the choice of representatives, so as not to leave the aggregate minority of the people in a state, though approaching perhaps to a majority, to be wholly overpowered by the combined action of the numerical majority, without any voice whatever in the national councils." 1 J. Kent, Commentaries on American Law *230-*231, n. (c) (12th ed. 1873). See also Mobile v. Bolden, 446 U.S. 55, 105, n. 3, 100 S.Ct. 1490, 1520, 64 L.Ed.2d 47 (MARSHALL, J., dissenting); Whitcomb v. Chavis, 403 U.S. 124, 158-160, 91 S.Ct. 1858, 1877, 29 L.Ed.2d 363. The challenge to multimember or at-large districts is, of course, quite different from the challenge to the value of individual votes considered in Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506. An at-large system is entirely consistent with the one-person, one-vote rule developed in that case. As Justice Stewart noted in Mobile, in considering the applicability of Reynolds and the cases that followed it: "Those cases established that the Equal Protection Clause guarantees the right of each voter to 'have his vote weighted equally with those of all other citizens.' 377 U.S., at 576, 84 S.Ct., at 1389. The Court recognized that a voter's right to 'have an equally effective voice' in the election of representatives is impaired where representation is not apportioned substantially on a population basis. In such cases, the votes of persons in more populous districts carry less weight than do those of persons in smaller districts. There can be, of course, no claim that the 'one person, one vote' principle has been violated in this case, because the city of Mobile is a unitary electoral district and the Commission elections are conducted at large. It is therefore obvious that nobody's vote has been 'diluted' in the sense in which that word was used in the Reynolds case." 446 U.S., at 77-78, 100 S.Ct., at 1505-1506 (plurality opinion). See also id., at 83, 100 S.Ct., at 1508 (STEVENS, J., concurring in judgment). 17 This feature distinguishes Burke County's at-large electoral system from the municipal commission form of government popularized by reformers shortly after the turn of the century and known as the Galveston Plan or the Des Moines Plan. See n. 19, infra. 18 Other features of certain at-large electoral schemes that make it more difficult for a minority group to elect a favored candidate when bloc voting occurs—prohibitions against cumulative and incomplete voting—are not involved in this case. Prohibitions against cumulative or partial voting are generally inapplicable in electoral schemes involving numbered posts. 19 "During its evolution as a progressive solution to municipal problems, the commission format was variously known as the Galveston plan, the Texas idea, and the Des Moines plan. Since Galveston invented the basic organization and Des Moines popularized the addition of related reform techniques, the new type of government is probably best described as the Galveston—Des Moines plan. So popular did the new idea become that towns could reap advertising benefits for being in the forefront of municipal innovation if they used the commission plan. Consequently, some cities boasted that they had the system, knowing full well that their charters had little resemblance to Galveston's. But there were certain essentials necessary before a city could claim commission status. Benjamin DeWitt, an early historian of the progressive movement, explained: " 'In every case, however, no matter how much charters may differ as to minor details, they have certain fundamental features in common. These fundamental features of commission charters are four: "1. Authority and responsibility are centralized. "2. The number of men in whom this authority and this responsibility are vested is small. "3. These few men are elected from the city at large and not by wards or districts. "4. Each man is at the head of a single department.' "The most radical departure the new scheme made was the combination of legislative and executive functions in one body. The plan disregarded the federal model of separation of powers. Sitting together, the commission was a typical policy- and ordinance-making council; but, separately, each commissioner administered a specific department on a day-to-day basis. The original Galveston charter provided for a mayor-president plus commissioners of finance and revenue, waterworks and sewerage, streets and public property, and fire and police. Later commission cities followed a similar division of responsibility." B. Rice, Progressive Cities: The Commission Government Movement in America, 1901-1920, pp. xiii-xiv (1977) (footnote omitted). 20 It is noteworthy that these features apparently characterize many governmental units in jurisdictions that have been subjected to the strictures of the Voting Rights Act as the result of prior practices that excluded black citizens from the electoral process. See generally The Voting Rights Act: Unfulfilled Goals, A Report of the United States Commission on Civil Rights 38-50 (1981). 21 No group has a right to proportional representation. See Mobile v. Bolden, 446 U.S., at 75-76, 100 S.Ct., at 1504-1505 (plurality opinion); id., at 122, 100 S.Ct., at 1530 (MARSHALL, J., dissenting). But in a representative democracy, meaningful participation by minority groups in the electoral process is essential to ensure that representative bodies are responsive to the entire electorate. For this reason, a challenged electoral procedure may not be justified solely on the ground that it serves to reduce the ability of a minority group to participate effectively in the electoral process. 22 The record nevertheless does indicate that the validity of the at-large system itself need not be decided in this case. For it is apparent that elimination of the majority runoff requirement and the numbered posts would enable a well-organized minority to elect one or two candidates to the County Board. That consequence could be achieved without replacing the at-large system itself with five single-member districts. In other words, minority access to the political process could be effected by invalidating specific rules that impede that access and without changing the basic structure of the local governmental unit. See Mobile v. Bolden, supra, at 80, 100 S.Ct., at 1507 (BLACKMUN, J., concurring in result). 23 It is worth repeating the statement of Professor Ely noted by Justice MARSHALL: "The danger I see is the somewhat different one that the Court, in its new-found enthusiasm for motivation analysis, will seek to export it to fields where it has no business. It therefore cannot be emphasized too strongly that analysis of motivation is appropriate only to claims of improper discrimination in the distribution of goods that are constitutionally gratuitous (that is, benefits to which people are not entitled as a matter of substantive constitutional right). In such cases the covert employment of a principle of selection that could not constitutionally be employed overtly is equally unconstitutional. However, where what is denied is something to which the complainant has a substantive constitutional right either because it is granted by the terms of the Constitution, or because it is essential to the effective functioning of a democratic government—the reasons it was denied are irrelevant. It may become important in court what justifications counsel for the state can articulate in support of its denial or non-provision, but the reasons that actually inspired the denial never can: To have a right to something is to have a claim on it irrespective of why it is denied. It would be a tragedy of the first order were the Court to expand its burgeoning awareness of the relevance of motivation into the thoroughly mistaken notion that a denial of a constitutional right does not count as such unless it was intentional." Ely, The Centrality and Limits of Motivation Analysis, 15 San Diego L.Rev. 1155, 1160-1161 (1978) (emphasis in original) (footnotes omitted). 24 In Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed.2d 597, I wrote: "Frequently the most probative evidence of intent will be objective evidence of what actually happened rather than evidence describing the subjective state of mind of the actor. For normally the actor is presumed to have intended the natural consequences of his deeds. This is particularly true in the case of governmental action which is frequently the product of compromise, of collective decisionmaking, and of mixed motivation. It is unrealistic, on the one hand, to require the victim of alleged discrimination to uncover the actual subjective intent of the decisionmaker or, conversely, to invalidate otherwise legitimate action simply because an improper motive affected the deliberation of a participant in the decisional process. A law conscripting clerics should not be invalidated because an atheist voted for it." Id., at 253, 96 S.Ct., at 2054 (concurring opinion). 25 See Miller, If "The Devil Himself Knows Not the Mind of Man," How Possibly Can Judges Know the Motivation of Legislators?, 15 San Diego L.Rev. 1167, 1170 (1978). 26 The following table shows a breakdown of the population of the districts in the plan selected by the District Court as to race and voting age: Voting Age Black Voting Age White Voting Age District Population Population (%) Population (%) 1 2,048 1,482 (72.4) 556 (27.6) 2 2,029 1,407 (69.3) 622 (30.7) 3 2,115 978 (46.2) 1,137 (53.8) 4 2,112 947 (44.6) 1,175 (55.4) 5 2,217 803 (36.2) 1,414 (63.8) See Lodge v. Buxton, 639 F.2d 1358, 1361, n. 4 (CA5 1981). 27 In Wright v. Rockefeller, 376 U.S. 52, 84 S.Ct. 603, 11 L.Ed.2d 512, a group of minority voters in New York City challenged a districting scheme that placed most minority voters in one of four districts. They sought "a more even distribution of minority groups among the four congressional districts." Id., at 58, 84 S.Ct., at 606. Congressman Adam Clayton Powell intervened in the lawsuit and argued strenuously "that the kind of districts for which appellants contended would be undesirable and, because based on race or place of origin, would themselves be unconstitutional." Ibid. 28 Professor Karst has strongly criticized motivational analysis on the ground that it is inadequate to protect black citizens from unconstitutional conduct: "[E]ven though the proof will center on the effects of what officials have done, the ultimate issue will be posed in terms of the goodness or the evil of the officials' hearts. Courts have long regarded such inquiries as unseemly, as the legislative investigation cases of the 1950's attest. The principal concern here is not that tender judicial sensitivities may be bruised, but that a judge's reluctance to challenge the purity of other officials' motives may cause her to fail to recognize valid claims of racial discrimination even when the motives for governmental action are highly suspect. Because an individual's behavior results from the interaction of a multitude of motives, and because racial attitudes often operate at the margin of consciousness, in any given case there almost certainly will be an opportunity for a governmental official to argue that his action was prompted by racially neutral considerations. When that argument is made, should we not expect the judge to give the official the benefit of the moral doubt? When the governmental action is the product of a group decision, will not that tendency toward generosity be heightened?" Karst, The Costs of Motive-Centered Inquiry, 15 San Diego L.Rev. 1163, 1164-1165 (1978) (footnote omitted). To reject an examination into subjective intent is not to rule that the reasons for legislative action are irrelevant. "In my opinion, customary indicia of legislative intent provide an adequate basis for ascertaining the purpose that a law is intended to achieve. The formal proceedings of the legislature and its committees, the effect of the measure as evidenced by its text, the historical setting in which it was enacted, and the public acts and deeds of its sponsors and opponents, provide appropriate evidence of legislative purpose." Cousins v. City Council of Chicago, 466 F.2d 830, 856 (CA7 1972) (Stevens, J., dissenting), cert. denied, 409 U.S. 893, 93 S.Ct. 85, 34 L.Ed.2d 151. If a challenged law disadvantages minority citizens and its justifications—as evidenced by customary indicia of legislative intent—are insufficient to persuade a neutral observer that the law was enacted for legitimate, nondiscriminatory reasons, it is, in my opinion, invalid. 29 As the Court of Appeals noted: "The general election laws in many jurisdictions were originally adopted at a time when Blacks had not receiv[ed] their franchise. No one disputes that such laws were not adopted to achieve an end, the exclusion of Black voting, that was the status quo. Other states' election laws, though adopted shortly after the enactment of the Fifteenth Amendment, are so old that whatever evidence of discriminatory intent may have existed, has long since disappeared. This case falls within that category. The focus then becomes the existence of a discriminatory purpose for the maintenance of such a system." 639 F.2d, at 1363, n. 7. 30 Apart from the lack of "state action," the very purpose of the secret ballot is to protect the individual's right to cast a vote without explaining to anyone for whom, or for what reason, the vote is cast. 31 A stereotypical reaction to particular characteristics of a disfavored group cannot justify discriminatory legislation. See, e.g., Mathews v. Lucas, 427 U.S. 495, 520-521, 96 S.Ct. 2755, 2769, 49 L.Ed.2d 651 (STEVENS, J., dissenting). It is nevertheless important to remember that the First Amendment protects an individual's right to entertain unsound and unpopular beliefs—including stereotypical beliefs about classes of persons and to expound those beliefs publicly. There is a vast difference between rejecting an irrational belief as a justification for discriminatory legislation and concluding that neutral legislation is invalid because it was motivated by an irrational belief. Fresh air and open discussion are better cures for vicious prejudice than are secrecy and dissembling. No matter how firmly I might disagree with a legislator's motivation in casting a biased vote, I not only must respect his right to form his own opinions, cf. Young v. American Mini Theatres, Inc., 427 U.S. 50, 63, 96 S.Ct. 2440, 2448, 49 L.Ed.2d 310 (opinion of STEVENS, J.), but also would prefer a candid explanation of those opinions to a litigation-oriented silence. 32 See United Jewish Organization v. Carey, 430 U.S. 144, 97 S.Ct. 996, 51 L.Ed.2d 229. 33 See Cousins v. City Council of Chicago, 466 F.2d 830 (CA7 1972), cert. denied, 409 U.S. 893, 93 S.Ct. 85, 34 L.Ed.2d 151. 34 See White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314. 35 See Rice, supra n. 19, at 78. 36 The term "gerrymander" arose from an election district that took the shape of a salamander—formed in Massachusetts by Governor Elbridge Gerry's Jeffersonian or Democratic-Republican Party. The phrase was coined by Gerry's opponents, the Federalists. 37 See 39 Congressional Quarterly, 758 (1981). 38 See id., at 941. 39 Since I do not understand the Court's opinion to rely on an affirmative-action rationale, I put that entire subject to one side. If that were the rationale for the Court's holding, however, there would be no need to inquire into subjective intent. 40 The Court does not address the statutory question whether the at-large system violates § 2 of the Voting Rights Act of 1965. Neither the District Court nor the Court of Appeals considered this issue. Since appellees have been granted full relief by the Court, I express no opinion on their statutory claims.

12

458 U.S. 670 102 S.Ct. 3304 73 L.Ed.2d 1057 FLORIDA DEPARTMENT OF STATE, Petitioner,v.TREASURE SALVORS, INC., etc. No. 80-1348. Argued Jan. 20, 1982. Decided July 1, 1982. Syllabus After respondents had located the wreck of a 17th-century Spanish galleon off the Florida coast, Florida immediately claimed ownership of the galleon pursuant to a Florida statute. Contracts were then entered into between the Florida Division of Archives, as owner of the galleon and its cargo, and respondents, whereby respondents agreed to conduct underwater salvage operations in exchange for the Division's agreement to transfer ownership of 75% of the appraised value of all material recovered from the galleon to respondents. The contracts did not purport to transfer ownership of any property to the Division. Ultimately, many valuable artifacts of the galleon were discovered. In the meantime, in proceedings unrelated to the salvage operations, it was held in United States v. Florida, 420 U.S. 531, 95 S.Ct. 1162, 43 L.Ed.2d 375, that, as against Florida, the United States was entitled to the lands, minerals, and other natural resources in the area in which the remains of the galleon had come to rest. Respondents thereafter filed an admiralty in rem action in the Federal District Court for the Southern District of Florida, naming the galleon as defendant but not the State of Florida and seeking a declaration of title to the galleon. Throughout the ensuing proceedings, in which the United States intervened and in which both the District Court and the Court of Appeals on appeal rejected the United States' claim to ownership of the galleon, some of the valuable artifacts remained in the custody of officials of the Florida Division of Archives in Tallahassee, which is located beyond the District Court's territorial jurisdiction. After the Court of Appeals' decision, respondents filed a motion in the District Court for an order commanding the United States Marshal to arrest and take custody of those artifacts and bring them within the court's jurisdiction. The District Court granted the motion and issued a warrant of arrest. Although the warrant was addressed to the state officials, the State itself filed a motion to quash the warrant, but the court denied this motion, ruling that the extraterritorial seizure was proper under Supplemental Admiralty Rule C(5), and issued an order to show cause why the State should not deliver the artifacts into the Marshal's custody. The State then argued that the Eleventh Amendment barred exercise of the District Court's jurisdiction, but the District Court rejected this argument, holding that the State had waived the Eleventh Amendment as to any claim to the property, and that, apart from any such claim, the Eleventh Amendment did not bar the seizure of the artifacts and subsequent transfer to the Marshal's custody. On the merits, the court also rejected the State's claim to the property based on the salvage contracts with respondents. The Court of Appeals affirmed. Held : The judgment is affirmed in part and reversed in part. 5 Cir., 621 F.2d 1340, affirmed in part and reversed in part. Justice STEVENS, joined by THE CHIEF JUSTICE, Justice MARSHALL, and Justice BLACKMUN, concluded that: 1 1. The Eleventh Amendment did not bar the process issued by the District Court to secure possession of the artifacts held by the state officials. Pp. 683-699. 2 (a) The Eleventh Amendment, while barring an action directly against the State itself or any agency thereof, does not bar an action against a state official that is based on the theory that the official acted beyond the scope of his statutory authority or, if within that authority, that such authority is unconstitutional. The Eleventh Amendment, however, limits the relief that may be recovered in the latter kind of action; the judgment may not compel the State to use its funds to compensate the plaintiff for his injury. Pp. 683-690. 3 (b) Here, the process at issue is not barred by the Eleventh Amendment as a direct action against the State, because it was directed only at state officials. Neither the fact that the State elected to defend on behalf of the officials, nor the fact that the District Court purported to adjudicate the State's rights, deprives that court of jurisdiction that had been properly invoked over other parties. Pp.691-692 4 (c) The state officials named in the warrant of arrest do not have a colorable claim to possession of the artifacts, and thus may not invoke the Eleventh Amendment to block execution of the warrant. The salvage contracts, whether valid or not, provide no authority for the officials' refusal to surrender possession of the artifacts, and no statutory provision that even arguably would authorize the officials to retain the artifacts has been advanced. Pp. 692-697. 5 (d) The relief sought by respondents is not barred by the Eleventh Amendment but is consistent with the principles of Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662. The warrant of arrest sought possession of specific property. It did not seek any attachment of state funds and would impose no burden on the state treasury. And respondents are not asserting a claim for damages against either the State or its officers. Pp. 697-699. 6 2. The proper resolution of the Eleventh Amendment issue does not require—or permit—a determination of the State's ownership of the artifacts, and hence the Court of Appeals improperly adjudicated the State's right to the artifacts. Pp.699-700 7 Justice BRENNAN, while agreeing with the opinion that the State of Florida has not established even a colorable claim to the artifacts, concluded that the Eleventh Amendment is inapplicable in this case because both respondents are Florida corporations and thus the suit was not "commenced or prosecuted against one of the United States by citizens of another State," as the Eleventh Amendment provides. Pp. 700-702. 8 Justice WHITE, joined by Justice POWELL, Justice REHNQUIST, and Justice O'CONNOR, concurred in the Court's judgment insofar as it reverses the Court of Appeals' determination of the State's ownership of the artifacts. P. 703, n. *. 9 Susan Gamble Smathers, Tallahassee, Fla., for petitioner, pro hac vice, by special leave of Court. 10 David Paul Horan, Key West, Fla., for respondent. 11 Justice STEVENS announced the judgment of the Court and delivered an opinion, in which THE CHIEF JUSTICE, Justice MARSHALL, and Justice BLACKMUN joined. 12 In this admiralty in rem action, a federal court attempted to arrest property held by two state officials and bring it within the jurisdiction of the court. The property—artifacts of the Nuestra Senora de Atocha, a 17th-century Spanish galleon—was discovered by respondents on the floor of the ocean in international waters. The question presented is whether the Eleventh Amendment immunized the property from the federal court's process. 13 * Battered by a tropical hurricane, the Nuestra Senora de Atocha, a Spanish galleon carrying a cargo of New World treasure to King Philip IV of Spain, sank in 1622, 40 nautical miles west of what is today Key West, Fla. After years of searching the ocean floor and studying Spanish archives in Seville, respondent Treasure Salvors1 located the wreck site in the spring of 1971 near shoals known as the "Quicksands," nine and one-half nautical miles west of the Marquesas Keys.2 14 The State of Florida immediately claimed that the Atocha belonged to the State. The State claimed ownership pursuant to Fla.Stat. § 267.061(1)(b) (1974), which then provided:3 15 "It is further declared to be the public policy of the state that all treasure trove, artifacts and such objects having intrinsic or historical and archeological value which have been abandoned on state-owned lands or state-owned sovereignty submerged lands shall belong to the state with the title thereto vested in the division of archives, history, and records management of the department of state for the purpose of administration and protection." (Emphasis added.) 16 Officials of the Florida Division of Archives threatened to arrest Mel Fisher, president of Treasure Salvors, and to confiscate the boats and equipment of Treasure Salvors if it commenced salvage operations on the Atocha without a salvage contract from the State. Under this threat of arrest, Treasure Salvors executed a one-year contract with the State that permitted it to conduct underwater salvage operations on the vessel.4 Similar contracts were executed during each of the three succeeding years. 17 Each of the contracts was expressly predicated on the assumption that the Atocha was the property of the State of Florida because it had been found on submerged lands within the boundaries of the State. The contracts permitted Treasure Salvors "to conduct underwater salvage from and upon certain submerged sovereignty lands of and belonging to the State of Florida." App. 20. After describing in metes and bounds an area claimed to be "lying and being in Monroe County, Florida," the contract provided that the shipwreck site "is to be worked for the purpose of salvaging abandoned vessels or the remains thereof including, but not limited to, relics, treasure trove and other materials related thereto and located thereupon and therein, which abandoned material is the property of the State of Florida." Id., at 22 (emphasis added). The contract further provided: "In payment for the Salvager's satisfactory performance and compliance with this Agreement, the Division will award to the Salvager seventy-five percent (75%) of the total appraised value of all material recovered hereunder, which payment shall be made at the time division of such material is made by the parties hereto. Said payment may be made in either recovered material or fair market value, or in a combination of both, at the option of the Division's director." Id., at 32-33. 18 The bargain, in brief, was between the Division of Archives, as the owner of the Atocha and its cargo, and Treasure Salvors, as a contractor that agreed to perform services for the Division. Treasure Salvors agreed to pay the Division $1,200 each year, to post a performance bond, and to perform its work in a specified manner, all in exchange for the Division's agreement to transfer ownership of 75% of the proceeds of the operation—or its equivalent—to Treasure Salvors. The contracts did not purport to transfer ownership of any property to the Division of Archives; the State's claim to the property was predicated entirely on a provision of state law. 19 In its attempt to salvage the lost treasure of the Atocha, Treasure Salvors was immensely successful. The salvager held some of the artifacts at its headquarters in Key West, while state officials held the remainder at the Division of Archives in Tallahassee. All of the property was deemed to belong to the State, however, subject to a subsequent distribution in which Treasure Salvors would receive its 75% contractual share. 20 In proceedings unrelated to the salvage operation, the United States and the State of Florida were engaged in litigation to determine the seaward boundary of submerged lands in the Atlantic Ocean and the Gulf of Mexico in which the State had rights to natural resources. In February 1974, a Special Master filed a Report that defined Florida's boundary landward of the site of the wreck of the Atocha. The State's objections to the Report were overruled. United States v. Florida, 420 U.S. 531, 95 S.Ct. 1162, 43 L.Ed.2d 375 (1975).5 A final decree was entered providing that, as against the State of Florida, the United States was entitled to the lands, minerals, and other natural resources in the area in which the remains of the Atocha had come to rest. United States v. Florida, 425 U.S. 791, 96 S.Ct. 1840, 48 L.Ed.2d 388 (1976).6 21 After this Court overruled Florida's exceptions to the Special Master's Report, Treasure Salvors filed a complaint in the Federal District Court for the Southern District of Florida demanding that "Plaintiffs be put into possession of the ATOCHA and other property and that all other persons, firms, and corporations or government agencies be enjoined from interfering with Plaintiffs title, possession, and property," and that "Plaintiffs title be confirmed against all claimants and all the world." App. 9. The complaint invoked the court's admiralty and maritime jurisdiction pursuant to Federal Rule of Civil Procedure 9(h) and, as an admiralty action in rem, named the Atocha as defendant. Items recovered from the Atocha in Treasure Salvors' possession were duly served with process and brought into the custody of the court. Most of the remainder of the wreck and its valuable cargo lay buried under sand in international waters; state officials held other artifacts in Tallahassee. No attempt was made at this time to serve the artifacts in Tallahassee. 22 The United States intervened in the action as a party-defendant and filed a counterclaim seeking a declaratory judgment that the United States was the proper owner of the Atocha.7 The District Court rejected the Government's claim of ownership and held that "possession and title are rightfully conferred upon the finder of the res derelictae." Treasure Salvors, Inc. v. Abandoned Sailing Vessel, 408 F.Supp. 907, 911 (1976). The court entered judgment in favor of Treasure Salvors "against the United States of America and all other claimants." Record 270.8 23 The Court of Appeals affirmed the judgment of the District Court as against the United States, but modified its decree. Treasure Salvors, Inc. v. Unidentified Wrecked and Abandoned Sailing Vessel, 569 F.2d 330 (CA5 1978). The United States had argued that the District Court lacked in rem jurisdiction to determine rights of the parties to that portion of the Atocha lying beyond the territorial jurisdiction of the court. The Court of Appeals agreed that the District Court lacked in rem jurisdiction over those portions of the res located outside the district; the court noted that for a court to exercise admiralty in rem jurisdiction the res itself must be brought within the district and seized by the court. Id., at 333. The appellate court held, however, that by intervening in the action and stipulating to the court's admiralty jurisdiction the Government had "waived the usual requirement that the res be present within the territorial jurisdiction of the court and consented to the court's jurisdiction to determine its interest in the extraterritorial portion of the vessel." Id., at 335. The court concluded that jurisdiction thus existed to determine claims of the United States to those portions of the Atocha lying beyond the territorial jurisdiction of the court, but not claims of other parties who had not appeared and submitted to the jurisdiction of the court.9 On the merits, the Court of Appeals rejected the statutory and common-law claims advanced by the United States. 24 Throughout these proceedings, valuable artifacts of the Atocha remained in the custody of officials of the Florida Division of Archives in Tallahassee. Since Tallahassee is located in the Northern District of Florida, these artifacts also were located beyond the territorial jurisdiction of the District Court. Immediately following the decision of the Court of Appeals, Treasure Salvors filed a motion in the District Court for an order commanding the United States Marshal to arrest and take custody of these artifacts and bring them within the jurisdiction of the court. Record 318. That motion forms the basis of the present controversy. 25 The District Court issued a warrant to arrest.10 Although the warrant was addressed to two officers of the Division of Archives, the State itself filed a motion to quash the warrant, contending that the State of Florida was not a party in the case and had not waived the requirement that the court could exercise in rem jurisdiction only over that portion of the res within the territorial boundaries of the court. App. 43.11 The State also sought and obtained an emergency stay from the Court of Appeals. Record 368. The District Court denied the motion to quash, ruling that the extraterritorial seizure was proper under Supplemental Admiralty Rule C(5). App. 51.12 Since the Court of Appeals had stayed execution of the warrant, the District Court issued an order to show cause why the State should not deliver the artifacts into the custody of the Marshal.13 26 In response to the order to show cause, the State raised several substantive issues in the District Court. Record 425. Contending that a supplemental complaint filed by Treasure Salvors, see n. 11, supra, demonstrated that the State of Florida was a defendant in the action, the State argued that the Eleventh Amendment barred an exercise of the court's jurisdiction. The State also repeated its arguments that the court lacked in rem jurisdiction in admiralty because the res was not present within the district and that the decision of this Court in United States v. Florida did not affect the State's "contractual" right to a share of the artifacts. Record 429-439. 27 The District Court rejected these arguments in a comprehensive memorandum. Treasure Salvors, Inc. v. Unidentified Wrecked and Abandoned Sailing Vessel, 459 F.Supp. 507 (1978). The court first held that, just as all claims of the United States had been resolved in the earlier proceeding, all claims of the State were barred because the State of Florida had acted in privity with the United States in that proceeding. Id., at 512; see n. 7, supra. Alternatively, the court held that the extraterritorial arrest of the salvaged articles was proper under Supplemental Admiralty Rule C(5) and that the court thus had obtained jurisdiction in rem to resolve ownership of the res. 459 F.Supp., at 518. On the merits, the court rejected on multiple grounds the State's contractual claim to the property. Id., at 521. 28 At the conclusion of its memorandum opinion, the court rejected the State's Eleventh Amendment defense. Id., at 526. The court first held that the State necessarily had waived the Amendment as to any claim to the property that it asserted in federal court. Ibid. The court then held that, apart from any claim advanced by the State, the Eleventh Amendment did not bar the seizure of the artifacts and subsequent transfer to the custody of the Marshal.14 29 The Court of Appeals affirmed. 621 F.2d 1340 (CA5 1980). As had the District Court, see n. 14, supra, the court concluded that the Eleventh Amendment did not prevent the court from resolving the controverted claims to ownership of the res, since resolution of that dispute was essential to a determination of whether the Eleventh Amendment in fact barred an exercise of jurisdiction by the federal court. 621 F.2d, at 1345.15 The court then held that the extraterritorial process issued pursuant to Supplemental Admiralty Rule C(5) was proper, id., at 1346, and that the State did not have a valid claim to the property. Id., at 1349.16 30 The Florida Department of State filed a petition for writ of certiorari, presenting only one question: "Whether the Eleventh Amendment to the United States Constitution bars an in rem admiralty action seeking to recover property owned by a state." Pet. for Cert. I. We granted the petition. 451 U.S. 982, 101 S.Ct. 2312, 68 L.Ed.2d 838. We hold that the federal court had jurisdiction to secure possession of the property from the named state officials, since they had no colorable basis on which to retain possession of the artifacts. The court did not have power, however, to adjudicate the State's interest in the property without the State's consent. II 31 Stripped of its procedural complexities and factual glamor, this case presents a narrow legal question. The District Court attempted to seize artifacts held by state officials and to bring the property within its admiralty in rem jurisdiction. Although the seizure in this case was extraterritorial, and thus involved an application of Supplemental Admiralty Rule C(5), the question presented for our decision would not be any different if the State merely resisted an attachment of property located within the district. 32 In response to the warrant of arrest, the State contended that it was immune from the federal process under the Eleventh Amendment.17 It argued that the contracts executed with Treasure Salvors "alone determined the rights and obligations of the contracting parties . . . ." App. 44. The difficult question presented in this case is whether a federal court exercising admiralty in rem jurisdiction may seize property held by state officials under a claim that the property belongs to the State.18 33 suit generally may not be maintained directly against the State itself, or against an agency or department of the State, unless the State has waived its sovereign immunity. Alabama v. Pugh, 438 U.S. 781, 98 S.Ct. 3057, 57 L.Ed.2d 1114. If the State is named directly in the complaint and has not consented to the suit, it must be dismissed from the action. Id., at 782, 98 S.Ct., at 3057.19 Of course, the fact that the State should have been dismissed from an action that has proceeded to judgment does not mean that the judgment may not stand against other parties who are not immune from suit.20 34 The Eleventh Amendment does not bar all claims against officers of the State, even when directed to actions taken in their official capacity and defended by the most senior legal officers in the executive branch of the state government. In Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714, the Court held that an action brought against a state official to enjoin the enforcement of an unconstitutional state statute is not a suit against a State barred by the Eleventh Amendment. In response to the argument that the official in such a case could act only as an officer of the State and that the suit therefore could be characterized only as an action against the State itself, the Court explained: 35 "The act to be enforced is alleged to be unconstitutional, and if it be so, the use of the name of the State to enforce an unconstitutional act to the injury of complainants is a proceeding without the authority of and one which does not affect the State in its sovereign or governmental capacity. It is simply an illegal act upon the part of a state official in attempting by the use of the name of the State to enforce a legislative enactment which is void because unconstitutional. If the act which the state Attorney General seeks to enforce is a violation of the Federal Constitution, the officer in proceeding under such enactment comes into conflict with the superior authority of that Constitution, and he is in that case stripped of his official or representative character and is subjected in his person to the consequences of his individual conduct. The State has no power to impart to him any immunity from responsibility to the supreme authority of the United States." Id., at 159-160, 28 S.Ct., at 454. 36 There is a well-recognized irony in Ex parte Young; unconstitutional conduct by a state officer may be "state action" for purposes of the Fourteenth Amendment yet not attributable to the State for purposes of the Eleventh. Nevertheless, the rule of Ex parte Young is one of the cornerstones of the Court's Eleventh Amendment jurisprudence. See Edelman v. Jordan, 415 U.S. 651, 663-664, 94 S.Ct. 1347, 1355-1356, 39 L.Ed.2d 662; Quern v. Jordan, 440 U.S. 332, 337, 99 S.Ct. 1139, 1143, 59 L.Ed.2d 358. 37 In Tindal v. Wesley, 167 U.S. 204, 17 S.Ct. 770, 42 L.Ed. 137, the Court applied the analysis later enshrined in Ex parte Young in a suit to recover property wrongfully held by state officials on behalf of the State of South Carolina. In Tindal, the plaintiff claimed title and a right of possession to certain real property held by a state official; the defendant answered that the property belonged to the State and asserted the Eleventh Amendment as a defense to the action. The Court described the issue presented for decision: 38 "So that the question is directly presented, whether an action brought against individuals to recover the possession of land of which they have actual possession and control, is to be deemed an action against the State within the meaning of the Constitution, simply because those individuals claim to be in rightful possession as officers or agents of the State, and assert title and right of possession in the State. Can the court, in such an action, decline to inquire whether the plaintiff is, in law, entitled to possession, and whether the individual defendants have any right, in law, to withhold possession? And if the court finds, upon due inquiry, that the plaintiff is entitled to possession, and that the assertion by the defendants of right of possession and title in the State is without legal foundation, may it not, as between the plaintiff and the defendants, adjudge that the plaintiff recover possession?" 167 U.S., at 212, 17 S.Ct., at 773. 39 Relying extensively on the earlier decision in United States v. Lee, 106 U.S. 196, 1 S.Ct. 240, 27 L.Ed.171,21 the Court in Tindal held that the "settled doctrine of this court wholly precludes the idea that a suit against individuals to recover possession of real property is a suit against the State simply because the defendant holding possession happens to be an officer of the State and asserts that he is lawfully in possession on its behalf." 167 U.S., at 221, 17 S.Ct., at 777. The Court refused to accept the proposition that the "doors of the courts of justice are . . . closed against one legally entitled to possession, by the mere assertion of the defendants that they are entitled to possession for the State." Id., at 222, 17 S.Ct., at 777. In explaining the extent of its decision, the Court stated: 40 "[T]he Eleventh Amendment gives no immunity to officers or agents of a State in withholding the property of a citizen without the authority of law. And when such officers or agents assert that they are in rightful possession, they must make good that assertion when it is made to appear in a suit against them as individuals that the legal title and right of possession is in the plaintiff. If a suit against officers of a State to enjoin them from enforcing an unconstitutional statute, whereby the plaintiff's property will be injured . . . be not one against the State, it is impossible to see how a suit against the same individuals to recover the possession of property belonging to the plaintiff and illegally withheld by the defendants can be deemed a suit against the State." Ibid.22 41 In holding that the action was not barred by the Eleventh Amendment, the Court in Tindal emphasized that any judgment awarding possession to the plaintiff would not subsequently bind the State. "It is a judgment to the effect only that, as between the plaintiff and the defendants, the former is entitled to possession of the property in question, the latter having shown no valid authority to withhold possession from the plaintiff," id., at 223, 17 S.Ct., at 777; "it will be open to the State to bring any action that may be appropriate to establish and protect whatever claim it has to the premises in dispute." Ibid. 42 The rule of law set forth in United States v. Lee and Tindal v. Wesley was clarified in Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 69 S.Ct. 1457, 93 L.Ed. 1628. In that case the plaintiff brought suit against a Government official to compel specific performance of a contract.23 The plaintiff theorized that by withholding delivery of property as required by the contract the agent had exceeded his official authority and could be sued in federal court. The Court in Larson stated that "the action of an officer of the sovereign (be it holding, taking or otherwise legally affecting the plaintiff's property) can be regarded as so 'illegal' as to permit a suit for specific relief against the officer as an individual only if it is not within the officer's statutory powers or, if within those powers, only if the powers, or their exercise in a particular case, are constitutionally void." Id., at 701-702, 69 S.Ct., at 1467. The Court held that the fact that an officer wrongfully withholds property belonging to another does not necessarily establish that he is acting beyond the permissible scope of his official capacity.24 Since in Larson it was not alleged that the Government official had exceeded his statutory authority—indeed, the plaintiff had affirmatively contended that the officer had authority to bind the Government on the contract at issue25—or that the exercise of such authority was unconstitutional,26 the Court held that the action was barred by sovereign immunity. 43 These cases make clear that the Eleventh Amendment does not bar an action against a state official that is based on a theory that the officer acted beyond the scope of his statutory authority or, if within that authority, that such authority is unconstitutional. In such an action, however, the Amendment places a limit on the relief that may be obtained by the plaintiff. If the action is allowed to proceed against the officer only because he acted without proper authority, the judgment may not compel the State to use its funds to compensate the plaintiff for the injury. In Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662, the Court made clear that "a suit by private parties seeking to impose a liability which must be paid from public funds in the state treasury is barred by the Eleventh Amendment." Id., at 663, 94 S.Ct., at 1356. See Ford Motor Co. v. Department of Treasury, 323 U.S. 459, 65 S.Ct. 347, 89 L.Ed. 389; Quern v. Jordan, 440 U.S., at 337, 99 S.Ct., at 1143.27 In determining the relief that may be granted if a state officer is found to have acted without valid statutory authority, the question is whether the relief "constitute[s] permissible prospective relief or a 'retroactive award which requires the payment of funds from the state treasury.' " Quern v. Jordan, supra, at 346-347, 99 S.Ct., at 1147-1148. III 44 In light of the principles set forth above, the proper resolution of the Eleventh Amendment issue raised in this case requires an answer to each of three specific questions: (a) Is this action asserted against officials of the State or is it an action brought directly against the State of Florida itself? (b) Does the challenged conduct of state officials constitute an ultra vires or unconstitutional withholding of property or merely a tortious interference with property rights? (c) Is the relief sought by Treasure Salvors permissible prospective relief or is it analogous to a retroactive award that requires "the payment of funds from the state treasury"? 45 * Treasure Salvors filed this admiralty in rem action in federal court, seeking a declaration of title to an abandoned sailing vessel that had been discovered on the ocean floor. The State of Florida was not named as a party and was not compelled to appear. Some of the property at issue, however, was held by officials of the Florida Division of Archives. Asserting that it was the rightful owner of the property, Treasure Salvors filed a motion "for an Order commanding the United States Marshal to arrest and take custody of those portions of the Plaintiffs' vessel now being held by L. Ross Morrell or James McBeth or being held under their custody, care or control." App. 11.28 As requested, the District Court issued a warrant of arrest commanding the Marshal of the United States for the Southern District of Florida "to take into your possession the portions of said vessel which have been in the possession or are in the possession of L. Ross Morrell and/or James McBeth, or under their custody, care or control and to bring said portions of said vessel within the jurisdiction of this Honorable Court and transfer possession of same to the substitute custodian appointed in this action." Id., at 41-42. It is this process from which the State contends it is immune under the Eleventh Amendment.29 46 It is clear that the process at issue was directed only at state officials and not at the State itself or any agency of the State.30 Neither the fact that the State elected to defend on behalf of its agents, nor the fact that the District Court purported to adjudicate the rights of the State, deprived the federal court of jurisdiction that had been properly invoked over other parties. See Alabama v. Pugh, 438 U.S. 781, 98 S.Ct. 3057, 57 L.Ed.2d 1114; n. 20, supra. The process thus is not barred by the Eleventh Amendment as a direct action against the State. B 47 The second question that must be considered is whether the state officials named in the warrant acted without legitimate authority in withholding the property at issue. In Treasure Salvors' first response to the State's Eleventh Amendment argument, it contended: 48 "If the Division of Archives were allowed to retain this property, its officials would be acting outside the scope of their authority under state law since the state statute under which they claim [does] not apply outside the states territory. The rationale of Home Tel. & Tel. Co. v. Los Angeles, [227 U.S. 278, 33 S.Ct. 312, 57 L.Ed. 510 (1913),] prohibits this result since to allow such action would be to deprive Treasure Salvors their property without due process in violation of the Fourteenth Amendment to the Constitution of the United States." Record 472. 49 Thus from the outset, Treasure Salvors has asserted that state officials do not have valid statutory authority to hold the property at issue. 50 In Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 69 S.Ct. 1457, 93 L.Ed. 1628, this Court held that the actions of a federal official in withholding the delivery of goods pursuant to his interpretation of a disputed provision of a contract constituted at most a tortious deprivation of property. The proper remedy for the plaintiff was not an action in district court to compel delivery, but a suit for breach of contract in the Court of Claims. Actions of the Government official pursuant to legitimate contractual authority were neither ultra vires nor unconstitutional. 51 From the outset of the proceedings at issue here, the State of Florida has advanced the contracts that it executed with Treasure Salvors as a defense to the federal court's attempt to secure possession of the artifacts held by the named state officials. It is noteworthy, however, that the State has never argued that the contracts conferred upon the State a right of ownership in the artifacts; the contracts simply "determined the rights and obligations of the contracting parties. . . ." App. 44. The State has argued that the contracts are valid and "in no way affected" by the decision of this Court in United States v. Florida, 420 U.S. 531, 95 S.Ct. 1162, 43 L.Ed.2d 375. App. 44.31 52 We are not called upon in this case to determine "the rights and obligations" of two parties to a contract. The issue presented is whether state officials had authority to refuse to surrender possession of the artifacts to the District Court. The salvage contracts are not relevant to that question unless they provide a basis upon which the officials may claim a right to withhold possession of the property. Unless the contracts determine rights of the parties to the property, they are collateral to the issue before us. 53 It is apparent that the State does not have even a colorable claim to the artifacts pursuant to these contracts. The contracts did not purport to transfer ownership of any artifacts to the State; they permitted Treasure Salvors "to conduct underwater salvage from and upon certain submerged sovereignty lands of and belonging to the State of Florida," id., at 20-21, "for the purpose of salvaging abandoned vessels or the remains thereof . . . which abandoned material is the property of the State of Florida." Id., at 22 (emphasis added). The contracts provided for the performance of services on property that was believed to belong in toto to the State of Florida, in exchange for which the State agreed to "award to the Salvager seventy-five percent (75%) of the total appraised value of all material recovered . . . ." Id., at 33. The State did not "yield" its claim to 75% of the artifacts in order to receive an undisputed right to the remaining 25%; the State agreed to pay Treasure Salvors the equivalent of 75% of the proceeds in compensation for the difficult and expensive work undertaken by Treasure Salvors in retrieving from the floor of the ocean property that was believed to belong to the State. 54 The salvage contracts might well provide a basis for a claim to the property by Treasure Salvors; for the contracts did purport to transfer a portion of the artifacts from the State to Treasure Salvors in compensation for the latter's services. Treasure Salvors does claim a right to ownership, but based entirely on the fact that it was the finder of abandoned property and therefore entitled to the property independently of the contracts.32 Thus neither party's rights to ownership is affected in any way by the salvage contracts; whether the contracts are valid or not, they provide no authority for the refusal of state officials to surrender possession of the artifacts. 55 The authority of state officials to claim the artifacts was derived solely from Fla.Stat. § 267.061(1)(b) (1974), which provided: 56 "It is further declared to be the public policy of the state that all treasure trove, artifacts and such objects having intrinsic or historical and archaeological value which have been abandoned on state-owned lands or state-owned sovereignty submerged lands shall belong to the state with the title thereto vested in the division of archives, history and records management of the department of state for the purpose of administration and protection." (Emphasis added.) 57 This Court has determined, however, that the Atocha was not found on "state-owned sovereignty submerged lands." Rather, it was discovered on the Outer Continental Shelf of the United States, beneath international waters.33 58 No statutory provision has been advanced that even arguably would authorize officials of the Division of Archives to retain the property at issue. Throughout this litigation, the State has relied solely on the contracts that it executed with Treasure Salvors as a defense to the federal court's process; those contracts were predicated entirely on a state statute that on its face is inapplicable in this case.34 Actions of state officials in holding property on the assumption that it was found on state land and for that reason belongs to the State—when it is undisputed that the property was not found on state land—is beyond the authority of any reasonable reading of any statute that has been cited to us by the State.35 59 As recognized in Larson, "action of an officer of the sovereign (be it holding, taking or otherwise legally affecting the plaintiff's property)" that is beyond the officer's statutory authority is not action of the sovereign, 337 U.S., at 701, 69 S.Ct., at 1467; a suit for specific relief against the officer is not barred by the Eleventh Amendment. This conclusion follows inevitably from Ex parte Young. If conduct of a state officer taken pursuant to an unconstitutional state statute is deemed to be unauthorized and may be challenged in federal court, conduct undertaken without any authority whatever is also not entitled to Eleventh Amendment immunity. 60 If a statute of the State of Florida were to authorize state officials to hold artifacts in circumstances such as those presented in this case, a substantial constitutional question would be presented. In essence, the State would have authorized state officials to retain property regardless of the manner in which it was acquired, with no duty to provide compensation for a public taking. If the Constitution provided no protection against such unbridled authority, all property rights would exist only at the whim of the sovereign. 61 Thus, since the state officials do not have a colorable claim to possession of the artifacts, they may not invoke the Eleventh Amendment to block execution of the warrant of arrest. Of course, the warrant itself merely secures possession of the property; its execution does not finally adjudicate the State's right to the artifacts. See Tindal v. Wesley, 167 U.S., at 223, 17 S.Ct., at 777. In ruling that the Eleventh Amendment does not bar execution of the warrant, we need not decide the extent to which a federal district court exercising admiralty in rem jurisdiction over property before the court may adjudicate the rights of claimants to that property as against sovereigns that did not appear and voluntarily assert any claim that they had to the res. C 62 Finally, it is clear that the relief sought in this case is consistent with the principles of Edelman v. Jordan, 415 U.S. 651, 94 S.Ct. 1347, 39 L.Ed.2d 662. The arrest warrant sought possession of specific property. It did not seek any attachment of state funds and would impose no burden on the state treasury. 63 This case is quite different from In re New York (I), 256 U.S. 490, 41 S.Ct. 588, 65 L.Ed. 1057, and In re New York (II), 256 U.S. 503, 41 S.Ct. 592, 65 L.Ed. 1063, relied on by the State. In In re New York (I), the plaintiff brought an action in federal court to recover damages caused by canal boats chartered by the State of New York. Pursuant to admiralty practice, the action was brought in rem against the vessels themselves. The owner of the vessels answered the complaint, contending that the action should be directed against the Superintendent of Public Works of the State of New York. The District Court agreed and ordered the Superintendent to appear and answer; in the event that he could not be found the court directed that "the goods and chattels of the State of New York used and controlled by him" should be attached. 256 U.S., at 496, 41 S.Ct., at 589. 64 The Attorney General of the State appeared on behalf of the Superintendent and asserted the Eleventh Amendment as a defense to the action. This Court held that the District Court lacked jurisdiction to proceed against the Superintendent. The Court noted that "the proceedings against which prohibition is here asked have no element of a proceeding in rem, and are in the nature of an action in personam against Mr. Walsh, not individually, but in his capacity as Superintendent of Public Works of the State of New York," id., at 501, 41 S.Ct., at 591, moreover, "[t]here is no suggestion that the Superintendent was or is acting under color of an unconstitutional law, or otherwise than in the due course of his duty under the constitution and laws of the State of New York." Id., at 502, 41 S.Ct., at 591. The Court concluded: "In the fullest sense, therefore, the proceedings are shown by the entire record to be in their nature and effect suits brought by individuals against the State of New York, and therefore—since no consent has been given—beyond the jurisdiction of the courts of the United States." Ibid. In In re New York (II), the plaintiff filed an action in admiralty to recover damages caused by the negligent operation of a canal boat owned by the State of New York. The action was brought in rem and the vessel was arrested. This Court held, as it had in In re New York (I), that the federal court lacked jurisdiction to adjudicate the claim. In broad language urged upon us here, the Court stated that property owned by a State and employed solely for governmental uses was exempt from seizure by admiralty process in rem. 256 U.S., at 511, 41 S.Ct., at 593. The force of the holding in In re New York (II), however, is that an action—otherwise barred as an in personam action against the State—cannot be maintained through seizure of property owned by the State. Otherwise, the Eleventh Amendment could easily be circumvented; an action for damages could be brought simply by first attaching property that belonged to the State and then proceeding in rem. 65 In these cases the plaintiff did not claim an ownership interest in the vessels and did not question the State's assertion of ownership. The sole purpose of the attempted arrests was to enable the court to acquire jurisdiction over a damages claim that was otherwise barred by the Eleventh Amendment. In this case Treasure Salvors is not asserting a claim for damages against either the State of Florida or its officials. The present action is not an in personam action brought to recover damages from the State. The relief sought is not barred by the Eleventh Amendment. IV 66 The Eleventh Amendment thus did not bar the process issued by the District Court to secure possession of artifacts of the Atocha held by the named state officials. The proper resolution of this issue, however, does not require—or permit—a determination of the State's ownership of the artifacts. 67 This resolution of the immunity issue is not consistent with the disposition of the Court of Appeals. The court properly held that the Eleventh Amendment did not bar execution of the warrant of arrest, in making that determination, however, the Court of Appeals improperly adjudicated the State's right to the artifacts. While such an adjudication would be justified if the State voluntarily advanced a claim to the artifacts, it may not be justified as part of the Eleventh Amendment analysis, the only issue before us. 68 For these reasons, the judgment of the Court of Appeals must be affirmed in part and reversed in part. To the extent that the court held that the Eleventh Amendment did not prohibit an execution of the warrant and transfer of the artifacts to Treasure Salvors, its judgment is affirmed. To the extent that the court determined the State's ownership of the artifacts as part of its Eleventh Amendment analysis, its judgment is reversed. 69 It is so ordered. 70 Justice BRENNAN, concurring in the judgment in part and dissenting in part. 71 I agree with the plurality that the Eleventh Amendment prohibited neither an execution of the warrant nor a transfer to respondents of the artifacts at issue in this case. See ante,at 699 this page. My rationale for this conclusion differs from the plurality's, however. Both respondents are corporations organized under the laws of the State of Florida. Thus this suit is not "commenced or prosecuted against one of the United States by citizens of another State." U.S.Const., Amdt. 11 (emphasis added). The plurality asserts that this constitutional provision "long has been held to govern" "actions brought against a State by its own citizens." Ante, at 683, n. 17 (emphasis added), citing Hans v. Louisiana, 134 U.S. 1, 10 S.Ct. 504, 33 L.Ed. 842 (1890). I have long taken the view that Hans did not rely upon the Eleventh Amendment, and that that Amendment does not bar federal court suits against a State when brought by its own citizens. See Employees v. Missouri Public Health Dept., 411 U.S. 279, 309-322, 93 S.Ct. 1614, 1630-1636, 36 L.Ed.2d 251 (1973) (dissenting opinion); Edelman v. Jordan, 415 U.S. 651, 687, 94 S.Ct. 1347, 1367, 39 L.Ed.2d 662 (1974) (dissenting opinion). I adhere to this view, and I therefore believe that the Eleventh Amendment is wholly inapplicable in the present case.* To this extent, I am in agreement with the plurality's disposition. 72 I disagree, however, with the plurality's conclusion that the courts below erred when they "determined the State's ownership of the artifacts as part of [their] Eleventh Amendment analysis." Ante, at 700. The record before us plainly indicates that the State had a full opportunity to present its arguments respecting ownership of the artifacts at issue in this case when the action was in the District Court, and that that court held a full evidentiary hearing on the merits of these arguments. See Treasure Salvors, Inc. v. Unidentified Wrecked and Abandoned Sailing Vessel, 459 F.Supp. 507, 521 (SD Fla.1978); 621 F.2d 1340, 1344 (CA5 1980). The State's arguments were rejected in the District Court, and that rejection was affirmed by the Court of Appeals. The plurality today appears to agree with the courts below that the arguments available to the State on the merits were, and are, insubstantial. Ante, at 694-697. "No statutory provision has been advanced that even arguably would authorize officials of the Division of Archives to retain the property at issue," ante, at 696 (emphasis added), and "the State does not have even a colorable claim to the artifacts" pursuant to its contracts with respondents, ante, at 694 (emphasis added). Given such legal conclusions, I fail to see any need to reverse the determination by the courts below of the State's ownership, as the plurality prescribes, ante, at 700. I do understand that the plurality does not remand this action for a determination of the State's ownership, and rather simply reverses the judgment below on this point. But the fact remains that the courts below have already determined the merits of the State's claim: Even if they were incorrect to make that determination at the time that they did, why should that fact invalidate that determination? Why should the State now get a second bite at the apple? 73 In sum, I would affirm the judgment of the Court of Appeals in its entirety. 74 Justice WHITE, with whom Justice POWELL, Justice REHNQUIST, and Justice O'CONNOR join, concurring in the judgment in part and dissenting in part. 75 The essence of this litigation is a dispute between the State of Florida and one of its citizens over ownership of treasure. The Eleventh Amendment precludes federal courts from entertaining such suits unless the State agrees to waive its Eleventh Amendment immunity. Because it is the State itself which purports to own the controverted treasure, and because the very nature of this suit, as defined in the complaint and recognized by both the District Court and Court of Appeals, is to determine the State's title to such property, this is not a case subject to the doctrine of Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908). In short, this is a suit against the State of Florida, without its permission. Moreover, were the suit to be characterized as one against only state agents, I would find that contract with the State provided a colorable basis upon which the agents could hold the property. 76 The Court of Appeals, like the District Court, thought that the jurisdictional issue raised by the State merged with a determination on the merits of the validity of the State's claim to the property. The appellate court believed that it had "jurisdiction to decide jurisdiction" and could therefore determine who owned the artifacts in order to ascertain whether the suit was, in fact, an action against the State. By holding that "[t]he court did not have power . . . to adjudicate the State's interest in the property without the State's consent," ante, at 682, the Court properly rejects this novel conception of the Eleventh Amendment.* The appellate court's approach to the jurisdictional issue is not consistent with our prior cases; it incorrectly assumes that a federal court may adjudicate a State's right to ownership of specific property within the possession of state officials without the State's consent. The approach is unsatisfactory because, as Judge Rubin noted in dissent, it "is equivalent to asserting that suits against a state are permitted by the eleventh amendment if the result is that the state loses." 621 F.2d 1340, 1351 (CA5 1980). Although disagreeing with the Court of Appeals' Eleventh Amendment holding, the plurality nevertheless proceeds to conclude that the "State does not have even a colorable claim to the artifacts pursuant to [its] contracts" with respondents, ante, at 694, and that the state officials "have [no] colorable claim to possession of the artifacts." Ante, at 694. This for all practical purposes adjudicates the State's title, thus repeating the Eleventh Amendment error of the Court of Appeals. 77 Justice STEVENS' plurality opinion rests precariously on two transparent fictions. First, it indulges in the fantasy that the enforcement of process by arrest of the res is somehow divorced from the action to determine the State's claim to the res—a position contradicted by our own most apposite precedents, the two In re New York cases, 256 U.S. 490, 41 S.Ct. 588, 65 L.Ed. 1057 (1921), and 256 U.S. 503, 41 S.Ct. 592, 65 L.Ed. 1063 (1921). That dubious proposition is parlayed by a second fiction—that Florida's Eleventh Amendment freedom from suit is meaningfully safeguarded by not formally rejecting the State's claim to the artifacts although federal agents may seize the contested property and federal courts may adjudicate its title. Neither of these novel propositions follows from Ex parte Young, supra. The rule of Ex parte Young is premised on the axiom that state officials cannot evade responsibility when their conduct "comes into conflict with the superior authority of [the] Constitution." Id., at 159, 28 S.Ct., at 454. Today, the plurality dilutes the probative force behind that cornerstone decision by extrapolating it to allow federal courts to decide a property dispute between a State and one of its citizens, without the State's consent. For these reasons, as explained below, I dissent in part. 78 * The Suit Is Against the State 79 The case is directly traceable to Treasure Salvors' filing of a motion in District Court for an order commanding the United States Marshal to arrest and take custody of the contested artifacts and to bring them within the jurisdiction of the court. Record 318. The roots of the case, however, rest in the earlier in rem action brought by Treasure Salvors to establish its title to the wreck and its bounty. The District Court held that possession and title rested with Treasure Salvors. Treasure Salvors, Inc. v. Abandoned Sailing Vessel, 408 F.Supp. 907, 911 (SD Fla.1976). The Court of Appeals affirmed Treasure Salvors' ownership of all objects within the District Court's jurisdiction and to those objects outside its territory with respect to the United States. Treasure Salvors, Inc. v. Unidentified Wrecked and Abandoned Sailing Vessel, 569 F.2d 330 (CA5 1978) (Treasure Salvors I ). 80 Treasure Salvors' subsequent request for an arrest warrant was predicated on this decision.1 The warrant was to issue because it had already been decided that Treasure Salvors had "sole title and right to possession of the Defendant vessel." App. 13. Notwithstanding the Court of Appeals' limitation of its opinion to artifacts within the District Court's jurisdiction and to rights in the treasure asserted by the United States, Treasure Salvors sought enforcement of the judgment against the State of Florida. It did so on grounds that this Court's decision in United States v. Florida, 420 U.S. 521, 95 S.Ct. 1162, 43 L.Ed.2d 375 (1975), removed Florida's right to the artifacts, and that Florida was privy to and bound by Treasure Salvors I. 81 "Inasmuch as the State of Florida [and its officers] were privy to this litigation, it is clear that [the district court] confirmed to the Plaintiffs' . . . title to and right to immediate and sole possession of the vessel . . . together with all her . . . cargo, wherever the same may be found." App. 18 (emphasis deleted). 82 In short, Treasure Salvors requested seizure of the artifacts in order to enforce an earlier judgment against the State. This is reason enough to conclude that the suit, and the accompanying warrant for arrest of the articles, were actions invoking federal judicial power against the State and not merely its agents. 83 But even if this were not so, subsequent events reveal that the case is one against the State. After the State filed a motion to quash the warrant, Treasure Salvors filed a supplemental complaint requesting that the contract be held void; it also requested that the District Court rule "[t]hat the State has no right, title or interest" in any portions of the Atocha in its possession. Record 371. The District Court then entered an order to show cause addressed directly to the State of Florida. App. 63. The State then argued that the Eleventh Amendment barred the suit. After rejecting all of the State's arguments, the District Court ordered that Treasure Salvors "have full right and title to articles arrested and that they are entitled to possession." Id., at 85. The Court of Appeals affirmed this judgment. 84 I find the inescapable conclusion to be that this suit, as filed, litigated, and decided, was an action to determine the title of the State of Florida to the artifacts.2 A suit of this type is at the heart of the Eleventh Amendment immunity. 85 The line of cases culminating in Ex parte Young, 209 U.S. 123, 28 S.Ct. 441, 52 L.Ed. 714 (1908), are not to the contrary. In both United States v. Lee, 106 U.S. 196, 1 S.Ct. 240, 27 L.Ed. 171 (1882), and Tindal v. Wesley, 167 U.S. 204, 17 S.Ct. 770, 42 L.Ed. 137 (1897), the suits were against individual agents and did not purport to conclude the rights of the Government. As the Court correctly notes, Tindal made plain that a judgment awarding possession to the plaintiff would not subsequently bind the Government. Here the entire point of the in rem proceeding is to apply the judgment in Treasure Salvors I to erase the State's claim to the treasure. This is the only basis for issuance of the arrest warrant; it was the relief expressly requested by the respondents, and the relief subsequently granted by the District Court and the Court of Appeals. 86 My position is supported by the precedents closest to the instant case: the In re New York cases, 256 U.S. 490, 41 S.Ct. 588, 65 L.Ed. 1057 (1921, and 256 U.S. 503, 41 S.Ct. 592, 65 L.Ed. 1063 (1921). The first In re New York decision arose from an in rem libel against the private owners of tugboats that had been at fault in collisions while chartered and operated by the State. The owners sought to bring in the Superintendent of Public Works who had entered into the charters on the State's behalf. The issue before this Court was whether the State could, without its consent, be impleaded in admiralty process in an action against private parties. The Court held that the "proceedings against which prohibition is here asked," i.e., the attempt to implead the State, "have no element of a proceeding in rem and are in the nature of an action in personam" against a state officer. The purpose of this distinction was not to suggest that in rem actions could be brought against the State, or even that the original libel was not a true in rem cause, but rather to highlight that impleading of a state official, no less than a direct action against the official, constituted a suit against a state officer in his "official capacity" and might require satisfaction out of the property of New York. 256 U.S., at 501, 41 S.Ct., at 591. 87 The second In re New York decision, a sovereign immunity case, made clear that a State's immunity extended to admiralty actions in rem. 88 "The principle so uniformly held to exempt the property of municipal corporations employed for public and governmental purposes from seizure by admiralty process in rem, applies with even greater force to exempt public property of a State used and employed for public and governmental purposes." 256 U.S., at 511, 41 S.Ct., at 593. 89 The plurality's reading of In re New York (II) is that an action "otherwise barred as an in personam action against the State—cannot be maintained through seizure of property owned by the State." Ante, at 699.3 Nothing in the language of Justice Pitney's opinion supports this interpretation. Moreover, the libel brought before the Court in that case was a true in rem action; an action in admiralty to recover damages caused by a ship is a classic in rem action, although after the owners of the vessel are identified the libel often will be amended to include an in personam claim as well. G. Gilmore & C. Black, Law of Admiralty 498 (2d ed. 1975) (Gilmore & Black). Therefore, In re New York (II) is as "true" an in rem action as the instant case. 90 The grounds of similarity between the cases are clear: in both cases in rem libels were filed and process by arrest was requested; in both suits the State by its Attorney General responded and indicated to the District Court that the property to be arrested was in the possession and ownership of the State, and therefore immune from seizure and attachment. In both cases, the District Court overruled the suggestion and awarded process in rem, authorizing the arrest of the res. When the seizure of the Queen City finally reached this forum, the Court stated that the property was exempt from seizure by admiralty process in rem.4 The plurality's distinction aside, the cases can be distinguished on but a single relevant point: the fact that ownership of the res is contested here. That, of course, is the grounds on which the Court of Appeals decided the case—a resolution which the plurality apparently rejects. 91 In re New York (I) indicates that the Eleventh Amendment will bar a suit that has the effect of proceeding against a state officer and involving the State's property. In re New York (II) squarely stands for the proposition that sovereign immunity bars process against a res in the hands of state officers. This is true even though an in rem action strictly proceeds against the vessel, and the owner of the vessel or artifacts is not an indispensable party. Significantly, In re New York (II) did not distinguish between the service of process to arrest the res and the thrust of the libel itself to determine the rights in the vessel. I follow that course in this case, and refuse to sever the attempt to arrest the artifacts from the attempt to decide their ownership. 92 The In re New York cases are particularly forceful because they reflect the special concern in admiralty that maritime property of the sovereign is not to be seized. This principle dates back to the English5 and has not been significantly altered in this country.6 The In re New York cases are but the most apposite examples of the line of cases concerning in rem actions brought against vessels in which an official of the State, the Federal Government, or a foreign government has asserted ownership of the res. The Court's consistent interpretation of the respective but related immunity doctrines pertaining to such vessels has been, upon proper presentation that the sovereign entity claims ownership of a res in its possession, to dismiss the suit or modify the judgment accordingly.7 93 Finally, the allowance of an in rem suit against arguably state-owned maritime property rests on the "personification" theory of the res—that the action runs against the Atocha and not the State of Florida. This distinction between in rem and in personam actions has been decisively rejected. As the fiction of the personality of the ship declined, Gilmore & Black 615, 804-805, in rem actions were given in personam effect, and in personam judgments barred subsequent in rem actions. Id., at 802, 613-614. See, e.g., Burns Bros. v. Central R. Co. of New Jersey, 202 F.2d 910 (CA2 1953) (L. Hand, J.). In short, under long-established admiralty law, arrest of sovereign maritime property is not tolerated, and an in rem suit directed at government property is an action against the State. II 94 Holding of the Treasure by State Officials Was Not Ultra Vires 95 Alternatively, if the arrest of the artifacts was not, without more, a suit against the State, the action was nevertheless against state agents acting within their authority and holding property for the State under a colorable claim of right. It is settled that the Eleventh Amendment bars actions which are in effect against the State, even though the State is not the nominal party. Louisiana v. Jumel, 107 U.S. 711, 719-723, 727-728, 2 S.Ct. 128, 134-137, 141-142, 27 L.Ed. 448 (1883). 96 Leaving aside other possible bases by which the state officials had authority to refuse to surrender possession of the artifacts, I address the salvage contracts entered into between the State and Treasure Salvors. Under the contracts, which were renewed annually, Treasure Salvors was to conduct underwater salvage on Florida lands. By the terms of the contract, Treasure Salvors received 75% of the artifacts recovered. The State was to retain 25% of the representative artifacts. This arrangement was renewed on three occasions, the last contract being entered into on December 3, 1974. It was during that contract's duration that we decided United States v. Florida, 420 U.S. 531, 95 S.Ct. 1162, 43 L.Ed.2d 375 (1975), which established Florida's boundaries along lines which placed the Atocha in international waters. 97 If it were not for this decision, it would be beyond cavil that Florida owned one-fourth of the artifacts pursuant to its ownership of the submerged land on which the Atocha rested as well as the contracts. It is also beyond reasonable dispute that the Eleventh Amendment bars a federal court from deciding the rights and obligations of a State in a contract unless the State consents. Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 69 S.Ct. 1457, 93 L.Ed. 1628 (1949). The plurality does not take issue with this proposition.8 98 The plurality treats this as a different case for two reasons. The first is that the State has never, in so many words, argued that the contracts conferred upon the State a right of ownership in the artifacts. Ante, at 693. While this may be true in the sense that Florida believed that it owned the artifacts even aside from the contracts, it is not true that Florida has not asserted that the contracts create an independent right to the treasure. Florida has repeatedly and expressly made precisely such a claim.9 99 The plurality's second argument is that the "state does not have even a colorable claim to the artifacts pursuant to these contracts." Ante, at 694. I disagree with this conclusion. The wording of the contract is reasonably interpretable as providing for a division of the recovered treasure. The intention of the parties upon the making of the contract, of course, governs the interpretation of the instrument. If United States v. Florida, supra, had placed the Atocha within Florida waters, it could not reasonably be argued that the contract did not constitute a valid basis for the State's claim to 25% of the artifacts. Both Treasure Salvors and the State entered into the contracts on the assumption that the Atocha rested in Florida waters. As it happened, the Florida decision upset that mutual assumption. This does not, however, inexorably mean that the contracts are so invalid as to render possession of the artifacts ultra vires.10 Admiralty law may provide that such a mistake is not grounds for rescission of fully performed contracts in these circumstances.11 The plurality's contention that the language of the contracts does not purport to transfer artifacts from Treasure Salvors to the State utterly ignores the concept of mistake. The notion of mistake would be read out of contract law if courts expected a contract, written under mistaken assumptions, to read as if the mistake had not occurred. 100 Whether the contracts are ultimately valid is beside the point. The existence of a colorable contractual claim to the artifacts, the presence of statutory authority for the State to enter into the contracts, and the ability to raise a mistake-of-law defense not rejectable on its face, is all that need be shown to indicate that possession of the artifacts by the state officials was not ultra vires. Although it would be too much to suggest that our Eleventh Amendment is crystal clear in all respects, this is, at least, the teaching of our most recent cases. 101 Larson v. Domestic & Foreign Commerce Corp., supra, is most directly apposite. There a private corporation brought suit in Federal District Court against the Administrator of the War Assets Administration, an agency of the United States Government, in his official capacity. The claim was that the Administration had sold certain surplus coal to the plaintiff, but had refused to deliver it and had made a new contract to sell it to others. A declaration was sought that the first contract was valid, the second contract invalid, and appropriate injunctive relief was requested. The Court held that the suit was against the United States and the District Court was therefore without jurisdiction to entertain it. The Court's decision rested on the Administrator's statutory authority to enter a binding contract to sell coal, and the absence of a claim that the failure to deliver the coal constituted a taking of private property. The Court refused to pass upon the validity of the contract itself, i.e., whether the initial contract with the plaintiff was breached.12 102 Larson established that where the officer's actions are limited by statute, actions beyond those limitations are to be considered individual and not sovereign actions. "The officer is not doing the business which the sovereign has empowered him to do. . . . His actions are ultra vires his authority and therefore may be made the object of specific relief." 337 U.S., at 689, 69 S.Ct., at 1461. Similarly, unconstitutional actions by state officers could not be considered the work of the sovereign and were not protected by the shield of sovereign immunity. The Larson Court rejected, however, a third proposed category of official actions amenable to suit.13 It was urged upon the Court that if an "officer . . . wrongly takes or holds specific property to which the plaintiff has title," then his action is illegal and the officer may be sued. The Court found the theory erroneous: 103 "The mere allegation that the officer, acting officially, wrongfully holds property to which the plaintiff has title does not meet that requirement. True, it establishes a wrong to the plaintiff. But it does not establish that the officer, in committing that wrong, is not exercising the powers delegated to him by the sovereign." Id., at 693, 69 S.Ct., at 1463. 104 This is a Larson case. Florida entered into the contract pursuant to an indisputably valid state statute, Fla.Stat. § 267.061(1)(b) (1974), providing title to treasure trove abandoned on state-owned submerged lands. The Court relies heavily, as it must, on the subsequent determination that the wreck of the Atocha was in international waters. This, of course, was not settled law at the time the contracts were entered into and executed. Before concluding that the state officials' exercise of rights under the contracts was ultra vires, it is necessary to reach the merits of the contract, and dispose of the mistake-of-law contention. Similarly, the scattershot reasoning of the District Court in refusing to honor the contract—characterization of the mistake as one of fact, treatment of the contract as void for coercion and lack of consideration—constitutes an adjudication of the merits of the contracts. At the time the contracts were entered into and executed they were not ultra vires or otherwise so plainly invalid as not to offer a colorable basis for possession of the artifacts. 105 It is significant that the analysis pursued by the plurality in this respect is little different from that of the Fifth Circuit in deciding the merits in order to ascertain jurisdiction over the matter. As indicated earlier, the plurality performs the task under a different rubric, but the result is equally objectionable. A colorable basis for the exercise of authority by state officials may not ultimately be a valid one, but it does serve to invoke the Eleventh Amendment. That is the lesson of Larson and we should adhere to it. III 106 The plurality begins by stating that "[s]tripped of its procedural complexities and factual glamor, this case presents a narrow legal question." Ante, at 683. Be that as it may, the answer supplied by the plurality is anything but narrow. If the plurality means all that it says today, the consequences will be unfortunate. Given that all property of the State must be held by its officers, and assuming a jurisdictional basis, there is no item within state possession whose ownership cannot be made the subject of federal litigation by the expedient of arrest or attachment. The State must then defend on the merits: it must persuade a federal court that its officers were justified in holding the controverted property. We see today that this inquiry will be tantamount to deciding the question of title itself. Moreover, the State's immunity from suit is stripped away on land as well as sea: the plurality notes that the question presented would not be any different if the State merely resisted an attachment of property. Ibid.. 107 The plurality hardly conceals its view of Florida's claim to the artifacts or the equities involved in this litigation. Yet the Eleventh Amendment teaches that a federal court has no right to offer its opinion on a local dispute between a State and its citizens unless the State consents. In sum, the disposition of this case can only be explained by "procedural complexities and factual glamor." If so, the decision has earned a fitting sobriquet: aberration. 1 The two respondents in this action, Treasure Salvors, Inc., and Armada Research Corp., were organized by the same parties. Throughout these proceedings they have been treated as a single entity referred to as "Treasure Salvors." 2 The story of the Atocha and its discovery is recounted in Lyon, The Trouble with Treasure, 149 National Geographic 787 (1976). 3 The statute since has been amended in a manner not relevant to this case. 4 The District Court found that the contract was entered into as a result of the "coercive acts of the Division of Archives in threatening arrest and confiscation." Treasure Salvors, Inc. v. Unidentified Wrecked and Abandoned Sailing Vessel, 459 F.Supp. 507, 522 (SD Fla.1978). The State admits that if Treasure Salvors had salvaged without a contract arrests would have been made. Tr. of Oral Arg. 9. 5 In its exceptions to the Special Master's Report, the State contended that the Master should have recognized that the boundaries of the State extended to the boundaries defined in the State's 1868 Constitution, rather than to the limits specified in the Submerged Lands Act of 1953. See 420 U.S., at 532, 95 S.Ct., at 1163. This Court considered that exception and held that the Master had properly rejected the State's argument. Id., at 533, 95 S.Ct., at 1163. 6 This area is on the Continental Shelf of the United States, in international waters. Treasure Salvors, Inc. v. Abandoned Sailing Vessel, 408 F.Supp. 907, 909 (SD Fla.1976). 7 The United States asserted a right of ownership under several federal statutes and the common-law doctrine of "sovereign prerogative." The State of Florida did not intervene at this time. It had notice of the litigation, however, and both assisted the United States in the lawsuit and entered into preliminary negotiations with the United States Department of the Interior regarding disposition of the Atocha § treasure in the event the Federal Government prevailed. See Florida Dept. of State v. Treasure Salvors, Inc., 621 F.2d 1340, 1343-1344 (CA5 1980). 8 The court explained: "General principles of maritime and international law dictate that an abandonment constitutes a repudiation of ownership, and that a party taking possession under salvage operations may be considered a finder under the doctrine of 'animus revertendi,' i.e., the owner has no intention of returning. Ownership in the vessel would then vest in the finder by operation of law." 408 F.Supp., at 909 (citation omitted). 9 The court stated: "[T]he district court properly adjudicated title to all those objects within its territorial jurisdiction and to those objects without its territory as between plaintiffs and the United States. In affirming the district court, we do not approve that portion of its order which may be construed as a holding that plaintiffs have exclusive title to, and the right to immediate and sole possession of, the vessel and cargo as to other claimants, if any there be, who are not parties or privies to this litigation." 569 F.2d, at 335-336 (footnote omitted). 10 The warrant provided: "WARRANT FOR ARREST IN REM FC "THE PRESIDENT OF THE UNITED STATES OF AMERICA FC "TO: THE MARSHAL OF THE UNITED STATES FOR THE SOUTHERN DISTRICT OF FLORIDA "GREETING: "WHEREAS, on the 18th day of July, 1975, Treasure Salvors, Inc., a corporation and Armada Research Corporation, a corporation, filed a Complaint under Rule 9(h) against the Unidentified Wrecked and Abandoned Sailing Vessel, her tackle, armament, apparel and cargo located with 2500 yards of a [sic] at coordinates 24x 31.5' North Latitude and 82x 20' West Longitude, said sailing vessel believed to be the NUESTRA SENORA DE ATOCHA for the reasons in said Complaint, and "WHEREAS, in November of 1975 Notice of said claim was published in a newspaper of general circulation within the District, and "WHEREAS, the State of Florida nor any of its agencies, agents, or employees, did appear in this cause to defend or prosecute any claim that they might have to any portions of said vessel that were in their possession, custody, care or control. "NOW, THEREFORE, you are hereby commanded to take into your possession the portions of said vessel which have been in the possession or are in the possession of L. Ross Morrell and/or James McBeth, or under their custody, care or control and to bring said portions of said vessel within the jurisdiction of this Honorable Court and transfer possession of same to the substitute custodian appointed in this action." App. 40-42. 11 The State also asserted: "A contract was entered into between Armada Research Corporation and the State of Florida on December 3, 1974 and was for a good and valid consideration. The contract alone determined the rights and obligations of the contracting parties and was in no way affected by [the decision of this Court in] United States v. Florida. This contract was fully executed and performed prior to the United States v. Florida [sic]." Id., at 44. In response to the State's assertion that the contracts determined the rights of the contracting parties, Treasure Salvors filed a supplemental complaint in federal court. Record 369. The complaint sought a declaratory judgment that the contracts between Treasure Salvors and the State were void. 12 The court also held that, in light of the State's claim that it had a contractual right to 25% of the res, "the State of Florida has waived the general requirement that the res be within the territorial jurisdiction of the court and, further, has consented to the court's jurisdiction over its interest in any portions of the vessel." App. 59. 13 The Court of Appeals then dissolved the emergency stay. Id., at 65. The court ordered: "The United States Marshal may execute the warrant of arrest and upon doing so shall forthwith deliver custody of all of the items in question to a custodian who will take possession of them in situ and shall place them under lock or seal at their present location and hold them secure." Id., at 68. The appellate court denied a motion for reconsideration that had contended that the District Court lacked jurisdiction. "The question of the jurisdiction of the District Court for the Southern District of Florida is for that court to determine in the first instance on the basis of such record as may be developed in that court." Id., at 69. To expedite the litigation, Treasure Salvors agreed to permit the State to serve as substitute custodian. The warrant was executed and, with the State serving as custodian, the artifacts came into the control of the United States Marshal. 14 The court asserted several grounds in support of this decision. Essentially, the court held: "There is no Eleventh Amendment bar to the mere arrest of articles of salvage unless the state is the owner. If the state is not the owner, the court may proceed." 459 F.Supp., at 527. The court concluded that ownership is thus a "jurisdictional" fact and, citing United States v. Mine Workers, 330 U.S. 258, 67 S.Ct. 677, 91 L.Ed. 884, noted that "[i]t is axiomatic that the federal courts have jurisdiction to determine jurisdiction." 459 F.Supp., at 527. The court held that no Eleventh Amendment bar existed because "[t]his Court finds as fact that the Division of Archives is not and never was the rightful owner of the articles of salvage from the ship Atocha that were seized by the ancillary warrant of arrest and which have been improperly removed and held by the Division of Archives; that the Division of Archives is not the owner of any right or interest in such property based upon the purported and invalid contract with Treasure Salvors; and that the Division of Archives was wrongfully withholding a portion of the res of the Atocha over which this Court was properly exercising in rem jurisdiction." Ibid. On the basis of its memorandum, the court "ORDERED and ADJUDGED and DECREED that Treasure Salvors, Inc. and Armada Research Corp. have full right and title to articles arrested and that they are entitled to possession and that the United States Marshal, who has possession and control of such articles, shall deliver them to Treasure Salvors, Inc. and Armada Research Corp." App. 85. Pursuant to this order, Treasure Salvors eventually received under certain restrictions—the artifacts that the State held as custodian for the court. Record 554. 15 The court noted that this result was particularly compelling in admiralty in rem actions. The court reasoned that, since federal courts have exclusive jurisdiction over such actions, if the mere assertion of ownership by a State were sufficient to invoke the Amendment, petitioners such as Treasure Salvors would be stranded without a forum in which to litigate their claim. 621 F.2d, at 1346, n. 19. 16 The court neither affirmed nor reversed the District Court's holding that Florida was in privity with the United States and therefore bound by the earlier decision of the Court of Appeals. Id., at 1344, n. 17. 17 The Eleventh Amendment provides: "The Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens of another State, or by Citizens or Subjects of any Foreign State." Although the Amendment does not literally apply to actions brought against a State by its own citizens, the Amendment long has been held to govern such actions. Hans v. Louisiana, 134 U.S. 1, 10 S.Ct. 504, 33 L.Ed. 842. See Employees v. Missouri Public Health Dept., 411 U.S. 279, 280, 93 S.Ct. 1614, 1615, 36 L.Ed.2d 251; Edelman v. Jordan, 415 U.S. 651, 662, 94 S.Ct. 1347, 1355, 39 L.Ed.2d 662. Nor does the Amendment literally apply to proceedings in admiralty. Again, however, the Court has found it to govern certain admiralty actions. See In re New York, 256 U.S. 490, 500, 41 S.Ct. 588, 590, 65 L.Ed. 1057. 18 The fact that the State appeared and offered defenses on the merits does not foreclose consideration of the Eleventh Amendment issue; "the Eleventh Amendment defense sufficiently partakes of the nature of a jurisdictional bar" that it may be raised at any point of the proceedings. Edelman v. Jordan, supra, at 678, 94 S.Ct., at 1363; see Ford Motor Co. v. Department of Treasury, 323 U.S. 459, 467, 65 S.Ct. 347, 352, 89 L.Ed. 389 ("The Eleventh Amendment declares a policy and sets forth an explicit limitation on federal judicial power of such compelling force that this Court will consider the issue arising under this Amendment in this case even though urged for the first time in this Court"). 19 But see Fitzpatrick v. Bitzer, 427 U.S. 445, 456, 96 S.Ct. 2666, 2671, 49 L.Ed.2d 614 ("Congress may, in determining what is 'appropriate legislation' for the purpose of enforcing the provisions of the Fourteenth Amendment, provide for private suits against States or state officials which are constitutionally impermissible in other contexts"); see also Hutto v. Finney, 437 U.S. 678, 98 S.Ct. 2565, 57 L.Ed.2d 522; Maher v. Gagne, 448 U.S. 122, 100 S.Ct. 2570, 65 L.Ed.2d 653. 20 Thus, in Alabama v. Pugh, our holding that the State of Alabama and the Alabama Board of Corrections should have been dismissed as parties did not affect the substance of the relief granted against a number of Alabama officials responsible for the administration of its prison system. 21 In Lee, the plaintiff brought an action in ejectment in federal court to recover the Virginia estate of General Robert E. Lee. The estate had been acquired by the United States for nonpayment of taxes, although the taxes in fact had been tendered by a third party. Once in possession, the Government had established a federal military installation and a national cemetery on the property. The plaintiff brought suit against the governmental custodians of the estate, who pleaded the sovereign immunity of the United States as a defense. This Court upheld a trial court judgment in favor of the plaintiff, on the ground that the defendants' possession of the estate was unlawful. The Court held that a suit against the federal officers under such circumstances was not a suit against the sovereign. Although Lee involved the sovereign immunity of the United States, the Court in Tindal stated that "it cannot be doubted that the question whether a particular suit is one against the State, within the meaning of the Constitution, must depend upon the same principles that determine whether a particular suit is one against the United States." 167 U.S., at 213, 17 S.Ct., at 773. 22 The Court continued: "Any other view leads to this result: That if a State, by its officers, acting under a void statute, should seize for public use the property of a citizen, without making or securing just compensation for him, and thus violate the constitutional provision declaring that no State shall deprive any person of property without due process of law, Chicago, Burlington &c. Railroad v. Chicago, 166 U.S. 226, 236, 241 [17 S.Ct. 581, 584, 586, 41 L.Ed. 979], the citizen is remediless so long as the State, by its agents, chooses to hold his property; for, according to the contention of the defendants, if such agents are sued as individuals, wrongfully in possession, they can bring about the dismissal of the suit by simply informing the court of the official character in which they hold the property thus illegally appropriated." Id., at 222, 17 S.Ct., at 777. 23 The plaintiff had contracted to purchase surplus coal from the War Assets Administration; the Administrator of that agency had withheld delivery and entered a new contract to sell the coal on the ground that the plaintiff had failed to perform a condition precedent to delivery. The plaintiff contended that title to the coal had passed at the time the contract was made, so that the Administrator was wrongfully withholding property that belonged to him. 24 The Court stated: "The mere allegation that the officer, acting officially, wrongfully holds property to which the plaintiff has title does not meet [the requirement that the action to be restrained or directed is not action of the sovereign]. True, it establishes a wrong to the plaintiff. But it does not establish that the officer, in committing that wrong, is not exercising the powers delegated to him by the sovereign. If he is exercising such powers, the action is the sovereign's and a suit to enjoin it may not be brought unless the sovereign has consented." 337 U.S., at 693, 69 S.Ct., at 1463. The Court explicitly rejected the argument that "the commission of a tort cannot be authorized by the sovereign." Ibid.; see also id., at 695, 69 S.Ct., at 1464. 25 The Court found that the Administrator "was empowered by the sovereign to administer a general sales program encompassing the negotiation of contracts, the shipment of goods and the receipt of payment." Id., at 692, 69 S.Ct., at 1462. "A normal concomitant of such powers, as a matter of general agency law, is the power to refuse delivery when, in the agent's view, delivery is not called for under a contract and the power to sell goods which the agent believes are still his principal's to sell." Ibid. The Court also noted that the "very basis of the respondent's action is that the Administrator was an officer of the Government, validly appointed to administer its sales program and therefore authorized to enter, through his subordinates, into a binding contract concerning the sale of the Government's coal." Id., at 703, 69 S.Ct., at 1468. 26 The Court held that there could be no claim that the Administrator's actions constituted an unconstitutional taking of property without compensation because the plaintiff had a remedy, in a suit for breach of contract, in the Court of Claims. Id., at 703, n. 27, 69 S.Ct., at 1468, n. 27. 27 This principle is not absolute. As noted, n. 19, supra, Congress may authorize a suit against a State—pursuant to § 5 of the Fourteenth Amendment—that would entail the payment of public funds from the state treasury. Fitzpatrick v. Bitzer, 427 U.S. 445, 96 S.Ct. 2666, 49 L.Ed.2d 614; Hutto v. Finney, 437 U.S. 678, 98 S.Ct. 2565, 57 L.Ed.2d 522. Moreover, a prospective decree that has an "ancillary effect" on the state treasury "is a permissible and often an inevitable consequence of the principle announced in Ex parte Young." Edelman v. Jordan 415 U.S., at 668, 94 S.Ct., at 1358; see also Milliken v. Bradley, 433 U.S. 267, 288, 97 S.Ct. 2749, 2761, 53 L.Ed.2d 745. Finally, "[w]hile it is clear that the doctrine of Ex parte Young is of no aid to a plaintiff seeking damages from the public treasury . . . damages against individual defendants are a permissible remedy in some circumstances notwithstanding the fact that they hold public office." Scheuer v. Rhodes, 416 U.S. 232, 238, 94 S.Ct. 1683, 1687, 40 L.Ed.2d 90. 28 The motion identified L. Ross Morrell as the Director of the Division of Archives and James McBeth as the Bureau Chief of the Historical Museum of the Division of Archives. App. 15. 29 As noted, the State immediately filed a motion to quash the warrant. Id., at 43. Although that effort failed, the State asserted an Eleventh Amendment defense in its attempt to defeat a transfer of the property—and thus ultimate execution of the warrant—to Treasure Salvors. Record 422. 30 As noted, n. 11, supra, Treasure Salvors filed a supplemental complaint seeking a declaratory judgment that its contracts with the State were void. This action might be characterized as an action against the State itself. The District Court emphasized, however, that "the warrant was not issued in response to Treasure Salvors' Supplemental Complaint for Declaratory Judgment and Other Relief which was filed April 17, 1978." 459 F.Supp., at 526 (emphasis in original). The order to show cause entered by the District Court was addressed directly to the State of Florida. See App. 63. That order was issued, however, only after the State itself had filed a motion to quash the warrant. Id., at 43 ("COMES NOW, the State of Florida, by and through the undersigned counsel, and moves this Court to set aside and quash the warrant for arrest in rem issued against the State of Florida at the request of Plaintiffs herein . . ."). The order to show cause did not alter the fact that the process resisted by the State on Eleventh Amendment grounds was directed only at state officials. 31 In this Court the State has asserted that the issue on the merits involves a determination of the validity of the contracts. See post, at 712, n. 9. But the State has not identified any language in the contracts that provides even a colorable basis for a claim that the State has an ownership interest in the artifacts. 32 This case is thus unlike Larson v. Domestic & Foreign Commerce Corp., 337 U.S. 682, 69 S.Ct. 1457, 93 L.Ed.1628, in which the plaintiff asserted a right to the property pursuant to the very contract that it contended the Government official had breached without authority. Treasure Salvors claims ownership of the res on the ground that the property was abandoned by the former owner, and discovered by Treasure Salvors, on the Continental Shelf of the United States in international waters. See n. 8, supra. 33 In this Court the State has advanced the argument that its boundaries for purposes of rightful ownership of sunken ships extend further than its boundaries for purposes of ownership of mineral resources. This argument was not raised in the petition for certiorari, is foreclosed by our prior determination of the State's boundaries, see n. 5, supra, and is refuted by the State's own conduct in this case. The State has never attempted to claim ownership of the property that Treasure Salvors has continued to recover since the expiration of the contracts. Given the State's vigorous defense of the relatively few artifacts at issue in this case, it is difficult to imagine that the State idly would permit Treasure Salvors to pirate other treasure that rightfully belonged to the State. 34 The fact that the contracts were executed on the basis of a mistaken understanding concerning the ownership of the Atocha cannot, of course, provide Florida with a colorable claim of ownership. For if the mistake had not occurred, it would have been apparent from the outset that Treasure Salvors had no reason to enter into a contract with Florida or any other stranger to the transaction. The State of Florida has never contended that it would benefit from a reformation of the contracts; Treasure Salvors' position does not depend on any change in the terms of the contracts. The Eleventh Amendment analysis in this case does not require any consideration of the doctrine of mistake. 35 Although the State in this case relies only on the disputed contracts—and not on any statutory provision—we note that Fla.Stat. § 267.061(2)(a) (1981) provides generally that it is the responsibility of the Division of Archives to "[l]ocate, acquire, protect, preserve, and promote the location, acquisition, and preservation of historic sites and properties, buildings, artifacts, treasure trove, and objects of antiquity which have scientific or historical value or are of interest to the public, including, but not limited to, monuments, memorials, fossil deposits, Indian habitations, ceremonial sites, abandoned settlements, caves, sunken or abandoned ships, or any part thereof." Surely this section does not authorize state officials, however, to seize and hold historical artifacts at will wherever they are found. * For this reason, I cannot agree with footnote 17 of the plurality's opinion. To the extent, however, that the plurality concludes that the judgment of the Court of Appeals should be affirmed because the State of Florida does not have even a colorable claim to the artifacts, I agree with its opinion. * I therefore concur in the judgment of the Court only insofar as it reverses the Court of Appeals' determination of the State's ownership of the artifacts. On this point, all Members of the Court, except Justice BRENNAN, are in agreement. 1 "[T]he plaintiffs . . . pursuant to the Final Judgment rendered by this Court February 19, 1976 and the Appellate Opinion rendered by the United States Court of Appeals for the Fifth Circuit No. 76-2151, March 13, 1978, move this Court for an Order commanding the United States Marshal to arrest and take custody of those portions of the Plaintiff's vessel now being held by L. Ross Morrell or James McBeth or being held under their custody, care or control." App. 11. 2 The fact that the District Court did not issue its arrest warrant in response to Treasure Salvors' amended complaint is of little significance. It is the complaint which defines the nature of an action, and once accepted, an amended complaint replaces the original. Moreover, the adjudication of title either reflects that the ownership claim followed from the original complaint or constituted action upon the amended complaint. 3 The plurality confuses the matter further by treating the cases as bearing on the question of whether a burden is imposed on the state treasury. The In re New York cases pertain instead to the initial issue of whether the action is against the State. 4 In re New York (II) was decided on straight sovereign immunity grounds: "[T]he record—aside from whether a suit in admiralty brought by private parties through process in rem against property owned by a State is not in effect a suit against the State, barred by the general principle applied in Ex parte New York, No. 1, No. 25, Original—presents the question whether the proceeding can be based upon the seizure of property owned by a State and used and employed solely for its governmental uses and purposes." The Court went on to decide the vessel was immune from admiralty process, based upon "the law of nations" and "general grounds of comity and policy." 256 U.S., at 510, 41 S.Ct., at 593. In re New York (II) § resolution on sovereign immunity grounds has several implications. First, as with other sovereign immunity decisions, it is direct support for determining what constitutes a suit against the State. Ante, at 686, n. 21. Cf. Tindal v. Wesley, 167 U.S. 204, 213, 17 S.Ct. 770, 773, 42 L.Ed. 137 (1897). Second, it undercuts the plurality's analysis that the case merely stops roundabout circumvention of In re New York (I) through "first attaching property that belonged to the State and then proceeding in rem." Ante, at 699. As the above quoted passage indicates, the In re New York (II) Court did not need to go so far in order to find the suit barred. 5 Under English law, no warrant for arrest will issue against any vessel in the actual service of a recognized foreign government. Significantly, this is so even if the suit itself is not barred. See, e.g., The Messicano, 32 T.L.R. 519 (1916). Where plaintiff sues in rem for possession "the writ will be dismissed, if a foreign recognized government claims the right to possession and is in the actual possession of the vessel, regardless of whether possession was rightfully or wrongfully obtained." Riesenfeld, Sovereign Immunity of Foreign Vessels in Anglo-American Law: The Evolution of a Legal Doctrine, 25 Minn.L.Rev. 1, 25 (1940). In The Parlement Belge, 5 P.D. 197, 220 (1980), the "leading authority" in England, it was held that "[i]f the remedy sought by an action in rem against public property is, as we think it is, an indirect mode of exercising the authority of the Court against the owner of the property, then the attempt to exercise such an authority is an attempt inconsistent with the independence and equality of the state which is represented by such owner." Moreover, after a ship was declared by the foreign sovereign "to be in his possession as sovereign and to be a public vessel of the state," it was "very difficult to say that any Court can inquire by contentious testimony whether that declaration is or is not correct." Id., at 219. 6 For early cases, see United States v. Peters, 3 Dall. 121, 1 L.Ed. 535 (1795); The Schooner Exchange v. McFaddon, 7 Cranch 116, 3 L.Ed. 287 (1812); L'Invincible, 1 Wheat. 238, 4 L.Ed. 80 (1816); The Santissima Trinidad, 7 Wheat. 283, 5 L.Ed. 454 (1822). In The Siren, 7 Wall. 152, 19 L.Ed. 129 (1869), the Court allowed a claim against the proceeds of the vessel when sold, but stressed that no claim could be enforced while the Government owned the vessel. In The Western Maid, 257 U.S. 419, 42 S.Ct. 159, 66 L.Ed. 299 (1922), the Court, per Justice Holmes, went further and refused to allow a claim against a Government-owned vessel as enforceable either during government ownership or thereafter. Shortly thereafter, sovereign immunity was expanded to embrace ships engaged solely in commerce. Berizzi Bros. Co. v. S.S. Pesaro, 271 U.S. 562, 46 S.Ct. 611, 70 L.Ed. 1088 (1926). 7 See Gilmore & Black 606-613. Only when a vessel is not in the sovereign's possession, is there controversy over the proper means by which the foreign government may assert its ownership. See Compania Espanola de Navegacion Maritima v. The Navemar, 303 U.S. 68, 58 S.Ct. 432, 82 L.Ed. 667 (1938). 8 "In Larson . . . this Court held that the actions of a federal official in withholding the delivery of goods pursuant to his interpretation of a disputed provision of a contract constituted at most a tortious deprivation of property. . . . Actions of the Government official pursuant to legitimate contractual authority were neither ultra vires nor unconstitutional." Ante, at 693. 9 "At issue in the present case is both a contract and property right of the State of Florida to the artifacts previously in its possession . . ." Brief for Petitioner 32; "The issue on the merits was whether the State had property rights to artifacts in its Archives—that is, whether the contract to which the state was a party was valid." Id., at 60. "The State of Florida has not claimed a lien on the artifacts; it has claimed ownership through fully executed contracts." Reply Brief for Petitioner 16-17. "The contract alone determined the rights and obligations of the contracting parties and was in no way affected by United States v. Florida." State's Motion to Quash Warrant for Arrest in Rem, App. 44. 10 The plurality also suggests that the contracts "were predicated entirely on a state statute that on its face is inapplicable in this case." Ante, at 696. This no more than restates the plurality's characterization of the contracts. But it does highlight that the contracts' validity is called into question only by a mistaken assumption of law—the statute's "inapplica[bility]" after United States v. Florida, 420 U.S. 531, 95 S.Ct. 1162, 43 L.Ed.2d 375 (1975). 11 The inherent uncertainty in contracts for salvage has led admiralty courts to find few reasons that would justify reformation of a contract. See The Elfrida, 172 U.S. 186, 196, 19 S.Ct. 146, 149, 43 L.Ed. 413 (1898) ("We do not think that a salvage contract should be sustained as an exception to the general rule, but rather that it should, prima facie, be enforced, and that it belongs to the defendant to establish the exception"). Gilmore & Black 582 ("Whether the gamble turns well or badly for the salvor, the 'no cure no pay' contract is everywhere recognized as enforceable, absent such invalidating causes as fraud and duress"). 12 The plurality's attempt to distinguish Larson is puzzling. It notes that while the plaintiff in Larson asserted a right to the property pursuant to the very contract it contended the Government official had breached, here Treasure Salvors claims ownership on grounds entirely independent of the contracts. This is a distinction without meaning: it is the State's claim to the property which is significant; the basis for Treasure Salvors' claim is quite beside the point. The relevant comparison is that the federal official in Larson was arguably without authority to enter a contract to sell coal that he had already sold just as the State was arguably without authority to enter a contract respecting salvage on lands outside its waters. 13 The plurality acknowledges that Larson clarified the understanding of earlier cases such as Tindal v. Wesley, 167 U.S. 204, 17 S.Ct. 770, 42 L.Ed. 137 (1897), and United States v. Lee, 106 U.S. 196, 1 S.Ct. 240, 27 L.Ed. 171 (1882). Dicta in both Tindal and Lee are cited by the Court to suggest that a federal court may adjudicate the validity of a title in order to determine whether the case is a suit against the State. It is precisely this aspect of the cases that Larson "clarified." A court may go only so far as to ascertain whether an official has a colorable basis for his action—to go farther is to, in effect, try the case on the jurisdictional issue and "is equivalent to asserting that suits against the state are permitted by the eleventh amendment if the result is that the state loses." 621 F.2d 1340, 1351 (CA5 1980) (Rubin, J., dissenting). The inapplicability of United States v. Lee was made clear in Malone v. Bowdoin, 369 U.S. 643, 82 S.Ct. 980, 8 L.Ed.2d 168 (1962), a case involving an attempt to eject a Forest Service Officer from land occupied by him solely in his official capacity under a claim of title in the United States. The plaintiffs argued they were the rightful owners of the land. The Court held that the suit was an impermissible action against the United States, and stated: "While not expressly overruling United States v. Lee, supra, the Court in Larson limited that decision in such a way as to make it inapplicable to the case before us. Pointing out that at the time of the Lee decision there was no remedy by which the plaintiff could have recovered compensation for the taking of his land, the Court interpreted Lee as simply 'a specific application of the constitutional exception to the doctrine of sovereign immunity.' 337 U.S., at 696 [69 S.Ct., at 1464]. So construed, the Lee case has continuing validity only 'where there is a claim that the holding constitutes an unconstitutional taking of property without just compensation.' Id., at 697 [69 S.Ct., at 1465]." Id., at 647-648, 82 S.Ct., at 983-984. An in rem admiralty action, like an ejectment suit, is an action to determine title to property, and, here, like in Bowdoin, there is no claim of an unconstitutional taking without adequate compensation. Indeed, Treasure Salvors may be able to bring an in personam action in state court to determine ownership of the treasure.

89

458 U.S. 654 102 S.Ct. 3294 73 L.Ed.2d 1042 Baxter RICE, Director, Department of Alcoholic Beverage Control of California, Petitioner,v.NORMAN WILLIAMS COMPANY et al. BOHEMIAN DISTRIBUTING COMPANY, Petitioner, v. NORMAN WILLIAMS COMPANY et al. WINE & SPIRITS WHOLESALERS OF CALIFORNIA, Petitioner, v. NORMAN WILLIAMS COMPANY et al. Nos. 80-1012, 80-1030 and 80-1052. Argued April 21, 1982. Decided July 1, 1982. Syllabus A provision of California's alcoholic beverage laws states that a "licensed importer shall not purchase or accept delivery of any brand of distilled spirits unless he is designated as an authorized importer of such brand by the brand owner or his authorized agent" (designation statute). The statute apparently was enacted in response to the perceived extraterritorial effects of Oklahoma's "open wholesaling" statutes, whereby a licensed California importer who was unable to obtain distilled spirits through the distiller's established distribution system could obtain them from Oklahoma wholesalers. Prior to the designation statute's effective date, respondents sought an extraordinary writ from the California Court of Appeal to enjoin the enforcement of the statute. The court entered judgment for respondents, holding that the conduct contemplated by the statute was per se illegal under § 1 of the Sherman Act because it gave distillers the unfettered power to restrain competition by merely deciding who may or may not compete in handling the distillers' brands, and that thus the statute on its face was invalid pursuant to the Supremacy Clause of the Federal Constitution. Held : 1. California's designation statute is not invalid on its face as being pre-empted by the Sherman Act. Pp. 659-662. (a). A state statute, when considered in the abstract, may be condemned under the antitrust laws only if it mandates or authorizes conduct that necessarily constitutes a violation of those laws in all cases, or if it places irresistible pressure on a private party to violate the antitrust laws in order to comply with the statute. Such condemnation will follow under § 1 of the Sherman Act when the conduct contemplated by the statute is in all cases a per se violation. If the activity addressed by the statute does not fall into that category, and therefore must be analyzed under the rule of reason, the statute cannot be condemned in the abstract. Pp. 659-661. (b). A distiller's invocation of California's statute would not be subject in all cases to a per se rule of illegality under the Sherman Act. A manufacturer's use of vertical nonprice restraints is not per se illegal. Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568. California's designation statute merely enforces the distiller's decision to restrain intrabrand competition, preventing an unauthorized wholesaler from obtaining the distiller's products from outside the distiller's established distribution chain. The effect of the statute is simply to counteract the perceived extraterritorial effects of Oklahoma's alcoholic beverage laws, thus restoring the distiller's ability to determine which wholesalers may import its products into California. While the manner in which a distiller utilizes the designation statute and the arrangements a distiller makes with its wholesalers will be subject to Sherman Act analysis under the rule of reason, there is no basis for condemning the statute itself by force of the Sherman Act. Pp. 661-662. 2. The California statute is not pre-empted by § 5(a) of the Federal Alcohol Administration Act, which prohibits a distiller or wholesaler from establishing exclusive retail outlets. California's statute in no way requires exclusive retail outlets, and by its terms does not even require exclusive wholesale arrangements. Pp.663-664 3. The designation statute does not deny respondents due process of law. Respondents do not possess any constitutionally protected liberty or property interest in obtaining the distiller's permission to deal in its products, and thus the Due Process Clause is not offended by the wholesaler's inability to challenge the distiller's decisionmaking. P. 664. 4. Nor does the designation statute violate the Equal Protection Clause because it discriminates between designated and nondesignated wholesalers. The statute is rationally related to its legitimate purposes, enabling the distiller to place restraints on intrabrand competition in order to foster interbrand competition. P. 665. 108 Cal.App.3d 348, 166 Cal.Rptr. 563, reversed and remanded. John R. McDonough, Los Angeles, Cal., for petitioners in 80-1030 and 80-1052. George J. Roth, Sacramento, Cal., for petitioner in 80-1012. George G. Weickhardt, San Francisco, Cal., for all respondents. Justice REHNQUIST delivered the opinion of the Court. 1 Respondents in these cases obtained from the California Court of Appeal an extraordinary writ prohibiting the California Department of Alcoholic Beverage Control from enforcing an amendment to the State's liquor statutes. That court held that because the conduct contemplated by the amendment was per se illegal under the Sherman Act, the statute on its face was invalid pursuant to the Supremacy Clause of the United States Constitution. 108 Cal.App.3d 348, 166 Cal.Rptr. 563 (1980). We conclude that the California Court of Appeal was mistaken in its application of antitrust and preemption principles, and we reverse its judgment. 2 * Alcoholic beverages may be brought into California from outside the State for delivery or use within the State only if the beverages are consigned to a licensed importer. Cal.Bus. & Prof.Code Ann. § 23661 (West Supp. 1982). In 1979, the California Legislature amended the State's alcoholic beverage control laws to provide that a "licensed importer shall not purchase or accept delivery of any brand of distilled spirits unless he is designated as an authorized importer of such brand by the brand owner or his authorized agent." § 23672. This challenged statute, which was to become effective on January 1, 1980, is understandably referred to as a "designation statute."1 3 California apparently enacted its designation statute in response to the effects of Oklahoma's alcoholic beverage laws. At the time, Oklahoma's statutes were understood to require any distiller or brand owner selling its products to Oklahoma wholesalers to sell to all wholesalers on a nondiscriminatory basis.2 Because of the perceived extraterritorial effect of Oklahoma's "open-wholesaling" statutes, a licensed California importer who was unable to obtain distilled spirits through the distiller's established distribution system could obtain them from Oklahoma wholesalers. As a result, a distiller who desired to sell its products to Oklahoma wholesalers was unable to rely on contractual undertakings to determine which California wholesalers would handle its products. California's designation statute, therefore, sought to close off the "Oklahoma connection" to California importers not authorized by the distiller to deal in its products.3 4 Prior to the effective date of the designation statute, respondents, liquor importers who were benefiting from the "Oklahoma connection," sought an extraordinary writ from the California Court of Appeal enjoining the enforcement of the designation statute. The Court of Appeal agreed with respondents that the designation statute on its face conflicted with § 1 of the Sherman Act, 26 Stat. 209, as amended, 15 U.S.C. § 1.4 According to that court, the designation statute would result, in all cases, in a per se violation of the Sherman Act, because it "gives brand owners the unfettered power to restrain competition . . . by merely deciding who may and who may not compete." 108 Cal.App.3d, at 356, 166 Cal.Rptr., at 569. The Court of Appeal distinguished Continental T. V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568 (1977), in which we held that vertical nonprice restraints are to be judged under the "rule of reason" rather than under a per se rule of illegality, on the ground that respondents did not attack the distiller's decision to refuse to do business with them, but "the state provided authority of the distillers to prohibit them from trading with others." 108 Cal.App.3d, at 357, 166 Cal.Rptr., at 570. 5 The Supreme Court of California denied review. We granted certiorari, 454 U.S. 1080, 102 S.Ct. 632, 70 L.Ed.2d 613 (1981), and now reverse. II A. 6 In determining whether the Sherman Act pre-empts a state statute, we apply principles similar to those which we employ in considering whether any state statute is pre-empted by a federal statute pursuant to the Supremacy Clause. As in the typical pre-emption case, the inquiry is whether there exists an irreconcilable conflict between the federal and state regulatory schemes. The existence of a hypothetical or potential conflict is insufficient to warrant the pre-emption of the state statute. A state regulatory scheme is not pre-empted by the federal antitrust laws simply because in a hypothetical situation a private party's compliance with the statute might cause him to violate the antitrust laws. A state statute is not preempted by the federal antitrust laws simply because the state scheme might have an anticompetitive effect. See,e.g., New Motor Vehicle Bd. of Cal. v. Orrin W. Fox Co., 439 U.S. 96, 110-111, 99 S.Ct. 403, 412-413, 58 L.Ed.2d 361 (1978); Exxon Corp. v. Governor of Maryland, 437 U.S. 117, 129-134, 98 S.Ct. 2207, 2215-2218, 57 L.Ed.2d 91 (1978); Joseph E. Seagram & Sons, Inc. v. Hostetter, 384 U.S. 35, 45-46, 86 S.Ct. 1254, 1260-1261, 16 L.Ed.2d 336 (1966). 7 A party may successfully enjoin the enforcement of a state statute only if the statute on its face irreconcilably conflicts with federal antitrust policy. In California Retail Liquor Dealers Assn. v. Midcal Aluminum, Inc., 445 U.S. 97, 100 S.Ct. 937, 63 L.Ed.2d 233 (1980), we examined a statute that required members of the California wine industry to file fair trade contracts or price schedules with the State, and provided that if a wine producer had not set prices through a fair trade contract, wholesalers must post a resale price schedule for that producer's brands. We held that the statute facially conflicted with the Sherman Act because it mandated resale price maintenance, an activity that has long been regarded as a per se violation5 of the Sherman Act. Id., at 102-103, 100 S.Ct., at 941-942; see Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 407-409, 31 S.Ct. 376, 384-385, 55 L.Ed. 502 (1911). 8 By contrast, in Joseph E. Seagram & Sons, Inc. v. Hostetter, supra, we rejected a facial attack upon § 9 of New York's Alcoholic Beverage Control Law,6 which required retailers and wholesalers to file monthly price schedules with the State Liquor Authority accompanied by an affirmation that the prices charged were no higher than the lowest price at which sales were made anywhere in the United States during the preceding month. Id., 384 U.S. at 39-40, 86 S.Ct., at 1257-1258. The Court found no clear repugnancy between § 9 and the federal antitrust laws: 9 "The bare compilation, without more, of price information on sales to wholesalers and retailers to support the affirmations filed with the State Liquor Authority would not of itself violate the Sherman Act. Section 9 imposes no irresistible economic pressure on the appellants to violate the Sherman Act in order to comply with the requirements of § 9. On the contrary, § 9 appears firmly anchored to the assumption that the Sherman Act will deter any attempts by the appellants to preserve their New York price level by conspiring to raise the prices at which liquor is sold elsewhere in the country. . . . 10 "Although it is possible to envision circumstances under which price discriminations proscribed by the Robinson-Patman Act might be compelled by § 9, the existence of such potential conflicts is entirely too speculative in the present posture of this case. . . ." Id., at 45-46, 86 S.Ct., at 1261 (citations omitted). 11 Our decisions in this area instruct us, therefore, that a state statute, when considered in the abstract, may be condemned under the antitrust laws only if it mandates or authorizes conduct that necessarily constitutes a violation of the antitrust laws in all cases, or if it places irresistible pressure on a private party to violate the antitrust laws in order to comply with the statute. Such condemnation will follow under § 1 of the Sherman Act when the conduct contemplated by the statute is in all cases a per se violation. If the activity addressed by the statute does not fall into that category, and therefore must be analyzed under the rule of reason, the statute cannot be condemned in the abstract. Analysis under the rule of reason requires an examination of the circumstances underlying a particular economic practice, and therefore does not lend itself to a conclusion that a statute is facially inconsistent with federal antitrust laws. 12 It remains for us to determine whether a distiller's invocation of the designation statute would be subject in all cases to a per se rule of illegality under the Sherman Act. B 13 We held in GTE Sylvania that a manufacturer's use of vertical nonprice restraints is not per se illegal. Because restraints on intrabrand competition may promote interbrand competition, we concluded that non-price vertical restraints should be scrutinized under the rule of reason. 433 U.S., at 57-59, 97 S.Ct., at 2561-2562. After our decision in GTE Sylvania, it cannot be said that every attempt by a manufacturer to restrain competition in its own products is illegal under the Sherman Act. 14 California's designation statute merely enforces the distiller's decision to restrain intrabrand competition. It permits the distiller to designate which wholesalers may import the distiller's products into the State. It prevents an unauthorized wholesaler from obtaining the distiller's products from outside the distiller's established distribution chain. The designation statute does not require the distiller to impose vertical restraints of any kind; that is a matter for it to determine. The number of importers which may be designated by the distiller is not limited; the designated importer is not required to sell the imported brand to retailers within a specified area or from a specified location within the State. 15 It is irrelevant for our purposes that the distiller's ability to restrict intrabrand competition in California has the imprimatur of a state statute. New Motor Vehicle Bd. of Cal. v. Orrin W. Fox Co., 439 U.S., at 110-111, 99 S.Ct., at 412-413.7 The effect of the statute is simply to counteract the perceived extraterritorial effects of Oklahoma's alcoholic beverage laws, which, as once understood, operated to deprive the distiller of control over its distribution system nationwide. Thus, California's designation statute merely restored what Oklahoma had taken away: the distiller's ability to determine which wholesalers may import its products into California. 16 In these respects, therefore, we find these cases to be much like Joseph E. Seagram & Sons, Inc. v. Hostetter, 384 U.S. 35, 86 S.Ct. 1254, 16 L.Ed.2d 336 (1966). As in Hostetter, upholding the validity of the designation statute will not insulate a distiller's invocation of the statute from scrutiny under the Sherman Act. The manner in which a distiller utilizes the designation statute and the arrangements a distiller makes with its wholesalers will be subject to Sherman Act analysis under the rule of reason.8 There is no basis, however, for condemning the statute itself by force of the Sherman Act.9 III 17 Respondents seek to support the judgment of the Court of Appeal on three federal grounds not considered by the court below. None of these contentions have merit. A. 18 Respondents contend that the California designation statute is pre-empted by § 5(a) of the Federal Alcohol Administration Act, 49 Stat. 981, as amended, 27 U.S.C. § 205(a).10 Section 5(a) prohibits a distiller or wholesaler from establishing exclusive retail outlets. See S.Rep.No.1215, 74th Cong., 1st Sess., 6-7 (1935); H.R.Rep.No.1542, 74th Cong., 1st Sess., 10-11 (1935). In other words, § 5(a) prohibits a distiller or wholesaler from requiring a retailer to buy only the distiller's or wholesaler's products to the exclusion of the products of other distillers or wholesalers. The statute does not prohibit a distiller from requiring its wholesalers to purchase the distiller's products from the distiller itself rather than from a third party.11 California's statute in no way requires exclusive retail outlets. By its terms, the designation statute does not even require exclusive wholesale arrangements. One might be able to hypothesize an arrangement enforced by the designation statute that might be prohibited by § 5(a), but this is insufficient to invalidate a state statute pursuant to the Supremacy Clause. "To hold otherwise would be to ignore the teaching of this Court's decisions which enjoin seeking out conflicts between state and federal regulation where none clearly exists." Huron Portland Cement Co. v. Detroit, 362 U.S. 440, 446, 80 S.Ct. 813, 817, 4 L.Ed.2d 852 (1960). B 19 Respondents contend that the designation statute denies them due process of law. According to respondents, California has established a "second tier of private licensing over the state's licensing process," and therefore procedural due process protections apply with regard to the distiller's designation decisions. Brief for Respondents 36. 20 We find this contention without merit. The designation statute merely enforces the distiller's decision to deny permission to a California wholesaler to deal in the distiller's products. We do not think that respondents possess any constitutionally protected liberty or property interest in obtaining the distiller's permission. Thus, the Due Process Clause is not offended by the wholesaler's inability to challenge the distiller's decisionmaking. What respondents are really challenging is the California Legislature's decision to give such a power to the distiller without establishing any criteria to govern the exercise of that power. The Due Process Clause does not authorize this Court to assess the wisdom of the California Legislature's decision. SeeFerguson v. Skrupa, 372 U.S. 726, 729-732, 83 S.Ct. 1028, 1030-1032, 10 L.Ed.2d 93 (1963). C 21 Finally, respondents contend that the designation statute violates the Equal Protection Clause because it discriminates between designated and nondesignated wholesalers. There can be little doubt but that the designation statute is rationally related to the statute's legitimate purposes. Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 461-470, 101 S.Ct. 715, 722-727, 66 L.Ed.2d 659 (1981). The designation statute enables the distiller to place restraints on intrabrand competition in order to foster interbrand competition. It is not our province to determine whether or not California consumers would be better off had the California Legislature decided not to close off the "Oklahoma connection." See Vance v. Bradley, 440 U.S. 93, 109, 99 S.Ct. 939, 411-412, 59 L.Ed.2d 171 (1979). 22 The judgment of the Court of Appeal is reversed, and these cases are remanded to that court for proceedings not inconsistent with this opinion. 23 It is so ordered. 24 Justice STEVENS, with whom Justice WHITE joins, concurring in the judgment. 25 Under the California designation statute, each distiller is empowered to decide whether to regulate its product distribution within California by designating those importers that may sell its product. The statute contemplates a private market decision but provides a nonmarket mechanism for enforcing the decision. Hybrid restraints of this character require analysis that is different from a public regulatory scheme on the one hand, see, e.g., Exxon Corp. v. Governor of Maryland, 437 U.S. 117, 98 S.Ct. 2207, 57 L.Ed.2d 91; Joseph E. Seagram & Sons, Inc. v. Hostetter, 384 U.S. 35, 86 S.Ct. 1254, 16 L.Ed.2d 336,1 and a purely private restraint on the other, see, e.g., Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568; Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S.Ct. 376, 55 L.Ed. 502. We have twice held that hybrid price-fixing restraints are prohibited by the Sherman Act. Schwegmann Bros. v. Calvert Distillers Corp., 341 U.S. 384, 71 S.Ct. 745, 95 L.Ed. 1035; California Retail Liquor Dealers Ass'n v. Midcal Aluminum, Inc., 445 U.S. 97, 100 S.Ct. 937, 63 L.Ed.2d 233. In both cases the private decision to fix prices was unsupervised by the State but made effective by state law. 26 The facts of Schwegmann, involving the Louisiana marketing practices of two out-of-state distributors of gin and whiskey, are particularly instructive. The distributors sought to control the retail prices of their products by obtaining from individual retailers a written agreement that they would comply with the minimum retail price schedules established by the distributors. These resale price maintenance agreements, which otherwise violated the Sherman Act, were rendered lawful by the Miller-Tydings Act and a Louisiana fair trade law. But a New Orleans retailer refused to sign such an agreement and sold the distributors' products at cutrate prices. The distributors responded by seeking to enjoin the retailer from selling the products at less than the minimum prices fixed by their schedules. The basis for their complaint was a Louisiana statute that condemned as unfair competition a sale at less than the price stipulated in a fair trade contract, even though the particular retailer was not a party to that contract. This Court held that the Sherman Act, as amended by the Miller-Tydings Act, precluded enforcement of the nonsigner provision. 27 Even though the private agreements to fix resale prices were not unlawful, Schwegmann held that the distributor could not place the same restraint on the market by using the state statute as a "club." 341 U.S., at 395, 71 S.Ct. at 751 (emphasis in original). The Court's holding teaches that a state statute that facilitates the manufacturer's decision to impose a vertical restriction is not lawful simply because the Sherman Act permits the manufacturer, if it has sufficient power in the private market, to impose that same restriction without the aid of the statute. In other words, a statute that gives distributors additional power over the wholesale or retail market to impose an otherwise permissible restraint might not pass muster under the Sherman Act.2 28 The inquiry in these cases therefore cannot simply be whether the Sherman Act would have been violated had the distillers obtained the control over their California distribution systems without the aid of the designation statute. For the distillers' power to impose resale restrictions on California importers has been drastically affected first by the Oklahoma "open wholesaling" and "free export" provisions and second by the California designation statute enacted as a response to the Oklahoma laws. It may be that the amount of distiller control over California importers under the two statutes is not significantly greater than the amount that would exist if neither State intervened in the private market. Contrary to the Court's perception, ante, at 662,3 however, the character of control is different. For the designation statute gives the distillers direct authority over California importers,4 whereas in the private market the distillers must persuade Oklahoma wholesalers not to resell to California importers. It is possible that, absent the state laws, the distillers would have insufficient market power to obtain and enforce such agreements. The designation statute therefore may give the distillers more power over California importers than was taken away by the Oklahoma laws. 29 The validity of the designation statute obviously presents a more difficult question than was presented in Schwegmann and Midcal.5 For in both cases the Court had the benefit of a conclusive presumption that resale price maintenance is anticompetitive. This case, however, not only involves a species of vertical nonprice restriction with respect to which there are no sure rules relating to effect on competition; it also involves a nonmarket enforcement mechanism that, according to Schwegmann, can make the difference between legality and illegality. The statute conceivably could create such an unacceptable and unnecessary risk of anticompetitive effect as to result in its invalidation. The removal of the Oklahoma legal obstacle to the purely private imposition of vertical restrictions in the California liquor market significantly enhances this possibility. 30 I agree with the Court that our price-fixing cases do not require the invalidation of the designation statute. The question on remand should be whether the statute's provision to distillers of an additional club over California importers affords distillers an unreasonable degree of unsupervised power to regulate their distribution practices that they would not otherwise enjoy under a free market. Because that question cannot be determined without a more sophisticated inquiry, I concur in the Court's judgment. 1 Section 23672 is actually an amended version of a statute invalidated by the California Supreme Court in Rice v. Alcoholic Beverage Control Appeals Bd., 21 Cal.3d 431, 146 Cal.Rptr. 585, 579 P.2d 476 (1978), because its minimum price system constituted resale price maintenance in violation of the Sherman Act. See California Retail Liquor Dealers Assn. v. Midcal Aluminum, Inc., 445 U.S. 97, 100-102, 100 S.Ct. 937, 940-941, 63 L.Ed.2d 233 (1980). 2 Okla.Stat., Tit. 37, § 533 (1981). 3 The Oklahoma Supreme Court, however, has recently closed off the "Oklahoma connection" by holding that the open-wholesaling statute does not apply to alcoholic beverages destined for consumption in other States. Central Liquor Co. v. Oklahoma Alcoholic Beverage Control Bd., Okl., 640 P.2d 1351 (1982). What made the "Oklahoma connection" particularly attractive to California wholesalers was that Oklahoma required distillers to sell to Oklahoma wholesalers at the lowest price charged for its products anywhere in the United States. See Okla.Stat., Tit. 37, § 536.1 (1981). The demise of the "Oklahoma connection," however, has no bearing on our disposition of the legal issues in these cases. 4 Although it did not phrase its conclusion in these terms, it is evident that the California Court of Appeal concluded that the designation statute was pre-empted by the Sherman Act. The court properly recognized that it had no jurisdiction to entertain a lawsuit brought pursuant to § 4 of the Clayton Act, 15 U.S.C. § 15. 108 Cal.App.3d 348, 354, n. 2, 166 Cal.Rptr. 563, 568, n. 2 (1980). Rather than seeking a private remedy against private parties, respondents in these cases sought to enjoin the enforcement of a state statute that they contend to be unconstitutional under the Supremacy Clause in its every application. Indeed, because respondents brought this suit prior to the effective date of the statute, respondents did not, and could not, challenge any vertical restraints actually employed by a distiller pursuant to the statute. Instead, respondents challenge the statute on its face without consideration of particular circumstances. 5 Under established antitrust principles, per se rules of illegality are appropriate only when they apply to practices " 'which because of their pernicious effect on competition and lack of any redeeming virtue are conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use.' " Continental T. V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 50, 97 S.Ct. 2549, 2557, 53 L.Ed.2d 568 (1977), quoting Northern Pacific R. Co. v. United States, 356 U.S. 1, 5, 78 S.Ct. 514, 518, 2 L.Ed.2d 545 (1958). It is not surprising, therefore, that a statute which requires practices per se illegal under the Sherman Act may be subject to a facial challenge under the Supremacy Clause. 6 As with the instant case, because the challenged statute had not as yet been put into effect, this Court in Hostetter was presented only with a facial challenge to its constitutionality. 7 This is merely another way of stating that the designation statute might have an anticompetitive effect when applied in concrete factual situations. See New Motor Vehicle Bd. of Cal. v. Orrin W. Fox Co., 439 U.S., at 110-111, 99 S.Ct., at 412-413. We have explained, however, that this is insufficient to declare the statute itself void on its face. 8 It is certainly conceivable, however, that particular conduct pursuant to the statute might be subject to a challenge under one or more of the established per se rules of illegality. 9 Because of our resolution of the pre-emption issue, it is not necessary for us to consider whether the statute may be saved from invalidation under the doctrine of Parker v. Brown, 317 U.S. 341, 63 S.Ct. 307, 87 L.Ed. 315 (1943), or under the Twenty-first Amendment. 10 Section 5(a) provides: "It shall be unlawful for any person engaged in business as a distiller, brewer, rectifier, blender, or other producer, or as an importer or wholesaler, of distilled spirits, wine, or malt beverages, or as a bottler, or warehouseman and bottler, of distilled spirits, directly or indirectly or through an affiliate: "(a) Exclusive outlet "To require, by agreement or otherwise, that any retailer engaged in the sale of distilled spirits, wine, or malt beverages, purchase any such products from such person to the exclusion in whole or in part of distilled spirits, wine, or malt beverages sold or offered for sale by other persons in interstate or foreign commerce, if such requirement is made in the course of interstate or foreign commerce, or if such person engages in such practice to such an extent as substantially to restrain or prevent transactions in interstate or foreign commerce in any such products, or if the direct effect of such requirement is to prevent, deter, hinder, or restrict other persons from selling or offering for sale any such products to such retailer in interstate or foreign commerce." 27 U.S.C. § 205(a). 11 See 27 CFR §§ 8.3, 8.11, 8.23 (1982). 1 The Court states that Seagram & Sons is "much like" these cases. Ante, at 662. Except for the fact that Seagram & Sons also involved a facial challenge against a state statute, the two cases are quite different. The New York statute involved in Seagram & Sons imposed a degree of public regulation of the market; it did not grant liquor distributors a degree of private regulatory power. The restraint on the market was, therefore, not of the hybrid character that distinguishes these cases from most antitrust cases. 2 Whereas Schwegmann best illustrates the different treatment accorded purely private restraints and hybrid restraints, perhaps Midcal best illustrates the different treatment accorded hybrid restraints and public regulation of the market. In Midcal a California statute required distributors of wine to file either fair trade contracts or price schedules with the State. All California retailers were required to sell wine at the price the distributors fixed. Relying upon Schwegmann, the Court invalidated the statute. Even though the State presumably could regulate the wine market by fixing retail prices itself, it could not empower private parties to undertake such regulation. 3 "Thus, California's designation statute merely restored what Oklahoma had taken away: the distiller's ability to determine which wholesalers may import its products into California." 4 Unless an importer is expressly designated, it may not lawfully sell the distiller's products within California. 5 Under the Sherman Act, a manufacturer may choose the wholesalers with whom it will do business. It may also place certain reasonable restrictions on each wholesaler as a condition of doing business. Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 97 S.Ct. 2549, 53 L.Ed.2d 568. Subject to the exception recognized in United States v. Colgate & Co., 250 U.S. 300, 39 S.Ct. 465, 63 L.Ed. 992, it may not, however, dictate the price at which the wholesaler must sell the product to retailers or to ultimate consumers. Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373, 31 S.Ct. 376, 55 L.Ed. 502. From these different rules governing the purely private decisions of manufacturers, it follows that a state statute that facilitates resale price maintenance and a state statute that facilitates other vertical restrictions are also subject to different antitrust analyses.

78

458 U.S. 718 102 S.Ct. 3331 73 L.Ed.2d 1090 MISSISSIPPI UNIVERSITY FOR WOMEN, et al., Petitioners,v.Joe HOGAN. No. 81-406. Argued March 22, 1982. Decided July 1, 1982. Syllabus Held : The policy of petitioner Mississippi University for Women (MUW), a state-supported university which has from its inception limited its enrollment to women, of denying otherwise qualified males (such as respondent) the right to enroll for credit in its School of Nursing violates the Equal Protection Clause of the Fourteenth Amendment. Pp. 723-733. (a) The party seeking to uphold a statute that classifies individuals on the basis of their gender must carry the burden of showing an "exceedingly persuasive justification" for the classification. Kirchberg v. Feenstra, 450 U.S. 455, 461, 101 S.Ct. 1195, 1199, 67 L.Ed.2d 428; Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 273, 99 S.Ct. 2282, 2293, 60 L.Ed.2d 870. The burden is met only by showing at least that the classification serves "important governmental objectives and that the discriminatory means employed" are "substantially related to the achievement of those objectives." Wengler v. Druggists Mutual Insurance Co., 446 U.S. 142, 150, 100 S.Ct. 1540, 1545, 64 L.Ed.2d 107. The test must be applied free of fixed notions concerning the roles and abilities of males and females. Pp. 723-727. (b) The single-sex admissions policy of MUW's School of Nursing cannot be justified on the asserted ground that it compensates for discrimination against women and, therefore, constitutes educational affirmative action. A State can evoke a compensatory purpose to justify an otherwise discriminatory classification only if members of the gender benefited by the classification actually suffer a disadvantage related to the classification. Rather than compensating for discriminatory barriers faced by women, MUW's policy tends to perpetuate the stereotyped view of nursing as an exclusively woman's job. Moreover, the State has not shown that the gender-based classification is substantially and directly related to its proposed compensatory objective. To the contrary, MUW's policy of permitting men to attend classes as auditors fatally undermines its claim that women, at least those in the School of Nursing, are adversely affected by the presence of men. Thus, the State has fallen far short of establishing the "exceedingly persuasive justification" needed to sustain the gender-based classification. Pp. 727-731. (c) Nor can the exclusion of men from MUW's School of Nursing be justified on the basis of the language of § 901(a)(5) of Title IX of the Education Amendments of 1972, which exempts from § 901(a)'s general prohibition of gender discrimination in federally funded education programs the admissions policies of public institutions of undergraduate higher education "that traditionally and continually from [their] establishment [have] had a policy of admitting only students of one sex." It is not clear that, as argued by the State, Congress enacted the statute pursuant to its power granted by § 5 of the Fourteenth Amendment to enforce that Amendment, and thus placed a limitation upon the broad prohibitions of the Equal Protection Clause. Rather, Congress apparently intended, at most, to create an exemption from Title IX's requirements. In any event, Congress' power under § 5 "is limited to adopting measures to enforce the guarantees of the Amendment; § 5 grants Congress no power to restrict, abrogate, or dilute these guarantees." Katzenbach v. Morgan, 384 U.S. 641, 651, n. 10, 86 S.Ct. 1717, 1724, n. 10, 16 L.Ed.2d 828. Pp. 731-733. 646 F.2d 1116 (5th Cir.) and 653 F.2d 222 (5th Cir.) affirmed. Hunter M. Gholson, Columbus, Miss., for petitioners. Wilbur O. Colom, Columbus, Miss., for respondent. Justice O'CONNOR delivered the opinion of the Court. 1 This case presents the narrow issue of whether a state statute that excludes males from enrolling in a state-supported professional nursing school violates the Equal Protection Clause of the Fourteenth Amendment. 2 * The facts are not in dispute. In 1884, the Mississippi Legislature created the Mississippi Industrial Institute and College for the Education of White Girls of the State of Mississippi, now the oldest state-supported all-female college in the United States. 1884 Miss.Gen.Laws, Ch. 30, § 6. The school, known today as Mississippi University for Women (MUW), has from its inception limited its enrollment to women.1 3 In 1971, MUW established a School of Nursing, initially offering a 2-year associate degree. Three years later, the school instituted a 4-year baccalaureate program in nursing and today also offers a graduate program. The School of Nursing has its own faculty and administrative officers and establishes its own criteria for admission.2 4 Respondent, Joe Hogan, is a registered nurse but does not hold a baccalaureate degree in nursing. Since 1974, he has worked as a nursing supervisor in a medical center in Columbus, the city in which MUW is located. In 1979, Hogan applied for admission to the MUW School of Nursing's baccalaureate program.3 Although he was otherwise qualified, he was denied admission to the School of Nursing solely because of his sex. School officials informed him that he could audit the courses in which he was interested, but could not enroll for credit. Tr. 26.4 5 Hogan filed an action in the United States District Court for the Northern District of Mississippi, claiming the single-sex admissions policy of MUW's School of Nursing violated the Equal Protection Clause of the Fourteenth Amendment. Hogan sought injunctive and declaratory relief, as well as compensatory damages. 6 Following a hearing, the District Court denied preliminary injunctive relief. App. to Pet. for Cert. A4. The court concluded that maintenance of MUW as a single-sex school bears a rational relationship to the State's legitimate interest "in providing the greatest practical range of educational opportunities for its female student population." Id., at A3. Furthermore, the court stated, the admissions policy is not arbitrary because providing single-sex schools is consistent with a respected, though by no means universally accepted, educational theory that single-sex education affords unique benefits to students. Ibid. Stating that the case presented no issue of fact, the court informed Hogan that it would enter summary judgment dismissing his claim unless he tendered a factual issue. When Hogan offered no further evidence, the District Court entered summary judgment in favor of the State. Record 73. 7 The Court of Appeals for the Fifth Circuit reversed, holding that, because the admissions policy discriminates on the basis of gender, the District Court improperly used a "rational relationship" test to judge the constitutionality of the policy. 646 F.2d 1116, 1118 (1981). Instead, the Court of Appeals stated, the proper test is whether the State has carried the heavier burden of showing that the gender-based classification is substantially related to an important governmental objective. Id., at 1118, 1119. Recognizing that the State has a significant interest in providing educational opportunities for all its citizens, the court then found that the State had failed to show that providing a unique educational opportunity for females, but not for males, bears a substantial relationship to that interest. Id., at 1119. Holding that the policy excluding Hogan because of his sex denies him equal protection of the laws, the court vacated the summary judgment entered against Hogan as to his claim for monetary damages, and remanded for entry of a declaratory judgment in conformity with its opinion and for further appropriate proceedings. Id., at 1119-1120. 8 On rehearing, the State contended that Congress, in enacting § 901(a)(5) of Title IX of the Education Amendments of 1972, Pub.L. 92-318, 86 Stat. 373, 20 U.S.C. § 1681 et seq., expressly had authorized MUW to continue its single-sex admissions policy by exempting public undergraduate institutions that traditionally have used single-sex admissions policies from the gender discrimination prohibition of Title IX.5 Through that provision, the State argued, Congress limited the reach of the Fourteenth Amendment by exercising its power under § 5 of the Amendment.6 The Court of Appeals rejected the argument, holding that § 5 of the Fourteenth Amendment does not grant Congress power to authorize States to maintain practices otherwise violative of the Amendment. 653 F.2d 222 (1981). 9 We granted certiorari, 454 U.S. 962, 102 S.Ct. 501, 70 L.Ed.2d 377 (1981), and now affirm the judgment of the Court of Appeals.7 II 10 We begin our analysis aided by several firmly established principles. Because the challenged policy expressly discriminates among applicants on the basis of gender, it is subject to scrutiny under the Equal Protection Clause of the Fourteenth Amendment. Reed v. Reed, 404 U.S. 71, 75, 92 S.Ct. 251, 253, 30 L.Ed.2d 225 (1971). That this statutory policy discriminates against males rather than against females does not exempt it from scrutiny or reduce the standard of review.8 Caban v. Mo- hammed, 441 U.S. 380, 394, 99 S.Ct. 1760, 1769, 60 L.Ed.2d 297 (1979); Orr v. Orr, 440 U.S. 268, 279, 99 S.Ct. 1102, 1111, 59 L.Ed.2d 306 (1979). Our decisions also establish that the party seeking to uphold a statute that classifies individuals on the basis of their gender must carry the burden of showing an "exceedingly persuasive justification" for the classification. Kirchberg v. Feenstra, 450 U.S. 455, 461, 101 S.Ct. 1195, 1199, 67 L.Ed.2d 428 (1981); Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 273, 99 S.Ct. 2282, 2293, 60 L.Ed.2d 870 (1979). The burden is met only by showing at least that the classification serves "important governmental objectives and that the discriminatory means employed" are "substantially related to the achievement of those objectives." Wengler v. Druggists Mutual Ins. Co., 446 U.S. 142, 150, 100 S.Ct. 1540, 1545, 64 L.Ed.2d 107 (1980).9 11 Although the test for determining the validity of a gender-based classification is straightforward, it must be applied free of fixed notions concerning the roles and abilities of males and females. Care must be taken in ascertaining whether the statutory objective itself reflects archaic and stereotypic notions. Thus, if the statutory objective is to exclude or "protect" members of one gender because they are presumed to suffer from an inherent handicap or to be innately inferior, the objective itself is illegitimate. See Frontiero v. Richardson, 411 U.S. 677, 684-685, 93 S.Ct. 1764, 1769-70, 36 L.Ed.2d 583 (1973) (plurality opinion).10 12 If the State's objective is legitimate and important, we next determine whether the requisite direct, substantial relationship between objective and means is present. The purpose of requiring that close relationship is to assure that the validity of a classification is determined through reasoned analysis rather than through the mechanical application of traditional, often inaccurate, assumptions about the proper roles of men and women.11 The need for the requirement is amply revealed by reference to the broad range of statutes already invalidated by this Court, statutes that relied upon the simplistic, outdated assumption that gender could be used as a "proxy for other, more germane bases of classification," Craig v. Boren, 429 U.S. 190, 198, 97 S.Ct. 451, 457, 50 L.Ed.2d 397 (1976), to establish a link between objective and classification.12 13 Applying this framework, we now analyze the arguments advanced by the State to justify its refusal to allow males to enroll for credit in MUW's School of Nursing. III A. 14 The State's primary justification for maintaining the single-sex admissions policy of MUW's School of Nursing is that it compensates for discrimination against women and, therefore, constitutes educational affirmative action. Brief for Petitioners 8.13 As applied to the School of Nursing, we find the State's argument unpersuasive. 15 In limited circumstances, a gender-based classification favoring one sex can be justified if it intentionally and directly assists members of the sex that is disproportionately burdened. See Schlesinger v. Ballard, 419 U.S. 498, 95 S.Ct. 572, 42 L.Ed.2d 610 (1975). However, we consistently have emphasized that "the mere recitation of a benign, compensatory purpose is not an automatic shield which protects against any inquiry into the actual purposes underlying a statutory scheme." Weinberger v. Wiesenfeld, 420 U.S. 636, 648, 95 S.Ct. 1225, 1233, 43 L.Ed.2d 514 (1975). The same searching analysis must be made, regardless of whether the State's objective is to eliminate family controversy, Reed v. Reed, 404 U.S. 71, 92 S.Ct. 251, 30 L.Ed.2d 225 (1971), to achieve administrative efficiency, Frontiero v. Richardson, 411 U.S. 677, 93 S.Ct. 1764, 36 L.Ed.2d 583 (1973), or to balance the burdens borne by males and females. 16 It is readily apparent that a State can evoke a compensatory purpose to justify an otherwise discriminatory classification only if members of the gender benefited by the classification actually suffer a disadvantage related to the classification. We considered such a situation in Califano v. Webster, 430 U.S. 313, 97 S.Ct. 1192, 51 L.Ed.2d 360 (1977), which involved a challenge to a statutory classification that allowed women to eliminate more low-earning years than men for purposes of computing Social Security retirement benefits. Although the effect of the classification was to allow women higher monthly benefits than were available to men with the same earning history, we upheld the statutory scheme, noting that it took into account that women "as such have been unfairly hindered from earning as much as men" and "work[ed] directly to remedy" the resulting economic disparity. Id., at 318, 97 S.Ct., at 1195. 17 A similar pattern of discrimination against women influenced our decision in Schlesinger v. Ballard, supra. There, we considered a federal statute that granted female Naval officers a 13-year tenure of commissioned service before mandatory discharge, but accorded male officers only a 9-year tenure. We recognized that, because women were barred from combat duty, they had had fewer opportunities for promotion than had their male counterparts. By allowing women an additional four years to reach a particular rank before subjecting them to mandatory discharge, the statute directly compensated for other statutory barriers to advancement. 18 In sharp contrast, Mississippi has made no showing that women lacked opportunities to obtain training in the field of nursing or to attain positions of leadership in that field when the MUW School of Nursing opened its door or that women currently are deprived of such opportunities. In fact, in 1970, the year before the School of Nursing's first class enrolled, women earned 94 percent of the nursing baccalaureate degrees conferred in Mississippi and 98.6 percent of the degrees earned nationwide. U.S. Dept. of Health, Education, and Welfare, Earned Degrees Conferred: 1969-1970, Institutional Data 388 (1972). That year was not an aberration; one decade earlier, women had earned all the nursing degrees conferred in Mississippi and 98.9 percent of the degrees conferred nationwide. U.S. Dept. of Health, Education, and Welfare, Earned Degrees Conferred, 1959-1960: Bachelor's and Higher Degrees 135 (1960). As one would expect, the labor force reflects the same predominance of women in nursing. When MUW's School of Nursing began operation, nearly 98 percent of all employed registered nurses were female.14 United States Bureau of Census, 1981 Statistical Abstract of the United States 402 (1981). 19 Rather than compensate for discriminatory barriers faced by women, MUW's policy of excluding males from admission to the School of Nursing tends to perpetuate the stereotyped view of nursing as an exclusively woman's job.15 By assuring that Mississippi allots more openings in its state-supported nursing schools to women than it does to men, MUW's admissions policy lends credibility to the old view that women, not men, should become nurses, and makes the assumption that nursing is a field for women a self-fulfilling prophecy. See Stanton v. Stanton, 421 U.S. 7, 95 S.Ct. 1373, 43 L.Ed.2d 688 (1975). Thus, we conclude that, although the State recited a "benign, compensatory purpose," it failed to establish that the alleged objective is the actual purpose underlying the discriminatory classification.16 20 The policy is invalid also because it fails the second part of the equal protection test, for the State has made no showing that the gender-based classification is substantially and directly related to its proposed compensatory objective. To the contrary, MUW's policy of permitting men to attend classes as auditors fatally undermines its claim that women, at least those in the School of Nursing, are adversely affected by the presence of men. 21 MUW permits men who audit to participate fully in classes. Additionally, both men and women take part in continuing education courses offered by the School of Nursing, in which regular nursing students also can enroll. Deposition of Dr. James Strobel 56-60 and Deposition of Dean Annette K. Barrar 24-26. The uncontroverted record reveals that admitting men to nursing classes does not affect teaching style, Deposition of Nancy L. Herban 4, that the presence of men in the classroom would not affect the performance of the female nursing students, Tr. 61 and Deposition of Dean Annette K. Barrar 7-8, and that men in coeducational nursing schools do not dominate the classroom. Deposition of Nancy Herban 6. In sum, the record in this case is flatly inconsistent with the claim that excluding men from the School of Nursing is necessary to reach any of MUW's educational goals. 22 Thus, considering both the asserted interest and the relationship between the interest and the methods used by the State, we conclude that the State has fallen far short of establishing the "exceedingly persuasive justification" needed to sustain the gender-based classification. Accordingly, we hold that MUW's policy of denying males the right to enroll for credit in its School of Nursing violates the Equal Protection Clause of the Fourteenth Amendment.17 B 23 In an additional attempt to justify its exclusion of men from MUW's School of Nursing, the State contends that MUW is the direct beneficiary "of specific congressional legislation which, on its face, permits the institution to exist as it has in the past." Brief for Petitioners 19. The argument is based upon the language of § 901(a) in Title IX of the Education Amendments of 1972, 20 U.S.C. § 1681(a). Although § 901(a) prohibits gender discrimination in education programs that receive federal financial assistance, subsection 5 exempts the admissions policies of undergraduate institutions "that traditionally and continually from [their] establishment [have] had a policy of admitting only students of one sex" from the general prohibition. See n. 5, supra. Arguing that Congress enacted Title IX in furtherance of its power to enforce the Fourteenth Amendment, a power granted by § 5 of that Amendment, the State would have us conclude that § 901(a)(5) is but "a congressional limitation upon the broad prohibitions of the Equal Protection Clause of the Fourteenth Amendment." Brief for Petitioners 20. 24 The argument requires little comment. Initially, it is far from clear that Congress intended, through § 901(a)(5), to exempt MUW from any constitutional obligation. Rather, Congress apparently intended, at most, to exempt MUW from the requirements of Title IX. 25 Even if Congress envisioned a constitutional exemption, the State's argument would fail. Section 5 of the Fourteenth Amendment gives Congress broad power indeed to enforce the command of the Amendment and "to secure to all persons the enjoyment of perfect equality of civil rights and the equal protection of the laws against State denial or invasion. . . ." Ex parte Virginia, 100 U.S. (10 Otto) 339, 346, 25 L.Ed. 676 (1880). Congress' power under § 5, however, "is limited to adopting measures to enforce the guarantees of the Amendment; § 5 grants Congress no power to restrict, abrogate, or dilute these guarantees." Katzenbach v. Morgan, 384 U.S. 641, 651, n. 10, 86 S.Ct. 1717, 1724, n. 10, 16 L.Ed.2d 828 (1966). Although we give deference to congressional decisions and classifications, neither Congress nor a State can validate a law that denies the rights guaranteed by the Fourteenth Amendment. See, e.g., Califano v. Goldfarb, 430 U.S. 199, 210, 97 S.Ct. 1021, 1028, 51 L.Ed.2d 270 (1977); Williams v. Rhodes, 393 U.S. 23, 29, 89 S.Ct. 5, 9, 21 L.Ed.2d 24 (1968). 26 The fact that the language of § 901(a)(5) applies to MUW provides the State no solace: "[A] statute apparently governing a dispute cannot be applied by judges, consistently with their obligations under the Supremacy Clause, when such an application of the statute would conflict with the Constitution. Marbury v. Madison, 1 Cranch 137 [2 L.Ed. 60] (1803)." Younger v. Harris, 401 U.S. 37, 52, 91 S.Ct. 746, 754, 27 L.Ed.2d 669 (1971). IV 27 Because we conclude that the State's policy of excluding males from MUW's School of Nursing violates the Equal Protection Clause of the Fourteenth Amendment, we affirm the judgment of the Court of Appeals. 28 It is so ordered. 29 Chief Justice BURGER, dissenting. 30 I agree generally with Justice POWELL's dissenting opinion. I write separately, however, to emphasize that the Court's holding today is limited to the context of a professional nursing school. Ante, at 723, n. 7, 3338. Since the Court's opinion relies heavily on its finding that women have traditionally dominated the nursing profession, see ante, at 729-731, it suggests that a State might well be justified in maintaining, for example, the option of an all-women's business school or liberal arts program. 31 Justice BLACKMUN, dissenting. 32 Unless Mississippi University for Women wished to preserve a historical anachronism, one only states the obvious when he observes that the University long ago should have replaced its original statement of purpose and brought its corporate papers into the 20th century. It failed to do so and, perhaps in partial consequence, finds itself in this litigation, with the Court's opinion, ante, at 719-720, and n.1, now taking full advantage of that failure, to MUW's embarrassment and discomfiture. 33 Despite that failure, times have changed in the intervening 98 years. What was once an "Institute and College" is now a genuine university, with a 2-year School of Nursing established 11 years ago and then expanded to a 4-year baccalaureate program in 1974. But respondent Hogan "wants in" at this particular location in his home city of Columbus. It is not enough that his State of Mississippi offers baccalaureate programs in nursing open to males at Jackson and at Hattiesburg. Mississippi thus has not closed the doors of its educational system to males like Hogan. Assuming that he is qualified—and I have no reason whatsoever to doubt his qualifications—those doors are open and his maleness alone does not prevent his gaining the additional education he professes to seek. 34 I have come to suspect that it is easy to go too far with rigid rules in this area of claimed sex discrimination, and to lose—indeed destroy—values that mean much to some people by forbidding the State to offer them a choice while not depriving others of an alternative choice. Justice POWELL in his separate opinion, post, p. 735, advances this theme well. 35 While the Court purports to write narrowly, declaring that it does not decide the same issue with respect to "separate but equal" undergraduate institutions for females and males, ante, at 720, n.1, or with respect to units of MUW other than its School of Nursing, ante, at 723, n.7, there is inevitable spillover from the Court's ruling today. That ruling, it seems to me, places in constitutional jeopardy any state-supported educational institution that confines its student body in any area to members of one sex, even though the State elsewhere provides an equivalent program to the complaining applicant. The Court's reasoning does not stop with the School of Nursing of the Mississippi University for Women. 36 I hope that we do not lose all values that some think are worthwhile (and are not based on differences of race or religion) and relegate ourselves to needless conformity. The ringing words of the Equal Protection Clause of the Fourteenth Amendment what Justice POWELL aptly describes as its "liberating spirit," post, at 741,—do not demand that price. 37 Justice POWELL, with whom Justice REHNQUIST joins, dissenting. 38 The Court's opinion bows deeply to conformity. Left without honor—indeed, held unconstitutional—is an element of diversity that has characterized much of American education and enriched much of American life. The Court in effect holds today that no State now may provide even a single institution of higher learning open only to women students. It gives no heed to the efforts of the State of Mississippi to provide abundant opportunities for young men and young women to attend coeducational institutions, and none to the preferences of the more than 40,000 young women who over the years have evidenced their approval of an all-women's college by choosing Mississippi University for Women (MUW) over seven coeducational universities within the State. The Court decides today that the Equal Protection Clause makes it unlawful for the State to provide women with a traditionally popular and respected choice of educational environment. It does so in a case instituted by one man, who represents no class, and whose primary concern is personal convenience. 39 It is undisputed that women enjoy complete equality of opportunity in Mississippi's public system of higher education. Of the State's 8 universities and 16 junior colleges, all except MUW are coeducational. At least two other Mississippi universities would have provided respondent with the nursing curriculum that he wishes to pursue.1 No other male has joined in his complaint. The only groups with any personal acquaintance with MUW to file amicus briefs are female students and alumnae of MUW. And they have emphatically rejected respondent's arguments, urging that the State of Mississippi be allowed to continue offering the choice from which they have benefited. 40 Nor is respondent significantly disadvantaged by MUW's all-female tradition. His constitutional complaint is based upon a single asserted harm: that he must travel to attend the state-supported nursing schools that concededly are available to him. The Court characterizes this injury as one of "inconvenience." Ante, at 724, n. 8. This description is fair and accurate, though somewhat embarrassed by the fact that there is, of course, no constitutional right to attend a state-supported university in one's home town. Thus the Court, to redress respondent's injury of inconvenience, must rest its invalidation of MUW's single-sex program on a mode of "sexual stereotype" reasoning that has no application whatever to the respondent or to the "wrong" of which he complains. At best this is anomalous. And ultimately the anomaly reveals legal error—that of applying a heightened equal protection standard, developed in cases of genuine sexual stereotyping, to a narrowly utilized state classification that provides an additional choice for women. Moreover, I believe that Mississippi's educational system should be upheld in this case even if this inappropriate method of analysis is applied. 41 * Coeducation, historically, is a novel educational theory. From grade school through high school, college, and graduate and professional training, much of the Nation's population during much of our history has been educated in sexually segregated classrooms. At the college level, for instance, until recently some of the most prestigious colleges and universities—including most of the Ivy League—had long histories of single-sex education. As Harvard, Yale, and Princeton remained all-male colleges well into the second half of this century, the "Seven Sister" institutions established a parallel standard of excellence for women's colleges. Of the Seven Sisters, Mount Holyoke opened as a female seminary in 1837 and was chartered as a college in 1888. Vassar was founded in 1865, Smith and Wellesley in 1875, Radcliffe in 1879, Bryn Mawr in 1885, and Barnard in 1889. Mount Holyoke, Smith, and Wellesley recently have made considered decisions to remain essentially single-sex institutions. See Carnegie Commission on Higher Education 70-75 (1973) Opportunities for Women in Higher Education 70-75 (1973) (Carnegie Report), excerpted in B. Babcock, A. Freedman, E. Norton, & S. Ross, Sex Discrimination and the Law 1013, 1014 (1975) (Babcock). Barnard retains its independence from Columbia, its traditional coordinate institution. Harvard and Radcliffe maintained separate admissions policies as recently as 1975.2 42 The sexual segregation of students has been a reflection of, rather than an imposition upon, the preference of those subject to the policy. It cannot be disputed, for example, that the highly qualified women attending the leading women's colleges could have earned admission to virtually any college of their choice.3 Women attending such colleges have chosen to be there, usually expressing a preference for the special benefits of single-sex institutions. Similar decisions were made by the colleges that elected to remain open to women only.4 43 The arguable benefits of single-sex colleges also continue to be recognized by students of higher education. The Carnegie Commission on Higher Education has reported that it "favor[s] the continuation of colleges for women. They provide an element of diversity . . . and [an environment in which women] generally . . . speak up more in their classes, . . . hold more positions of leadership on campus, . . . and . . . have more role models and mentors among women teachers and administrators." Carnegie Report, quoted in K. Davidson, R. Ginsburg, & H. Kay, Sex-Based Discrimination 814 (1975 ed.). A 10-year empirical study by the Cooperative Institutional Research Program of the American Counsel of Education and the University of California, Los Angeles, also has affirmed the distinctive benefits of single-sex colleges and universities. As summarized in A. Astin, Four Critical Years 232 (1977), the data established that 44 "[b]oth [male and female] single-sex colleges facilitate student involvement in several areas: academic, interaction with faculty, and verbal aggressiveness. . . . Men's and women's colleges also have a positive effect on intellectual self-esteem. Students at single-sex colleges are more satisfied than students at coeducational colleges with virtually all aspects of college life . . . . The only area where students are less satisfied is social life."5 45 Despite the continuing expressions that single-sex institutions may offer singular advantages to their students, there is no doubt that coeducational institutions are far more numerous. But their numerical predominance does not establish—in any sense properly cognizable by a court—that individual preferences for single-sex education are misguided or illegitimate, or that a State may not provide its citizens with a choice.6 II 46 The issue in this case is whether a State transgresses the Constitution when—within the context of a public system that offers a diverse range of campuses, curricula, and educational alternatives—it seeks to accommodate the legitimate personal preferences of those desiring the advantages of an all-women's college. In my view, the Court errs seriously by assuming—without argument or discussion—that the equal protection standard generally applicable to sex discrimination is appropriate here. That standard was designed to free women from "archaic and overbroad generalizations . . . ." Schlesinger v. Ballard, 419 U.S. 498, 508, 95 S.Ct. 572, 577, 42 L.Ed.2d 610 (1975). In no previous case have we applied it to invalidate state efforts to expand women's choices. Nor are there prior sex discrimination decisions by this Court in which a male plaintiff, as in this case, had the choice of an equal benefit. 47 The cases cited by the Court therefore do not control the issue now before us. In most of them women were given no opportunity for the same benefit as men.7 Cases involving male plaintiffs are equally inapplicable. In Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976), a male under 21 was not permitted to buy beer anywhere in the State, and women were afforded no choice as to whether they would accept the "statistically measured but loose-fitting generalities concerning the drinking tendencies of aggregate groups." Id., at 209, 97 S.Ct., at 463. A similar situation prevailed in Orr v. Orr, 440 U.S. 268, 279, 99 S.Ct. 1102, 1111, 59 L.Ed.2d 306 (1979), where men had no opportunity to seek alimony from their divorced wives, and women had no escape from the statute's stereotypical announcement of "the State's preference for an allocation of family responsibilities under which the wife plays a dependent role . . . ."8 48 By applying heightened equal protection analysis to this case,9 the Court frustrates the liberating spirit of the Equal Protection Clause. It prohibits the States from providing women with an opportunity to choose the type of university they prefer. And yet it is these women whom the Court regards as the victims of an illegal, stereotyped perception of the role of women in our society. The Court reasons this way in a case in which no woman has complained, and the only complainant is a man who advances no claims on behalf of anyone else. His claim, it should be recalled, is not that he is being denied a substantive educational opportunity, or even the right to attend an all-male or a coeducational college. See Brief for Respondent 24.10 It is only that the colleges open to him are located at inconvenient distances.11 III 49 The Court views this case as presenting a serious equal protection claim of sex discrimination. I do not, and I would sustain Mississippi's right to continue MUW on a rational-basis analysis. But I need not apply this "lowest tier" of scrutiny. I can accept for present purposes the standard applied by the Court: that there is a gender-based distinction that must serve an important governmental objective by means that are substantially related to its achievement. E.g., Wengler v. Druggists Mutual Ins. Co., 446 U.S. 142, 150, 100 S.Ct. 1540, 1545, 64 L.Ed.2d 107 (1980). The record in this case reflects that MUW has a historic position in the State's educational system dating back to 1884. More than 2,000 women presently evidence their preference for MUW by having enrolled there. The choice is one that discriminates invidiously against no one.12 And the State's purpose in preserving that choice is legitimate and substantial. Generations of our finest minds, both among educators and students, have believed that single-sex, college-level institutions afford distinctive benefits. There are many persons, of course, who have different views. But simply because there are these differences is no reason—certainly none of constitutional dimension—to conclude that no substantial state interest is served when such a choice is made available. 50 In arguing to the contrary, the Court suggests that the MUW is so operated as to "perpetuate the stereotyped view of nursing as an exclusively women's job." Ante, at 729. But as the Court itself acknowledges, ante, at 720, MUW's School of Nursing was not created until 1971—about 90 years after the single-sex campus itself was founded. This hardly supports a link between nursing as a woman's profession and MUW's single-sex admission policy. Indeed, MUW's School of Nursing was not instituted until more than a decade after a separate School of Nursing was established at the coeducational University of Mississippi at Jackson. See University of Mississippi, 1982 Undergraduate Catalog 162. The School of Nursing makes up only one part—a relatively small part13—of MUW's diverse modern university campus and curriculum. The other departments on the MUW campus offer a typical range of degrees14 and a typical range of subjects.15 There is no indication that women suffer fewer opportunities at other Mississippi state campuses because of MUW's admission policy.16 51 In sum, the practice of voluntarily chosen single-sex education is an honored tradition in our country, even if it now rarely exists in state colleges and universities. Mississippi's accommodation of such student choices is legitimate because it is completely consensual and is important because it permits students to decide for themselves the type of college education they think will benefit them most. Finally, Mississippi's policy is substantially related to its long-respected objective.17 IV 52 A distinctive feature of America's tradition has been respect for diversity. This has been characteristic of the peoples from numerous lands who have built our country. It is the essence of our democratic system. At stake in this case as I see it is the preservation of a small aspect of this diversity. But that aspect is by no means insignificant, given our heritage of available choice between single-sex and coeducational institutions of higher learning. The Court answers that there is discrimination—not just that which may be tolerable, as for example between those candidates for admission able to contribute most to an educational institution and those able to contribute less—but discrimination of constitutional dimension. But, having found "discrimination," the Court finds it difficult to identify the victims. It hardly can claim that women are discriminated against. A constitutional case is held to exist solely because one man found it inconvenient to travel to any of the other institutions made available to him by the State of Mississippi. In essence he insists that he has a right to attend a college in his home community. This simply is not a sex discrimination case. The Equal Protection Clause was never intended to be applied to this kind of case.18 1 The charter of MUW, basically unchanged since its founding, now provides: "The purpose and aim of the Mississippi State College for Women is the moral and intellectual advancement of the girls of the state by the maintenance of a first-class institution for their education in the arts and sciences, for their training in normal school methods and kindergarten, for their instruction in bookkeeping, photography, stenography, telegraphy, and typewriting, and in designing, drawing, engraving, and painting, and their industrial application, and for their instruction in fancy, general and practical needlework, and in such other industrial branches as experience, from time to time, shall suggest as necessary or proper to fit them for the practical affairs of life." Miss.Code Ann. § 37-117-3 (1972). Mississippi maintains no other single-sex public university or college. Thus, we are not faced with the question of whether States can provide "separate but equal" undergraduate institutions for males and females. Cf. Vorchheimer v. School District of Philadelphia, 532 F.2d 880 (CA3 1975), aff'd by an equally divided Court, 430 U.S. 703, 97 S.Ct. 1671, 51 L.Ed.2d 750 (1977). 2 Record, Exhibit 1, 1980-1981 Bulletin of Mississippi University for Women 31-34, 212-229. 3 With a baccalaureate degree, Hogan would be able to earn a higher salary and would be eligible to obtain specialized training as an anesthetist. Tr. 18. 4 Dr. James Strobel, President of MUW, verified that men could audit the equivalent of a full classload in either night or daytime classes. Id., at 39-40. 5 Section 901(a) of Title IX, Education Amendments of 1972, Pub.L. 92-318, 86 Stat. 373, 20 U.S.C. § 1681(a), provides in part: "(a) No person in the United States shall, on the basis of sex, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any education program or activity receiving Federal financial assistance, except that: "(1) . . . in regard to admissions to educational institutions, this section shall apply only to institutions of vocational education, professional education, and graduate higher education, and to public institutions of undergraduate higher education; * * * * * "(5) . . . in regard to admissions this section shall not apply to any public institution of undergraduate higher education which is an institution that traditionally and continually from its establishment has had a policy of admitting only students of one sex. . . ." 6 Section 5 of the Fourteenth Amendment provides: "The Congress shall have power to enforce, by appropriate legislation, the provisions of this article." 7 Although some statements in the Court of Appeals' decision refer to all schools within MUW, see 646 F.2d, at 1119, the factual underpinning of Hogan's claim for relief involved only his exclusion from the nursing program, Complaint ¶ 8-10, and the Court of Appeals' holding applies only to Hogan's individual claim for relief. 646 F.2d, at 1119-1120. Additionally, during oral argument, counsel verified that Hogan sought only admission to the School of Nursing. Tr. of Oral Arg. 24. Because Hogan's claim is thus limited, and because we review judgments, not statements in opinions, Black v. Cutter Laboratories, 351 U.S. 292, 76 S.Ct. 824, 100 L.Ed. 1188 (1956), we decline to address the question of whether MUW's admissions policy, as applied to males seeking admission to schools other than the School of Nursing, violates the Fourteenth Amendment. 8 Without question, MUW's admissions policy worked to Hogan's disadvantage. Although Hogan could have attended classes and received credit in one of Mississippi's state-supported coeducational nursing programs, none of which was located in Columbus, he could attend only by driving a considerable distance from his home. Tr. 19-20, 63-65. A similarly situated female would not have been required to choose between forgoing credit and bearing that inconvenience. Moreover, since many students enrolled in the School of Nursing hold full-time jobs, Deposition of Dean Annette K. Barrar 29-30, Hogan's female colleagues had available an opportunity, not open to Hogan, to obtain credit for additional training. The policy of denying males the right to obtain credit toward a baccalaureate degree thus imposed upon Hogan "a burden he would not bear were he female." Orr v. Orr, 440 U.S. 268, 273, 99 S.Ct. 1102, 1108, 59 L.Ed.2d 306 (1979). 9 In his dissenting opinion, Justice POWELL argues that a less rigorous test should apply because Hogan does not advance a "serious equal protection claim." Post, at 742. Justice BLACKMUN, without proposing an alternative test, labels the test applicable to gender-based discrimination as "rigid" and productive of "needless conformity." Post, at 734, 735. Our past decisions establish, however, that when a classification expressly discriminates on the basis of gender, the analysis and level of scrutiny applied to determine the validity of the classification do not vary simply because the objective appears acceptable to individual Members of the Court. While the validity and importance of the objective may affect the outcome of the analysis, the analysis itself does not change. Thus, we apply the test previously relied upon by the Court to measure the constitutionality of gender-based discrimination. Because we conclude that the challenged statutory classification is not substantially related to an important objective, we need not decide whether classifications based upon gender are inherently suspect. See Stanton v. Stanton, 421 U.S. 7, 13, 95 S.Ct. 1373, 1377, 43 L.Ed.2d 688 (1975). 10 History provides numerous examples of legislative attempts to exclude women from particular areas simply because legislators believed women were less able than men to perform a particular function. In 1873, this Court remained unmoved by Myra Bradwell's argument that the Fourteenth Amendment prohibited a State from classifying her as unfit to practice law simply because she was female. Bradwell v. Illinois, 16 Wall. 130, 21 L.Ed. 442 (1873). In his opinion concurring in the judgment, Justice Bradley described the reasons underlying the State's decision to determine which positions only men could fill: "It is the prerogative of the legislator to prescribe regulations founded on nature, reason, and experience for the due admission of qualified persons to professions and callings demanding special skill and confidence. This fairly belongs to the police power of the State; and, in my opinion, in view of the peculiar characteristics, destiny, and mission of woman, it is within the province of the legislature to ordain what offices, positions, and callings shall be filled and discharged by men, and shall receive the benefit of those energies and responsibilities, and that decision and firmness which are presumed to predominate in the sterner sex." Id., 16 Wall., at 142. In a similar vein, the Court in Goesaert v. Cleary, 335 U.S. 464, 466, 69 S.Ct. 198, 199, 93 L.Ed. 163 (1948), upheld a legislature's right to preclude women from bartending, except under limited circumstances, on the ground that the legislature could devise preventive measures against "moral and social problems" that result when women, but apparently not men, tend bar. Similarly, the many protective labor laws enacted in the late 19th and early 20th centuries often had as their objective the protection of weaker workers, which the laws assumed meant females. See generally B. Brown, A. Freedman, H. Katz, & A. Price, Women's Rights and the Law 209-210 (1977). 11 For instance, in Stanton v. Stanton, supra, this Court invalidated a state statute that specified a greater age of majority for males than for females and thereby affected the period during which a divorced parent was responsible for supporting his children. We did not question the importance or validity of the State's interest in defining parents' obligation to support children during their minority. On analysis, however, we determined that the purported relationship between that objective and the gender-based classification was based upon traditional assumptions that "the female [is] destined solely for the home and the rearing of the family, and only the male for the marketplace and the world of ideas. . . . If a specified age of minority is required for the boy in order to assure him parental support while he attains his education and training, so, too, is it for the girl." 421 U.S., at 14-15, 95 S.Ct., at 1377-78. Once those traditional notions were abandoned, no basis for finding a substantial relationship between classification and objective remained. 12 See, e.g., Kirchberg v. Feenstra, 450 U.S. 455, 101 S.Ct. 1195, 67 L.Ed.2d 428 (1981) (statute granted only husbands the right to manage and dispose of jointly owned property without the spouse's consent); Wengler v. Druggists Mutual Ins. Co., 446 U.S. 142, 100 S.Ct. 1540, 64 L.Ed.2d 107 (1980) (statute required a widower, but not a widow, to show he was incapacitated from earning to recover benefits for a spouse's death under workers' compensation laws); Orr v. Orr, supra (only men could be ordered to pay alimony following divorce); Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976) (women could purchase "nonintoxicating" beer at a younger age than could men); Stanton v. Stanton, supra (women reached majority at an earlier age than did men); Weinberger v. Wiesenfeld, 420 U.S. 636, 95 S.Ct. 1225, 43 L.Ed.2d 514 (1975) (widows, but not widowers, could collect survivors' benefits under the Social Security Act); Frontiero v. Richardson, 411 U.S. 677, 93 S.Ct. 1764, 36 L.Ed.2d 583 (1973) (determination of spouse's dependency based upon gender of member of Armed Forces claiming dependency benefits); Reed v. Reed, 404 U.S. 71, 92 S.Ct. 251, 30 L.Ed.2d 225 (1971) (statute preferred men to women as administrators of estates). 13 In the reply brief, the State understandably retreated from its contention that MUW was founded to provide opportunities for women which were not available to men. Reply Brief for Petitioners 4. Apparently, the impetus for founding MUW came not from a desire to provide women with advantages superior to those offered men, but rather from a desire to provide white women in Mississippi access to state-supported higher learning. In 1856, Sally Reneau began agitating for a college for white women. Those initial efforts were unsuccessful, and, by 1870, Mississippi provided higher education only for white men and black men and women. E. Mayes, History of Education in Mississippi 178, 228, 245, 259, 266, 270 (1899) (hereinafter Mayes). See also S. Neilson, The History of Mississippi State College for Women 4-5 (unpublished manuscript, 1952) (hereinafter Neilson). In 1882, two years before MUW was chartered, the University of Mississippi opened its doors to women. However, the institution was in those early years not "extensively patronized by females; most of those who come being such as desire to qualify themselves to teach." Id., at 178. By 1890, the largest number of women in any class at the University had been 23, while nearly 350 women enrolled in the first session of MUW. Id., at 178, 253. Because the University did not solicit the attendance of women until after 1920, and did not accept women at all for a time between 1907 and 1920, most Mississippi women who attended college attended MUW. Neilson, at 86. Thus, in Mississippi, as elsewhere in the country, women's colleges were founded to provide some form of higher education for the academically disenfranchised. See generally 2 T. Woody, A History of Women's Education in the United States 137-223 (1929); L. Baker, I'm Radcliffe! Fly Me! The Seven Sisters and the Failure of Women's Education 22, 136-141 (1976). 14 Relatively little change has taken place during the past 10 years. In 1980, women received more than 94 percent of the baccalaureate degrees conferred nationwide, National Center for Education Statistics, 1981 Digest of Education Statistics 121 (1981), and constituted 96.5 percent of the registered nurses in the labor force. United States Bureau of the Census, 1981 Statistical Abstract of the United States 402 (1981). 15 Officials of the American Nurses Association have suggested that excluding men from the field has depressed nurses' wages. Hearings before the United States Equal Employment Opportunity Commission on Job Segregation and Wage Discrimination 510-511, 517-518, 523 (Apr.1980). To the extent the exclusion of men has that effect, MUW's admissions policy actually penalizes the very class the State purports to benefit. Cf. Weinberger v. Wiesenfeld, 420 U.S. 636, 95 S.Ct. 1225, 43 L.Ed.2d 514 (1975). 16 Even were we to assume that discrimination against women affects their opportunity to obtain an education or to obtain leadership roles in nursing, the challenged policy nonetheless would be invalid, for the State has failed to establish that the legislature intended the single-sex policy to compensate for any perceived discrimination. Cf. Califano v. Webster, 430 U.S. 313, 318, 97 S.Ct. 1192, 1195, 51 L.Ed.2d 360 (1977) (legislative history of the compensatory statute revealed that Congress "directly addressed the justification for differing treatment of men and women" and "purposely enacted the more favorable treatment for female wage earners. . ."). The State has provided no evidence whatever that the Mississippi Legislature has ever attempted to justify its differing treatment of men and women seeking nurses' training. Indeed, the only statement of legislative purpose is that in § 37-117-3 of the Mississippi Code, see n. 1, supra, a statement that relies upon the very sort of archaic and overbroad generalizations about women that we have found insufficient to justify a gender-based classification. E.g., Orr v. Orr, 440 U.S. 268, 99 S.Ct. 1102, 59 L.Ed.2d 306 (1979); Stanton v. Stanton, 421 U.S. 7, 95 S.Ct. 1373, 43 L.Ed.2d 688 (1975). 17 Justice POWELL's dissent suggests that a second objective is served by the gender-based classification in that Mississippi has elected to provide women a choice of educational environments. Post, at -742-744. Since any gender-based classification provides one class a benefit or choice not available to the other class, however, that argument begs the question. The issue is not whether the benefited class profits from the classification, but whether the State's decision to confer a benefit only upon one class by means of a discriminatory classification is substantially related to achieving a legitimate and substantial goal. 1 "[T]wo other Mississippi universities offered coeducational programs leading to a Bachelor of Science in Nursing the University of Southern Mississippi in Hattiesburg, 178 miles from Columbus; and the University of Mississippi in Jackson, 147 miles from Columbus . . . ." Brief for Respondent 3. See also Tr. of Oral Arg. 8. 2 The history, briefly summarized above, of single-sex higher education in the Northeast is duplicated in other States. I mention only my State of Virginia, where even today Hollins College, Mary Baldwin College, Randolph Macon Woman's College, and Sweet Briar College remain all women's colleges. Each has a proud and respected reputation of quality education. 3 It is true that historically many institutions of higher education—particularly in the East and South—were single-sex. To these extents, choices were by no means universally available to all men and women. But choices always were substantial, and the purpose of relating the experience of our country with single-sex colleges and universities is to document what should be obvious: generations of Americans, including scholars, have thought—wholly without regard to any discriminatory animus—that there were distinct advantages in this type of higher education. 4 In announcing Wellesley's decision in 1973 to remain a women's college, President Barbara Newell said that "[t]he research we have clearly demonstrates that women's colleges produce a disproportionate number of women leaders and women in responsible positions in society; it does demonstrate that the higher proportion of women on the faculty the higher the motivation for women students." Carnegie Report, in Babcock, at 1014. Similarly rejecting coeducation in 1971, the Mount Holyoke Trustees Committee on Coeducation reported that "the conditions that historically justified the founding of women's colleges" continued to justify their remaining in that tradition. Ibid. 5 In this Court the benefits of single-sex education have been asserted by the students and alumnae of MUW. One would expect the Court to regard their views as directly relevant to this case: "[I]n the aspect of life known as courtship or mate-pairing, the American female remains in the role of the pursued sex, expected to adorn and groom herself to attract the male. Without comment on the common sense or equities of this social arrangement, it remains a sociological fact. "An institution of collegiate higher learning maintained exclusively for women is uniquely able to provide the education atmosphere in which some, but not all, women can best attain maximum learning potential. It can serve to overcome the historic repression of the past and can orient a woman to function and achieve in the still male dominated economy. It can free its students of the burden of playing the mating game while attending classes, thus giving academic rather than sexual emphasis. Consequently, many such institutions flourish and their graduates make significant contributions to the arts, professions and business." Brief for Mississippi University for Women Alumnae Association as Amicus Curiae 2-3. 6 "[T]he Constitution does not require that a classification keep abreast of the latest in educational opinion, especially when there remains a respectable opinion to the contrary . . . . Any other rule would mean that courts and not legislatures would determine all matters of public policy." Williams v. McNair, 316 F.Supp. 134, 137 (SC 1970) (footnote omitted), summarily aff'd, 401 U.S. 951, 91 S.Ct. 976, 28 L.Ed.2d 235 (1971). 7 See Kirchberg v. Feenstra, 450 U.S. 455, 456, 101 S.Ct. 1195, 1197, 67 L.Ed.2d 428 (1981) (invalidating statute "that gave husband, as 'head and master' of property jointly owned with his wife, the unilateral right to dispose of such property without his spouse's consent"); Wengler v. Druggists Mutual Ins. Co., 446 U.S. 142, 147, 100 S.Ct. 1540, 1544, 64 L.Ed.2d 107 (1980) (invalidating law under which the benefits "that the working woman can expect to be paid to her spouse in the case of her work-related death are less than those payable to the spouse of the deceased male wage earner"); Stanton v. Stanton, 421 U.S. 7, 95 S.Ct. 1373, 43 L.Ed.2d 688 (1975) (invalidating statute that provided a shorter period of parental support obligation for female children than for male children); Weinberger v. Wiesenfeld, 420 U.S. 636, 645, 95 S.Ct. 1225, 1232, 43 L.Ed.2d 514 (1975) (invalidating statute that failed to grant a woman worker "the same protection which a similarly situated male worker would have received"); Frontiero v. Richardson, 411 U.S. 677, 683, 93 S.Ct. 1764, 1768, 36 L.Ed.2d 583 (1973) (invalidating statute containing a "mandatory preference for male applicants"); Reed v. Reed, 404 U.S. 71, 74, 92 S.Ct. 251, 253, 30 L.Ed.2d 225 (1971) (invalidating an "arbitrary preference established in favor of males" in the administration of decedent's estates). 8 See also Caban v. Mohammed, 441 U.S. 380, 99 S.Ct. 1760, 60 L.Ed.2d 297 (1979) (invalidating law that both denied men the opportunity—given to women—of blocking the adoption of his illegitimate child by means of withholding his consent, and did not permit women to counter the statute's generalization that the maternal role is more important to women than the paternal role is to men). 9 Even the Court does not argue that the appropriate standard here is "strict scrutiny"—a standard that none of our "sex discrimination" cases ever has adopted. Sexual segregation in education differs from the tradition, typified by the decision in Plessy v. Ferguson, 163 U.S. 537, 16 S.Ct. 1138, 41 L.Ed. 256 (1896), of "separate but equal" racial segregation. It was characteristic of racial segregation that segregated facilities were offered, not as alternatives to increase the choices available to blacks, but as the sole alternative. MUW stands in sharp contrast. Of Mississippi's 8 public universities and 16 public junior colleges, only MUW considers sex as a criterion for admission. Women consequently are free to select a coeducational education environment for themselves if they so desire; their attendance at MUW is not a matter of coercion. 10 The Court says that "any gender-based classification provides one class a benefit or choice not available to the other class . . . ." Ante, at 731, n. 17. It then states that the issue "is not whether the benefited class profits from the classification, but whether the State's decision to confer a benefit only upon one class by means of a discriminatory classification is substantially related to achieving a legitimate and substantial goal." Ibid. (emphasis added). This is not the issue in this case. Hogan is not complaining about any benefit conferred upon women. Nor is he claiming discrimination because Mississippi offers no all-male college. As his brief states: "Joe Hogan does not ask to attend an all-male college which offers a Bachelor of Science in Nursing; he asks only to attend MUW." Brief for Respondent 24. And he asks this only for his personal convenience. 11 Students in respondent's position, in "being denied the right to attend the State college in their home town, are treated no differently than are other students who reside in communities many miles distant from any State supported college or university. The location of any such institution must necessarily inure to the benefit of some and to the detriment of others, depending upon the distance the affected individuals reside from the institution." Heaton v. Bristol, 317 S.W.2d 86, 99 (Tex.Civ.App.1958), cert. denied, 359 U.S. 230, 79 S.Ct. 802, 3 L.Ed.2d 765 (1959), quoted in Williams v. McNair, 316 F.Supp., at 137. 12 " 'Such a plan (i.e., giving the student a choice of a "single-sex" and coeducational institutions) exalts neither sex at the expense of the other, but to the contrary recognizes the equal rights of both sexes to the benefit of the best, most varied system of higher education that the State can supply.' " Williams v. McNair, supra, at 138, n. 15, quoting Heaton v. Bristol, supra, at 100. 13 For instance, the School of Nursing takes up 15 pages of MUW's 234-page course catalog. See Mississippi University for Women, 81/82 Bulletin 185-200. 14 E.g., Bachelor of Arts; Bachelor of Science; Master of Arts; Master of Science. See id., at 40. MUW also offers special preprofessional programs in law, dentistry, medicine, pharmacy, physical therapy, and veterinary medicine. Ibid. 15 MUW's Bulletin in its Table of Contents lists the following subjects (offered in its School of Arts and Sciences): Air Force ROTC; Art; Behavioral Sciences; Biological Sciences; Business and Economics; Cooperative Education; English and Foreign Languages; Health, Physical Education, Recreation, and Dance; History, Journalism and Broadcasting; Mathematics; Music; Physical Sciences; and Speech Communication. See id., at 3. 16 For instance, the catalog for the coeducational University of Mississippi lists in its general description the "Sarah Isom Center for Women's Studies," which is described as "dedicated to the development of curriculum and scholarship about women, the dissemination of information about their expanding career opportunities, and the establishment of mutual support networks for women of all ages and backgrounds." University of Mississippi, 1982 Undergraduate Catalog 13-14. This listing precedes information about the University's Law and Medical Centers. Id., at 14-15. 17 The Court argues that MUW's means are not sufficiently related to its goal because it has allowed men to audit classes. The extent of record information is that men have audited 138 courses in the last 10 years. Brief for Respondent 21. On average, then, men have audited 14 courses a year. MUW's current annual catalog lists 913 courses offered in one year. See Mississippi University for Women, 81/82 Bulletin passim. It is understandable that MUW might believe that it could allow men to audit courses without materially affecting its environment. MUW charges tuition but gives no academic credit for auditing. The University evidently is correct in believing that few men will choose to audit under such circumstances. This deviation from a perfect relationship between means and ends is insubstantial. 18 The Court, in the opening and closing sentences and note 7 of its opinion, states the issue in terms only of a "professional nursing school" and "decline[s] to address the question of whether MUW's admissions policy, as applied to males seeking admission to schools other than the School of Nursing, violates the Fourteenth Amendment." This would be a welcome limitation if, in fact, it leaves MUW free to remain an all-women's university in each of its other schools and departments—which include four schools and more than a dozen departments. Cf. nn. 13-15, supra. The question the Court does not answer is whether MUW may remain a women's university in every respect except its School of Nursing. This is a critical question for this University and its responsible board and officials. The Court holds today that they have deprived Hogan of constitutional rights because MUW is adjudged guilty of sex discrimination. The logic of the Court's entire opinion, apart from its statements mentioned above, appears to apply sweepingly to the entire University. The exclusion of men from the School of Nursing is repeatedly characterized as "gender-based discrimination," subject to the same standard of analysis applied in previous sex discrimination cases of this Court. Nor does the opinion anywhere deny that this analysis applies to the entire University. The Court nevertheless purports to decide this case "narrow[ly]." Normally and properly we decide only the question presented. It seems to me that in fact the issue properly before us is the single-sex policy of the University, and it is this issue that I have addressed in this dissent. The Court of Appeals so viewed this case, and unambiguously held that a single-sex state institution of higher education no longer is permitted by the Constitution. I see no principled way—in light of the Court's rationale—to reach a different result with respect to other MUW schools and departments. But given the Court's insistence that its decision applies only to the School of Nursing, it is my view that the Board and officials of MUW may continue to operate the remainder of the University on a single-sex basis without fear of personal liability. The standard of such liability is whether the conduct of the official "violate[s] clearly established statutory or constitutional rights of which a reasonable person would have known." Harlow v. Fitzgerald, 457 U.S. 800, 818, 102 S.Ct. 2727, 2738, 73 L.Ed.2d 396 (1982). The Court today leaves in doubt the reach of its decision.

12

458 U.S. 592 102 S.Ct. 3260 73 L.Ed.2d 995 ALFRED L. SNAPP & SON, INC., et al., Petitionersv.PUERTO RICO, ex rel., Pedro BAREZ, Secretary of Labor and Human Resources. No. 80-1305. Argued April 20, 1982. Decided July 1, 1982. Syllabus Respondent Commonwealth of Puerto Rico filed suit in Federal District Court against petitioners, individuals and companies engaged in the apple industry in Virginia, alleging that petitioners had violated related provisions of the Wagner-Peyser Act and the Immigration and Nationality Act of 1952, and implementing regulations. The purposes of this statutory and regulatory scheme are to give United States workers, including citizens of Puerto Rico, a preference over temporary foreign workers for jobs that become available within this country, to ensure that working conditions of domestic employees are not adversely affected when foreign workers are brought in, and to prohibit discrimination against United States workers in favor of foreign workers. It was alleged that pursuant to the federal laws petitioners had reported 787 job openings for temporary farm labor to pick the 1978 apple crop, and that in violation of such laws petitioners had discriminated against Puerto Rican workers by failing to provide employment for qualified Puerto Rican migrant farmworkers, by subjecting those Puerto Rican workers that were employed to working conditions more burdensome than those established for temporary foreign workers, and by improperly terminating employment of Puerto Rican workers. Seeking declaratory and injunctive relief in its capacity as parens patriae, Puerto Rico asserted that this alleged discrimination deprived the Commonwealth of its right "to effectively participate in the benefits of the Federal Employment Service System of which it is a part" and thereby caused irreparable injury to the Commonwealth's efforts "to promote opportunities for profitable employment for Puerto Rican laborers and to reduce unemployment in the Commonwealth." The District Court dismissed the complaint, holding that Puerto Rico lacked standing to bring the action in view of the small number of individuals directly involved and the slight impact upon Puerto Rico's general economy that the loss of 787 temporary jobs could have. The Court of Appeals reversed. Held: Puerto Rico has parens patriae standing to maintain this suit. Pp. 600-610. (a) In order to maintain a parens patriae action, a State must articulate an interest apart from the interests of particular private parties, that is, the State must be more than a nominal party. The State must express a "quasi-sovereign" interest, such as its interest in the health and well-being—both physical and economic—of its residents in general. Although more must be alleged than injury to an identifiable group of individual residents, the indirect effects of the injury must be considered as well in determining whether the State has alleged injury to a sufficiently substantial segment of its population. A State also has a quasi-sovereign interest in not being discriminatorily denied its rightful status within the federal system—that is, in ensuring that the State and its general population are not excluded from the benefits that are to flow from participation in the federal system. Pp. 600-608. (b) Under the above principles, Puerto Rico's allegations that petitioners discriminated against Puerto Ricans in favor of foreign laborers falls within the Commonwealth's quasi-sovereign interest in the general well-being of its citizens. A State's interest in the well-being of its residents, which extends beyond mere physical interests to economic and commercial interests, also includes the State's substantial interest in securing its residents from the harmful effects of discrimination. This interest is peculiarly strong in the case of Puerto Rico simply because of the fact that invidious discrimination frequently occurs along ethnic lines. Alternatively, Puerto Rico has parens patriae standing to pursue its residents' interests in the Commonwealth's full and equal participation in the federal employment service scheme established by the laws involved here. Pp.608-610 4th Cir., 632 F.2d 365, affirmed. Thomas J. Bacas, Washington, D. C., for petitioners. Paul A. Lenzini, Washington, D. C., for respondent. Justice WHITE delivered the opinion of the Court. 1 In this case, the Commonwealth of Puerto Rico seeks to bring suit in its capacity as parens patriae against petitioners for their alleged violations of federal law. Puerto Rico contends that those violations discriminated against Puerto Ricans and injured the Puerto Rican economy. The question presented here is whether Puerto Rico has standing to maintain this suit. 2 * A. 3 The factual background of this case involves the interaction of two federal statutes, the Wagner-Peyser Act, 48 Stat. 113, 29 U.S.C. § 49 et seq., and the Immigration and Nationality Act of 1952, 66 Stat. 163, as amended, 8 U.S.C. § 1101 et seq. (1976 ed. and Supp.IV). The Wagner-Peyser Act was passed in 1933 in order to deal with the massive problem of unemployment resulting from the Depression. The Act establishes the United States Employment Service within the Department of Labor "[i]n order to promote the establishment and maintenance of a national system of public employment offices." 29 U.S.C. § 49. State agencies, which have been approved by the Secretary of Labor, are authorized to participate in the nationwide employment service.1 § 49g. The Secretary is authorized to make "such rules and regulations as may be necessary" to accomplish the ends of the Act. § 49k. Federal regulations issued pursuant to that authority have established an interstate clearance system to provide employers a means of recruiting nonlocal workers, when the supply of local workers is inadequate. 20 CFR § 602.2(c) (1981). If local workers are not available, a "clearance order" is sent through the Employment and Training Administration of the Department of Labor to other state agencies in order to give them an opportunity to meet the request. 4 Some of petitioners' obligations under the employment system established by the Wagner-Peyser Act stem from the Immigration and Nationality Act of 1952, insofar as it regulates the admission of nonimmigrant aliens into the United States. The latter Act authorizes the admission of temporary foreign workers into the United States only "if unemployed persons capable of performing such service or labor cannot be found in this country." 8 U.S.C. § 1101(a)(15)(H)(ii). The Attorney General is charged with determining whether entry of foreign workers would meet this standard, "upon petition of the importing employer." 8 U.S.C. § 1184(c). He is to make this determination "after consultation with appropriate agencies of the Government." Ibid. The Attorney General has delegated this responsibility to the Commissioner of Immigration and Naturalization, 8 CFR § 2.1 (1982), who, in turn, relies on the Secretary of Labor for the initial determinations. 8 CFR § 214.2(h)(3) (1982).2 To meet this responsibility, the Secretary of Labor relies upon the employment referral system established under the Wagner-Peyser Act. 5 Any employer who wants to employ temporary foreign agricultural laborers must first seek domestic laborers for the openings through use of the interstate clearance system. The employer who anticipates a need for foreign laborers must file an application with the local public employment office, including a copy of the job offer. 20 CFR §§ 655.201(a)(1), (b)(1) (1981). The application must be filed in sufficient time to allow the agency to recruit through the interstate clearance system for 60 days prior to the estimated date of the start of employment. § 655.201(c). The regulations further provide that the employer must include assurances that the job opportunity is "open to all qualified U. S. workers without regard to race, color, national origin, sex, or religion, and is open to United States workers with handicaps who are qualified to perform the work," and that the employer will continue to seek United States workers until the foreign workers have departed for the employer's place of employment. §§ 655.203(c), (d).3 Finally, the regulations require that "each employer's job offer to U. S. workers must offer U. S. workers at least the same benefits which the employer is offering, intends to offer, or will afford, to temporary foreign workers." § 655.202(a). Similarly, the employer may not impose obligations or restrictions on domestic workers that are not, or will not be, imposed on foreign workers. Ibid. 6 The obvious point of this somewhat complicated statutory and regulatory framework is to provide two assurances to United States workers, including the citizens of Puerto Rico. First, these workers are given a preference over foreign workers for jobs that become available within this country. Second, to the extent that foreign workers are brought in, the working conditions of domestic employees are not to be adversely affected, nor are United States workers to be discriminated against in favor of foreign workers. B 7 The particular facts of this case involve the 1978 apple harvest on the east coast. That was apparently a good year for apples, resulting in a substantial need for temporary farm laborers to pick the crop. To meet this need the apple growers filed clearance orders with their state employment agencies. Through the system described above, a total of 2,318 job openings were transmitted to Puerto Rico on August 2, 1978. As of August 14, which marked the end of the 60-day "availability" period, supra, at 596, the Commonwealth Department of Labor had recruited 1,094 Puerto Rican workers. Puerto Rican workers for the remaining openings were subsequently recruited. As stated in Puerto Rico's complaint: 8 "Of this total number of 2,318 Puerto Rican workers, only 992 actually arrived on the mainland. The remainder never left Puerto Rico because of oral advice from the United States Department of Labor requesting cancellation of remaining flights because many of the defendant growers had refused to employ Puerto Rican workers who had already arrived. Of the 992 workers who arrived at the orchards, 420 came to Virginia orchards. Of these 420 workers, fewer than 30 had employment three weeks later, the growers having refused to employ most of these workers and having dismissed most of the rest within a brief time for alleged unproductivity." App. 17-18. 9 Puerto Rico filed this suit on January 11, 1979, naming as defendants numerous individuals and companies engaged in the apple industry in Virginia.4 Of the 2,318 job requests forwarded to Puerto Rico, respondent alleged that 787 of these had come from the named Virginia growers. In three counts, the complaint alleged that the defendants had violated the Wagner-Peyser Act, the Immigration and Nationality Act of 1952, and various federal regulations implementing those statutes, by failing to provide employment for qualified Puerto Rican migrant farmworkers, by subjecting those Puerto Rican workers that were employed to working conditions more burdensome than those established for temporary foreign workers,5 and by improperly terminating employment of Puerto Rican workers. Alleging that this discrimination against Puerto Rican farmworkers deprived "the Commonwealth of Puerto Rico of its right to effectively participate in the benefits of the Federal Employment Service System of which it is a part" and thereby caused irreparable injury to the Commonwealth's efforts "to promote opportunities for profitable employment for Puerto Rican laborers and to reduce unemployment in the Commonwealth," respondent sought declaratory relief with respect to the past practices of petitioners and injunctive relief requiring petitioners to conform to the relevant federal statutes and regulations in the future. 10 Petitioners responded with a motion to dismiss, asserting that respondent lacked standing to bring this action. Although the District Court, 469 F.Supp. 928, held that the Commonwealth of Puerto Rico is capable of asserting parens patriae interests in general, it agreed with petitioners' contention that no such action could be maintained under the circumstances of this case. In particular, the District Court relied upon the relatively small number of individuals directly involved—some 787 out of a total population of close to 3 million—and the slight impact upon the general economy of Puerto Rico that the loss of this number of temporary jobs could have. 11 A divided panel of the Court of Appeals for the Fourth Circuit reversed.6 632 F.2d 365 (1980). The majority held that the District Court had focused too narrowly on those directly involved, ignoring those that were indirectly affected by petitioners' alleged actions. Noting the serious dimensions of the unemployment problem in Puerto Rico and the general condition of its economy,7 the court stated that "[d]eliberate efforts to stigmatize the labor force as inferior carry a universal sting" and the "inability of the United States government . . . to grant Puerto Ricans equal treatment with other citizens or even with foreign temporary workers must certainly have an effect which permeates the entire island of Puerto Rico." Id., at 370. These indirect effects on the interests of "a substantial portion of its citizenry" were sufficient, in its view, to support a parens patriae action. Ibid. 12 We granted certiorari to determine whether Puerto Rico could maintain a parens patriae action here, despite the small number of individuals directly involved. 454 U.S. 1079, 102 S.Ct. 631, 70 L.Ed.2d 612 (1981). II 13 Parens patriae means literally "parent of the country."8 The parens patriae action has its roots in the common-law concept of the "royal prerogative."9 The royal prerogative included the right or responsibility to take care of persons who "are legally unable, on account of mental incapacity, whether it proceed from 1st. nonage: 2. idiocy: or 3. lunacy: to take proper care of themselves and their property."10 At a fairly early date, American courts recognized this common-law concept, but now in the form of a legislative prerogative: "This prerogative of parens patriae is inherent in the supreme power of every State, whether that power is lodged in a royal person or in the legislature [and] is a most beneficent function . . . often necessary to be exercised in the interests of humanity, and for the prevention of injury to those who cannot protect themselves." Mormon Church v. United States, 136 U.S. 1, 57, 10 S.Ct. 792, 808, 34 L.Ed. 481 (1890). 14 This common-law approach, however, has relatively little to do with the concept of parens patriae standing that has developed in American law. That concept does not involve the States stepping in to represent the interests of particular citizens who, for whatever reason, cannot represent themselves. In fact, if nothing more than this is involved—i.e., if the State is only a nominal party without a real interest of its own—then it will not have standing under the parens patriae doctrine. See Pennsylvania v. New Jersey, 426 U.S. 660, 96 S.Ct. 2333, 49 L.Ed.2d 124 (1976); Oklahoma ex rel. Johnson v. Cook, 304 U.S. 387, 58 S.Ct. 954, 82 L.Ed. 1416 (1938); Oklahoma v. Atchison, T. & S.F.R. Co., 220 U.S. 277, 31 S.Ct. 434, 55 L.Ed. 465 (1911). Rather, to have such standing the State must assert an injury to what has been characterized as a "quasi-sovereign" interest, which is a judicial construct that does not lend itself to a simple or exact definition. Its nature is perhaps best understood by comparing it to other kinds of interests that a State may pursue and then by examining those interests that have historically been found to fall within this category. 15 Two sovereign interests are easily identified: First, the exercise of sovereign power over individuals and entities within the relevant jurisdiction—this involves the power to create and enforce a legal code, both civil and criminal; second, the demand for recognition from other sovereigns—most frequently this involves the maintenance and recognition of borders. The former is regularly at issue in constitutional litigation. The latter is also a frequent subject of litigation, particularly in this Court: 16 "The original jurisdiction of this Court is one of the mighty instruments which the framers of the Constitution provided so that adequate machinery might be available for the peaceful settlement of disputes between States and between a State and citizens of another State. . . . The traditional methods available to a sovereign for the settlement of such disputes were diplomacy and war. Suit in this Court was provided as an alternative." Georgia v. Pennsylvania R. Co., 324 U.S. 439, 450, 65 S.Ct. 716, 722, 89 L.Ed. 1051 (1945). 17 Not all that a State does, however, is based on its sovereign character. Two kinds of nonsovereign interests are to be distinguished. First, like other associations and private parties, a State is bound to have a variety of proprietary interests. A State may, for example, own land or participate in a business venture. As a proprietor, it is likely to have the same interests as other similarly situated proprietors. And like other such proprietors it may at times need to pursue those interests in court. Second, a State may, for a variety of reasons, attempt to pursue the interests of a private party, and pursue those interests only for the sake of the real party in interest. Interests of private parties are obviously not in themselves sovereign interests, and they do not become such simply by virtue of the State's aiding in their achievement. In such situations, the State is no more than a nominal party. 18 Quasi-sovereign interests stand apart from all three of the above: They are not sovereign interests, proprietary interests, or private interests pursued by the State as a nominal party. They consist of a set of interests that the State has in the well-being of its populace. Formulated so broadly, the concept risks being too vague to survive the standing requirements of Art. III: A quasi-sovereign interest must be sufficiently concrete to create an actual controversy between the State and the defendant. The vagueness of this concept can only be filled in by turning to individual cases. 19 That a parens patriae action could rest upon the articulation of a "quasi-sovereign" interest was first recognized by this Court in Louisiana v. Texas, 176 U.S. 1, 20 S.Ct. 251, 44 L.Ed. 347 (1900). In that case, Louisiana unsuccessfully sought to enjoin a quarantine maintained by Texas officials, which had the effect of limiting trade between Texas and the port of New Orleans. The Court labeled Louisiana's interest in the litigation as that of parens patriae, and went on to describe that interest by distinguishing it from the sovereign and proprietary interests of the State: 20 "Inasmuch as the vindication of the freedom of interstate commerce is not committed to the State of Louisiana, and that State is not engaged in such commerce, the cause of action must be regarded not as involving any infringement of the powers of the State of Louisiana, or any special injury to her property, but as asserting that the State is entitled to seek relief in this way because the matters complained of affect her citizens at large." Id., at 19, 20 S.Ct., at 257.11 21 Although Louisiana was unsuccessful in that case in pursuing the commercial interests of its residents, a line of cases followed in which States successfully sought to represent the interests of their citizens in enjoining public nuisances. North Dakota v. Minnesota, 263 U.S. 365, 44 S.Ct. 138, 68 L.Ed. 342 (1923); Wyoming v. Colorado, 259 U.S. 419, 42 S.Ct. 552, 66 L.Ed. 999 (1922); New York v. New Jersey, 256 U.S. 296, 41 S.Ct. 492, 65 L.Ed. 937 (1921); Kansas v. Colorado, 206 U.S. 46, 27 S.Ct. 655, 51 L.Ed. 956 (1907); Georgia v. Tennessee Copper Co., 206 U.S. 230, 27 S.Ct. 618, 51 L.Ed. 1038 (1907); Kansas v. Colorado, 185 U.S. 125, 22 S.Ct. 552, 46 L.Ed. 838 (1902); Missouri v. Illinois, 180 U.S. 208, 21 S.Ct. 331, 45 L.Ed. 497 (1901). 22 In the earliest of these, Missouri v. Illinois, Missouri sought to enjoin the defendants from discharging sewage in such a way as to pollute the Mississippi River in Missouri. The Court relied upon an analogy to independent countries in order to delineate those interests that a State could pursue in federal court as parens patriae, apart from its sovereign and proprietary interests:12 23 "It is true that no question of boundary is involved, nor of direct property rights belonging to the complainant State. But it must surely be conceded that, if the health and comfort of the inhabitants of a State are threatened, the State is the proper party to represent and defend them. If Missouri were an independent and sovereign State all must admit that she could seek a remedy by negotiation, and, that failing, by force. Diplomatic powers and the right to make war having been surrendered to the general government, it was to be expected that upon the latter would be devolved the duty of providing a remedy and that remedy, we think, is found in the constitutional provisions we are considering." Id., at 241, 21 S.Ct., at 343. 24 This analogy to an independent country was also articulated in Georgia v. Tennessee Copper Co., supra, at 237, 27 S.Ct., at 619, a case involving air pollution in Georgia caused by the discharge of noxious gasses from the defendant's plant in Tennessee. Justice Holmes, writing for the Court, described the State's interest under these circumstances as follows: 25 "[T]he State has an interest independent of and behind the titles of its citizens, in all the earth and air within its domain. It has the last word as to whether its mountains shall be stripped of their forests and its inhabitants shall breathe pure air. It might have to pay individuals before it could utter that word, but with it remains the final power. . . . 26 ". . . When the States by their union made the forcible abatement of outside nuisances impossible to each, they did not thereby agree to submit to whatever might be done. They did not renounce the possibility of making reasonable demands on the ground of their still remaining quasi -sovereign interests." 27 Both the Missouri case and the Georgia case involved the State's interest in the abatement of public nuisances, instances in which the injury to the public health and comfort was graphic and direct. Although there are numerous examples of such parens patriae suits, e.g., North Dakota v. Minnesota, supra (flooding); New York v. New Jersey, supra (water pollution); Kansas v. Colorado, 185 U.S. 125, 22 S.Ct. 552, 46 L.Ed. 838 (1902) (diversion of water), parens patriae interests extend well beyond the prevention of such traditional public nuisances. 28 In Pennsylvania v. West Virginia, 262 U.S. 553, 43 S.Ct. 658, 67 L.Ed. 1117 (1923), for example, Pennsylvania was recognized as a proper party to represent the interests of its residents in maintaining access to natural gas produced in West Virginia: 29 "The private consumers in each State . . . constitute a substantial portion of the State's population. Their health, comfort and welfare are seriously jeopardized by the threatened withdrawal of the gas from the interstate stream. This is a matter of grave public concern in which the State, as representative of the public, has an interest apart from that of the individuals affected. It is not merely a remote or ethical interest but one which is immediate and recognized by law." Id., at 592, 43 S.Ct., at 663. 30 The public nuisance and economic well-being lines of cases were specifically brought together in Georgia v. Pennsylvania R. Co., 324 U.S. 439, 65 S.Ct. 716, 89 L.Ed. 1051 (1945), in which Georgia alleged that some 20 railroads had conspired to fix freight rates in a manner that discriminated against Georgia shippers in violation of the federal antitrust laws: 31 "If the allegations of the bill are taken as true, the economy of Georgia and the welfare of her citizens have seriously suffered as the result of this alleged conspiracy. . . . [Trade barriers] may cause a blight no less serious than the spread of noxious gas over the land or the deposit of sewage in the streams. They may affect the prosperity and welfare of a State as profoundly as any diversion of waters from the rivers. . . . Georgia as a representative of the public is complaining of a wrong which, if proven, limits the opportunities of her people, shackles her industries, retards her development, and relegates her to an inferior economic position among her sister States. These are matters of grave public concern in which Georgia has an interest apart from that of particular individuals who may be affected." Id., at 450-451, 65 S.Ct., at 722-723.13 32 This summary of the case law involving parens patriae actions leads to the following conclusions. In order to maintain such an action, the State must articulate an interest apart from the interests of particular private parties, i.e., the State must be more than a nominal party. The State must express a quasi-sovereign interest. Although the articulation of such interests is a matter for case-by-case development—neither an exhaustive formal definition nor a definitive list of qualifying interests can be presented in the abstract—certain characteristics of such interests are so far evident. These characteristics fall into two general categories. First, a State has a quasi-sovereign interest in the health and well-being—both physical and economic of its residents in general. Second, a State has a quasi-sovereign interest in not being discriminatorily denied its rightful status within the federal system. 33 The Court has not attempted to draw any definitive limits on the proportion of the population of the State that must be adversely affected by the challenged behavior. Although more must be alleged than injury to an identifiable group of individual residents, the indirect effects of the injury must be considered as well in determining whether the State has alleged injury to a sufficiently substantial segment of its population. One helpful indication in determining whether an alleged injury to the health and welfare of its citizens suffices to give the State standing to sue as parens patriae is whether the injury is one that the State, if it could, would likely attempt to address through its sovereign lawmaking powers.14 34 Distinct from but related to the general well-being of its residents, the State has an interest in securing observance of the terms under which it participates in the federal system. In the context of parens patriae actions, this means ensuring that the State and its residents are not excluded from the benefits that are to flow from participation in the federal system. Thus, the State need not wait for the Federal Government to vindicate the State's interest in the removal of barriers to the participation by its residents in the free flow of interstate commerce. See Pennsylvania v. West Virginia, 262 U.S. 553, 43 S.Ct. 658, 67 L.Ed. 1117 (1923). Similarly, federal statutes creating benefits or alleviating hardships create interests that a State will obviously wish to have accrue to its residents. See Georgia v. Pennsylvania R. Co., 324 U.S. 439, 65 S.Ct. 716, 89 L.Ed. 1051 (1945) (federal antitrust laws); Maryland v. Louisiana, 451 U.S. 725, 101 S.Ct. 2114, 68 L.Ed.2d 576 (1981) (Natural Gas Act). Once again, we caution that the State must be more than a nominal party. But a State does have an interest, independent of the benefits that might accrue to any particular individual, in assuring that the benefits of the federal system are not denied to its general population. 35 We turn now to the allegations of the complaint to determine whether they satisfy either or both of these criteria.15 III 36 The complaint presents two fundamental contentions. First, it alleges that the petitioners discriminated against Puerto Ricans in favor of foreign laborers. Second, it alleges that Puerto Ricans were denied the benefits of access to domestic work opportunities that the Wagner-Peyser Act and the Immigration and Nationality Act of 1952 were designed to secure for United States workers. We find each of these allegations to fall within the Commonwealth's quasi-sovereign interests and, therefore, each will support a parens patriae action. 37 Petitioners contend that at most there were only 787 job opportunities at stake in Virginia and that this number of temporary jobs could not have a substantial direct or indirect effect on the Puerto Rican economy. We believe that this is too narrow a view of the interests at stake here. Just as we have long recognized that a State's interests in the health and well-being of its residents extend beyond mere physical interests to economic and commercial interests, we recognize a similar state interest in securing residents from the harmful effects of discrimination. This Court has had too much experience with the political, social, and moral damage of discrimination not to recognize that a State has a substantial interest in assuring its residents that it will act to protect them from these evils. This interest is peculiarly strong in the case of Puerto Rico simply because of the unfortunate fact that invidious discrimination frequently occurs along ethnic lines. Puerto Rico's situation differs somewhat from the States in this regard—not in theory but in fact—simply because this country has for the most part been spared the evil of invidious discrimination based on state lines. Were this to come to pass, however, we have no doubt that a State could seek, in the federal courts, to protect its residents from such discrimination to the extent that it violates federal law. Puerto Rico claims that it faces this problem now. Regardless of the possibly limited effect of the alleged financial loss at issue here, we agree with the Court of Appeals that "[d]eliberate efforts to stigmatize the labor force as inferior carry a universal sting." 632 F.2d, at 370. 38 Alternatively, we find that Puerto Rico does have parens patriae standing to pursue the interests of its residents in the Commonwealth's full and equal participation in the federal employment service scheme established pursuant to the Wagner-Peyser Act and the Immigration and Nationality Act of 1952. Unemployment among Puerto Rican residents is surely a legitimate object of the Commonwealth's concern. Just as it may address that problem through its own legislation, it may also seek to assure its residents that they will have the full benefit of federal laws designed to address this problem. The Commonwealth's position in this respect is not distinguishable from that of Georgia when it sought the protection of the federal antitrust laws in order to eliminate freight rates that discriminated against Georgia shippers, Georgia v. Pennsylvania R. Co., supra, or from that of Maryland when it sought to secure the benefits of the Natural Gas Act for its residents, Maryland v. Louisiana, supra. Indeed, the fact that the Commonwealth participates directly in the operation of the federal employment scheme makes even more compelling its parens patriae interest in assuring that the scheme operates to the full benefit of its residents.16 39 For these reasons, the judgment of the Court of Appeals is 40 Affirmed. 41 Justice POWELL took no part in the decision of this case. 42 Justice BRENNAN, with whom Justice MARSHALL, Justice BLACKMUN, and Justice STEVENS join, concurring. 43 As the Court notes, ante, at 603, n. 12, the question whether a State can bring a parens patriae action within the original jurisdiction of this Court may well turn on considerations quite different from those implicated where the State seeks to press a parens patriae claim in the district courts. The Framers, in establishing original jurisdiction in this Court for suits "in which a State shall be a Party," Art. III, § 2, cl. 2, and Congress, in implementing the grant of original jurisdiction with respect to suits between States, 28 U.S.C. § 1251(a) (1976 ed., Supp.IV), may well have conceived of a somewhat narrower category of cases as presenting issues appropriate for initial determination in this Court than the full range of cases to which a State may have an interest cognizable by a federal court. The institutional limits on the Court's ability to accommodate such suits accentuates the need for more restrictive access to the original docket. In addition, because the judicial power of the United States does not extend to suits "commenced or prosecuted against one of the United States by Citizens of another State," U.S.Const., Amdt. 11, where one State brings a suit parens patriae against another State, a more circumspect inquiry may be required in order to ensure that the provisions of the Eleventh Amendment are not being too easily circumvented by the device of the State's bringing suit on behalf of some private party. Of course, none of the concerns that might counsel for a restrictive approach to the question of parens patriae standing is present in this case. 44 In cases such as the present one, I can discern no basis either in the Constitution or in policy for denying a State the opportunity to vindicate the federal rights of its citizens. At the very least, the prerogative of a State to bring suits in federal court should be commensurate with the ability of private organizations. A private organization may bring suit to vindicate its own concrete interest in performing those activities for which it was formed. E.g., Havens Realty Corp. v. Coleman, 455 U.S. 363, 378-379, 102 S.Ct. 1114, 1124-1125, 71 L.Ed.2d 214 (1982);1 Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 263, 97 S.Ct. 555, 562, 50 L.Ed.2d 450 (1977); N.A.A.C.P. v. Button, 371 U.S. 415, 428, 83 S.Ct. 328, 335, 9 L.Ed.2d 405 (1963). See also Gladstone, Realtors v. Village of Bellwood, 441 U.S. 91, 109-111, 99 S.Ct. 1601, 1612-1613, 60 L.Ed.2d 66 (1979) (standing of municipality premised on diminished tax base and other "harms flowing from the realities of a racially segregated community"). Cf. Sierra Club v. Morton, 405 U.S. 727, 739, 92 S.Ct. 1361, 1368, 31 L.Ed.2d 636 (1972).2 There is no doubt that Puerto Rico's interest in this litigation compares favorably to interests of the private organizations, and municipality, in the cases cited above. 45 More significantly, a State is no ordinary litigant. As a sovereign entity, a State is entitled to assess its needs, and decide which concerns of its citizens warrant its protection and intervention. I know of nothing—except the Constitution or overriding federal law—that might lead a federal court to superimpose its judgment for that of a State with respect to the substantiality or legitimacy of a State's assertion of sovereign interest. 46 With these considerations in mind, I join the opinion of the Court. 1 As used in the Act, the word "State" includes Puerto Rico. 29 U.S.C. § 49b(b). Puerto Rico's Department of Labor and Human Resources has been approved by the Secretary of Labor and participates in the federal-state system established by the Act. 2 "Either a certification from the Secretary of Labor or his designated representative stating that qualified persons in the United States are not available and that the employment of the beneficiary will not adversely affect the wages and working conditions of workers in the United States similarly employed, or a notice that such a certification cannot be made, shall be attached to every nonimmigrant visa petition to accord an alien a classification under section 101(a)(15)(H)(ii)." 8 CFR § 214.2(h)(3)(i) (1982). 3 There is a further requirement that the employer continue to provide an opportunity for employment to any qualified United States worker who applies for a position from the time the foreign workers depart for the employer's place of employment until the time that 50 percent of the period of the work contract has elapsed. 20 CFR § 655.203(e) (1981). 4 The complaint named 51 defendants: 32 apple growers and 19 officers, partners or employees of the apple growers. 5 The theory of the complaint was that the apple growers were discriminating against the Puerto Ricans in favor of Jamaican workers. In August 1978, apple growers in several States, including Virginia, filed suit in Federal District Court seeking an injunction against the United States Secretary of Labor, the Commissioner of the Immigration and Naturalization Service, and their subordinates, to permit the recruitment and employment of foreign workers. Puerto Rico was allowed to intervene in this suit to represent the interests of its residents in these work opportunities. The growers complained that the federal employment service had not produced sufficient laborers to assure that the harvest, which was about to begin, could be successfully accomplished with sufficient speed. The District Court issued a preliminary injunction ordering that a certain number of foreign workers be allowed to enter this country to pick apples. Frederick County Fruit Growers Assn., Inc. v. Marshall, No. 78-0086(H) (WD Va., Aug. 31, 1978). The Jamaicans secured entry under this order. Prior to issuing this injunction, however, the court was assured by the apple growers that they recognized their obligation to give priority to Puerto Rican workers, notwithstanding the court order. Puerto Rico's complaint was founded on the charge that the apple growers failed to meet this commitment and, thus, failed to meet their obligations under federal law. 6 The dissenting judge agreed with the analysis of the District Court. 7 In September 1978, 18.5% of the adults in the Puerto Rican labor force were unemployed. Rural unemployment stood at 23%. 8 " 'Parens patriae,' literally 'parent of the country,' refers traditionally to role of state as sovereign and guardian of persons under legal disability." Black's Law Dictionary 1003 (5th ed. 1979). 9 See Hawaii v. Standard Oil Co., 405 U.S. 251, 257, 92 S.Ct. 885, 888, 31 L.Ed.2d 184 (1972); G. Curtis, The Checkered Career of Parens Patriae, 25 DePaul L.Rev. 895, 896 (1976); Black's, supra. 10 J. Chitty, Prerogatives of the Crown 155 (1820), quoted in Curtis, supra, at 896. 11 Justice Harlan, in a concurring opinion, specifically rejected the idea that Louisiana had standing to pursue more than its sovereign and proprietary interests: "I am of opinion that the State of Louisiana, in its sovereign or corporate capacity, cannot bring any action in this court on account of the matters set forth in its bill. The case involves no property interest of that State. Nor is Louisiana charged with any duty, nor has it any power, to regulate interstate commerce." 176 U.S., at 24, 20 S.Ct., at 259. 12 Admittedly, the discussion here and in the other cases discussed below focused on the parens patriae question in the context of a suit brought in the original jurisdiction of this Court. There may indeed be special considerations that call for a limited exercise of our jurisdiction in such instances; these considerations may not apply to a similar suit brought in federal district court. 13 The Court also said, 324 U.S., at 450, 451-452, 65 S.Ct., at 722, 723-724: "It seems to us clear that under the authority of these cases Georgia may maintain this suit as parens patriae acting on behalf of her citizens though here, as in Georgia v. Tennessee Copper Co., [206 U.S., at] 237 [27 S.Ct., at 619], we treat the injury to the State as proprietor merely as a 'makeweight.' The original jurisdiction of this Court is one of the mighty instruments which the framers of the Constitution provided so that adequate machinery might be available for the peaceful settlement of disputes between States and between a State and citizens of another State. See Missouri v. Illinois, [180 U.S., at] 221-224 [21 S.Ct., at 336-337]; Virginia v. West Virginia, 246 U.S. 565, 599, 38 S.Ct. 400, 404, 59 L.Ed. 1272. Trade barriers, recriminations, intense commercial rivalries had plagued the colonies. The traditional methods available to a sovereign for the settlement of such disputes were diplomacy and war. Suit in this Court was provided as an alternative. Missouri v. Illinois, supra, 180 U.S., p. 241, 21 S.Ct., at 343; Georgia v. Tennessee Copper Co., supra, 206 U.S., p. 237, 27 S.Ct., p. 619. * * * * * "Oklahoma v. Atchison, T. & S. F. R. Co., [220 U.S. 277, 31 S.Ct. 434, 55 L.Ed. 465 (1911)], is not opposed to this view. In that case, the defendant railroad company had obtained a grant from Congress to locate and maintain a railway line through the Indian Territory out of which the State of Oklahoma was later formed. The federal act provided certain maximum transportation rates which the company might charge. Oklahoma sued to cancel the grant, to have the property granted decreed to be in the State of Oklahoma as cestui que trust, to enjoin the defendant from operating a railroad in the State, and to enjoin pendente lite the exaction of greater rates than the maximum rates specified. The Court construed the Act of Congress as subjecting the rates to federal control until the territory became a part of a State, at which time the rates became subject to state control. The Court held that our original jurisdiction could not be invoked by a State merely because its citizens were injured. We adhere to that decision. It does not control the present one. This is no attempt to utilize our original jurisdiction in substitution for the established methods of enforcing local law. This is not a suit in which a State is a mere nominal plaintiff, individual shippers being the real complainants. This is a suit in which Georgia asserts claims arising out of federal laws and the gravamen of which runs far beyond the claim of damage to individual shippers." 14 Obviously, a State might make use of "private bills" in order to use its legislative power to aid particular individuals. If the analogy spoken of above is to this form of legislative action, then the State remains merely a nominal party from the perspective of a federal court; it has failed to articulate any general interest, apart from that of the individual involved. 15 Although we have spoken throughout of a "State's" standing as parens patriae, we agree with the lower courts and the parties that the Commonwealth of Puerto Rico is similarly situated to a State in this respect: It has a claim to represent its quasi-sovereign interests in federal court at least as strong as that of any State. 16 A State does not have standing as parens patriae to bring an action against the Federal Government. Massachusetts v. Mellon, 262 U.S. 447, 485-486, 43 S.Ct. 597, 600-601, 67 L.Ed. 1078 (1923) ("While the State, under some circumstances, may sue in that capacity for the protection of its citizens (Missouri v. Illinois, 180 U.S. 208, 241, 21 S.Ct. 331, 343, 45 L.Ed. 497), it is no part of its duty or power to enforce their rights in respect of their relations with the Federal Government. In that field it is the United States, and not the State, which represents them as parens patriae"). Here, however, the Commonwealth is seeking to secure the federally created interests of its residents against private defendants. Indeed, the Secretary of Labor has represented that he has no objection to Puerto Rico's standing as parens patriae under these circumstances. See Brief for the Secretary of Labor as Amicus Curiae in Puerto Rico v. Bramkamp, No. 724, Docket 79-7777 (CA2). 1 Indeed, in Havens we held that interference with HOME's "ability to provide counseling and referral services," 455 U.S., at 379, 102 S.Ct., at 1124, provided it with standing to vindicate claims under the Fair Housing Act of 1968. In this case, the alleged violations of the Wagner-Peyser Act, 29 U.S.C. § 49 et seq., directly interfere with Puerto Rico's ability to perform the job referral service that it has undertaken as part of its sovereign responsibility to its citizens. 2 A private organization may also maintain a federal-court action on behalf of its members. E.g., N.A.A.C.P. v. Button, 371 U.S. 415, 428, 83 S.Ct. 328, 335, 9 L.Ed.2d 405 (1963); National Motor Freight Assn. v. United States, 372 U.S. 246, 83 S.Ct. 688, 9 L.Ed.2d 709 (1963) (per curiam ). See Hunt v. Washington State Apple Advertising Comm'n, 432 U.S. 333, 341-345, 97 S.Ct. 2434, 2440-2442, 53 L.Ed.2d 383 (1977).

89

458 U.S. 858 102 S.Ct. 3440 73 L.Ed.2d 1193 UNITED STATES, Petitionerv.Ricardo VALENZUELA-BERNAL. No. 81-450. Argued April 20, 1982. Decided July 2, 1982. Syllabus Respondent was indicted in Federal District Court for transporting one Romero-Morales in violation of 8 U.S.C. § 1324(a)(2), which prohibits the knowing transportation of an alien illegally in the United States who last entered the country within three years prior to the date of the transportation. Two other illegal aliens—who, with Romero-Morales, were passengers in the car being driven by respondent and were apprehended with respondent—were deported after an Assistant United States Attorney concluded that they possessed no evidence material to respondent's prosecution. Romero-Morales was detained to provide a nonhearsay basis for establishing that respondent had violated § 1324(a)(2). The District Court denied respondent's motion to dismiss the indictment on the asserted ground that the deportation of the other passengers deprived him of the opportunity to interview them to determine whether they could aid in his defense and thus violated his Fifth Amendment right to due process and his Sixth Amendment right to compulsory process for obtaining witnesses. Following a bench trial respondent was convicted, but the Court of Appeals reversed, holding that although a constitutional violation occurs only when "the alien's testimony could conceivably benefit the defendant," the "conceivable benefit" test was satisfied without requiring the defendant to explain what beneficial evidence would have been provided by the alien—whenever, as here, the deported alien was an eyewitness to the crime. Held: Respondent failed to establish a violation of the Fifth or Sixth Amendment. Pp. 863-874. (a) In cases like this, the Executive Branch's responsibility faithfully to execute Congress' immigration policy of prompt deportation of illegal aliens justifies deportation of illegal-alien witnesses upon the Executive's good-faith determination that they possess no evidence favorable to the defendant in a criminal prosecution. In addition to satisfying such policy, the prompt deportation of such witnesses is justified by practical considerations, including the financial and physical burdens imposed upon the Government in detaining alien eyewitnesses. Pp. 863-866. (b) Respondent cannot establish a violation of the Sixth Amendment, which guarantees a criminal defendant the right to compulsory process for obtaining witnesses "in his favor," merely by showing that deportation of the aliens deprived him of their testimony. He must at least make some plausible showing of how their testimony would have been both material and favorable to his defense. Cf. Washington v. Texas, 388 U.S. 14, 87 S.Ct. 1920, 18 L.Ed.2d 1019. While a relaxation of the specificity required in showing materiality may be supported by the fact that, because the witnesses were deported, neither respondent nor his attorney had an opportunity to interview the witnesses to determine what favorable information they possessed, this does not afford a basis for wholly dispensing with a showing of materiality. Cf. Roviaro v. United States, 353 U.S. 53, 77 S.Ct. 623, 1 L.Ed.2d 639. Moreover, respondent was present throughout the commission of the crime, and no one knew better than he what the deported witnesses said in his presence that might bear upon whether he knew that Romero-Morales was an illegal alien who had entered the country within the past three years. Pp. 867-871. (c) At least the same materiality requirement obtains with respect to a due process claim. In order to establish a denial of due process, the acts complained of must be of such quality as necessarily prevents a fair trial. Such an absence of fairness is not made out by the Government's deportation of the witnesses here unless there is some explanation of how their testimony would have been favorable and material. P. 872. (d) Sanctions against the Government are warranted for deportation of alien witnesses only if there is a reasonable likelihood that the testimony could have affected the judgment of the trier of fact. In this case, respondent made no effort to explain what material, favorable evidence the deported aliens would have provided for his defense. Pp. 872-874. 647 F.2d 72, reversed. Carter G. Phillips, Washington, D. C., for petitioner. Eugene G. Iredale, San Diego, Cal., for respondent. Justice REHNQUIST delivered the opinion of the Court. 1 Respondent, a citizen of Mexico, was indicted in the United States District Court for the Southern District of California for transporting one Romero-Morales in violation of 8 U.S.C. § 1324(a)(2). That section generally prohibits the knowing transportation of an alien illegally in the United States who last entered the country within three years prior to the date of the transportation.1 Respondent was found guilty after a bench trial, but his conviction was overturned by the United States Court of Appeals for the Ninth Circuit. That court held that the action of the Government in deporting two aliens other than Romero-Morales violated respondent's right under the Sixth Amendment to the United States Constitution to compulsory process, and his right under the Fifth Amendment to due process of law. We granted certiorari in order to review the Court of Appeals' application of these constitutional provisions to this case, 454 U.S. 963, 102 S.Ct. 501, 70 L.Ed.2d 377 (1981),2 and we now reverse. 2 * Respondent entered the United States illegally on March 23, 1980, and was taken by smugglers to a house in Escondido, Cal. Six days later, in exchange for his not having to pay the smugglers for bringing him across the border, respondent agreed to drive himself and five other passengers to Los Angeles. When the car which respondent was driving approached the Border Patrol checkpoint at Temecula, agents noticed the five passengers lying down inside the car and motioned to respondent to stop. Respondent accelerated through the checkpoint and was chased at high speed for approximately one mile before stopping the car and fleeing on foot along with the five passengers. Three of the passengers and respondent were apprehended by the Border Patrol agents. 3 Following their arrest, respondent and the other passengers were interviewed by criminal investigators. Respondent admitted his illegal entry into the country and explained his reason for not stopping at the checkpoint: "I was bringing the people [and] I already knew I had had it—too late—it was done." App. 27. The three passengers also admitted that they were illegally in the country and each identified respondent as the driver of the car. Id., at 66. An Assistant United States Attorney concluded that the passengers possessed no evidence material to the prosecution or defense of respondent for transporting illegal aliens, and two of the passengers were deported to Mexico. The third, Enrique Romero-Morales, was detained to provide a nonhearsay basis for establishing that respondent had transported an illegal alien in violation of 8 U.S.C. § 1324(a)(2). 4 Respondent moved in the District Court to dismiss the indictment, claiming that the Government's deportation of the two passengers other than Romero-Morales violated his Fifth Amendment right to due process of law and his Sixth Amendment right to compulsory process for obtaining favorable witnesses. He claimed that the deportation had deprived him of the opportunity to interview the two remaining passengers to determine whether they could aid in his defense. Although he had been in their presence throughout the allegedly criminal activity, respondent made no attempt to explain how the deported passengers could assist him in proving that he did not know that Romero-Morales was an illegal alien who had last entered the United States within the preceding three years. 5 At least one evidentiary hearing was held on respondent's motion, at which Romero-Morales testified that he had not spoken to respondent during the entire time that they were together. At the same hearing the Government offered, without obtaining agreement by respondent, to stipulate that none of the passengers in the car told respondent that they were in the United States illegally. The District Court denied respondent's motion and, following a bench trial on stipulated evidence, found respondent guilty as charged.3 6 The Court of Appeals reversed the conviction. The court relied upon the rule, first stated in United States v. Mendez-Rodriguez, 450 F.2d 1 (CA9 1971), that the Government violates the Fifth and Sixth Amendments when it deports alien witnesses before defense counsel has an opportunity to interview them. 647 F.2d 72, 73-75 (1981). Although it stated that a constitutional violation occurs only when "the alien's testimony could conceivably benefit the defendant," id., at 74, the court's application of the "conceivable benefit" test demonstrated that the test will be satisfied whenever the deported aliens were eyewitnesses to the crime.4 Respondent's failure to explain what beneficial evidence would have been provided by the two passengers was thus inapposite, for "the deported aliens were eyewitnesses to, and active participants in, the crime charged, thus establishing a strong possibility that they could have provided material and relevant information concerning the events constituting the crime." Id., at 75. Accordingly, the Court of Appeals held that respondent's motion to dismiss the indictment should have been granted by the District Court. II 7 We think that the decision of the Court of Appeals in this case, and some of the additional arguments made in support of it by respondent, misapprehend the varied nature of the duties assigned to the Executive Branch by Congress. The Constitution imposes on the President the duty to "take Care that the Laws be faithfully executed." U.S.Const., Art. II, § 3. One of the duties of the Executive Branch, and a vitally important one, is that of apprehending and obtaining the conviction of those who have violated criminal statutes of the United States. The prosecution of respondent is of course one example of the Executive's effort to discharge that responsibility. 8 But the Government is charged with a dual responsibility when confronted with incidents such as that which resulted in the apprehension of respondent. One or more of the persons in the car may have violated the criminal laws enacted by Congress; but some or all of the persons in the car may also be subject to deportation as provided by Congress. The Government may, therefore, find itself confronted with the obligation of prosecuting persons in the position of respondent on criminal charges, and at the same time obligated to deport other persons involved in the event in order to carry out the immigration policies that Congress has enacted. 9 The power to regulate immigration—an attribute of sovereignty essential to the preservation of any nation—has been entrusted by the Constitution to the political branches of the Federal Government. See Mathews v. Diaz, 426 U.S. 67, 81, 96 S.Ct. 1883, 1892, 48 L.Ed.2d 478 (1976). "The Court without exception has sustained Congress' 'plenary power to make rules for the admission of aliens.' " Kleindienst v. Mandel, 408 U.S. 753, 766, 92 S.Ct. 2576, 2583, 33 L.Ed.2d 683 (1972) (quoting Boutilier v. INS, 387 U.S. 118, 123, 87 S.Ct. 1563, 1566, 18 L.Ed.2d 661 (1967)). In exercising this power, Congress has adopted a policy of apprehending illegal aliens at or near the border and deporting them promptly. Border Patrol agents are authorized by statute to make warrantless arrests of aliens suspected of "attempting to enter the United States in violation of . . . law," 8 U.S.C. § 1357(a)(2), and are directed to examine them without "unnecessary delay" to determine whether "there is prima facie evidence establishing" their attempted illegal entry. 8 CFR § 287.3 (1982). Aliens against whom such evidence exists may be granted immediate voluntary departure from the country. See 8 U.S.C. § 1252(b); 8 CFR § 242.5(a)(2)(i) (1982). Thus, Congress has determined that prompt deportation, such as occurred in this case, constitutes the most effective method for curbing the enormous flow of illegal aliens across our southern border.5 10 In addition to satisfying immigration policy, the prompt deportation of alien witnesses who are determined by the Government to possess no material evidence relevant to a criminal trial is justified by several practical considerations. During fiscal year 1979, almost one-half of the more than 11,000 inmates incarcerated in federal facilities in the Southern District of California were material witnesses who had neither been charged with nor convicted of a criminal offense. App. 18. The average period of detention for such witnesses exceeded 5 days, and many were detained for more than 20 days. Id., at 20. The resulting overcrowded conditions forced the Government to house many detainees in federal facilities located outside the Southern District of California or in state-operated jails. Id., at 21-22; Brief for United States 19. Thus, the detention of alien eyewitnesses imposes substantial financial and physical burdens upon the Government, not to mention the human cost to potential witnesses who are incarcerated though charged with no crime. In addition, the rule adopted by the Court of Appeals significantly constrains the Government's prosecutorial discretion. As explained by the United States: 11 "Because of budget limitations and the unavailability of adequate detention facilities, it is simply impossible as a practical matter to prosecute many cases involving the transportation or harboring of large numbers of illegal aliens, where all the aliens must be incarcerated for a substantial period of time to avoid dismissal of the charges, even though the prosecution's case may be overwhelming. As a consequence, many valid and appropriate prosecutions are foregone." Id., at 21-22. 12 It simply will not do, therefore, to minimize the Government's dilemma in cases like this with statements such as "[t]he prosecution may not deny access to a witness by hiding him out. See Freeman v. State of Georgia, 599 F.2d 65 (5th Cir. 1979) (police detective concealed location of witness)." Brief for Respondent 35. Congress' immigration policy and the practical considerations discussed above demonstrate that the Government had good reason to deport respondent's passengers once it concluded that they possessed no evidence relevant to the prosecution or the defense of respondent's criminal charge. No onus, in the sense of "hiding out" or "concealing" witnesses, attached to the Government by reason of its discharge of the obligations imposed upon it by Congress; its exercise of these manifold responsibilities is not to be judged by standards which might be appropriate if the Government's only responsibility were to prosecute criminal offenses. III 13 Viewing the Government's conduct in this light, we turn to the evaluation of the Court of Appeals' "conceivable benefit" test. There seems to us to be little doubt that this test is a virtual "per se" rule which requires little if any showing on the part of the accused defendant that the testimony of the absent witness would have been either favorable or material. As we said with respect to a similar test—phrased in terms of information "that might affect the jury's verdict"—for determining when a prosecutor must disclose information to a criminal defendant: 14 "If everything that might influence a jury must be disclosed, the only way a prosecutor could discharge his constitutional duty would be to allow complete discovery of his files as a matter of routine practice." United States v. Agurs, 427 U.S. 97, 109, 96 S.Ct. 2392, 2400, 49 L.Ed.2d 342 (1976). 15 So it is with the "conceivable benefit" test. Given the vagaries of a typical jury trial, it would be a bold statement indeed to say that the testimony of any missing witness could not have "conceivably benefited" the defense. To us, the number of situations which will satisfy this test is limited only by the imaginations of judges or defense counsel.6 16 The only recent decision of this Court dealing with the right to compulsory process guaranteed by the Sixth Amendment suggests that more than the mere absence of testimony is necessary to establish a violation of the right. See Washington v. Texas, 388 U.S. 14, 87 S.Ct. 1920, 18 L.Ed.2d 1019 (1967). Indeed, the Sixth Amendment does not by its terms grant to a criminal defendant the right to secure the attendance and testimony of any and all witnesses: it guarantees him "compulsory process for obtaining witnesses in his favor." U.S.Const., Amdt. 6 (emphasis added). In Washington, this Court found a violation of this Clause of the Sixth Amendment when the defendant was arbitrarily deprived of "testimony [that] would have been relevant and material, and . . . vital to the defense." 388 U.S., at 16, 87 S.Ct., at 1922 (emphasis added). This language suggests that respondent cannot establish a violation of his constitutional right to compulsory process merely by showing that deportation of the passengers deprived him of their testimony. He must at least make some plausible showing of how their testimony would have been both material and favorable to his defense.7 17 When we turn from Washington to other cases in what might loosely be called the area of constitutionally guaranteed access to evidence, we find Washington' § intimation of a materiality requirement more than borne out. Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), held "that the suppression by the prosecution of evidence favorable to an accused upon request violates due process where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution." Id., at 87, 83 S.Ct., at 1196. This materiality requirement was emphasized inMoore v. Illinois, 408 U.S. 786, 92 S.Ct. 2562, 33 L.Ed.2d 706 (1972), where we stated that a defendant will prevail upon a Brady claim "where the evidence is favorable to the accused and is material either to guilt or to punishment." Id., at 794, 92 S.Ct., at 2568. And in United States v. Agurs, supra, we noted that "[a] fair analysis of the holding in Brady indicates that implicit in the requirement of materiality is a concern that the suppressed evidence might have affected the outcome of the trial." Id., at 104, 96 S.Ct., at 2398. We further explained: 18 "The proper standard of materiality must reflect our overriding concern with the justice of the finding of guilt. . . . This means that the omission must be evaluated in the context of the entire record. If there is no reasonable doubt about guilt whether or not the additional evidence is considered, there is no justification for a new trial. On the other hand, if the verdict is already of questionable validity, additional evidence of relatively minor importance might be sufficient to create a reasonable doubt." Id., at 112-113, 96 S.Ct., at 2401-2402 (footnotes omitted). 19 Similarly, when the Government has been responsible for delay resulting in a loss of evidence to the accused, we have recognized a constitutional violation only when loss of the evidence prejudiced the defense. In United States v. Marion, 404 U.S. 307, 92 S.Ct. 455, 30 L.Ed.2d 468 (1971), for example, the Court held that pre-indictment delay claims were governed by the Due Process Clause of the Fifth Amendment, not by the speedy-trial guarantee of the Sixth Amendment. Elaborating on the nature of the guarantee provided by the Due Process Clause in such cases, the Court emphasized the requirement of materiality: 20 "Nor have appellees adequately demonstrated that the pre-indictment delay by the Government violated the Due Process Clause. No actual prejudice to the conduct of the defense is alleged or proved, and there is no showing that the Government intentionally delayed to gain some tactical advantage over appellees or to harass them." Id., at 325, 92 S.Ct., at 466. 21 Five Terms later, in United States v. Lovasco, 431 U.S. 783, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977), we summarized this aspect of Marion: 22 "Thus Marion makes clear that proof of prejudice is generally a necessary but not sufficient element of a due process claim, and that the due process inquiry must consider the reasons for the delay as well as the prejudice to the accused." Id., at 790, 97 S.Ct., at 2048-2049. 23 The same "prejudice" requirement has been applied to cases of postindictment delay. In Barker v. Wingo, 407 U.S. 514, 92 S.Ct. 2182, 33 L.Ed.2d 101 (1972), the Court set forth several factors to be considered in determining whether an accused has been denied his Sixth Amendment right to a speedy trial by the Government's pretrial delay. One of the four factors identified by the Court, and a factor more fully discussed in United States v. MacDonald, 435 U.S. 850, 858-859, 98 S.Ct. 1547, 1551-52, 56 L.Ed.2d 18 (1978), was whether there had been any "prejudice to the defendant from the delay." Id., at 858, 98 S.Ct., at 1551. Although the Court recognized that prejudice may take the form of " 'oppressive pretrial incarceration' " or " 'anxiety and concern of the accused,' " the " 'most serious' " consideration, analogous to considerations in this case, was impairment of the ability to mount a defense. See ibid. (quoting Barker v. Wingo, supra, at 532, 92 S.Ct., at 2193). Thus, other interests protected by the Sixth Amendment look to the degree of prejudice incurred by a defendant as a result of governmental action or inaction. 24 The principal difference between these cases in related areas of the law and the present case is that respondent simply had no access to the witnesses who were deported after he was criminally charged. Respondent contends that requiring him to show materiality is unreasonable in light of the fact that neither he nor his attorney was afforded an opportunity to interview the deported witnesses to determine what favorable information they possessed. But while this difference may well support a relaxation of the specificity required in showing materiality, we do not think that it affords the basis for wholly dispensing with such a showing. 25 The closest case in point is Roviaro v. United States, 353 U.S. 53, 77 S.Ct. 623, 1 L.Ed.2d 639 (1957). While Roviaro was not decided on the basis of constitutional claims, its subsequent affirmation in McCray v. Illinois, 386 U.S. 300, 87 S.Ct. 1056, 18 L.Ed.2d 62 (1967), where both due process and confrontation claims were considered by the Court, suggests that Roviaro would not have been decided differently if those claims had actually been called to the Court's attention. 26 Roviaro deals with the obligation of the prosecution to disclose to the defendant the name of an informer-eyewitness, and was cast in terms of the traditional governmental privilege to refuse disclosure of such an identity. The Roviaro Court held that the informer's identity had to be disclosed, but only after it concluded that the informer's testimony would be highly relevant: 27 "This is a case where the Government's informer was the sole participant, other than the accused, in the transaction charged. The informer was the only witness in a position to amplify or contradict the testimony of government witnesses. Moreover, a government witness testified that [the informer] denied knowing petitioner or ever having seen him before. We conclude that, under these circumstances, the trial court committed prejudicial error in permitting the Government to withhold the identity of its undercover employee in the face of repeated demands by the accused for his disclosure." 353 U.S., at 64-65, 77 S.Ct., at 630. 28 "What Roviaro thus makes clear is that this Court was unwilling to impose any absolute rule requiring disclosure of an informer's identity," McCray v. Illinois, supra, at 311, 87 S.Ct., at 1062, despite the fact that criminal defendants otherwise have no access to such informers to determine what relevant information they possess. Roviaro supports the conclusion that while a defendant who has not had an opportunity to interview a witness may face a difficult task in making a showing of materiality, the task is not an impossible one. In such circumstances it is of course not possible to make any avowal of how a witness may testify. But the events to which a witness might testify, and the relevance of those events to the crime charged, may well demonstrate either the presence or absence of the required materiality. 29 In addition, it should be remembered that respondent was present throughout the commission of this crime. No one knows better than he what the deported witnesses actually said to him, or in his presence, that might bear upon whether he knew that Romero-Morales was an illegal alien who had entered the country within the past three years. And, in light of the actual charge made in the indictment, it was only the status of Romero-Morales which was relevant to the defense. Romero-Morales, of course, remained fully available for examination by the defendant and his attorney. We thus conclude that the respondent can establish no Sixth Amendment violation without making some plausible explanation of the assistance he would have received from the testimony of the deported witnesses.8 B 30 Having borrowed much of our reasoning with respect to the Compulsory Process Clause of the Sixth Amendment from cases involving the Due Process Clause of the Fifth Amendment, we have little difficulty holding that at least the same materiality requirement obtains with respect to a due process claim. Due process guarantees that a criminal defendant will be treated with "that fundamental fairness essential to the very concept of justice. In order to declare a denial of it we must find that the absence of that fairness fatally infected the trial; the acts complained of must be of such quality as necessarily prevents a fair trial." Lisenba v. California, 314 U.S. 219, 236, 62 S.Ct. 280, 290, 86 L.Ed. 166 (1941). In another setting, we recognized that Jencks Act violations, wherein the Government withholds evidence required by statute to be disclosed, rise to the level of due process violations only when they so infect the fairness of the trial as to make it "more a spectacle or trial by ordeal than a disciplined contest." United States v. Augenblick, 393 U.S. 348, 356, 89 S.Ct. 528, 534, 21 L.Ed.2d 537 (1969) (citations omitted). Such an absence of fairness is not made out by the Government's deportation of the witnesses in this case unless there is some explanation of how their testimony would have been favorable and material. See United States v. Lovasco, 431 U.S. 783, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977); United States v. Marion, 404 U.S. 307, 92 S.Ct. 455, 30 L.Ed.2d 468 (1971). IV 31 To summarize, the responsibility of the Executive Branch faithfully to execute the immigration policy adopted by Congress justifies the prompt deportation of illegal-alien witnesses upon the Executive's good-faith determination that they possess no evidence favorable to the defendant in a criminal prosecution. The mere fact that the Government deports such witnesses is not sufficient to establish a violation of the Compulsory Process Clause of the Sixth Amendment or the Due Process Clause of the Fifth Amendment. A violation of these provisions requires some showing that the evidence lost would be both material and favorable to the defense. 32 Because prompt deportation deprives the defendant of an opportunity to interview the witnesses to determine precisely what favorable evidence they possess, however, the defendant cannot be expected to render a detailed description of their lost testimony. But this does not, as the Court of Appeals concluded, relieve the defendant of the duty to make some showing of materiality. Sanctions may be imposed on the Government for deporting witnesses only if the criminal defendant makes a plausible showing that the testimony of the deported witnesses would have been material and favorable to his defense, in ways not merely cumulative to the testimony of available witnesses. In some cases such a showing may be based upon agreed facts, and will be in the nature of a legal argument rather than a submission of additional facts. In other cases the criminal defendant may advance additional facts, either consistent with facts already known to the court or accompanied by a reasonable explanation for their inconsistency with such facts, with a view to persuading the court that the testimony of a deported witness would have been material and favorable to his defense.9 Because in the latter situation the explanation of materiality is testimonial in nature, and constitutes evidence of the prejudice incurred as a result of the deportation, it should be verified by oath or affirmation of either the defendant or his attorney. See Fed.Rule Evid. 603; Fed.Rule Crim.Proc. 47. 33 As in other cases concerning the loss of material evidence, sanctions will be warranted for deportation of alien witnesses only if there is a reasonable likelihood that the testimony could have affected the judgment of the trier of fact. See Giglio v. United States, 405 U.S. 150, 154, 92 S.Ct. 763, 766, 31 L.Ed.2d 104 (1972). In making such a determination, courts should afford some leeway for the fact that the defendant necessarily proffers a description of the material evidence rather than the evidence itself. Because determinations of materiality are often best made in light of all of the evidence adduced at trial, judges may wish to defer ruling on motions until after the presentation of evidence.10 34 In this case the respondent made no effort to explain what material, favorable evidence the deported passengers would have provided for his defense. Under the principles set forth today, he therefore failed to establish a violation of the Fifth or Sixth Amendment, and the District Court did not err in denying his motion to dismiss the indictment. Accordingly, the judgment of the Court of Appeals is 35 Reversed. 36 Justice BLACKMUN, concurring in the judgment. 37 I concur in the judgment of the Court essentially for the reasons set forth by Judge Roney, in writing for a panel of the former Fifth Circuit, in United States v. Avila-Dominguez, 610 F.2d 1266, 1269-1270, cert. denied sub nom. Perez v. United States, 449 U.S. 887, 101 S.Ct. 242, 66 L.Ed.2d 113 (1980). At least a "plausible theory" of how the testimony of the deported witnesses would be helpful to the defense must be offered. None was advanced here; therefore, the motion to dismiss the indictment was properly denied by the District Court. 38 Justice O'CONNOR, concurring in the judgment. 39 "The right to offer the testimony of witnesses, and to compel their attendance, if necessary, is in plain terms the right to present a defense, the right to present the defendant's version of the facts as well as the prosecution's to the jury so it may decide where the truth lies." Washington v. Texas, 388 U.S. 14, 19, 87 S.Ct. 1920, 1923, 18 L.Ed.2d 1019 (1967). 40 In short, the right to compulsory process is essential to a fair trial. Today's decision, I fear, may not protect adequately the interests of the prosecution and the defense in a fair trial, and may encourage litigation over whether the defendant has made a "plausible showing that the testimony of the deported witnesses would have been material and favorable to his defense." Ante, at 873. A preferable approach would be to accommodate both the Government's interest in prompt deportation of illegal aliens and the defendant's need to interview alien witnesses in order to decide which of them can provide material evidence for the defense. Through a suitable standard, imposed on the federal courts under our supervisory powers, a practical accommodation can be reached without any increase in litigation. 41 * One cannot discount the importance of the Federal Government's role in the regulation of immigration.1 As the Court points out, Congress and the Immigration and Naturalization Service, the agency authorized to make such policy decisions, have decided that prompt deportation is the appropriate response to the tremendous influx of illegal aliens. Ante, at 864. The Court is also correct that the Federal Government has legitimate reasons for reducing the number of illegal aliens detained for possible use as material witnesses. Particularly because most of the detained aliens are never called to testify, we should be careful not to permit either needless human suffering or excessive burdens on the Federal Government. Under these circumstances, courts should be especially circumspect about interfering with congressional judgments. 42 Nevertheless, the constitutional obligation of the Executive to "take Care that the Laws be faithfully executed," U.S.Const., Art. II, § 3, including the immigration laws, does not lessen the importance of affording the defendant the "fundamental fairness" inherent in due process, Lisenba v. California, 314 U.S. 219, 236, 62 S.Ct. 280, 290, 86 L.Ed. 166 (1941). Moreover, the defendant's express right in the Sixth Amendment to compel the testimony of "witnesses in his favor," requires recognition of the importance, both to the individual defendant and to the integrity of the criminal justice system, of permitting the defendant the opportunity to interview eyewitnesses to the alleged crime. A governmental policy of deliberately putting potential defense witnesses beyond the reach of compulsory process is not easily reconciled with the spirit of the Compulsory Process Clause. II 43 The Court's solution to this apparent conflict between the Executive's duty to enforce the immigration laws and its duty not to impair the defendant's rights to due process and compulsory process is to permit the Government to deport potential alien witnesses, and to put the burden on the defendant of making a plausible showing that the deported aliens would have provided material and relevant evidence. The Court's approach thus permits the Government to make a practice of deporting alien witnesses immediately, taking only the risk that the defendant will be able to show that the deported witnesses, whom the defendant's counsel never will be able to interview, would have provided useful testimony. In effect, to the extent that the Government has conflicting obligations, the defendant is selected to carry the burden of their resolution. 44 As the Court poses the issue today, the only alternatives are either to (1) permit routine deportation of witnesses and require the defendant to make some showing of prejudice, or (2) delay deportation so that defense counsel can interview the potential witnesses, and provide for automatic dismissal of the indictment if the witnesses are deported. There is, however, another alternative that would avoid unduly burdening either the Government or the defendant. The Court could require that deportation of potential alien witnesses be delayed for a very brief interval to allow defense counsel, as well as the Government, to interview them. That approach is somewhat similar to the Ninth Circuit's practice, originally described in United States v. Mendez-Rodriguez, 450 F.2d 1 (1971). Under the holding in that case, illegal alien witnesses were held in custody for a short period, an average of five days, following the appointment of counsel. During that time, defense counsel had the opportunity to interview the witnesses and determine whether any of them might provide material and relevant evidence. Following the interviews, a Federal Magistrate held a hearing to determine whether any of the witnesses could provide material evidence, and ordered deportation of those aliens who could not provide such testimony. On those occasions when the Government nevertheless deported potential witnesses before the materiality hearing was held, the District Court determined whether the deported witnesses could have been of some "conceivable benefit" to the defendant. If the defendant met that standard, the court dismissed the indictment. 45 The principal difficulty with the Ninth Circuit's approach was, as the Court notes, ante, at 866-867, that it required virtually no evidence that the deported witness' testimony would have been material to the defense. Under the Ninth Circuit's formulation, the Government's deportation of an alien witness resulted in virtually an automatic dismissal of the indictment. 46 In adopting a standard requiring brief detention of potential alien witnesses, the Court need not take so extreme a position. In United States v. Avila-Dominguez, 610 F.2d 1266 (1980), for example, the Fifth Circuit followed the Ninth Circuit's rationale in concluding that a defendant's constitutional rights are violated if the Government deports an alien witness before the defendant has had an opportunity to interview him. The court nevertheless affirmed the defendant's conviction because he could not offer a "plausible theory" explaining how the witness' testimony would have been helpful to the defense. Id., at 1270. The court thus adopted a more stringent test than the Ninth Circuit's "conceivable benefit" test. 47 The standard I propose is an amalgam of the approaches used by the Fifth and Ninth Circuits.2 As a matter of course, the deportable aliens who are potential witnesses should be detained for a very brief period to afford Government and defense counsel the opportunity to interview them. If, within that period, the defendant requests that certain aliens not be deported, a federal magistrate should hold a hearing to determine whether deportation of any of the witnesses should be deferred until after trial. As evidenced by the statistics provided by the respondent, similar procedures in the Ninth Circuit have produced very little litigation. See Brief for Respondent 30. Of course, the Government could be expected to abide by such a rule, but in the occasional event that it deports alien witnesses without affording the defendant any opportunity to interview them, the defendant should not be entitled to an automatic dismissal of the indictment; nor should the defendant be expected to prove prejudice—after all, the Government has deported his potential witnesses. Instead, I agree with the Court that sanctions should be available against the Government if the defendant sets forth some plausible theory explaining how the deported witnesses would have provided material evidence that was not simply cumulative of evidence readily available to the defendant. III 48 In the case before us, the respondent made no plausible suggestion that the deported aliens possessed any material evidence that was not merely cumulative of other evidence. Under the standard I have proposed, the District Court properly denied the respondent's motion to dismiss the indictment. Accordingly, I concur in the judgment of the Court. 49 Justice BRENNAN, with whom Justice MARSHALL joins, dissenting. 50 Today's holding flaunts a transparent contradiction. On the one hand, the Court recognizes respondent's constitutional right, under the Compulsory Process Clause of the Sixth Amendment, to the production of all witnesses whose testimony would be relevant and material to his defense. Ante, at 867-869. But on the other hand, the Court holds that the Government may deport illegal-alien eyewitnesses to respondent's alleged crime immediately upon their apprehension, before respondent or his attorney have had any opportunity to interview them—thus depriving respondent of the surest and most obvious means by which he could establish the materiality and relevance of such witnesses' testimony. Ante, at 872-873. Truly, the Court giveth, and the Court taketh away. But surely a criminal defendant has a constitutional right to interview eyewitnesses to his alleged crime before they are whisked out of the country by his prosecutor. The Court's decision today makes a mockery of that right. Accordingly, I dissent. 51 The premise of the Court's holding is that "the responsibility of the Executive Branch faithfully to execute the immigration policy adopted by Congress justifies the prompt deportation of illegal-alien witnesses," ante, at 872; this governmental power is conditioned only upon the Executive's "good-faith determination" that those witnesses possess "no evidence favorable to the defendant in a criminal prosecution," ibid. The Court sets up this asserted "responsibility" of the Executive Branch as a counterweight to its responsibility for "apprehending and obtaining the conviction of those who have violated criminal statutes of the United States." Id., at 863. Thus the Court presents this case as involving a governmental "dilemma," id., at 865, in which the Executive Branch is caught between the conflicting demands of its "dual responsibility," id., at 864. This supposed "dilemma" is a pure figment of the Court's imagination, repudiated by our precedents and by common sense. 52 The Executive Branch has many responsibilities, any of which may conflict with its duty to enforce the federal criminal law. For example, the Executive Branch has an obvious and imperative obligation to preserve the national security. But when the Executive Branch chooses to prosecute a violation of federal law, it incurs a constitutional responsibility manifestly superior to its other duties: namely, the responsibility to ensure that the accused receives the due process of law. The Government simply cannot be heard to argue that the criminal defendant's rights may be infringed because of the Executive Branch's "other responsibilities": Given the vast and manifold character of those responsibilities, to accept such an argument would be to accede to the rapid evisceration of the constitutional rights of the accused. 53 This point is hardly a novel one. In Jencks v. United States, 353 U.S. 657, 77 S.Ct. 1007, 1 L.Ed.2d 1103 (1957), we noted that "the protection of vital national interests may militate against public disclosure of documents in the Government's possession." Id., at 670, 77 S.Ct., at 1014. But at the same time we noticed: 54 "[I]n criminal causes, '. . . the Government can invoke its evidentiary privileges only at the price of letting the defendant go free. The rationale of the criminal cases is that, since the Government which prosecutes an accused also has the duty to see that justice is done, it is unconscionable to allow it to undertake prosecution and then invoke its governmental privileges to deprive the accused of anything which might be material to his defense. . . .' " Id., at 671, 77 S.Ct., at 1014-15, quoting United States v. Reynolds, 345 U.S. 1, 12, 73 S.Ct. 528, 534, 97 L.Ed. 727 (1953). 55 We also quoted with approval from the opinion of the Court of Appeals for the Second Circuit in United States v. Andolschek, 142 F.2d 503 (1944), in which Judge Learned Hand said: 56 "While we must accept it as lawful for a department of the government to suppress documents, even when they will help determine controversies between third persons, we cannot agree that this should include their suppression in a criminal prosecution, founded upon those very dealings to which the documents relate, and whose criminality they will, or may, tend to exculpate. So far as they directly touch the criminal dealings, the prosecution necessarily ends any confidential character the documents may possess; it must be conducted in the open, and will lay bare their subject matter. The government must choose; either it must leave the transactions in the obscurity from which a trial will draw them, or it must expose them fully." Id., at 506.1 57 The principle affirmed in these precedents is directly applicable to this case. Of course, the Government has a responsibility to execute our national immigration policy. But that responsibility does not conflict in the smallest degree with the Government's "duty to see that justice is done" to the criminal defendant whom it has chosen to prosecute. If the Government wishes to pursue criminal remedies against the accused, then its other "responsibilities" must yield before the rights to which an accused is constitutionally entitled. 58 Of course, the Government's duty to enforce the immigration laws should not be deferred indefinitely. But no inordinate delay is necessary in cases such as the one before us. The Southern District of California long ago adopted a procedure to enforce the Mendez-Rodriguez doctrine announced by the Court of Appeals for the Ninth Circuit in 1971.2 The Southern District's procedure represents a practical and sensitive accommodation between a criminal defendant's constitutional rights under the Compulsory Process Clause and the Government's policy of prompt deportation of illegal aliens. Under that procedure, illegal-alien eyewitnesses are held in custody for a short period of time—about 10 days—after appointment of counsel for the criminal defendant. At the end of that period, the United States magistrate holds a material witness bail review hearing, pursuant to 18 U.S.C. § 3149. In the intervening time, counsel for the defendant may interview the witnesses, and determine whether they can provide testimony material to the defense. At the hearing, both prosecution and defense are required to show the materiality of each of the detained witnesses, or they are released and deported. Brief for Respondent 6-7; Brief for United States 13-14, 18. If this traditional Southern District procedure had been adhered to in the present case, the Government would have clearly discharged its constitutional obligation to afford respondent an opportunity to develop evidence bearing upon the materiality of the testimony of the witnesses to his alleged offense. In contrast, the Court permits the Government to adopt a wholly unilateral procedure that deprives respondent and future criminal defendants of any such opportunity. 59 The Court suggests that a criminal defendant should be able to "demonstrate either the presence or absence of the required materiality" even without having had an opportunity to interview the detained eyewitnesses. Ante, at 871. But this notion has been flatly rejected by our precedents. Roviaro v. United States, 353 U.S. 53, 77 S.Ct. 623, 1 L.Ed.2d 639 (1957), denied the Government's claimed privilege to withhold the identity of its informer, "John Doe," from the petitioner.3 Roviaro, like respondent in the present case, was "present throughout the commission of this crime." Ante, at 871; see 353 U.S., at 64 ("So far as [Roviaro] knew, he and John Doe were alone and unobserved during the crucial occurrence for which he was indicted"). But the Court in Roviaro refused to say, as the Court does today, that a criminal defendant "can establish no Sixth Amendment violation without making some plausible explanation of the assistance he would have received from the testimony" that he seeks. Ante, at 871. Rather, the Court in Roviaro required disclosure simply because John Doe's testimony "might have been helpful to the defense." 353 U.S., at 63-64, 77 S.Ct., at 629 (emphasis added). 60 "Doe had helped to set up the criminal occurrence and had played a prominent part in it. His testimony might have disclosed an entrapment. He might have thrown doubt upon petitioner's identity or the identity of the package [of heroin]. He was the only witness who might have testified to petitioner's possible lack of knowledge of the contents of the package that he 'transported' . . . to John Doe's car. The desirability of calling John Doe as a witness, or at least interviewing him in preparation for trial, was a matter for the accused rather than the Government to decide." Id., at 64, 77 S.Ct., at 629 (emphasis added). 61 Like Doe in Roviaro, the illegal aliens deported by the Government in the present case "played a prominent part" in respondent's alleged offense—if, indeed, they did not help to set it up without the knowledge of respondent. And they, like Doe, might have testified to respondent's "possible lack of knowledge" respecting essential elements of the crime charged against him.4 Under Roviaro, respondent, not the Government, was entitled to decide whether or not the illegal-alien eyewitnesses in this case could give testimony material and relevant to the defense. 62 I dissent. 1 Section 1324(a)(2) applies to "[a]ny person" who "transports, or moves, or attempts to transport or move," "any alien," "knowing that [the alien] is in the United States in violation of law, and knowing or having reasonable grounds to believe that his last entry into the United States occurred less than three years prior" to the transportation or attempted transportation with which the person is charged. 2 Other Courts of Appeals have adopted slight variations of the position held by the Court of Appeals for the Ninth Circuit. See, e.g., United States v. Armijo-Martinez, 669 F.2d 1131 (CA6 1982); United States v. Rose, 669 F.2d 23 (CA1 1982); United States v. Avila-Dominguez, 610 F.2d 1266 (CA5 1980); United States v. Calzada, 579 F.2d 1358 (CA7 1978). 3 The joint appendix contains excerpts of transcribed testimony from a hearing on June 2, 1980, at which the District Court heard arguments of counsel and the testimony of Romero-Morales. At the conclusion of this testimony, counsel for respondent proposed the highly unusual step of calling the Assistant United States Attorney as a witness. App. 45. The attorney testified at further proceedings held on June 12, 1980, and was interrogated, inter alia, about his understanding of various decisions of the Court of Appeals for the Ninth Circuit and about the Government's litigating strategy in these cases. Id., at 63-64. This procedure seems to us highly unusual, if not bizarre; ordinarily the litigating strategies of the United States Attorney are no more the subject of permissible inquiry by his opponent than would be the litigating strategies of the Public Defender by his opponent. 4 As the Court of Appeals explained: "The conceivable benefit in Mendez-Rodriguez stemmed from the fact that the deported aliens were eyewitnesses to, and active participants in, the crime charged, so that there was a strong possibility that they could have provided material and relevant evidence concerning the events constituting the crime. Conversely, where a missing deported alien was not an eyewitness to the offense, we have been unwilling to assume that the alien's testimony could conceivably benefit the defendant." 647 F.2d, at 74 (citation and footnotes omitted). As described by the Court of Appeals, the "conceivable benefit" test "impose[s] no requirement of government misconduct or negligence before dismissal of an indictment is warranted. Nor is a defendant required to show specific prejudice caused by the unavailability of the alien eyewitnesses." Ibid. (citation omitted). Other Courts of Appeals have recognized the Ninth Circuit rule as requiring no showing of prejudice, United States v. Calzada, 579 F.2d, at 1362, and as permitting dismissal of the indictment even when the " 'record is completely devoid of anything which would suggest that the testimony of any one, or more, of the deported persons would have been helpful' to the defendants." United States v. Avila-Dominguez, supra, 610 F.2d, at 1269-1270 (quoting United States v. Mendez-Rodriguez, 450 F.2d 1, 6 (CA9 1971) (Kilkenny, J., dissenting)). 5 As evidence of the effectiveness of Congress' policy and of the colossal problem presented by illegal entries from Mexico, the United States notes that approximately one million illegal aliens were detained by Border Patrol officials during each of the three years preceding 1981. Brief for United States 19; see U. S. Department of Justice, Internal Audit Report, U. S. Border Patrol Management of the Mexican Border 1, 6 (Jan. 1981). 6 See n. 4, supra. 7 That the Sixth Amendment does not guarantee criminal defendants the right to compel the attendance of any and all witnesses is reflected in the Federal Rules of Criminal Procedure. Rule 17(b) requires the Government to subpoena witnesses on behalf of indigent defendants, but only "upon a satisfactory showing . . . that the presence of the witness is necessary to an adequate defense." See also Isaacs v. United States, 159 U.S. 487, 489, 16 S.Ct. 51, 52, 40 L.Ed. 229 (1895); Crumpton v. United States, 138 U.S. 361, 364-365, 11 S.Ct. 355, 356-57, 34 L.Ed. 958 (1891). 8 Respondent's knowledge of the truth distinguishes this case from United States v. Burr, 25 F.Cas. 187 (No. 14,694) (C.C.D.Va.1807), a case cited by respondent in support of his argument that it is unreasonable to require him to explain the relevance of the missing testimony. In Burr, Chief Justice Marshall found it unreasonable to require Aaron Burr to explain the relevancy of General Wilkinson's letter to President Jefferson, upon which the President's allegations of treason were based, precisely because Burr had never read the letter and was unaware of its contents. In this case, respondent observed the passengers, heard their comments, and is fully aware of the ways in which they influenced his knowledge about the status of Romero-Morales. 9 In adopting this standard, we express no opinion on the showing which a criminal defendant must make in order to obtain compulsory process for securing the attendance at his criminal trial of witnesses within the United States. 10 The counsel of United States v. Agurs, 427 U.S. 97, 112-113, 96 S.Ct. 2392, 2402, 49 L.Ed.2d 342 (1976), is helpful here: "[T]he omission must be evaluated in the context of the entire record. If there is no reasonable doubt about guilt whether or not additional evidence is considered, there is no justification for a new trial. On the other hand, if the verdict is already of questionable validity, additional evidence of relatively minor importance might be sufficient to create a reasonable doubt." 1 Article I, § 8, cl. 4, states that Congress shall have the power "To establish an uniform Rule of Naturalization." See Mathews v. Diaz, 426 U.S. 67, 81, 96 S.Ct. 1883, 1892, 48 L.Ed.2d 478 (1976) ("For reasons long recognized as valid, the responsibility for regulating the relationship between the United States and our alien visitors has been committed to the political branches of the Federal Government"); Galvan v. Press, 347 U.S. 522, 531, 74 S.Ct. 737, 743, 98 L.Ed. 911 (1954) ("that the formulation of [immigration] policies is entrusted exclusively to Congress has become about as firmly imbedded in the legislative and judicial tissues of our body politic as any aspect of our government"). 2 This Court has not hesitated to use its supervisory power over federal courts to set standards to ensure the fair administration of justice. For example, in McCarthy v. United States, 394 U.S. 459, 468-472, 89 S.Ct. 1166, 1172-73, 22 L.Ed.2d 418 (1969), this Court, under its supervisory power, held that when a district court does not comply fully with Federal Rule of Criminal Procedure 11 in accepting a guilty plea, the plea must be set aside and the case remanded for the defendant to enter a new plea. The Court expressly rejected the rule, adopted by some Circuits, of holding a hearing to determine whether the defendant had entered his plea voluntarily with an understanding of the charge. See also Marshall v. United States, 360 U.S. 310, 313, 79 S.Ct. 1171, 1173, 3 L.Ed.2d 1250 (1959) (using the Court's "supervisory power to formulate and apply proper standards for enforcement of the criminal law in the federal courts" in setting aside a criminal conviction because several jurors had read inadmissible news accounts of the defendant's past activities). 1 See United States v. Beekman, 155 F.2d 580, 583-584 (CA2 1946). See also United States v. Burr, 25 F.Cas. 187, 191 (No. 14,694) (CC Va. 1807) ("If this might be likened to a civil case, the law is express on the subject. It is that either party may require the other to produce books or writings in their possession or power, which contain evidence pertinent to the issue. . . . [I]f the order be disobeyed by the plaintiff, judgment as in the case of a nonsuit may be entered against him"); United States v. Nixon, 418 U.S. 683, 709, 94 S.Ct. 3090, 3108, 41 L.Ed.2d 1039 (1974) ("To ensure that justice is done, it is imperative to the function of courts that compulsory process be available for the production of evidence needed either by the prosecution or the defense"). 2 See United States v. Mendez-Rodriguez, 450 F.2d 1 (CA9 1971). 3 Roviaro represented an exercise of our supervisory jurisdiction. See McCray v. Illinois, 386 U.S. 300, 309, 87 S.Ct. 1056, 1061, 18 L.Ed.2d 62 (1967). But as the Court concedes, ante, at 870, Roviaro would not have been decided differently if the Due Process and Confrontation Clause claims implicit in that case had been brought to the fore. 4 In order to obtain a conviction under 8 U.S.C. § 1324(a)(2), quoted ante, at 860, n. 1, the Government was required to show (1) that respondent transported an alien within the United States, (2) that the alien had not been lawfully admitted or was not lawfully entitled to enter, (3) that this was known to respondent, (4) that respondent knew that the alien's last entry was within three years, and (5) that respondent acted willfully in furtherance of the alien's violation of the law. United States v. Gonzalez-Hernandez, 534 F.2d 1353, 1354 (CA9 1976). Since the third and fourth elements of this statutory requirement bear upon respondent's state of mind, it is plain that the illegal aliens whom respondent was transporting might very well have been able to testify to his lack of knowledge on these critical points.

01

458 U.S. 941 102 S.Ct. 3456 73 L.Ed.2d 1254 Joy SPORHASE and Delmer Moss, etc., Appellantsv.NEBRASKA, ex rel. Paul L. DOUGLAS, Attorney General. No. 81-613. Argued March 30, 1982. Decided July 2, 1982. Syllabus A Nebraska statute provides that any person who intends to withdraw ground water from any well located in the State and transport it for use in an adjoining State must obtain a permit from the Nebraska Department of Water Resources. If the Director of Water Resources finds that such withdrawal is reasonable, not contrary to the conservation and use of ground water, and not otherwise detrimental to the public welfare, he will grant the permit if the State in which the water is to be used grants reciprocal rights to withdraw and transport ground water from that State for use in Nebraska. Appellants jointly own contiguous tracts of land in Nebraska and Colorado, on which a well on the Nebraska tract pumps ground water for irrigation of both the Nebraska and Colorado tracts, but they never applied for the permit required by the statute. Appellee brought an action in a Nebraska state court to enjoin appellants from transferring the water across the border without a permit. Rejecting the defense that the statute imposed an undue burden on interstate commerce, the trial court granted the injunction. The Nebraska Supreme Court affirmed. Held: 1. Ground water is an article of commerce and therefore subject to congressional regulation. Pp. 945-954. (a) Although appellee's claimed greater ownership interest in ground water than in certain other natural resources may not be irrelevant to Commerce Clause analysis, it does not remove Nebraska ground water from such scrutiny, since appellee's argument is still based on the legal fiction of state ownership. Pp. 945-952. (b) The States' interests in conserving and preserving scarce water resources in the arid Western States clearly have an interstate dimension. The agricultural markets supplied by irrigated farms provide the archtypical example of commerce among the States for which the Framers of the Constitution intended to authorize federal regulation. Here, the multistate character of the aquifer underlying appellants' tracts of land, as well as parts of Texas, New Mexico, Oklahoma, and Kansas, demonstrates that there is a significant federal interest in conservation as well as in fair allocation of diminishing water resources. Pp. 952-954. 2. The reciprocity requirement of the Nebraska statute violates the Commerce Clause as imposing an impermissible burden on interstate commerce. While the first three conditions set forth in the statute for granting a permit—that the withdrawal of the ground water be reasonable, not contrary to the conservation and use of ground water, and not otherwise detrimental to the public welfare do not on their faces impermissibly burden interstate commerce, the reciprocity provision operates as an explicit barrier to commerce between Nebraska and its adjoining States. Nebraska therefore has the initial burden of demonstrating a close fit between the reciprocity requirement and its asserted local purpose. Such requirement, when superimposed on the first three restrictions, fails to clear this initial hurdle, since there is no evidence that it is narrowly tailored to the conservation and preservation rationale. Thus, it does not survive the "strictest scrutiny" reserved for facially discriminatory legislation. Pp. 954-958. 3. Congress has not granted the States permission to engage in ground water regulation that would otherwise be impermissible. Although there are 37 federal statutes and a number of interstate compacts demonstrating Congress' deference to state water law, they do not indicate that Congress wished to remove federal constitutional restraints on such state law. Neither the fact that Congress has chosen not to create a federal water law to govern water rights involved in federal water projects nor the fact that Congress has been willing to let the States settle their differences over water rights through mutual agreement, constitutes persuasive evidence that Congress consented to the unilateral imposition of unreasonable burdens on commerce. Pp. 958-960. 208 Neb. 703, 305 N.W.2d 614, reversed and remanded. Richard A. Dudden, Ogallala, Neb., for appellants. George Roderic Anderson, Lincoln, Neb., for appellee. Justice STEVENS delivered the opinion of the Court. 1 Appellants challenge the constitutionality of a Nebraska statutory restriction on the withdrawal of ground water from any well within Nebraska intended for use in an adjoining State. The challenge presents three questions under the Commerce Clause:1 (1) whether ground water is an article of commerce and therefore subject to congressional regulation; (2) whether the Nebraska restriction on the interstate transfer of ground water imposes an impermissible burden on commerce; and (3) whether Congress has granted the States permission to engage in ground water regulation that otherwise would be impermissible. 2 Appellants jointly own contiguous tracts of land in Chase County, Nebraska, and Phillips County, Colorado. A well physically located on the Nebraska tract pumps ground water for irrigation of both the Nebraska tract and the Colorado tract. Previous owners of the land registered the well with the State of Nebraska in 1971, but neither they nor the present owners applied for the permit required by Neb.Rev.Stat. § 46-613.01 (1978). That section provides: 3 "Any person, firm, city, village, municipal corporation or any other entity intending to withdraw ground water from any well or pit located in the State of Nebraska and transport it for use in an adjoining state shall apply to the Department of Water Resources for a permit to do so. If the Director of Water Resources finds that the withdrawal of the ground water requested is reasonable, is not contrary to the conservation and use of ground water, and is not otherwise detrimental to the public welfare, he shall grant the permit if the state in which the water is to be used grants reciprocal rights to withdraw and transport ground water from that state for use in the State of Nebraska." 4 Appellee brought this action to enjoin appellants from transferring the water across the border without a permit.2 The trial court rejected the defense that the statute imposed an undue burden on interstate commerce and granted the injunction. The Nebraska Supreme Court affirmed. 208 Neb. 703, 305 N.W.2d 614 (1981). It held that, under Nebraska law, ground water is not "a market item freely transferable for value among private parties, and therefore [is] not an article of commerce." Id., at 705, 305 N.W.2d, at 616.3 The Chief Justice, while agreeing that the statutory criteria governing the transfer of water to an adjoining State did not violate the Commerce Clause, dissented on the narrow ground that appellee violated both the Federal and Nebraska Constitutions by attempting "to absolutely prohibit the transfer of water, without regard to its need or availability, based solely upon the acts of another state over which citizens of this state have no control." Id., at 713, 305 N.W.2d, at 620. 5 * In holding that ground water is not an article of commerce, the Nebraska Supreme Court and appellee cite as controlling precedent Hudson County Water Co. v. McCarter, 209 U.S. 349, 28 S.Ct. 529, 52 L.Ed. 828 (1908). In that case a New Jersey statute prohibited the interstate transfer of any surface water located within the State.4 The Hudson County Water Co. nevertheless contracted with New York City to supply one of its boroughs with water from the Passaic River in New Jersey. The State Attorney General sought from the New Jersey courts an injunction against fulfillment of the contract. Over the water company's objections that the statute impaired the obligation of contract, took property without just compensation, interfered with interstate commerce, denied New York citizens the privileges afforded New Jersey citizens, and denied New York citizens the equal protection of the laws, the injunction was granted. This Court, in an opinion by Justice Holmes, affirmed. 6 Most of the Court's opinion addresses the just compensation claim. Justice Holmes refused to ground the Court's holding, as did the New Jersey state courts,5 on "the more or less attenuated residuum of title that the State may be said to possess." Id., at 355, 28 S.Ct., at 530. For the statute was justified as a regulatory measure that, on balance, did not amount to a taking of property that required just compensation. Putting aside the "problems of irrigation," the State's interest in preserving its waters was well within its police power.6 That interest was not dependent on any demonstration that the State's water resources were inadequate for present or future use. The State "finds itself in possession of what all admit to be a great public good, and what it has it may keep and give no one a reason for its will." Id., at 357, 28 S.Ct., at 531. 7 Having disposed of the just compensation claim, Justice Holmes turned very briefly to the other constitutional challenges. In one paragraph, he rejected the Contract Clause claim. In the remaining paragraph of the opinion, he rejected all the other defenses. His treatment of the Commerce Clause challenge consists of three sentences: "A man cannot acquire a right to property by his desire to use it in commerce among the States. Neither can he enlarge his otherwise limited and qualified right to the same end. The case is covered in this respect by Geer v. Connecticut, 161 U.S. 519 [, 16 S.Ct. 600, 40 L.Ed. 793] [ (1896) ]." Ibid. 8 While appellee relies upon Hudson County, appellants rest on our summary affirmance of a three-judge District Court judgment in City of Altus v. Carr, 255 F.Supp. 828 (WD Tex.), summarily aff'd, 385 U.S. 35, 87 S.Ct. 240, 17 L.Ed.2d 34 (1966). The city of Altus is located near the southern border of Oklahoma. Large population increases rendered inadequate its source of municipal water. It consequently obtained from the owners of land in an adjoining Texas county the contractual right to pump the ground water underlying that land and to transport it across the border. The Texas Legislature thereafter enacted a statute that forbade the interstate exportation of ground water without the approval of that body.7 The city filed suit in Federal District Court, claiming that the statute violated the Commerce Clause. 9 The city relied upon West v. Kansas Natural Gas Co., 221 U.S. 229, 31 S.Ct. 564, 55 L.Ed. 716 (1911), which invalidated an Oklahoma statute that prevented the interstate transfer of natural gas produced within the State,8 and Pennsylvania v. West Virginia, 262 U.S. 553, 43 S.Ct. 658, 67 L.Ed. 1117 (1923), which invalidated a West Virginia statute that accorded a preference to the citizens of that State in the purchase of natural gas produced therein.9 The Texas Attorney General defended the statute on two grounds. First, he asserted that its purpose was to conserve and protect the State's water resources by regulating the withdrawal of ground water. The District Court rejected that defense because similar conservation claims had met defeat in West v. Kansas Natural Gas Co., supra, and Pennsylvania v. West Virginia, supra.10 Second, the State argued that the statute regulated ground water and that ground water is not an article of commerce, citing Geer v. Connecticut, 161 U.S. 519, 16 S.Ct. 600, 40 L.Ed. 793 (1896), and Hudson County Water Co. v. McCarter, 209 U.S. 349, 28 S.Ct. 529, 52 L.Ed. 828 (1908). The court rejected this argument since the statute directly regulated the interstate transportation of water that had been pumped from the ground, and under Texas law such water was an article of commerce. The court then had little difficulty in concluding that the statute imposed an impermissible burden on interstate commerce.11 10 In summarily affirming the District Court in City of Altus, we did not necessarily adopt the court's reasoning. Our affirmance indicates only our agreement with the result reached by the District Court. Metromedia, Inc. v. San Diego, 453 U.S. 490, 499, 101 S.Ct. 2882, 2888, 69 L.Ed.2d 800 (1981). That result is not necessarily inconsistent with the Nebraska Supreme Court's holding in this case. For Texas law differs significantly from Nebraska law regarding the rights of a surface owner to ground water that he has withdrawn. According to the District Court in City of Altus, the "rule in Texas was that an owner of land could use all of the percolating water he could capture from the wells on his land for whatever beneficial purposes he needed it, on or off the land, and could likewise sell it to others for use on or off the land and outside the basin where produced, just as he could sell any other species of property." 255 F.Supp., at 833, n. 8. Since ground water, once withdrawn, may be freely bought and sold in States that follow this rule, in those States ground water is appropriately regarded as an article of commerce. In Nebraska the surface owner has no comparable interest in ground water. As explained by the Nebraska Supreme Court, " 'the owner of land is entitled to appropriate subterranean waters found under his land, but he cannot extract and appropriate them in excess of a reasonable and beneficial use upon the land which he owns, especially if such use is injurious to others who have substantial rights to the waters, and if the natural underground supply is insufficient for all owners, each is entitled to a reasonable proportion of the whole.' " 208 Neb., at 705, 305 N.W.2d, at 617 (quoting Olson v. City of Wahoo, 124 Neb. 802, 811, 248 N.W. 304, 308 (1933)). 11 City of Altus, however, is inconsistent with Hudson County. For in the latter case the Court found Geer v. Connecticut, supra, to be controlling on the Commerce Clause issue. Geer, which sustained a Connecticut ban on the interstate transportation of game birds captured in that State, was premised on the theory that the State owned its wild animals and therefore was free to qualify any ownership interest it might recognize in the persons who capture them. One such restriction is a prohibition against interstate transfer of the captured animals. This theory of public ownership was advanced as a defense in City of Altus. The State argued that it owned all subterranean water and therefore could recognize ownership in the surface owner who withdraws the water, but restrict that ownership to use of the water within the State. That theory, upon which the Commerce Clause issue in Hudson County was decided, was rejected by the District Court in City of Altus.12 In expressly overruling Geer three years ago, this Court traced the demise of the public ownership theory and definitively recast it as " 'but a fiction expressive in legal shorthand of the importance to its people that a State have power to preserve and regulate the exploitation of an important resource.' " Hughes v. Oklahoma, 441 U.S. 322, 334, 99 S.Ct. 1727, 1735, 60 L.Ed.2d 250 (1979) (quoting Toomer v. Witsell, 334 U.S. 385, 402, 68 S.Ct. 1156, 1165, 92 L.Ed. 1460 (1948)). See also Baldwin v. Montana Fish and Game Comm'n, 436 U.S. 371, 384-387, 98 S.Ct. 1852, 1860-1862, 56 L.Ed.2d 354 (1978); Douglas v. Seacoast Products, Inc., 431 U.S. 265, 284-285, 97 S.Ct. 1740, 1751-1752, 52 L.Ed.2d 304 (1977). In Hughes the Court found the State's interests insufficient to sustain a ban on the interstate transfer of natural minnows seined from waters within the State. 12 Appellee insists, however, that Nebraska water is distinguishable from other natural resources. The surface owner who withdraws Nebraska ground water enjoys a lesser ownership interest in the water than the captor of game birds in Connecticut or minnows in Oklahoma or ground water in Texas, for in Geer, Hughes, and City of Altus the States permitted intrastate trade in the natural resources once they were captured. Although appellee's greater ownership interest may not be irrelevant to Commerce Clause analysis, it does not absolutely remove Nebraska ground water from such scrutiny. For appellee's argument is still based on the legal fiction of state ownership. The fiction is illustrated by municipal water supply arrangements pursuant to which ground water is withdrawn from rural areas and transferred to urban areas. Such arrangements are permitted in Nebraska, see Metropolitan Utilities District v. Merritt Beach Co., 179 Neb. 783, 140 N.W.2d 626 (1966), but the Nebraska Supreme Court distinguished them on the ground that the transferor was only permitted to charge as a price for the water his costs of distribution and not the value of the water itself. 208 Neb., at 708, 305 N.W.2d, at 618. Unless demand is greater than supply, however, this reasoning does not distinguish minnows, the price of which presumably is derived from the costs of seining and of transporting the catch to market. Even in cases of shortage, in which the seller of the natural resource can demand a price that exceeds his costs, the State's rate structure that requires the price to be cost-justified is economically comparable to price regulation. A State's power to regulate prices or rates has never been thought to depend on public ownership of the controlled commodity. It would be anomalous if federal power to regulate economic transactions in natural resources depended on the characterization of the payment as compensation for distribution services, on the one hand, or as the price of goods, on the other. Cf. In re Rahrer, 140 U.S. 545, 558, 11 S.Ct. 865, 868, 35 L.Ed. 572 (1891). 13 The second asserted distinction is that water, unlike other natural resources, is essential for human survival. Appellee, and the amici curiae that are vitally interested in conserving and preserving scarce water resources in the arid Western States, have convincingly demonstrated the desirability of state and local management of ground water.13 But the States' interests clearly have an interstate dimension. Although water is indeed essential for human survival, studies indicate that over 80% of our water supplies is used for agricultural purposes.14 The agricultural markets supplied by irrigated farms are worldwide. They provide the archtypical example of commerce among the several States for which the Framers of our Constitution intended to authorize federal regulation. The multistate character of the Ogallala aquifer—underlying appellants' tracts of land in Colorado and Nebraska, as well as parts of Texas, New Mexico, Oklahoma, and Kansas15—confirms the view that there is a significant federal interest in conservation as well as in fair allocation of this diminishing resource. Cf. Arizona v. California, 373 U.S. 546, 83 S.Ct. 1468, 10 L.Ed.2d 542 (1963). 14 The Western States' interests, and their asserted superior competence, in conserving and preserving scarce water resources are not irrelevant in the Commerce Clause inquiry. Nor is appellee's claim to public ownership without significance. Like Congress' deference to state water law, see infra, at 958-960, these factors inform the determination whether the burdens on commerce imposed by state ground water regulation are reasonable or unreasonable. But appellee's claim that Nebraska ground water is not an article of commerce goes too far: it would not only exempt Nebraska ground water regulation from burden-on-commerce analysis, it would also curtail the affirmative power of Congress to implement its own policies concerning such regulation. SeePhiladelphia v. New Jersey, 437 U.S. 617, 621-623, 98 S.Ct. 2531, 2534-2535, 57 L.Ed.2d 475 (1978). If Congress chooses to legislate in this area under its commerce power, its regulation need not be more limited in Nebraska than in Texas and States with similar property laws. Ground water overdraft is a national problem and Congress has the power to deal with it on that scale. II 15 Our conclusion that water is an article of commerce raises, but does not answer, the question whether the Nebraska statute is unconstitutional. For the existence of unexercised federal regulatory power does not foreclose state regulation of its water resources, of the uses of water within the State, or indeed, of interstate commerce in water. Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U.S. 761, 766-767, 65 S.Ct. 1515, 1518-1519, 89 L.Ed. 1915 (1945); United States v. South-Eastern Underwriters Assn., 322 U.S. 533, 548-549, 64 S.Ct. 1162, 1171, 88 L.Ed. 1440 (1944); Cooley v. Board of Wardens, 12 How. 299, 319, 13 L.Ed. 996 (1852). Determining the validity of state statutes affecting interstate commerce requires a more careful inquiry: 16 "Where the statute regulates evenhandedly to effectuate a legitimate local public interest, and its effects on interstate commerce are only incidental, it will be upheld unless the burden imposed on such commerce is clearly excessive in relation to the putative local benefits. If a legitimate local purpose is found, then the question becomes one of degree. And the extent of the burden that will be tolerated will of course depend on the nature of the local interest involved, and on whether it could be promoted as well with a lesser impact on interstate activities." Pike v. Bruce Church, Inc., 397 U.S. 137, 142, 90 S.Ct. 844, 847, 25 L.Ed.2d 174 (1970) (citation omitted). 17 The only purpose that appellee advances for § 46-613.01 is to conserve and preserve diminishing sources of ground water. The purpose is unquestionably legitimate and highly important,16 and the other aspects of Nebraska's ground water regulation demonstrate that it is genuine. Appellants' land in Nebraska is located within the boundaries of the Upper Republican Ground Water Control Area, which was designated as such by the Director of the Nebraska Department of Water Resources based upon a determination that there is "[a]n inadequate ground water supply to meet present or reasonably foreseeable needs for beneficial use of such water supply." Neb.Rev.Stat. § 46-658(1) (Supp. 1981); see App. 56-60. Pursuant to § 46-666(1), the Upper Republican Natural Resources District has promulgated special rules and regulations governing ground water withdrawal and use. See App. 61-82. The rules and regulations define as "critical" those townships in the control area in which the annual decline of the ground water table exceeds a fixed percentage; appellants' Nebraska tract is located within a critical township. The rules and regulations require the installation of flow meters on every well within the control area, specify the amount of water per acre that may be used for irrigation, and set the spacing that is required between wells. They also strictly limit the intrastate transfer of ground water: transfers are only permitted between lands controlled by the same groundwater user, and all transfers must be approved by the District Board of Directors. Id., at 68-69. 18 The State's interest in conservation and preservation of ground water is advanced by the first three conditions in § 46-613.01 for the withdrawal of water for an interstate transfer. Those requirements are "that the withdrawal of the ground water requested is reasonable, is not contrary to the conservation and use of ground water, and is not otherwise detrimental to the public welfare." Although Commerce Clause concerns are implicated by the fact that § 46-613.01 applies to interstate transfers but not to intrastate transfers, there are legitimate reasons for the special treatment accorded requests to transport ground water across state lines. Obviously, a State that imposes severe withdrawal and use restrictions on its own citizens is not discriminating against interstate commerce when it seeks to prevent the uncontrolled transfer of water out of the State. An exemption for interstate transfers would be inconsistent with the ideal of evenhandedness in regulation. At least in the area in which appellants' Nebraska tract is located, the first three standards of § 46-613.01 may well be no more strict in application than the limitations upon intrastate transfers imposed by the Upper Republican Natural Resources District. 19 Moreover, in the absence of a contrary view expressed by Congress, we are reluctant to condemn as unreasonable, measures taken by a State to conserve and preserve for its own citizens this vital resource in times of severe shortage. Our reluctance stems from the "confluence of [several] realities." Hicklin v. Orbeck, 437 U.S. 518, 534, 98 S.Ct. 2482, 2491, 57 L.Ed.2d 397 (1978). First, a State's power to regulate the use of water in times and places of shortage for the purpose of protecting the health of its citizens—and not simply the health of its economy—is at the core of its police power. For Commerce Clause purposes, we have long recognized a difference between economic protectionism, on the one hand, and health and safety regulation, on the other. See H. P. Hood & Sons v. Du Mond, 336 U.S. 525, 533, 69 S.Ct. 657, 662, 93 L.Ed. 865 (1949). Second, the legal expectation that under certain circumstances each State may restrict water within its borders has been fostered over the years not only by our equitable apportionment decrees, see, e.g., Wyoming v. Colorado, 353 U.S. 953, 77 S.Ct. 865, 1 L.Ed.2d 906 (1957), but also by the negotiation and enforcement of interstate compacts. Our law therefore has recognized the relevance of state boundaries in the allocation of scarce water resources. Third, although appellee's claim to public ownership of Nebraska ground water cannot justify a total denial of federal regulatory power, it may support a limited preference for its own citizens in the utilization of the resource. See Hicklin v. Orbeck, supra, 437 U.S., at 533-534, 98 S.Ct., at 2491. In this regard, it is relevant that appellee's claim is logically more substantial than claims to public ownership of other natural resources. See supra, at 950-951. Finally, given appellee's conservation efforts, the continuing availability of ground water in Nebraska is not simply happenstance; the natural resource has some indicia of a good publicly produced and owned in which a State may favor its own citizens in times of shortage. See Reeves, Inc. v. Stake, 447 U.S. 429, 100 S.Ct. 2271, 65 L.Ed.2d 244 (1980); cf. Philadelphia v. New Jersey, 437 U.S., at 627-628, and n. 6, 98 S.Ct., at 2537, and n. 6; Baldwin v. Montana Fish and Game Comm'n, 436 U.S. 371, 98 S.Ct. 1852, 56 L.Ed.2d 354 (1978). A facial examination of the first three conditions set forth in § 46-613.01 does not, therefore, indicate that they impermissibly burden interstate commerce. Appellants, indeed, seem to concede their reasonableness. 20 Appellants, however, do challenge the requirement that "the state in which the water is to be used grants reciprocal rights to withdraw and transport ground water from that state for use in the State of Nebraska"—the reciprocity provision that troubled the Chief Justice of the Nebraska Supreme Court. Because Colorado forbids the exportation of its ground water,17 the reciprocity provision operates as an explicit barrier to commerce between the two States. The State therefore bears the initial burden of demonstrating a close fit between the reciprocity requirement and its asserted local purpose. Hughes v. Oklahoma, 441 U.S., at 336, 99 S.Ct., at 1736; Dean Milk Co. v. City of Madison, 340 U.S. 349, 354, 71 S.Ct. 295, 297, 95 L.Ed. 329 (1951). 21 The reciprocity requirement fails to clear this initial hurdle. For there is no evidence that this restriction is narrowly tailored to the conservation and preservation rationale. Even though the supply of water in a particular well may be abundant, or perhaps even excessive, and even though the most beneficial use of that water might be in another State, such water may not be shipped into a neighboring State that does not permit its water to be used in Nebraska. If it could be shown that the State as a whole suffers a water shortage, that the intrastate transportation of water from areas of abundance to areas of shortage is feasible regardless of distance, and that the importation of water from adjoining States would roughly compensate for any exportation to those States, then the conservation and preservation purpose might be credibly advanced for the reciprocity provision. A demonstrably arid State conceivably might be able to marshal evidence to establish a close means-end relationship between even a total ban on the exportation of water and a purpose to conserve and preserve water. Appellee, however, does not claim that such evidence exists. We therefore are not persuaded that the reciprocity requirement—when superimposed on the first three restrictions in the statute significantly advances the State's legitimate conservation and preservation interest; it surely is not narrowly tailored to serve that purpose. The reciprocity requirement does not survive the "strictest scrutiny" reserved for facially discriminatory legislation. Hughes v. Oklahoma, supra, 441 U.S., at 337, 99 S.Ct., at 1736.18 III 22 Appellee's suggestion that Congress has authorized the States to impose otherwise impermissible burdens on interstate commerce in ground water is not well founded. The suggestion is based on 37 statutes in which Congress has deferred to state water law, and on a number of interstate compacts dealing with water that have been approved by Congress. 23 Abstracts of the relevant sections of the 37 statutes relied upon by appellee were submitted in connection with the Hearings on S. 1275 before the Subcommittee on Irrigation and Reclamation of the Senate Committee on Interior and Insular Affairs, 88th Cong., 2d Sess., 302-310 (1964). Appellee refers the Court to that submission but only discusses § 8 of the Reclamation Act of 1902, 32 Stat. 390. That section, it turns out, is typical of the other 36 statutes. It contains two parts. The first provides that "nothing in this Act shall be construed as affecting or intended to affect or to in any way interfere with the laws of any State or Territory relating to the control, appropriation, use, or distribution of water used in irrigation." Such language defines the extent of the federal legislation's pre-emptive effect on state law. New England Power Co. v. New Hampshire, 455 U.S. 331, 341, 102 S.Ct. 1096, 1102, 71 L.Ed.2d 188 (1982); Lewis v. BT Investment Managers, Inc., 447 U.S. 27, 49, 100 S.Ct. 2009, 2022, 64 L.Ed.2d 702 (1980). The second part provides that "the Secretary of the Interior, in carrying out the provisions of this Act, shall proceed in conformity with such laws." Such language mandates that questions of water rights that arise in relation to a federal project are to be determined in accordance with state law. See California v. United States, 438 U.S. 645, 98 S.Ct. 2985, 57 L.Ed.2d 1018 (1978). 24 The interstate compacts to which appellee refers are agreements among States regarding rights to surface water. See The Council of State Governments, Interstate Compacts and Agencies 25-29, 31-32 (1979). Appellee emphasizes a compact between Nebraska and Colorado involving water rights to the South Platte River, see 44 Stat. (part 2) 195, and a compact among Nebraska, Colorado, and Kansas involving water rights to the Republican River, see 57 Stat. 86. 25 Although the 37 statutes and the interstate compacts demonstrate Congress' deference to state water law,19 they do not indicate that Congress wished to remove federal constitutional constraints on such state laws. The negative implications of the Commerce Clause, like the mandates of the Fourteenth Amendment, are ingredients of the valid state law to which Congress has deferred. Neither the fact that Congress has chosen not to create a federal water law to govern water rights involved in federal projects, nor the fact that Congress has been willing to let the States settle their differences over water rights through mutual agreement,20 constitutes persuasive evidence that Congress consented to the unilateral imposition of unreasonable burdens on commerce. In the instances in which we have found such consent, Congress' " 'intent and policy' to sustain state legislation from attack under the Commerce Clause" was " 'expressly stated.' " New England Power Co. v. New Hampshire, supra, 455 U.S., at 343, 102 S.Ct., at 1102 (quoting Prudential Ins. Co. v. Benjamin, 328 U.S. 408, 427, 66 S.Ct. 1142, 1153, 90 L.Ed. 1343 (1946)).21 Cf. Merrion v. Jicarilla Apache Tribe, 455 U.S. 130, 155, n. 21, 102 S.Ct. 894, 911, n. 21, 71 L.Ed.2d 21 (1982). 26 The reciprocity requirement of Neb.Rev.Stat. § 46-613.01 (1978) violates the Commerce Clause. We leave to the state courts the question whether the invalid portion is severable. The judgment of the Nebraska Supreme Court is reversed, and the case is remanded for proceedings not inconsistent with this opinion. 27 It is so ordered. Justice REHNQUIST, with whom Justice O'CONNOR joins, dissenting. 28 The issue presented by this case, and the only issue, is whether the existence of the Commerce Clause of the United States Constitution by itself, in the absence of any action by Congress, invalidates some or all of Neb.Rev.Stat. § 46-613.01 (1978), which relates to groundwater. But instead of confining its opinion to this question, the Court first quite gratuitously undertakes to answer the question of whether the authority of Congress to regulate interstate commerce, conferred by the same provision of the Constitution, would enable it to legislate with respect to ground-water overdraft in some or all of the States. 29 That these two questions are quite distinct leaves no room for doubt. Congress may regulate not only the stream of commerce itself, but also activities which affect interstate commerce, including wholly intrastate activities. See, e.g., Kirschbaum Co. v. Walling, 316 U.S. 517, 62 S.Ct. 1116, 86 L.Ed. 1638 (1942); United States v. Darby, 312 U.S. 100, 61 S.Ct. 451, 85 L.Ed. 609 (1941); Houston & Texas R. Co. v. United States, 234 U.S. 342, 34 S.Ct. 833, 58 L.Ed. 1341 (1914). The activity upon which the regulatory effect of the congressional statute falls in many of these cases does not directly involve articles of commerce at all. For example, in Kirschbaum, the employees were engaged in the operation and maintenance of a loft building in which large quantities of goods for interstate commerce were produced; no one contended that these employees themselves, or the work which they actually performed, dealt with articles of commerce. Nonetheless, the provisions of the Fair Labor Standards Act were applied to them because Congress extended the terms of the Act not only to those who were "engaged in commerce" but also to those who were engaged "in the production of goods for commerce." 316 U.S., at 522, 62 S.Ct., at 1119. 30 Thus, the authority of Congress under the power to regulate interstate commerce may reach a good deal further than the mere negative impact of the Commerce Clause in the absence of any action by Congress. Upon a showing that ground-water overdraft has a substantial economic effect on interstate commerce, for example, Congress arguably could regulate ground-water overdraft, even if ground water is not an "article of commerce" itself. See, e.g., Hodel v. Virginia Surface Mining & Reclamation Assn., 452 U.S. 264, 281-283, 101 S.Ct. 2352, 2363, 69 L.Ed.2d 1 (1981); id., at 310-313, 101 S.Ct., at 2390-2392 (REHNQUIST, J., concurring in judgment); Wickard v. Filburn, 317 U.S. 111, 63 S.Ct. 82, 87 L.Ed. 122 (1942). It is therefore wholly unnecessary to decide whether Congress could regulate ground-water overdraft in order to decide this case; since Congress has not undertaken such a regulation, I would leave the determination of its validity until such time as it is necessary to decide that question. 31 The question actually involved in this case is whether Neb.Rev.Stat. § 46-613.01 (1978) runs afoul of the unexercised authority of Congress to regulate interstate commerce. While the Court apparently agrees that our equitable apportionment decrees in cases such as Wyoming v. Colorado, 353 U.S. 953, 77 S.Ct. 865, 1 L.Ed.2d 906 (1957), and the execution and approval of interstate compacts apportioning water have given rise to "the legal expectation that under certain circumstances each State may restrict water within its borders," ante, at 956, it insists on an elaborate balancing process in which the State's "interest" is weighed under traditional Commerce Clause analysis. 32 I think that in more than one of our cases in which a State has invoked our original jurisdiction, the unsoundness of the Court's approach is manifest. For example, in Georgia v. Tennessee Copper Co., 206 U.S. 230, 237, 27 S.Ct. 618, 619, 51 L.Ed. 1038 (1907), the Court said: 33 "This is a suit by a State for an injury to it in its capacity of quasi -sovereign. In that capacity the State has an interest independent of and behind the titles of its citizens, in all the earth and air within its domain. It has the last word as to whether its mountains shall be stripped of their forests and its inhabitants shall breathe pure air." Five years earlier, in Kansas v. Colorado, 185 U.S. 125, 142, 145-146, 22 S.Ct. 552, 558, 559, 46 L.Ed. 838 (1902), the Court had made clear that a State's quasi-sovereign interest in the flow of surface and subterranean water within its borders was of the same magnitude as its interest in pure air or healthy forests. 34 In my view, these cases appropriately recognize the traditional authority of a State over resources within its boundaries which are essential not only to the well-being but often to the very lives of its citizens. In the exercise of this authority, a State may so regulate a natural resource as to preclude that resource from attaining the status of an "article of commerce" for the purposes of the negative impact of the Commerce Clause. It is difficult, if not impossible, to conclude that "commerce" exists in an item that cannot be reduced to possession under state law and in which the State recognizes only a usufructuary right. "Commerce" cannot exist in a natural resource that cannot be sold, rented, traded, or transferred, but only used. 35 Of course, a State may not discriminate against interstate commerce when it regulates even such a resource. If the State allows indiscriminate intrastate commercial dealings in a particular resource, it may have a difficult task proving that an outright prohibition on interstate commercial dealings is not such a discrimination. I had thought that this was the basis for this Court's decisions in Hughes v. Oklahoma, 441 U.S. 322, 99 S.Ct. 1727, 60 L.Ed.2d 250 (1979), Pennsylvania v. West Virginia, 262 U.S. 553, 43 S.Ct. 658, 67 L.Ed. 1117 (1923), and West v. Kansas Natural Gas Co., 221 U.S. 229, 31 S.Ct. 564, 55 L.Ed. 716 (1911). In each case, the State permitted a natural resource to be reduced to private possession, permitted an intrastate market to exist in that resource, and either barred interstate commerce entirely or granted its residents a commercial preference.1 36 By contrast, Nebraska so regulates ground water that it cannot be said that the State permits any "commerce," intrastate or interstate, to exist in this natural resource. As with almost all of the Western States, Nebraska does not recognize an absolute ownership interest in ground water, but grants landowners only a right to use ground water on the land from which it has been extracted. Moreover, the landowner's right to use ground water is limited. Nebraska landowners may not extract ground water "in excess of a reasonable and beneficial use upon the land in which he owns, especially if such use is injurious to others who have substantial rights to the waters, and if the natural underground supply is insufficient for all owners, each is entitled to a reasonable proportion of the whole." Olson v. City of Wahoo, 124 Neb. 802, 811, 248 N.W. 304, 308 (1933). With the exception of municipal water systems, Nebraska forbids any transportation of ground water off the land owned or controlled by the person who has appropriated the water from its subterranean source. 208 Neb. 703, 710, 305 N.W.2d 614, 619 (1981). See App. 68-69. 37 Nebraska places additional restrictions on ground water users within certain areas, such as the portion of appellant's land situated in Nebraska, where the shortage of ground water is determined to be critical. Water users in appellants' district are permitted only to irrigate the acreage irrigated in 1977, or the average number of acres irrigated between 1972 and 1976, whichever is greater, and must obtain permission from the water district's board before any additional acreage may be placed under irrigation. The amount of ground water that may be extracted is strictly limited on an acre-inch-per-irrigated-acre basis. There are also detailed regulations as to the spacing of wells and the use and operation of flow meters. Id., at 71-82. 38 Since Nebraska recognizes only a limited right to use ground water on land owned by the appropriator, it cannot be said that "commerce" in ground water exists as far as Nebraska is concerned. Therefore, it cannot be said that Neb.Rev.Stat. § 46-613.01 (1978) either discriminates against, or "burdens," interstate commerce. Section 46-613.01 is simply a regulation of the landowner's right to use ground water extracted from lands he owns within Nebraska.2 Unlike the Court, I cannot agree that Nebraska's limitation upon a landowner's right to extract water from his land situated in Nebraska for his own use on land he owns in an adjoining State runs afoul of Congress' unexercised authority to regulate interstate commerce.3 1 Article I, § 8, cl. 3, of the United States Constitution provides: "The Congress shall have Power . . .. To regulate Commerce with foreign Nations, and among the several States, and with the Indian Tribes." For general explanations of Commerce Clause analysis, see, e.g., Western & Southern Life Insurance Co. v. State Board of Equalization, 451 U.S. 648, 652-653, 101 S.Ct. 2070, 2074-2075, 68 L.Ed.2d 514 (1981); Southern Pacific Co. v. Arizona ex rel. Sullivan, 325 U.S. 761, 766-770, 65 S.Ct. 1515, 1518-1521, 89 L.Ed. 1915 (1945). 2 Because of the reciprocity requirement of § 46-613.01, appellants would not have been granted a permit had they applied for one. Their failure to submit an application therefore does not deprive them of standing to challenge the legality of the reciprocity requirement. Cf. Larson v. Valente, 456 U.S. 228, 102 S.Ct. 1673, 72 L.Ed.2d 33 (1982). 3 The Nebraska Supreme Court also rejected appellants' equal protection and due process challenges. Appellants renew those challenges before this Court, but we need not reach these issues in light of our disposition of the Commerce Clause claim. 4 The Court quoted the statute: " 'It shall be unlawful for any person or corporation to transport or carry, through pipes, conduits, ditches or canals, the waters of any fresh water lake, pond, brook, creek, river or stream of this State into any other State, for use therein.' " 209 U.S., at 353, 28 S.Ct., at 530. 5 "The Courts below assumed or decided and we shall assume that the defendant represents the rights of a riparian proprietor, and on the other hand, that it represents no special chartered powers that give it greater rights than those. On these assumptions the Court of Errors and Appeals pointed out that a riparian proprietor has no right to divert waters for more than a reasonable distance from the body of the stream or for other than the well-known ordinary uses, and that for any purpose anywhere he is narrowly limited in amount. It went on to infer that his only right in the body of the stream is to have the flow continue, and that there is a residuum of public ownership in the State. It reinforced the State's rights by the State's title to the bed of the stream where flowed by the tide, and concluded from the foregoing and other considerations that, as against the rights of riparian owners merely as such, the State was warranted in prohibiting the acquisition of the title to water on a larger scale." Id., at 354, 28 S.Ct., at 530. 6 "The problems of irrigation have no place here. Leaving them on one side, it appears to us that few public interests are more obvious, indisputable and independent of particular theory than the interest of the public of a State to maintain the rivers that are wholly within it substantially undiminished, except by such drafts upon them as the guardian of the public welfare may permit for the purpose of turning them to a more perfect use." Id., at 356, 28 S.Ct., at 531. 7 The District Court quoted the statute: " 'No one shall withdraw water from any underground source in this State for use in any other state by drilling a well in Texas and transporting the water outside the boundaries of the State unless the same be specifically authorized by an Act of the Texas Legislature and thereafter as approved by it.' " 255 F.Supp., at 830. 8 Justice Holmes, the author of the Court's opinion in Hudson County, noted his dissent. See 221 U.S., at 262, 31 S.Ct., at 574. 9 Justice Holmes dissented, expressing the view that the Court's decision was inconsistent with Hudson County. See 262 U.S., at 603, 43 S.Ct., at 667. 10 The District Court opinion, 255 F.Supp., at 839, included these quotations from the two cases: "The statute of Oklahoma recognizes [natural gas] to be a subject of intrastate commerce, but seeks to prohibit it from being the subject of interstate commerce, and this is the purpose of its conservation. In other words, the purpose of its conservation is in a sense commercial—the business welfare of the State, as coal might be, or timber. Both of these products might be limited in amount, and the same consideration of the public welfare which would confine gas to the use of the inhabitants of a State would confine them to the inhabitants of the State. If the States have such power a singular situation might result. Pennsylvania might keep its coal, the Northwest its timber, the mining States their minerals. And why may not the products of the field be brought within the principle?" West v. Kansas Natural Gas Co., 221 U.S., at 255, 31 S.Ct., at 571. "Another consideration advanced to the same end is that natural gas is a natural product of the State and has become a necessity therein, that the supply is waning and no longer sufficient to satisfy local needs and be used abroad, and that the act is therefore a legitimate measure of conservation in the interest of the people of the State. If the situation be as stated, it affords no ground for the assumption by the State of the power to regulate interstate commerce, which is what the act attempts to do. That power is lodged elsewhere." Pennsylvania v. West Virginia, 262 U.S., at 598, 43 S.Ct., at 665. 11 "Considering the statute in question only with regard to whether it regulates the transportation and use of water after it has been withdrawn from a well and becomes personal property, such statute constitutes an unreasonable burden upon and interference with interstate commerce. Moreover, on the facts of this case it appear[s] to us that [the Texas statute] does not have for its purpose, nor does it operate to conserve water resources of the State of Texas except in the sense that it does so for her own benefit to the detriment of her sister States as in the case of West v. Kansas Natural Gas Co. In the name of conservation, the statute seeks to prohibit interstate shipments of water while indulging in the substantial discrimination of permitting the unrestricted intrastate production and transportation of water between points within the State, no matter how distant; for example, from Wilbarger County to El Paso County, Texas. Obviously, the statute had little relation to the cause of conservation." 255 F.Supp., at 839-840. 12 "This statute, however, seeks to prohibit the production of underground water for the purpose of transporting same in interstate commerce, and has the effect of prohibiting the interstate transportation of such water after it has become personal property. Whether a statute by its phraseology prohibits the interstate transportation of an article of commerce after it has become the personal property of someone as in the Pennsylvania and West cases, or prohibits the withdrawal of such substance where the intent is to transport such in interstate commerce, the result upon interstate commerce is the same. In both situations, the purpose and intent of the statute and the end result thereof is to prohibit the interstate transportation of an article of commerce." Id., at 840. 13 In California v. United States, 438 U.S. 645, 648, 98 S.Ct. 2985, 2987, 57 L.Ed.2d 1018 (1978), we explained some of the circumstances that support a general policy of local water management under differing legal systems: "The very vastness of our territory as a Nation, the different times at which it was acquired and settled, and the varying physiographic and climate regimes which obtain in its different parts have all but necessitated the recognition of legal distinctions corresponding to these differences. Those who first set foot in North America from ships sailing the tidal estuaries of Virginia did not confront the same problems as those who sailed flat boats down the Ohio River in search of new sites to farm. Those who cleared the forests in the old Northwest Territory faced totally different physiographic problems from those who built sod huts on the Great Plains. The final expansion of our Nation in the 19th century into the arid lands beyond the hundredth meridian of longitude, which had been shown on early maps as the 'Great American Desert,' brought the participants in that expansion face to face with the necessity for irrigation in a way that no previous territorial expansion had." 14 Soil Conservation Service, U. S. Dept. of Agriculture, America's Soil and Water: Conditions and Trends 21 (1980). 15 Comptroller General, Report to Congress, Ground Water Overdrafting Must Be Controlled 7-8 (1980). 16 See Cities Service Gas Co. v. Peerless Oil & Gas Co., 340 U.S. 179, 188, 71 S.Ct. 215, 220, 95 L.Ed. 190 (1950) ("Insofar as conservation is concerned, the national interest and the interest of producing states may well tend to coincide"). 17 Colorado Rev.Stat. § 37-90-136 (1973) provides as follows: "For the purpose of aiding and preserving unto the state of Colorado and all its citizens the use of all ground waters of this state, whether tributary or nontributary to a natural stream, which waters are necessary for the health and prosperity of all the citizens of the state of Colorado, and for the growth, maintenance, and general welfare of the state, it is unlawful for any person to divert, carry, or transport by ditches, canals, pipelines, conduits, or any other manner any of the ground waters of this state, as said waters are in this section defined, into any other state for use therein." 18 The reciprocity requirement cannot, of course, be justified as a response to another State's unreasonable burden on commerce. Great Atlantic & Pacific Tea Co. v. Cottrell, 424 U.S. 366, 379-381, 96 S.Ct. 923, 931-932, 47 L.Ed.2d 55 (1976). 19 "The history of the relationship between the Federal Government and the States in the reclamation of the arid lands of the Western States is both long and involved, but through it runs the consistent thread of purposeful and continued deference to state water law by Congress." California v. United States, 438 U.S., at 653, 98 S.Ct., at 2989. 20 Similarly, this Court has encouraged States to resolve their water disputes through interstate compacts rather than by equitable apportionment adjudication. See Colorado v. Kansas, 320 U.S. 383, 392, 64 S.Ct. 176, 180, 88 L.Ed. 116 (1943). 21 See, e.g., Prudential Ins. Co. v. Benjamin, 328 U.S. 408, 66 S.Ct. 1142, 90 L.Ed. 1343 (1946) (McCarran-Ferguson Act, 59 Stat. 33); International Shoe Co. v. Washington, 326 U.S. 310, 66 S.Ct. 154, 90 L.Ed. 95 (1945) (§ 1606(a) of the Internal Revenue Code, 53 Stat. 1391); Whitfield v. Ohio, 297 U.S. 431, 56 S.Ct. 532, 80 L.Ed. 778 (1936) (Hawes-Cooper Act, 45 Stat. 1084); In re Rahrer, 140 U.S. 545, 11 S.Ct. 865, 35 L.Ed. 572 (1891) (Wilson Act, 26 Stat. 313). 1 Similarly, in City of Altus v. Carr, 255 F.Supp. 828 (WD Tex.), summarily aff'd, 385 U.S. 35, 87 S.Ct. 240, 17 L.Ed.2d 34 (1966), Texas placed no restrictions upon the use or the intrastate sale of ground water. The "rule in Texas was that an owner of land could use all of the percolating water he could capture from the wells on his land for whatever beneficial purposes he needed it, on or off the land, and could likewise sell it to others for use on or off the land and outside the basin where produced, just as he could sell any other species of property." 255 F.Supp., at 833, n. 8. Texas' absolute ownership rule is an anomaly among the Western States. See 5 R. Clark, Waters and Water Rights § 441 (1972 and 1978 Supp.). In Nebraska, as in most of the Western States, ground water is not treated as "any other species of property." 2 Unlike several other Western States, Nebraska does not entirely forbid ground water extracted in Nebraska to be used in other States. See Brief for City of El Paso as Amicus Curiae 2, n. 3. As noted by the Court, Nebraska merely places conditions on such a use of the State's ground water. A permit must be obtained from the Nebraska Department of Water Resources. If the requested withdrawal of ground water is determined to be "reasonable, . . . not contrary to the conservation and use of ground water, and . . . not otherwise detrimental to the public welfare," a permit will be issued so long as the "state in which the water is to be used grants reciprocal rights to withdraw and transport ground water from that state for use in the State of Nebraska." Neb.Rev.Stat. § 46-613.01 (1978). 3 The Court today invalidates only the reciprocity provision in § 46-613.01. Ante, at 957-958. Appellants, however, have never applied for the permit required by the Nebraska statute. I see nothing in the Court's opinion that would preclude the Nebraska Department of Water Resources from prohibiting appellants from transporting ground water into the Colorado portion of their land until they obtain the permit required by the statute. I also see nothing in the Court's opinion that would preclude the Department of Water Resources from denying appellants a permit because of a failure to satisfy the remaining conditions in the statute.

910

458 U.S. 886 102 S.Ct. 3409 73 L.Ed.2d 1215 NATIONAL ASSOCIATION FOR the ADVANCEMENT OF COLORED PEOPLE, et al., Petitionersv.CLAIBORNE HARDWARE COMPANY et al. No. 81-202. Argued March 3, 1982. Decided July 2, 1982. Rehearing Denied Oct. 4, 1982. See 459 U.S. 898, 103 S.Ct. 199. Syllabus In 1966, a boycott of white merchants in Claiborne County, Miss., was launched at a meeting of a local branch of the National Association for the Advancement of Colored People (NAACP) attended by several hundred black persons. The purpose of the boycott was to secure compliance by both civic and business leaders with a lengthy list of demands for equality and racial justice. The boycott was largely supported by speeches encouraging nonparticipants to join the common cause and by nonviolent picketing, but some acts and threats of violence did occur. In 1969, respondent white merchants filed suit in Mississippi Chancery Court for injunctive relief and damages against petitioners (the NAACP, the Mississippi Action for Progress, and a number of individuals who had participated in the boycott, including Charles Evers, the field secretary of the NAACP in Mississippi and a principal organizer of the boycott). Holding petitioners jointly and severally liable for all of respondents' lost earnings during a 7-year period from 1966 to the end of 1972 on three separate conspiracy theories, including the tort of malicious interference with respondents' businesses, the Chancery Court imposed damages liability and issued a permanent injunction. The Mississippi Supreme Court rejected two theories of liability but upheld the imposition of liability on the basis of the common-law tort theory. Based on evidence that fear of reprisals caused some black citizens to withhold their patronage from respondents' businesses, the court held that the entire boycott was unlawful and affirmed petitioners' liability for all damages "resulting from the boycott" on the ground that petitioners had agreed to use force, violence, and "threats" to effectuate the boycott. Held: 1. The nonviolent elements of petitioners' activities are entitled to the protection of the First Amendment. Pp. 907-915. (a) Through exercise of their First Amendment rights of speech, assembly, association, and petition, rather than through riot or revolution, petitioners sought to bring about political, social, and economic change. Pp. 907-912. (b) While States have broad power to regulate economic activities, there is no comparable right to prohibit peaceful political activity such as that found in the boycott in this case. Pp. 912-915. 2. Petitioners are not liable in damages for the consequences of their nonviolent, protected activity. Pp. 915-920. (a) While the State legitimately may impose damages for the consequences of violent conduct, it may not award compensation for the consequences of nonviolent, protected activity; only those losses proximately caused by the unlawful conduct may be recovered. Pp. 915-918. (b) Similarly, the First Amendment restricts the ability of the State to impose liability on an individual solely because of his association with another. Civil liability may not be imposed merely because an individual belonged to a group, some members of which committed acts of violence. For liability to be imposed by reason of association alone, it is necessary to establish that the group itself possessed unlawful goals and that the individual held a specific intent to further those illegal aims. Pp. 918-920. 3. The award for all damages "resulting from the boycott" cannot be sustained, where the record discloses that all of the respondents' business losses were not proximately caused by violence or threats of violence. Pp. 920-932. (a) To the extent that the Mississippi Supreme Court's judgment rests on the ground that "many" black citizens were "intimidated" by "threats" of "social ostracism, vilification, and traduction," it is flatly inconsistent with the First Amendment. The court's ambiguous findings are inadequate to assure the "precision of regulation" demanded by that Amendment. Pp. 920-924. (b) Regular attendance and participation at the meetings of the Claiborne County Branch of the NAACP is an insufficient predicate on which to impose liability on the individual petitioners. Nor can liability be imposed on such individuals simply because they were either "store watchers" who stood outside the boycotted merchants' stores to record the names of black citizens who patronized the stores or members of a special group of boycott "enforcers." Pp. 924-926. (c) For similar reasons, the judgment against Evers cannot be separately justified, nor can liability be imposed upon him on the basis of speeches that he made, because those speeches did not incite violence or specifically authorize the use of violence. His acts, being insufficient to impose liability on him, may not be used to impose liability on the NAACP, his principal. Moreover, there is no finding that Evers or any other NAACP member had either actual or apparent authority from the NAACP to commit acts of violence or to threaten violent conduct or that the NAACP ratified unlawful conduct. To impose liability on the NAACP without such a finding would impermissibly burden the rights of political association that are protected by the First Amendment. Pp. 926-932. 393 So.2d 1290 (Miss.), reversed and remanded. Lloyd N. Cutler, Washington, D. C., for petitioners. Grover Rees, III, Austin, Tex., for respondents. Justice STEVENS delivered the opinion of the Court. 1 The term "concerted action" encompasses unlawful conspiracies and constitutionally protected assemblies. The "looseness and pliability" of legal doctrine applicable to concerted action led Justice Jackson to note that certain joint activities have a "chameleon-like" character.1 The boycott of white merchants in Claiborne County, Miss., that gave rise to this litigation had such a character; it included elements of criminality and elements of majesty. Evidence that fear of reprisals caused some black citizens to withhold their patronage from respondents' businesses convinced the Supreme Court of Mississippi that the entire boycott was unlawful and that each of the 92 petitioners was liable for all of its economic consequences. Evidence that persuasive rhetoric, determination to remedy past injustices, and a host of voluntary decisions by free citizens were the critical factors in the boycott's success presents us with the question whether the state court's judgment is consistent with the Constitution of the United States. 2 * In March 1966, black citizens of Port Gibson, Miss., and other areas of Claiborne County presented white elected officials with a list of particularized demands for racial equality and integration.2 The complainants did not receive a satisfactory response and, at a local National Association for the Advancement of Colored People (NAACP) meeting at the First Baptist Church, several hundred black persons voted to place a boycott on white merchants in the area. On October 31, 1969, several of the merchants filed suit in state court to recover losses caused by the boycott and to enjoin future boycott activity. We recount first the course of that litigation and then consider in more detail the events that gave rise to the merchants' claim for damages. 3 The complaint was filed in the Chancery Court of Hinds County by 17 white merchants.3 The merchants named two corporations and 146 individuals as defendants: the NAACP, a New York membership corporation; Mississippi Action for Progress (MAP), a Mississippi corporation that implemented the federal "Head Start" program; Aaron Henry, the President of the Mississippi State Conference of the NAACP; Charles Evers, the Field Secretary of the NAACP in Mississippi; and 144 other individuals who had participated in the boycott.4 The complaint sought injunctive relief and an attachment of property, as well as damages. Although it alleged that the plaintiffs were suffering irreparable injury from an ongoing conspiracy, no preliminary relief was sought. 4 Trial began before a chancellor in equity on June 11, 1973.5 The court heard the testimony of 144 witnesses during an 8-month trial. In August 1976, the chancellor issued an opinion and decree finding that "an overwhelming preponderance of the evidence" established the joint and several liability of 130 of the defendants on three separate conspiracy theories.6 First, the court held that the defendants were liable for the tort of malicious interference with the plaintiffs' businesses, which did not necessarily require the presence of a conspiracy.7 Second, the chancellor found a violation of a state statutory prohibition against secondary boycotts, on the theory that the defendants' primary dispute was with the governing authorities of Port Gibson and Claiborne County and not with the white merchants at whom the boycott was directed.8 Third, the court found a violation of Mississippi's antitrust statute, on the ground that the boycott had diverted black patronage from the white merchants to black merchants and to other merchants located out of Claiborne County and thus had unreasonably limited competition between black and white merchants that had traditionally existed.9 The chancellor specifically rejected the defendants' claim that their conduct was protected by the First Amendment.10 5 Five of the merchants offered no evidence of business losses. The chancellor found that the remaining 12 had suffered lost business earnings and lost goodwill during a 7-year period from 1966 to 1972 amounting to $944,699. That amount, plus statutory antitrust penalties of $6,000 and a $300,000 award of attorney's fees, produced a final judgment of $1,250,699, plus interest from the date of judgment and costs. As noted, the chancellor found all but 18 of the original 148 defendants jointly and severally liable for the entire judgment. The court justified imposing full liability on the national organization of the NAACP on the ground that it had failed to "repudiate" the actions of Charles Evers, its Field Secretary in Mississippi. 6 In addition to imposing damages liability, the chancellor entered a broad permanent injunction. He permanently enjoined petitioners from stationing "store watchers" at the respondents' business premises; from "persuading" any person to withhold his patronage from respondents; from "using demeaning and obscene language to or about any person" because that person continued to patronize the respondents; from "picketing or patroling" the premises of any of the respondents; and from using violence against any person or inflicting damage to any real or personal property.11 7 In December 1980, the Mississippi Supreme Court reversed significant portions of the trial court's judgment. 393 So.2d 1290. It held that the secondary boycott statute was inapplicable because it had not been enacted until "the boycott had been in operation for upward of two years."12 The court declined to rely on the restraint of trade statute, noting that the "United States Supreme Court has seen fit to hold boycotts to achieve political ends are not a violation of the Sherman Act, 15 U.S.C. § 1 (1970), after which our statute is patterned."13 Thus, the court rejected two theories of liability that were consistent with a totally voluntary and nonviolent withholding of patronage from the white merchants. 8 The Mississippi Supreme Court upheld the imposition of liability, however, on the basis of the chancellor's common-law tort theory. After reviewing the chancellor's recitation of the facts, the court quoted the following finding made by the trial court: 9 "In carrying out the agreement and design, certain of the defendants, acting for all others, engaged in acts of physical force and violence against the persons and property of certain customers and prospective customers. Intimidation, threats, social ostracism, vilification, and traduction were some of the devices used by the defendants to achieve the desired results. Most effective, also, was the stationing of guards ('enforcers,' 'deacons,' or 'black hats') in the vicinity of white-owned businesses. Unquestionably, the evidence shows that the volition of many black persons was overcome out of sheer fear, and they were forced and compelled against their personal wills to withhold their trade and business intercourse from the complainants." App. to Pet. for Cert. 39b (quoted 393 So.2d, at 1300). 10 On the basis of this finding, the court concluded that the entire boycott was unlawful. "If any of these factors—force, violence, or threats—is present, then the boycott is illegal regardless of whether it is primary, secondary, economical, political, social or other."14 In a brief passage, the court rejected petitioners' reliance on the First Amendment: 11 "The agreed use of illegal force, violence, and threats against the peace to achieve a goal makes the present state of facts a conspiracy. We know of no instance, and our attention has been drawn to no decision, wherein it has been adjudicated that free speech guaranteed by the First Amendment includes in its protection the right to commit crime." Id., at 1301. 12 The theory of the Mississippi Supreme Court, then, was that petitioners had agreed to use force, violence, and "threats" to effectuate the boycott.15 To the trial court, such a finding had not been necessary.16 13 Although the Mississippi Supreme Court affirmed the chancellor's basic finding of liability, the court held that respondents "did not establish their case" with respect to 38 of the defendants.17 The court found that MAP was a victim, rather than a willing participant, in the conspiracy and dismissed—without further explanation—37 individual defendants for lack of proof. Finally, the court ruled that certain damages had been improperly awarded and that other damages had been inadequately proved. The court remanded for further proceedings on the computation of damages.18 14 We granted a petition for certiorari. 454 U.S. 1030, 102 S.Ct. 565, 70 L.Ed.2d 473. At oral argument, a question arose concerning the factual basis for the judgment of the Mississippi Supreme Court. As noted, that court affirmed petitioners' liability for damages on the ground that each of the petitioners had agreed to effectuate the boycott through force, violence, and threats. Such a finding was not necessary to the trial court's imposition of liability and neither state court had identified the evidence actually linking the petitioners to such an agreement. In response to a request from this Court, respondents filed a supplemental brief "specifying the acts committed by each of the petitioners giving rise to liability for damages." Supplemental Brief for Respondents 1. That brief helpfully places the petitioners in different categories; we accept respondents' framework for analysis and identify these classes as a preface to our review of the relevant incidents that occurred during the 7-year period for which damages were assessed.19 15 First, respondents contend that liability is justified by evidence of participation in the "management" of the boycott.20 Respondents identify two groups of persons who may be found liable as "managers": 79 individuals who regularly attended Tuesday night meetings of the NAACP at the First Baptist Church; and 11 persons who took "leadership roles" at those meetings.21 16 Second, respondents contend that liability is justified by evidence that an individual acted as a boycott "enforcer."22 In this category, respondents identify 22 persons as members of the "Black Hats"—a special group organized during the boycott—and 19 individuals who were simply "store watchers." 17 Third, respondents argue that those petitioners "who themselves engaged in violent acts or who threatened violence have provided the best possible evidence that they wanted the boycott to succeed by coercion whenever it could not succeed by persuasion." Id., at 10. They identify 16 individuals for whom there is direct evidence of participation in what respondents characterize as violent acts or threats of violence. 18 Fourth, respondents contend that Charles Evers may be held liable because he "threatened violence on a number of occasions against boycott breakers." Id., at 13. Like the chancellor, respondents would impose liability on the national NAACP because Evers "was acting in his capacity as Field Secretary of the NAACP when he committed these tortious and constitutionally unprotected acts." Ibid. 19 Finally, respondents state that they are "unable to determine on what record evidence the state courts relied in finding liability on the part of seven of the petitioners." Id., at 16. With these allegations of wrongdoing in mind, we turn to consider the factual events that gave rise to this controversy. B 20 The chancellor held petitioners liable for all of respondents' lost earnings during a 7-year period from 1966 to December 31, 1972. We first review chronologically the principal events that occurred during that period, describe some features of the boycott that are not in dispute, and then identify the most significant evidence of violent activity. 21 In late 1965 or early 1966, Charles Evers, the Field Secretary of the NAACP, helped organize the Claiborne County Branch of the NAACP. The pastor of the First Baptist Church, James Dorsey, was elected president of the Branch; regular meetings were conducted each Tuesday evening at the church. At about the same time, a group of black citizens formed a Human Relations Committee and presented a petition for redress of grievances to civic and business leaders of the white community. In response, a biracial committee—including five of the petitioners and several of the respondents—was organized and held a series of unproductive meetings. 22 The black members of the committee then prepared a further petition entitled "Demands for Racial Justice." This petition was presented for approval at the local NAACP meeting conducted on the first Tuesday evening in March. As described by the chancellor, "the approximately 500 people present voted their approval unanimously."23 On March 14, 1966, the petition was presented to public officials of Port Gibson and Claiborne County. 23 The petition included 19 specific demands. It called for the desegregation of all public schools and public facilities, the hiring of black policemen, public improvements in black residential areas, selection of blacks for jury duty, integration of bus stations so that blacks could use all facilities, and an end to verbal abuse by law enforcement officers. It stated that "Negroes are not to be addressed by terms as 'boy,' 'girl,' 'shine,' 'uncle,' or any other offensive term, but as 'Mr.,' 'Mrs.,' or 'Miss,' as is the case with other citizens."24 As described by the chancellor, the purpose of the demands "was to gain equal rights and opportunities for Negro citizens."25 The petition further provided that black leaders hoped it would not be necessary to resort to the "selective buying campaigns" that had been used in other communities.26 On March 23, two demands that had been omitted from the original petition were added, one of which provided: "All stores must employ Negro clerks and cashiers."27 This supplemental petition stated that a response was expected by April 1. 24 A favorable response was not received. On April 1, 1966, the Claiborne County NAACP conducted another meeting at the First Baptist Church. As described by the chancellor: 25 "Several hundred black people attended the meeting, and the purpose was to decide what action should be taken relative to the twenty-one demands. Speeches were made by Evers and others, and a vote was taken. It was the unanimous vote of those present, without dissent, to place a boycott on the white merchants of Port Gibson and Claiborne County." App. to Pet. for Cert. 15b. The boycott was underway.28 26 In September 1966, Mississippi Action for Progress, Inc. (MAP), was organized to develop community action programs in 20 counties of Mississippi. One of MAP's programs—known as Head Start—involved the use of federal funds to provide food for young children. Originally, food purchases in Claiborne County were made alternately from white-owned and black-owned stores, but in February 1967 the directors of MAP authorized their Claiborne County representatives to purchase food only from black-owned stores. Since MAP bought substantial quantities of food, the consequences of this decision were significant. A large portion of the trial was devoted to the question whether MAP participated in the boycott voluntarily and under the chancellor's theories of liability—could be held liable for the resulting damages. The chancellor found MAP a willing participant, noting that "during the course of the trial, the only Head Start cooks called to the witness stand testified that they refused to go into white-owned stores to purchase groceries for the children in the program for the reason that they were in favor of the boycott and wanted to honor it."29 27 Several events occurred during the boycott that had a strong effect on boycott activity. On February 1, 1967, Port Gibson employed its first black policeman. During that month, the boycott was lifted on a number of merchants. On April 4, 1968, Dr. Martin Luther King, Jr., was assassinated in Memphis. The chancellor found that this tragic event had a depressing effect on the black community and, as a result, the boycott "tightened."30 28 One event that occurred during the boycott is of particular significance. On April 18, 1969, a young black man named Roosevelt Jackson was shot and killed during an encounter with two Port Gibson police officers.31 Large crowds immediately gathered, first at the hospital and later at the church. Tension in the community neared a breaking point. The local police requested reinforcements from the State Highway Patrol and sporadic acts of violence ensued. The Mayor and Board of Aldermen placed a dawn-to-dusk curfew into effect. 29 On April 19, Charles Evers spoke to a group assembled at the First Baptist Church and led a march to the courthouse where he demanded the discharge of the entire Port Gibson Police Force. When this demand was refused, the boycott was reimposed on all white merchants. One of Evers' speeches on this date was recorded by the police. In that speech—significant portions of which are reproduced in an Appendix to this opinion—Evers stated that boycott violators would be "disciplined" by their own people and warned that the Sheriff could not sleep with boycott violators at night. 30 On April 20, Aaron Henry came to Port Gibson, spoke to a large gathering, urged moderation, and joined local leaders in a protest march and a telegram sent to the Attorney General of the United States. On April 21, Evers gave another speech to several hundred people, in which he again called for a discharge of the police force and for a total boycott of all white-owned businesses in Claiborne County. Although this speech was not recorded, the chancellor found that Evers stated: "If we catch any of you going in any of them racist stores, we're gonna break your damn neck."32 31 As noted, this lawsuit was filed in October 1969. No significant events concerning the boycott occurred after that time. The chancellor identified no incident of violence that occurred after the suit was brought. He did identify, however, several significant incidents of boycott-related violence that occurred some years earlier. 32 Before describing that evidence, it is appropriate to note that certain practices generally used to encourage support for the boycott were uniformly peaceful and orderly. The few marches associated with the boycott were carefully controlled by black leaders. Pickets used to advertise the boycott were often small children. The police made no arrests—and no complaints are recorded—in connection with the picketing and occasional demonstrations supporting the boycott. Such activity was fairly irregular, occurred primarily on weekends, and apparently was largely discontinued around the time the lawsuit was filed.33 33 One form of "discipline" of black persons who violated the boycott appears to have been employed with some regularity. Individuals stood outside of boycotted stores and identified those who traded with the merchants. Some of these "store watchers" were members of a group known as the "Black Hats" or the "Deacons."34 The names of persons who violated the boycott were read at meetings of the Claiborne County NAACP and published in a mimeographed paper entitled the "Black Times." As stated by the chancellor, those persons "were branded as traitors to the black cause, called demeaning names, and socially ostracized for merely trading with whites."35 34 The chancellor also concluded that a quite different form of discipline had been used against certain violators of the boycott. He specifically identified 10 incidents that "strikingly" revealed the "atmosphere of fear that prevailed among blacks from 1966 until 1970."36 The testimony concerning four incidents convincingly demonstrates that they occurred because the victims were ignoring the boycott. In two cases, shots were fired at a house; in a third, a brick was thrown through a windshield; in the fourth, a flower garden was damaged. None of these four victims, however, ceased trading with white merchants.37 35 The evidence concerning four other incidents is less clear, but again it indicates that an unlawful form of discipline was applied to certain boycott violators. In April 1966, a black couple named Cox asked for a police escort to go into a white-owned dry cleaner and, a week later, shots were fired into their home. In another incident, an NAACP member took a bottle of whiskey from a black man who had purchased it in a white-owned store. The third incident involved a fight between a commercial fisherman who did not observe the boycott and four men who "grabbed me and beat me up and took a gun off me."38 In a fourth incident, described only in hearsay testimony, a group of young blacks apparently pulled down the overalls of an elderly brick mason known as "Preacher White" and spanked him for not observing the boycott.39 36 Two other incidents discussed by the chancellor are of less certain significance. Jasper Coleman testified that he participated in an all-night poker game at a friend's house on Christmas Eve 1966. The following morning he discovered that all four tires of his pickup truck had been slashed with a knife. Coleman testified that he did not participate in the boycott but was never threatened for refusing to do so. Record 13791. Finally, Willie Myles testified that he and his wife received a threatening phone call and that a boy on a barge told him that he would be whipped for buying his gas at the wrong place. 37 Five of these incidents occurred in 1966. The other five are not dated. The chancellor thus did not find that any act of violence occurred after 1966.40 In particular, he made no reference to any act of violence or threat of violence—with the exception, of course, of Charles Evers' speeches—after the shootings of Martin Luther King, Jr., in 1968 or Roosevelt Jackson in 1969. The chancellor did not find that any of the incidents of violence was discussed at the Tuesday evening meetings of the NAACP.41 II 38 This Court's jurisdiction to review the judgment of the Mississippi Supreme Court is, of course, limited to the federal questions necessarily decided by that court.42 We consider first whether petitioners' activities are protected in any respect by the Federal Constitution and, if they are, what effect such protection has on a lawsuit of this nature. 39 The boycott of white merchants at issue in this case took many forms. The boycott was launched at a meeting of a local branch of the NAACP attended by several hundred persons. Its acknowledged purpose was to secure compliance by both civic and business leaders with a lengthy list of demands for equality and racial justice. The boycott was supported by speeches and nonviolent picketing. Participants repeatedly encouraged others to join in its cause. 40 Each of these elements of the boycott is a form of speech or conduct that is ordinarily entitled to protection under the First and Fourteenth Amendments.43 The black citizens named as defendants in this action banded together and collectively expressed their dissatisfaction with a social structure that had denied them rights to equal treatment and respect. As we so recently acknowledged in Citizens Against Rent Control Coalition for Fair Housing v. Berkeley, 454 U.S. 290, 294, 102 S.Ct. 434, 436, 70 L.Ed.2d 492, "the practice of persons sharing common views banding together to achieve a common end is deeply embedded in the American political process." We recognized that "by collective effort individuals can make their views known, when, individually, their voices would be faint or lost." Ibid. In emphasizing "the importance of freedom of association in guaranteeing the right of people to make their voices heard on public issues," id., at 295, 102 S.Ct., at 437, we noted the words of Justice Harlan, writing for the Court in NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 460, 78 S.Ct. 1163, 1170, 2 L.Ed.2d 1488: 41 "Effective advocacy of both public and private points of view, particularly controversial ones, is undeniably enhanced by group association, as this Court has more than once recognized by remarking upon the close nexus between the freedoms of speech and assembly." 42 THE CHIEF JUSTICE stated for the Court in Citizens Against Rent Control : "There are, of course, some activities, legal if engaged in by one, yet illegal if performed in concert with others, but political expression is not one of them." 454 U.S., at 296, 102 S.Ct., at 437. 43 The right to associate does not lose all constitutional protection merely because some members of the group may have participated in conduct or advocated doctrine that itself is not protected. In De Jonge v. Oregon, 299 U.S. 353, 57 S.Ct. 255, 81 L.Ed. 278, the Court unanimously held that an individual could not be penalized simply for assisting in the conduct of an otherwise lawful meeting held under the auspices of the Communist Party, an organization that advocated "criminal syndicalism." After reviewing the rights of citizens "to meet peaceably for consultation in respect to public affairs and to petition for a redress of grievances," id., at 364, 57 S.Ct., at 259, Chief Justice Hughes, writing for the Court, stated: 44 "It follows from these considerations that, consistently with the Federal Constitution, peaceable assembly for lawful discussion cannot be made a crime. The holding of meetings for peaceable political action cannot be proscribed. Those who assist in the conduct of such meetings cannot be branded as criminals on that score. The question, if the rights of free speech and peaceable assembly are to be preserved, is not as to the auspices under which the meeting is held but as to its purpose; not as to the relations of the speakers, but whether their utterances transcend the bounds of the freedom of speech which the Constitution protects. If the persons assembling have committed crimes elsewhere, if they have formed or are engaged in a conspiracy against the public peace and order, they may be prosecuted for their conspiracy or other violation of valid laws. But it is a different matter when the State, instead of prosecuting them for such offenses, seizes upon mere participation in a peaceable assembly and a lawful public discussion as the basis for a criminal charge." Id., at 365, 57 S.Ct., at 260. 45 Of course, the petitioners in this case did more than assemble peaceably and discuss among themselves their grievances against governmental and business policy. Other elements of the boycott, however, also involved activities ordinarily safeguarded by the First Amendment. In Thornhill v. Alabama, 310 U.S. 88, 60 S.Ct. 736, 84 L.Ed. 1093, the Court held that peaceful picketing was entitled to constitutional protection, even though, in that case, the purpose of the picketing "was concededly to advise customers and prospective customers of the relationship existing between the employer and its employees and thereby to induce such customers not to patronize the employer." Id., at 99, 60 S.Ct., at 742. Cf. Chauffeurs v. Newell, 356 U.S. 341, 78 S.Ct. 779, 2 L.Ed.2d 809. In Edwards v. South Carolina, 372 U.S. 229, 83 S.Ct. 680, 9 L.Ed.2d 697, we held that a peaceful march and demonstration was protected by the rights of free speech, free assembly, and freedom to petition for a redress of grievances. 46 Speech itself also was used to further the aims of the boycott. Nonparticipants repeatedly were urged to join the common cause, both through public address and through personal solicitation. These elements of the boycott involve speech in its most direct form. In addition, names of boycott violators were read aloud at meetings at the First Baptist Church and published in a local black newspaper. Petitioners admittedly sought to persuade others to join the boycott through social pressure and the "threat" of social ostracism. Speech does not lose its protected character, however, simply because it may embarrass others or coerce them into action. As Justice Rutledge, in describing the protection afforded by the First Amendment, explained: 47 "It extends to more than abstract discussion, unrelated to action. The First Amendment is a charter for government, not for an institution of learning. 'Free trade in ideas' means free trade in the opportunity to persuade to action, not merely to describe facts." Thomas v. Collins, 323 U.S. 516, 537, 65 S.Ct. 315, 325, 89 L.Ed. 430. 48 In Organization for a Better Austin v. Keefe, 402 U.S. 415, 91 S.Ct. 1575, 29 L.Ed.2d 1, the Court considered the validity of a prior restraint on speech that invaded the "privacy" of the respondent. Petitioner, a racially integrated community organization, charged that respondent, a real estate broker, had engaged in tactics known as "blockbusting" or "panic peddling."44 Petitioner asked respondent to sign an agreement that he would not solicit property in their community. When he refused, petitioner distributed leaflets near respondent's home that were critical of his business practices.45 A state court enjoined petitioner from distributing the leaflets; an appellate court affirmed on the ground that the alleged activities were coercive and intimidating, rather than informative, and therefore not entitled to First Amendment protection. Id., at 418, 91 S.Ct., at 1577. This Court reversed. THE CHIEF JUSTICE explained: 49 "This Court has often recognized that the activity of peaceful pamphleteering is a form of communication protected by the First Amendment. E.g., Martin v. City of Struthers, 319 U.S. 141, 63 S.Ct. 862, 87 L.Ed. 1313 (1943); Schneider v. State, 308 U.S. 147, 60 S.Ct. 146, 84 L.Ed. 155 (1939); Lovell v. Griffin, 303 U.S. 444, 58 S.Ct. 666, 82 L.Ed. 949 (1938). In sustaining the injunction, however, the Appellate Court was apparently of the view that petitioners' purpose in distributing their literature was not to inform the public, but to 'force' respondent to sign a no-solicitation agreement. The claim that the expressions were intended to exercise a coercive impact on respondent does not remove them from the reach of the First Amendment. Petitioners plainly intended to influence respondent's conduct by their activities; this is not fundamentally different from the function of a newspaper. See Schneider v. State, supra; Thornhill v. Alabama, 310 U.S. 88, 60 S.Ct. 736, 84 L.Ed. 1093 (1940). Petitioners were engaged openly and vigorously in making the public aware of respondent's real estate practices. Those practices were offensive to them, as the views and practices of petitioners are no doubt offensive to others. But so long as the means are peaceful, the communication need not meet standards of acceptability." Id., at 419, 91 S.Ct., at 1577. 50 In dissolving the prior restraint, the Court recognized that "offensive" and "coercive" speech was nevertheless protected by the First Amendment.46 51 In sum, the boycott clearly involved constitutionally protected activity. The established elements of speech, assembly, association, and petition, "though not identical, are inseparable." Thomas v. Collins, supra, at 530, 65 S.Ct., at 322. Through exercise of these First Amendment rights, petitioners sought to bring about political, social, and economic change. Through speech, assembly, and petition—rather than through riot or revolution—petitioners sought to change a social order that had consistently treated them as second-class citizens. 52 The presence of protected activity, however, does not end the relevant constitutional inquiry. Governmental regulation that has an incidental effect on First Amendment freedoms may be justified in certain narrowly defined instances. See United States v. O'Brien, 391 U.S. 367, 88 S.Ct. 1673, 20 L.Ed.2d 672.47 A nonviolent and totally voluntary boycott may have a disruptive effect on local economic conditions. This Court has recognized the strong governmental interest in certain forms of economic regulation, even though such regulation may have an incidental effect on rights of speech and association. See Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 69 S.Ct. 684, 93 L.Ed. 834; NLRB v. Retail Store Employees, 447 U.S. 607, 100 S.Ct. 2372, 65 L.Ed.2d 377. The right of business entities to "associate" to suppress competition may be curtailed. National Society of Professional Engineers v. United States, 435 U.S. 679, 98 S.Ct. 1355, 55 L.Ed.2d 637. Unfair trade practices may be restricted. Secondary boycotts and picketing by labor unions may be prohibited, as part of "Congress' striking of the delicate balance between union freedom of expression and the ability of neutral employers, employees, and consumers to remain free from coerced participation in industrial strife." NLRB v. Retail Store Employees, supra, at 617-618, 100 S.Ct., at 2378 (BLACKMUN, J., concurring in part). See Longshoremen v. Allied International, Inc., 456 U.S. 212, 222-223, and n. 20, 102 S.Ct. 1656, 1662-1663, and n. 20, 72 L.Ed.2d 21. 53 While States have broad power to regulate economic activity, we do not find a comparable right to prohibit peaceful political activity such as that found in the boycott in this case. This Court has recognized that expression on public issues "has always rested on the highest rung of the hierarchy of First Amendment values." Carey v. Brown, 447 U.S. 455, 467, 100 S.Ct. 2286, 2293, 65 L.Ed.2d 263. "[S]peech concerning public affairs is more than self-expression; it is the essence of self-government." Garrison v. Louisiana, 379 U.S. 64, 74-75, 85 S.Ct. 209, 215, 13 L.Ed.2d 125. There is a "profound national commitment" to the principle that "debate on public issues should be uninhibited, robust, and wide-open." New York Times Co. v. Sullivan, 376 U.S. 254, 270, 84 S.Ct. 710, 720, 11 L.Ed.2d 686. 54 In Eastern Railroad Presidents Conference v. Noerr Motor Freight Inc., 365 U.S. 127, 81 S.Ct. 523, 5 L.Ed.2d 464, the Court considered whether the Sherman Act prohibited a publicity campaign waged by railroads against the trucking industry that was designed to foster the adoption of laws destructive of the trucking business, to create an atmosphere of distaste for truckers among the general public, and to impair the relationships existing between truckers and their customers. Noting that the "right of petition is one of the freedoms protected by the Bill of Rights, and we cannot, of course, lightly impute to Congress an intent to invade these freedoms," the Court held that the Sherman Act did not proscribe the publicity campaign. Id., at 137-138, 81 S.Ct., at 529. The Court stated that it could not see how an intent to influence legislation to destroy the truckers as competitors "could transform conduct otherwise lawful into a violation of the Sherman Act." Id., at 138-139, 81 S.Ct., at 530. Noting that the right of the people to petition their representatives in government "cannot properly be made to depend on their intent in doing so," the Court held that "at least insofar as the railroads' campaign was directed toward obtaining governmental action, its legality was not at all affected by any anticompetitive purpose it may have had." Id., at 139-140, 81 S.Ct., at 530. This conclusion was not changed by the fact that the railroads' anticompetitive purpose produced an anticompetitive effect ; the Court rejected the truckers' Sherman Act claim despite the fact that "the truckers sustained some direct injury as an incidental effect of the railroads' campaign to influence governmental action." Id., at 143, 81 S.Ct., at 532. 55 It is not disputed that a major purpose of the boycott in this case was to influence governmental action. Like the railroads in Noerr, the petitioners certainly foresaw—and directly intended—that the merchants would sustain economic injury as a result of their campaign. Unlike the railroads in that case, however, the purpose of petitioners' campaign was not to destroy legitimate competition. Petitioners sought to vindicate rights of equality and of freedom that lie at the heart of the Fourteenth Amendment itself. The right of the States to regulate economic activity could not justify a complete prohibition against a nonviolent, politically motivated boycott designed to force governmental and economic change and to effectuate rights guaranteed by the Constitution itself.48 56 In upholding an injunction against the state supersedeas bonding requirement in this case, Judge Ainsworth of the Court of Appeals for the Fifth Circuit cogently stated: 57 "At the heart of the Chancery Court's opinion lies the belief that the mere organization of the boycott and every activity undertaken in support thereof could be subject to judicial prohibition under state law. This view accords insufficient weight to the First Amendment's protection of political speech and association. There is no suggestion that the NAACP, MAP or the individual defendants were in competition with the white businesses or that the boycott arose from parochial economic interests. On the contrary, the boycott grew out of a racial dispute with the white merchants and city government of Port Gibson and all of the picketing, speeches, and other communication associated with the boycott were directed to the elimination of racial discrimination in the town. This differentiates this case from a boycott organized for economic ends, for speech to protest racial discrimination is essential political speech lying at the core of the First Amendment." Henry v. First National Bank of Clarksdale, 595 F.2d 291, 303 (1979) (footnote omitted). 58 We hold that the nonviolent elements of petitioners' activities are entitled to the protection of the First Amendment.49 B 59 The Mississippi Supreme Court did not sustain the chancellor's imposition of liability on a theory that state law prohibited a nonviolent, politically motivated boycott. The fact that such activity is constitutionally protected, however, imposes a special obligation on this Court to examine critically the basis on which liability was imposed.50 In particular, we consider here the effect of our holding that much of petitioners' conduct was constitutionally protected on the ability of the State to impose liability for elements of the boycott that were not so protected.51 60 The First Amendment does not protect violence. "Certainly violence has no sanctuary in the First Amendment, and the use of weapons, gunpowder, and gasoline may not constitutionally masquerade under the guise of 'advocacy.' " Samuels v. Mackell, 401 U.S. 66, 75, 91 S.Ct. 764, 769, 27 L.Ed.2d 688 (Douglas, J., concurring). Although the extent and significance of the violence in this case are vigorously disputed by the parties, there is no question that acts of violence occurred. No federal rule of law restricts a State from imposing tort liability for business losses that are caused by violence and by threats of violence. When such conduct occurs in the context of constitutionally protected activity, however, "precision of regulation" is demanded. NAACP v. Button, 371 U.S. 415, 438, 83 S.Ct. 328, 340, 9 L.Ed.2d 405.52 Specifically, the presence of activity protected by the First Amendment imposes restraints on the grounds that may give rise to damages liability and on the persons who may be held accountable for those damages. 61 In Mine Workers v. Gibbs, 383 U.S. 715, 86 S.Ct. 1130, 16 L.Ed.2d 218, the Court considered a case in many respects similar to the one before us. The case grew out of the rivalry between the United Mine Workers (UMW) and the Southern Labor Union (SLU) over representation of workers in the southern Appalachian coal fields. A coal company laid off 100 miners of UMW's Local 5881 when it closed one of its mines. That same year, a subsidiary of the coal company hired Gibbs as mine superintendent to attempt to open a new mine on nearby property through use of members of the SLU. Gibbs also received a contract to haul the mine's coal to the nearest railroad loading point. When he attempted to open the mine, however, he was met by armed members of Local 5881 who threatened Gibbs and beat an SLU organizer. These incidents occurred on August 15 and 16. Thereafter, there was no further violence at the mine site and UMW members maintained a peaceful picket line for nine months. No attempts to open the mine were made during that period. 62 Gibbs lost his job as superintendent and never began performance of the haulage contract. Claiming to have suffered losses as a result of the union's concerted plan against him, Gibbs filed suit in federal court against the international UMW. He alleged an unlawful secondary boycott under the federal labor laws and, as a pendent state-law claim, "an unlawful conspiracy and an unlawful boycott aimed at him . . . to maliciously, wantonly and willfully interfere with his contract of employment and with his contract of haulage." Id., at 720, 86 S.Ct., at 1135. The federal claim was dismissed on the ground that the dispute was "primary" and therefore not cognizable under the federal prohibition of secondary labor boycotts. Damages were awarded against the UMW, however, on the state claim of interference with an employment relationship. 63 This Court reversed. The Court found that the pleadings, arguments of counsel, and jury instructions had not adequately defined the compass within which damages could be awarded under state law. The Court noted that it had "consistently recognized the right of States to deal with violence and threats of violence appearing in labor disputes" and had sustained "a variety of remedial measures against the contention that state law was pre-empted by the passage of federal labor legislation." Id., at 729, 86 S.Ct., at 1140. To accommodate federal labor policy, however, the Court in Gibbs held: "the permissible scope of state remedies in this area is strictly confined to the direct consequences of such [violent] conduct, and does not include consequences resulting from associated peaceful picketing or other union activity." Ibid. The Court noted that in Construction Workers v. Laburnum Construction Corp., 347 U.S. 656, 74 S.Ct. 833, 98 L.Ed. 1025, damages were restricted to those directly and proximately caused by wrongful conduct chargeable to the defendants. " 'Thus there [was] nothing in the measure of damages to indicate that state power was exerted to compensate for anything more than the direct consequences of the violent conduct.' " 383 U.S., at 730, 86 S.Ct., at 1141 (quoting San Diego Building Trades Council v. Garmon, 359 U.S. 236, 249, n. 6, 79 S.Ct. 773, 781, n. 6, 3 L.Ed.2d 775). 64 The careful limitation on damages liability imposed in Gibbs resulted from the need to accommodate state law with federal labor policy. That limitation is no less applicable, however, to the important First Amendment interests at issue in this case. Petitioners withheld their patronage from the white establishment of Claiborne County to challenge a political and economic system that had denied them the basic rights of dignity and equality that this country had fought a Civil War to secure. While the State legitimately may impose damages for the consequences of violent conduct, it may not award compensation for the consequences of nonviolent, protected activity. Only those losses proximately caused by unlawful conduct may be recovered. 65 The First Amendment similarly restricts the ability of the State to impose liability on an individual solely because of his association with another. In Scales v. United States, 367 U.S. 203, 229, 81 S.Ct. 1469, 1486, 6 L.Ed.2d 782, the Court noted that a "blanket prohibition of association with a group having both legal and illegal aims" would present "a real danger that legitimate political expression or association would be impaired." The Court suggested that to punish association with such a group, there must be "clear proof that a defendant 'specifically intend[s] to accomplish [the aims of the organization] by resort to violence.' " Ibid. (quoting Noto v. United States, 367 U.S. 290, 299, 81 S.Ct. 1517, 1521, 6 L.Ed.2d 836).53 Moreover, inNoto v. United States the Court emphasized that this intent must be judged "according to the strictest law,"54 for "otherwise there is a danger that one in sympathy with the legitimate aims of such an organization, but not specifically intending to accomplish them by resort to violence, might be punished for his adherence to lawful and constitutionally protected purposes, because of other and unprotected purposes which he does not necessarily share." Id., at 299-300, 81 S.Ct., at 1521. 66 In Healy v. James, 408 U.S. 169, 92 S.Ct. 2338, 33 L.Ed.2d 266, the Court applied these principles in a noncriminal context. In that case the Court held that a student group could not be denied recognition at a state-supported college merely because of its affiliation with a national organization associated with disruptive and violent campus activity. It noted that "the Court has consistently disapproved governmental action imposing criminal sanctions or denying rights and privileges solely because of a citizen's association with an unpopular organization." Id., at 185-186, 92 S.Ct., at 2348. The Court stated that "it has been established that 'guilt by association alone, without [establishing] that an individual's association poses the threat feared by the Government,' is an impermissible basis upon which to deny First Amendment rights." Id., at 186, 92 S.Ct., at 2348 (quoting United States v. Robel, 389 U.S. 258, 265, 88 S.Ct. 419, 424, 19 L.Ed.2d 508). "The government has the burden of establishing a knowing affiliation with an organization possessing unlawful aims and goals, and a specific intent to further those illegal aims." 408 U.S., at 186, 92 S.Ct., at 2348 (footnote omitted).55 67 The principles announced in Scales, Noto, and Healy are relevant to this case. Civil liability may not be imposed merely because an individual belonged to a group, some members of which committed acts of violence. For liability to be imposed by reason of association alone, it is necessary to establish that the group itself possessed unlawful goals and that the individual held a specific intent to further those illegal aims.56 "In this sensitive field, the State may not employ 'means that broadly stifle fundamental personal liberties when the end can be more narrowly achieved.' Shelton v. Tucker, 364 U.S. 479, 488, 81 S.Ct. 247, 252, 5 L.Ed.2d 231 (1960)." Carroll v. Princess Anne, 393 U.S. 175, 183-184, 89 S.Ct. 347, 353, 21 L.Ed.2d 325. III 68 The chancellor awarded respondents damages for all business losses that were sustained during a 7-year period beginning in 1966 and ending December 31, 1972.57 With the exception of Aaron Henry, all defendants were held jointly and severally liable for these losses. The chancellor's findings were consistent with his view that voluntary participation in the boycott was a sufficient basis on which to impose liability. The Mississippi Supreme Court properly rejected that theory; it nevertheless held that petitioners were liable for all damages "resulting from the boycott."58 In light of the principles set forth above, it is evident that such a damages award may not be sustained in this case. 69 The opinion of the Mississippi Supreme Court itself demonstrates that all business losses were not proximately caused by the violence and threats of violence found to be present. The court stated that "coercion, intimidation, and threats" formed "part of the boycott activity" and "contributed to its almost complete success."59 The court broadly asserted—without differentiation—that " '[i]ntimidation, threats, social ostracism, vilification, and traduction' " were devices used by the defendants to effectuate the boycott.60 The court repeated the chancellor's finding that "the volition of many black persons was overcome out of sheer fear."61 These findings are inconsistent with the court's imposition of all damages "resulting from the boycott." To the extent that the court's judgment rests on the ground that "many" black citizens were "intimidated" by "threats" of "social ostracism, vilification, and traduction," it is flatly inconsistent with the First Amendment. The ambiguous findings of the Mississippi Supreme Court are inadequate to assure the "precision of regulation" demanded by that constitutional provision. 70 The record in this case demonstrates that all of respondents' losses were not proximately caused by violence or threats of violence. As respondents themselves stated at page 12 of their brief in the Mississippi Supreme Court: 71 "Most of the witnesses testified that they voluntarily went along with the NAACP and their fellow black citizens in honoring and observing the boycott because they wanted the boycott." 72 This assessment is amply supported by the record.62 It is indeed inconceivable that a boycott launched by the unanimous vote of several hundred persons succeeded solely through fear and intimidation. Moreover, the fact that the boycott "intensified" following the shootings of Martin Luther King, Jr., and Roosevelt Jackson demonstrates that factors other than force and violence (by the petitioners) figured prominently in the boycott's success. The chancellor made no finding that any act of violence occurred after 1966. While the timing of the acts of violence was not important to the chancellor's imposition of liability, it is a critical factor under the narrower rationale of the Mississippi Supreme Court. That court has completely failed to demonstrate that business losses suffered in 1972—three years after this lawsuit was filed were proximately caused by the isolated acts of violence found in 1966.63 It is impossible to conclude that state power has not been exerted to compensate respondents for the direct consequences of nonviolent, constitutionally protected activity. 73 This case is not like Milk Wagon Drivers v. Meadowmoor Dairies, Inc., 312 U.S. 287, 61 S.Ct. 552, 85 L.Ed. 836, in which the Court held that the presence of violence justified an injunction against both violent and nonviolent activity.64 The violent conduct present in that case was pervasive.65 The Court in Meadowmoor stated that "utterance in a context of violence can lose its significance as an appeal to reason and become part of an instrument of force." Id., at 293, 61 S.Ct., at 555. The Court emphasized, however: "Still it is of prime importance that no constitutional freedom, least of all the guarantees of the Bill of Rights, be defeated by insubstantial findings of fact screening reality. That is why this Court has the ultimate power to search the records in the state courts where a claim of constitutionality is effectively made. And so the right of free speech cannot be denied by drawing from a trivial rough incident or a moment of animal exuberance the conclusion that otherwise peaceful picketing has the taint of force." Ibid. 74 Such "insubstantial findings" were not present in Meadowmoor. But in this case, the Mississippi Supreme Court has relied on isolated acts of violence during a limited period to uphold respondents' recovery of all business losses sustained over a 7-year span. No losses are attributed to the voluntary participation of individuals determined to secure "justice and equal opportunity."66 The court's judgment "screens reality" and cannot stand.67 75 Respondents' supplemental brief also demonstrates that on the present record no judgment may be sustained against most of the petitioners. Regular attendance and participation at the Tuesday meetings of the Claiborne County Branch of the NAACP is an insufficient predicate on which to impose liability. The chancellor's findings do not suggest that any illegal conduct was authorized, ratified, or even discussed at any of the meetings. The Sheriff testified that he was kept informed of what transpired at the meetings; he made no reference to any discussion of unlawful activity.68 To impose liability for presence at weekly meetings of the NAACP would—ironically—not even constitute "guilt by association," since there is no evidence that the association possessed unlawful aims. Rather, liability could only be imposed on a "guilt for association" theory. Neither is permissible under the First Amendment.69 76 Respondents also argue that liability may be imposed on individuals who were either "store watchers" or members of the "Black Hats." There is nothing unlawful in standing outside a store and recording names. Similarly, there is nothing unlawful in wearing black hats, although such apparel may cause apprehension in others. As established above, mere association with either group—absent a specific intent to further an unlawful aim embraced by that group—is an insufficient predicate for liability. At the same time, the evidence does support the conclusion that some members of each of these groups engaged in violence or threats of violence. Unquestionably, these individuals may be held responsible for the injuries that they caused; a judgment tailored to the consequences of their unlawful conduct may be sustained. 77 Respondents have sought separately to justify the judgment entered against Charles Evers and the national NAACP. As set forth by the chancellor, Evers was specially connected with the boycott in four respects. First, Evers signed the March 23 supplemental demand letter and unquestionably played the primary leadership role in the organization of the boycott. Second, Evers participated in negotiations with MAP and successfully convinced MAP to abandon its practice of purchasing food alternately from white-owned and black-owned stores. Third, he apparently presided at the April 1, 1966, meeting at which the vote to begin the boycott was taken; he delivered a speech to the large audience that was gathered on that occasion. See n. 28, supra. Fourth, Evers delivered the speeches on April 19 and 21, 1969, which we have discussed previously. See supra, at 902; Appendix to this opinion. 78 For the reasons set forth above, liability may not be imposed on Evers for his presence at NAACP meetings or his active participation in the boycott itself. To the extent that Evers caused respondents to suffer business losses through his organization of the boycott, his emotional and persuasive appeals for unity in the joint effort, or his "threats" of vilification or social ostracism, Evers' conduct is constitutionally protected and beyond the reach of a damages award. Respondents point to Evers' speeches, however, as justification for the chancellor's damages award. Since respondents would impose liability on the basis of a public address—which predominantly contained highly charged political rhetoric lying at the core of the First Amendment—we approach this suggested basis of liability with extreme care. 79 There are three separate theories that might justify holding Evers liable for the unlawful conduct of others. First, a finding that he authorized, directed, or ratified specific tortious activity would justify holding him responsible for the consequences of that activity. Second, a finding that his public speeches were likely to incite lawless action could justify holding him liable for unlawful conduct that in fact followed within a reasonable period. Third, the speeches might be taken as evidence that Evers gave other specific instructions to carry out violent acts or threats. 80 While many of the comments in Evers' speeches might have contemplated "discipline" in the permissible form of social ostracism, it cannot be denied that references to the possibility that necks would be broken and to the fact that the Sheriff could not sleep with boycott violators at night implicitly conveyed a sterner message. In the passionate atmosphere in which the speeches were delivered, they might have been understood as inviting an unlawful form of discipline or, at least, as intending to create a fear of violence whether or not improper discipline was specifically intended. 81 It is clear that "fighting words"—those that provoke immediate violence—are not protected by the First Amendment. Chaplinsky v. New Hampshire, 315 U.S. 568, 572, 62 S.Ct. 766, 769, 86 L.Ed. 1031. Similarly, words that create an immediate panic are not entitled to constitutional protection. Schenck v. United States, 249 U.S. 47, 39 S.Ct. 247, 63 L.Ed. 470.70 This Court has made clear, however, that mere advocacy of the use of force or violence does not remove speech from the protection of the First Amendment. In Brandenburg v. Ohio, 395 U.S. 444, 89 S.Ct. 1827, 23 L.Ed.2d 430, we reversed the conviction of a Ku Klux Klan leader for threatening "revengeance" if the "suppression" of the white race continued; we relied on "the principle that the constitutional guarantees of free speech and free press do not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action." Id., at 447, 89 S.Ct., at 1829. See Noto v. United States, 367 U.S., at 297-298, 81 S.Ct., at 1520 ("the mere abstract teaching . . . of the moral propriety or even moral necessity for a resort to force and violence, is not the same as preparing a group for violent action and steeling it to such action"). See also Whitney v. California, 274 U.S. 357, 372, 47 S.Ct. 641, 647, 71 L.Ed. 1095 (Brandeis, J., concurring). 82 The emotionally charged rhetoric of Charles Evers' speeches did not transcend the bounds of protected speech set forth in Brandenburg. The lengthy addresses generally contained an impassioned plea for black citizens to unify, to support and respect each other, and to realize the political and economic power available to them. In the course of those pleas, strong language was used. If that language had been followed by acts of violence, a substantial question would be presented whether Evers could be held liable for the consequences of that unlawful conduct. In this case, however—with the possible exception of the Cox incident—the acts of violence identified in 1966 occurred weeks or months after the April 1, 1966, speech; the chancellor made no finding of any violence after the challenged 1969 speech. Strong and effective extemporaneous rhetoric cannot be nicely channeled in purely dulcet phrases. An advocate must be free to stimulate his audience with spontaneous and emotional appeals for unity and action in a common cause. When such appeals do not incite lawless action, they must be regarded as protected speech. To rule otherwise would ignore the "profound national commitment" that "debate on public issues should be uninhibited, robust, and wide-open." New York Times Co. v. Sullivan, 376 U.S., at 270, 84 S.Ct., at 720.71 83 For these reasons, we conclude that Evers' addresses did not exceed the bounds of protected speech. If there were other evidence of his authorization of wrongful conduct, the references to discipline in the speeches could be used to corroborate that evidence. But any such theory fails for the simple reason that there is no evidence—apart from the speeches themselves—that Evers authorized, ratified, or directly threatened acts of violence.72 The chancellor's findings are not sufficient to establish that Evers had a duty to "repudiate" the acts of violence that occurred.73 The findings are constitutionally inadequate to support the damages judgment against him. 84 The liability of the NAACP derived solely from the liability of Charles Evers.74 The chancellor found: 85 "The national NAACP was well-advised of Evers' actions, and it had the option of repudiating his acts or ratifying them. It never repudiated those acts, and therefore, it is deemed by this Court to have affirmed them." App. to Pet. for Cert. 42b-43b. 86 Of course, to the extent that Charles Evers' acts are insufficient to impose liability upon him, they may not be used to impose liability on his principal. On the present record, however, the judgment against the NAACP could not stand in any event. 87 The associational rights of the NAACP and its members have been recognized repeatedly by this Court.75 The NAACP—like any other organization—of course may be held responsible for the acts of its agents throughout the country that are undertaken within the scope of their actual or apparent authority.76 Cf. American Society of Mechanical Engineers, Inc. v. Hydrolevel Corp., 456 U.S. 556, 102 S.Ct. 1935, 72 L.Ed.2d 330. Moreover, the NAACP may be found liable for other conduct of which it had knowledge and specifically ratified. 88 The chancellor made no finding that Charles Evers or any other NAACP member had either actual or apparent authority to commit acts of violence or to threaten violent conduct. The evidence in the record suggests the contrary. Aaron Henry, president of the Mississippi State Conference of the NAACP and a member of the Board of Directors of the national organization, testified that the statements attributed to Evers were directly contrary to NAACP policy. Record 4930.77 Similarly, there is no evidence that the NAACP ratifieds or even had specific knowledge of—any of the acts of violence or threats of discipline associated with the boycott. Henry testified that the NAACP never authorized, and never considered taking, any official action with respect to the boycott. Id., at 4896. The NAACP supplied no financial aid to the boycott. Id., at 4940. The chancellor made no finding that the national organization was involved in any way in the boycott.78 89 To impose liability without a finding that the NAACP authorized—either actually or apparently—or ratified unlawful conduct would impermissibly burden the rights of political association that are protected by the First Amendment. As Justice Douglas noted in NAACP v. Overstreet, 384 U.S. 118, 86 S.Ct. 1306, 16 L.Ed.2d 409, dissenting from a dismissal of a writ of certiorari found to have been improvidently granted: 90 "To equate the liability of the national organization with that of the Branch in the absence of any proof that the national authorized or ratified the misconduct in question could ultimately destroy it. The rights of political association are fragile enough without adding the additional threat of destruction by lawsuit. We have not been slow to recognize that the protection of the First Amendment bars subtle as well as obvious devices by which political association might be stifled. See Bates v. Little Rock, 361 U.S. 516, 523, 80 S.Ct. 412, 416, 4 L.Ed.2d 480. Thus we have held that forced disclosure of one's political associations is, at least in the absence of a compelling state interest, inconsistent with the First Amendment's guaranty of associational privacy. E.g., DeGregory v. New Hampshire, 383 U.S. 825, 86 S.Ct. 1148, 16 L.Ed.2d 292; Gibson v. Florida Legislative Comm., 372 U.S. 539, 543-546, 83 S.Ct. 889, 892, 9 L.Ed.2d 929; Shelton v. Tucker, 364 U.S. 479, 81 S.Ct. 247, 5 L.Ed.2d 231; N.A.A.C.P. v. Alabama, 357 U.S. 449, 462-463, 78 S.Ct. 1163, 1171, 2 L.Ed.2d 1488. Recognizing that guilt by association is a philosophy alien to the traditions of a free society (see Schware v. Board of Bar Examiners, 353 U.S. 232, 245-246, 77 S.Ct. 752, 759, 1 L.Ed.2d 796) and the First Amendment itself, we have held that civil or criminal disabilities may not be imposed on one who joins an organization which has among its purposes the violent overthrow of the Government, unless the individual joins knowing of the organization's illegal purposes (Wieman v. Updegraff, 344 U.S. 183, 73 S.Ct. 215, 97 L.Ed. 216) and with the specific intention to further those purposes. See Elfbrandt v. Russell, [384 U.S., at] 11 [86 S.Ct., at 1238]; Aptheker v. Secretary of State, 378 U.S. 500, 84 S.Ct. 1659, 12 L.Ed.2d 992." Id., at 122, 86 S.Ct., at 1308. 91 The chancellor's findings are not adequate to support the judgment against the NAACP. IV 92 In litigation of this kind the stakes are high. Concerted action is a powerful weapon. History teaches that special dangers are associated with conspiratorial activity.79 And yet one of the foundations of our society is the right of individuals to combine with other persons in pursuit of a common goal by lawful means.80 93 At times the difference between lawful and unlawful collective action may be identified easily by reference to its purpose. In this case, however, petitioners' ultimate objectives were unquestionably legitimate. The charge of illegality—like the claim of constitutional protection—derives from the means employed by the participants to achieve those goals. The use of speeches, marches, and threats of social ostracism cannot provide the basis for a damages award. But violent conduct is beyond the pale of constitutional protection. 94 The taint of violence colored the conduct of some of the petitioners. They, of course, may be held liable for the consequences of their violent deeds. The burden of demonstrating that it colored the entire collective effort, however, is not satisfied by evidence that violence occurred or even that violence contributed to the success of the boycott. A massive and prolonged effort to change the social, political, and economic structure of a local environment cannot be characterized as a violent conspiracy simply by reference to the ephemeral consequences of relatively few violent acts. Such a characterization must be supported by findings that adequately disclose the evidentiary basis for concluding that specific parties agreed to use unlawful means, that carefully identify the impact of such unlawful conduct, and that recognize the importance of avoiding the imposition of punishment for constitutionally protected activity. The burden of demonstrating that fear rather than protected conduct was the dominant force in the movement is heavy. A court must be wary of a claim that the true color of a forest is better revealed by reptiles hidden in the weeds than by the foliage of countless freestanding trees. The findings of the chancellor, framed largely in the light of two legal theories rejected by the Mississippi Supreme Court, are constitutionally insufficient to support the judgment that all petitioners are liable for all losses resulting from the boycott. 95 The judgment is reversed. The case is remanded for further proceedings not inconsistent with this opinion. 96 It is so ordered. 97 Justice REHNQUIST concurs in the result. 98 Justice MARSHALL took no part in the consideration or decision of this case. APPENDIX TO OPINION OF THE COURT 99 Portions of speech delivered by Charles Evers on April 19, 1969 (Record 1092- 1108): 100 "Thank you very much. We want our white friends here to know what we tell them happens to be so. Thank you for having the courage to walk down those streets with us. We thank you for letting our white brethren know that guns and bullets ain't gonna stop us. (No) (No) We thank you for letting our white brothers know that Port Gibson ain't none of their town. (Amen) (Applause) That Port Gibson is all of our town. (Applause) That black folks, red folks, Chinese and Japanese alike (Yeah) (That's right.), that we are going to have our share. (Yeah, we are.) 101 * * * * * "We are going to beat you because we know you can't trick us no more. (yea) You are not going to be able to fool us by getting somebody to give us a drink of whiskey no more. (Applause) You ain't gonna be able to fool us by somebody giving us a few dollars no more. (Applause) We are gonna take your money and drink with you and then we're gonna (Applause) vote against you. Then we are going to elect a sheriff in this county and a sheriff that is responsible, that won't have to run and grab the telephone and call up the blood-thirsty highway patrol when he gets ready (Yeah) to come in and beat innocent folks down to the ground for no cause. (That's right) (Applause) (Boo) We are going to elect a sheriff that can call his deputies and represent black leaders in the community and stop whatever problem there is. (Yeah) (That's right.) 102 "Then we are going to do more than that. The white merchants of this town are so wrapped up in the power structure here, since you love your Police Department so well, since you support them so well (Yeah), we are going to let them buy your dirty clothes and your filthy, rotten groceries. 103 "Oh, no, white folks, we ain't going to shoot you with no bullet. (That's right.) We are going to shoot you with our ballots and with our bucks. (Yea) (That's right.) We are going to take away from you the thing that you have had over us all these years. (Yeah) Political power and economic power. While you kill our brothers and our sisters and rape our wives and our friends. (Yeah) You're guilty. You're guilty because you don't care a thing about anybody. (Yes.) And when you go and let a big, black burly nigger like you get on the police force (Yea) go down and grab another black brother's arm and hold it while a white racist stole him from us, and he's a liar if he says he didn't hold him. 104 * * * * * 105 "We mean what we are saying. We are not playing. (Right) We better not even think one of us is black. You better not even be caught near one of these stores. (Applause) "We don't want you caught in Piggly-Wiggly. You remember how he grinned at us four years ago? (Yeah) You know how when he took office he grinned at us? (Yeah) He ain't hired nobody yet. (That's right) (No) And you know old Jitney Jungle down there with those funny letters down on the end? (That's right) (Applause) He haven't hired nobody in there yet. (No) Do you know poor ole M & M or whatever it stands for, mud and mush, I guess. (Applause) They're out here on the highway and they haven't hired none of us yet. 106 "Do you know Ellis who had a part-time boy all his life? He ain't hired nobody, is he, yet? (No) Then we got ole Stampley, and ninety-nine and three-fourths of his sales are black folks business. He got the nerve to tell me he ain't gonna put no nigger ringing his cash register. I got news for you, Brother Stampley. You can ring it your damn self. (Extra loud applause.) I want some of you fat cats after this meeting who wants three of our young boys who ain't a'scar'd of white folks (Applause) (Me) and we want you that's willing to follow the rules now to go down by Brother Stampley's and serve notice on him with our placards that we ain't coming no more. 107 "Then we are going to tell all the young men that drive Piggly-Wiggly trucks now (Yeah) (Be careful, Son.) because the soul brothers and the spirit is watching you. (Extra loud applause.) 108 "All right, Brother Wolf, you're next. (Applause) We got a couple of 'em to come down by Brother Wolf's. We mean business, white folks. We ain't gonna shoot you all, we are going to hit you where it hurts most. (In the pocketbook) (Applause) In the pocketbook and in the ballot box. (Applause) We may as well tell our friends at Alcorn to stay away from up here. (Yea) Now, you say, 'What's wrong with you niggers?' I'll tell you what is wrong with us niggers; We are tired of you white folks, you racists and you bigots mistreating us. (Yeah) We are tired of paying you to deny us the right to even exist. (Tell'em about it.) And we ain't coming back, white folks. (We ain't.) 109 "You all put a curfew on us at eight o'clock tonight. We are going to do you better than that. We are going to leave at one-thirty. (Loud applause) We are going to leave at one-thirty and we ain't coming back, white folks. 110 * * * * * 111 "We are going to have Brother McCay; we are going to have our newly elected mayor who we elected, we are going to have him around here, too. Come on back, my dear friend. He say, 'Naw, brother, we ain't coming.' 'Have you got rid of all those bigots you got on your police force?' 'No.' 'Have you hired Negroes in all them stores?' 'No.' 'Well, we ain't coming back.' (Right) That's all we gonna do. You know, what they don't realize is you put on this curfew, that is all we needed. Let me just give them some instructions. We are going to buy gas only from the Negro-owned service stations. We agreed on it, remember? Now, don't back upon your agreement. (Yea) I don't care how many Negroes working on it, that's too bad. We are going only to Negro-owned service stations. And we are going only—the only time you will see us around on this street, now listen good, you are going to Lee's Grocery and other stores on this end. Is that clear? (Yeah) (Applause) 112 "We don't want to go to none of them drugstores. They get us confused. Now, who am I going to get my medicine from? Let us know in time and we will be glad to furnish a car free to carry you anywhere you have to go to get a prescription filled. You can't beat this. (No) It won't cost you a dime. You go to any of the local black businessmen and tell them you have got to go to Vicksburg to get your stuff. And then if they don't carry you, let us know. We'll take care of them later. (Applause) Now, you know, we have got a little song that says, 'This is your thing, do what you want to do.' (Applause) This is our thing, let's do what we want to do with it. Let's make sure now—if you be disobedient now you are going to be in trouble. Remember that, now, listen. Listen good. They are going to start saying, 'You know what, Evers is down there with his goon squad, . . .' Now, we know Claiborne County,—'with his goon squad harassing poor ole niggers.' 113 "Well, good white folks you have been harassing us all our lives. (Applause) And if we decided to harass you that's our business. (That's right) They are our children and we are going to discipline them the way we want to. Now, be sure you get all this right on all these tape recorders. Whatever I say on this trip I will say it in Jackson. (Amen) (Glory) And I will say it in Washington and New York. White folks ain't gonna never control us no more. (Applause) 114 * * * * * 115 "Now, my dear friends, the white folks have got the message. I hope you have got the message and tell every one of our black brothers until all these people are gone, you voted on this in the church, don't let me down, and don't let yourself down. We agreed in the church that we would vacate this town until they have met those requests, the white folks don't demand nothing out of us. All right, white folks, we are just saying until you decide when you want to do these little things we beg of you, we are not coming back. (No way) 116 "None of us better not be caught up here. (Yea) I don't care how old you are, I don't care how sick you are, I don't care how crazy you are, you better not be caught on these streets shopping in these stores until these demands are met. (Applause) 117 "Now, let's get together. Are you for this or against it? (Applause) (For it.) Remember you voted this. We intend to enforce it. You needn't go calling the chief of police, he can't help you none. You needn't go calling the sheriff, he can't help you none. (That's right.) He ain't going to offerto sleep with none of us men, I can tell you that. (Applause) Let's don't break our little rules that you agreed upon here. 118 * * * * * 119 "Let's go to the funeral of our young son whenever the funeral is. I don't want you to come with hate because that is not going to solve our problems. (No hate.) We don't want you to hate the white folks here in Port Gibson. That is not going to solve it. If you hate what they have done, I hate to get personal, I hate what they did so much to Medgar, (I know.) I ain't going to ever stop hating them for that. But I am going to chase them in the way what I know is right and just. I am not going to lay out in the bushes and shoot no white folks. That's wrong. I am not gonna go out here and bomb none of them's home. (No) That's not right. But I am going to do everything in my power to take away all the power, political power, legal power that they possess anywhere I live. We are going to compete against them. When we blacks learn to support and respect each other, then and not until then, will white folks respect us. (Applause) 120 "Now, you know I trust white folks and I mean every word I say. But it comes a time when we got to make up in our mind individually, are we going to make those persons worthwhile. We done talked and raised all kind of sand all day here, now, what is really going to prove it, are we going to live up to what we have said? (Applause) Now if there is any one of us breaks what we agreed upon, you are just as guilty as that little trigger-happy, blood-thirsty rascal. (Tell 'em about it.) 121 * * * * * 122 "I go all over this country, and I ought not to tell you white folks this, and I tell other white folks that some day we are going to get together in Mississippi, black and white, and work out our problems. And we are ready to start whenever you are. If you are ready to start, we are. We ain't going to let you push us, not one inch. (That's right.) If you come on beating us, we are going to fight back. (Right) We got our understanding. We are all God's children. The same man that brought you all here brought us. You could have been black just like we are. We could have been white and baldheaded just like you are. (Laughter) (Inaudible) We are going to work hard at this, Dan. We are going to be organized this time. We ain't going to be bought off and talked off. We are going to elect the county sheriff here this next time that don't need the highway patrol. Now, you see, Dan had a good chance to set himself up right, but he goofed it. He goofed. (Yeah) He blew it. (Laughter) Don't forget that, heah. (Right) It brings back memories like you know you remember things we do. 123 "Now, if you don't think it is necessary, we don't have to go back to the church. If you want to go back there, we can. I want you to make sure here that we are going to leave this town to our white brothers and we ain't coming back no more until all our requests have been met. Is that the common consent of all of you here? (Applause) (Let's go back to the church.) All right. Are we willing to make sure that everyone of us will be sure that none of the rest of our black brothers violate our . . . (Yea) We are all saying it now. Let's not say it now so much on my part. You know, I'm just sort of leading, you know, how these lawyers are, leading our folks on to say what has to be said. And that's the case. Let's make us a white town. We would like for you to start it. Be courteous now. Don't mistreat nobody. Tell them, in a nice forceful way, the curfew is going to be on until they do what we ask them." 1 See Krulewitch v. United States, 336 U.S. 440, 447-449, 69 S.Ct. 716, 720, 93 L.Ed. 790 (concurring opinion). 2 Port Gibson is the county seat and largest municipality in Claiborne County. 3 The affected businesses represented by the merchants included four grocery stores, two hardware stores, a pharmacy, two general variety stores, a laundry, a liquor store, two car dealers, two auto parts stores, and a gas station. Many of the owners of these boycotted stores were civic leaders in Port Gibson and Claiborne County. Respondents Allen and Al Batten were Aldermen in Port Gibson, Record 15111; Robert Vaughan, part owner and operator of one of the boycotted stores, represented Claiborne County in the Mississippi House of Representatives, id., at 15160; respondents Abraham and Hay had served on the school board, id., at 14906, 14678; respondent Hudson served on the Claiborne County Democratic Committee, id., at 840. 4 The complaint also named 52 banks as "attachment defendants." The banks answered that the NAACP had $16,800 on deposit in Mississippi. 5 As a result of the plaintiffs' prayer for an attachment in equity, jurisdiction existed in Chancery Court. The trial judge ruled: "It was incumbent upon this court to hear the case in full once jurisdiction was assumed. To have heard the portions of this matter sounding in equity, only, and to have transferred the questions of tort liability and damages to the circuit court would have been contrary to the maxim 'equity delights to do complete justice, and not by halves.' " App. to Pet. for Cert. 56b. The defendants thus were denied a jury trial on the liability issues. Although the court recognized that it had power to empanel a jury, it declined to exercise its discretion to do so. Ibid. The Mississippi nonresident attachment statute that provided the basis for equitable jurisdiction has since been declared unconstitutional by both Federal District Courts in Mississippi. MPI, Inc. v. McCullough, 463 F.Supp. 887 (ND Miss.1978); Mississippi Chem. Corp. v. Chemical Constr. Corp., 444 F.Supp. 925 (SD Miss.1977). Commencement of trial was delayed by collateral proceedings in federal court. See Henry v. First National Bank of Clarksdale, 50 F.R.D. 251 (ND Miss.1970), rev'd, 444 F.2d 1300 (CA5 1971), cert. denied, 405 U.S. 1019, 92 S.Ct. 1284, 31 L.Ed.2d 483. The District Court entered a preliminary injunction restraining the state proceedings on the theory that the merchants sought to infringe the defendants' First Amendment rights. The Court of Appeals reversed, holding that the mere commencement of a private tort suit did not itself involve "state action" for purposes of 28 U.S.C. § 1343(3). 6 App. to Pet. for Cert. 2g. Of the original 148 named defendants, 16 were dismissed by stipulation of counsel (12 had died, 2 were minors, 1 was non compos mentis, and 1—the Reverend Dominic Cangemi—was dismissed by agreement without explanation). One defendant was dismissed because he had been misidentified in the complaint. The chancellor dismissed one defendant—state NAACP leader Aaron Henry—because "the complainants failed to meet the burden of proof as to [his] wrongdoing." Id., at 28b. Thus, except for the defendants dismissed by stipulation or because of misidentification, the plaintiffs prevailed on the merits in the trial court against all but one of the defendants. 7 Although the bulk of the court's discussion of the defendants' common-law tort liability focused on the presence of a civil conspiracy, the chancellor did not appear to hold that a concerted refusal to deal—without more—was actionable under the common law of Mississippi. The court apparently based its first theory of liability on the ground that the "malicious interference by the defendants with the businesses of the complainants as shown by the evidence in this case is tortious per se, and this would be true even without the element of conspiracy." Id., at 42b (footnote omitted). In Mississippi, "[e]ither an individual or a corporation, whether acting in conjunction with others, or not," may be liable in an action for "malicious interference with a trade or calling." Memphis Laundry-Cleaners v. Lindsey, 192 Miss. 224, 239, 5 So.2d 227, 232 (1941). The chancellor in this case stated that the necessary element of malice is established by proof of "the intentional performance of an act harmful to another without just or lawful cause or excuse." App. to Pet. for Cert. 42b, n. 8. The repeated references to the presence of a conspiracy might be explained by the court's finding that each of the defendants with the exception of Aaron Henry—was jointly and severally liable for the plaintiffs' losses. As noted, an element of the plaintiffs' common-law action was the defendants' intentional performance of an "unprivileged" act harmful to another. The chancellor stated that the evidence clearly established that "certain defendants" had committed "overt acts which were injurious to the trade and business of complainants." Id., at 39b. The court continued: "Where two or more persons conspire together, the conspiracy makes the wrongful act of each person the joint acts of them all," id., at 41b; "[i]t follows that each act done in pursuance of the conspiracy by one of several conspirators is, in contemplation of the law, an act for which each is jointly and severally liable." Ibid. Thus, the presence of a conspiracy rendered all of the "conspirators" liable for the wrongful acts of any member of that conspiracy. 8 See Miss.Code Ann. § 97-23-85 (1972). The chancellor found: "The testimony in the case at bar clearly shows that the principal objective of the boycott was to force the white merchants of Port Gibson and Claiborne County to bring pressure upon governing authorities to grant defendants' demands or, in the alternative, to suffer economic ruin." App. to Pet. for Cert. 51b. As noted, however, many of the merchants themselves were civic leaders. See n.3, supra. 9 See Miss.Code Ann. § 75-21-9 (1972). The court made clear that under this theory intentional participation in the concerted action rendered each defendant directly liable for all resulting damages. "As a legal principle, it is sufficient to show that the concert of action on the part of the defendants was deliberately invited, and that the defendants gave their adherence to the scheme and participated in it." App. to Pet. for Cert. 54b. The same was true of the court's secondary boycott theory; "since an illegal boycott is an invasion of a property right, the members of the boycotting combination are liable for the resulting damages." Id., at 53b. 10 In its discussion of the secondary boycott statute, the court rejected an argument that the statute was unconstitutional under the First and Fourteenth Amendments. Noting as a "basic premise" that "secondary boycotts are unlawful under both United States and Mississippi law," the court stated that "conduct and communication which are illegal are not protected by the constitutional provisions relating to freedom of speech." Id., at 46b. In imposing liability under the state restraint of trade statute, the chancellor added: "After a careful consideration of the constitutional claims of defendants, the Court finds that none of the acts or conduct of defendants was shielded or protected by the Constitution of the United States or the Constitution of the State of Mississippi." Id., at 55b-56b. Finally, in assessing damages, the court stated: "Defendants base their defense on the concept that the right to boycott and inflict losses on complainants was a legally protected right afforded them under the laws and Constitution of the United States. This Court has hereinbefore found that the conduct of the defendants was unlawful and unprotected." Id., at 62b. 11 Id., at 19g. Following the entry of judgment, the defendants moved for relief from Mississippi's 125-percent supersedeas bonding requirement. Although the Mississippi Supreme Court denied the motion, a federal court enjoined execution of the Chancery Court judgment pending appeal. Henry v. First National Bank of Clarksdale, 424 F.Supp. 633 (ND Miss.1976), aff'd, 595 F.2d 291 (CA5 1979), cert. denied, 444 U.S. 1074, 100 S.Ct. 1020, 62 L.Ed.2d 756. 12 393 So.2d, at 1300. 13 Id., at 1301. 14 Ibid. 15 The court did not specifically identify the evidence linking any of the defendants to such an agreement. 16 As noted, liability under the secondary boycott and restraint of trade statutes could be found on the basis of an entirely voluntary and nonviolent agreement to withhold patronage. See n. 9, supra. It is not clear whether—in its imposition of tort liability—the trial court rested on a theory similar to that ultimately advanced by the Mississippi Supreme Court. In finding an unlawful civil conspiracy—which rendered each conspirator liable for the actions of others, see n. 7, supra—the chancellor arguably believed that it was necessary to connect all defendants to an agreement to use force or violence to effectuate the conspiracy. See App. to Pet. for Cert. 40b-41b. The chancellor made no factual finding, however, that such an agreement existed. 17 393 So.2d, at 1302. 18 Concerning the permanent injunction entered by the chancellor, the court stated: "Although the granting of injunction has been assigned as error, the error has not been argued, and NAACP, et al. say, at the conclusion of their brief '. . . the injunctive aspects of the case are now moot. . . .' " Id., at 1293. Despite this finding, the court did not vacate the injunction. 19 Respondents acknowledge that "[t]he basis on which the Mississippi Supreme Court held that petitioners were liable for damages was 'the agreed use of illegal force, violence and threats.' " Supplemental Brief for Respondents 1-2. 20 Respondents argue that anyone "who participates in the decisionmaking functions of an enterprise, with full knowledge of the tactics by which the enterprise is being conducted, manifests his assent to those tactics. . . ." Id., at 2. Respondents thus would impose liability for the managers' failure to act; respondents argue that, despite evidence that boycott "enforcers" caused fear of injury to persons and property, "they were not taken from their posts and replaced by a system of voluntary compliance; there is no evidence that any of the petitioners even admonished them for their enforcement methods; the successful system of paramilitary enforcers on the streets and 'rhetorical' threats of violence by boycott leaders was left in place for the duration." Id., at 5. 21 These groups are not meant to be exclusive. 22 "Once the pattern had been established—warnings to prospective customers, destruction of goods purchased at boycotted stores, public displays of weapons and of military discipline, denunciation of names gathered by the store-watchers, and subsequent violence against the persons and property of boycott breakers—store-watching in Port Gibson became the sort of activity from which a court could reasonably infer an intention to frighten people away from the stores." Id., at 8. 23 App. to Pet. for Cert. 15b. 24 Id., at 10b. 25 Id., at 12b. 26 The petition stated: "We hope it will not be necessary to resort to the kind of peaceful demonstrations and selective buying campaigns which have had to be used in other communities. It takes manpower, time and energy which could be better directed at solving these problems which exist in Port Gibson and Claiborne County by mutual cooperation and efforts at tolerant understanding. "No one likes to have to resort to picketing and other kinds of demonstration—just as no one likes to be the target of this kind of demonstration. But this sort of thing is inevitable unless there can be real progress toward giving all citizens their equal rights. There seems sometimes to be no other alternative. "Objectives of Negro citizens of Port Gibson and Claiborne County are, simply put, to have equality of opportunity, in every aspect of life, and to end the white supremacy which has pervaded community life. This implies many long-range objectives such as participation in decision-making at every level of community, civic, business and political affairs." Id., at 9b. 27 Id., at 13b. 28 Although Evers' speech on April 1, 1966, was not recorded, the chancellor found: "Evers told his audience that they would be watched and that blacks who traded with white merchants would be answerable to him. According to Sheriff Dan McKay, who was present during the speech, Evers told the assembled black people that any 'uncle toms' who broke the boycott would 'have their necks broken' by their own people. Evers' remarks were directed to all 8,000-plus black residents of Claiborne County, and not merely the relatively few members of the Claiborne NAACP." Id., at 17b-18b (footnote omitted). 29 Id., at 22b (emphasis in original). The chancellor also noted that MAP's Board of Directors "did not seek help from local law-enforcement officers, nor did they complain to United States authorities for protection of their cooks from possible reprisals arising from trade with the white merchants"; and that "MAP employees in Claiborne County continued to take an active part in the NAACP activities and to support the boycott by picketing and marching." Id., at 23b. The Mississippi Supreme Court rejected the chancellor's findings and concluded that MAP was not a willing participant in the boycott, thus absolving it from liability. 30 Id., at 25b. One of the respondents awarded the most in damages, Barbara Ellis—a partner in Ellis Variety Store—testified that the store was boycotted from April 1, 1966, until January 27, 1967. On the latter date, the store agreed—apparently at the urging of a biracial committee—to hire a black cashier. Record 1183. The boycott was reimposed on April 17, 1968, after the death of Martin Luther King, Jr., but again was lifted on May 1, 1968. Id., at 1184. The boycott finally was reimposed on April 19, 1969, the day following the shooting of Roosevelt Jackson. Ibid. 31 The officers had gone to Jackson's home to arrest him. A scuffle ensued and Jackson was shot by a white officer allegedly while being held by a black officer. 32 App. to Pet. for Cert. at 27b. 33 Record 1146. The Sheriff of Claiborne County testified: "There were pickets off and on from April, 1966 to 1970." Id., at 1060. When asked to describe "how they conducted themselves, what they did, what they went about doing," he stated: "Most of them carried or either had signs on their shoulders and they walked up and down the streets in front of the stores. They wouldn't always picket the same stores at the same time. At different times they might picket M&M then they would move up and picket Claiborne Hardware down Market Street to other businesses. Most of the time it was teenagers and at the last it was little bitty fellows, as young as about six years old. That was '69 and '70." Ibid. The Sheriff also testified that the boycott was "tight" in April 1966, April 1968, and April 1969. Id., at 1152. 34 Evidence concerning the aims and practices of the "Black Hats" is contradictory. Respondents describe them as a "paramilitary organization." Petitioner Elmo Scott, a member of the group, testified concerning instructions that were given to him: "It was given to the Deacons to give respect to the people that was on the street and, regardless of what they say back to you, for you not to use bad language to them or not to curse them or no kind of way, just talk to them in the right manner of way." Id., at 2985. It is undisputed that the "Black Hats" were formed during the boycott, that members of the organization engaged in "store watching" and other "enforcement" activities, and that some individuals who belonged to the group committed acts of violence. 35 App. to Pet. for Cert. 19b. 36 Id., at 35b. 37 On August 22, 1966, birdshot was fired into the home of James Gilmore, a black man who ignored the boycott. He immediately grabbed a shotgun, leapt into his car, pursued the vehicle from which he believed the shots had come, forced it to the side of the road, and apprehended three young black men who were active supporters of the boycott. They were indicted, tried, and convicted, but the convictions were set aside on appeal. Whitney v. State, 205 So.2d 284 (Miss.1967). Gilmore continued to patronize white merchants after the incident. In June 1966, while Murriel Cullens was having a beer in Wolf's Store, a brick was thrown through the windshield of his parked car. He had been patronizing white merchants and continued to do so thereafter. Record 14049. In November 1966, shotgun pellets were fired into the wall of his mother's home. She had received a number of threatening telephone calls criticizing her for patronizing white stores. She continued to do so after the incident. Id., at 14003. At trial, Laura Cullens testified, in response to a question whether she had been scared: "No indeed. I haven't had a bone in me scared in my life from nobody. And I have always told them, they say, 'You're just an uncle tom.' And I say, 'Well, uncle tom can be blue, black, green or purple or white. If I feel I am in the right, I stand in that right and nobody tells me what to do.' " Id., at 14017. James Bailey, who was a teenager at the time of the incident, testified that he had noticed that an elderly black lady named Willie Butler traded with a white merchant and had groceries delivered to her home. He testified that he destroyed flowers in her garden to punish her for violating the boycott. Id., at 3656. He stated that he acted on his own initiative and that Mrs. Butler continued to trade with the merchant. Id., at 3660, 3741. 38 Id., at 13868. One of his assailants testified that the incident resulted from an automobile accident, rather than the boycott. Id., at 3656. 39 "Preacher White" had died by the time of trial. No witness admitted being present at what respondents' counsel characterized as "the spanking of Preacher White." Id., at 3696. The Port Gibson Chief of Police testified, however, that White had come in and complained that a group of young blacks had pulled his overalls down and whipped him. Id., at 2176. In describing this incident, the chancellor stated that Preacher White "was stripped of his clothing and whipped by a group of young blacks because he refused to honor the boycott." App. to Pet. for Cert. 37b. 40 In describing the "atmosphere of fear" existing during the boycott, the chancellor emphasized the participation of petitioner Rudy Shields. He stated: "Defendant Rudolph J. (Rudy) Shields, formerly of Chicago, was the principal figure in several altercations. He boasted that he was 'the most jailed person in the Claiborne County boycott.' This man was the acknowledged leader of the 'Deacons.' " Id., at 35b. See also Supplemental Brief for Respondents 10-13. The record indicates that Shields was in Port Gibson for approximately eight months during 1966. Record 4993. 41 The chancellor did find—and apparently believed this fact to be significant—that the NAACP provided attorneys to black persons arrested in connection with acts arising from the boycott. App. to Pet. for Cert. 38b. The NAACP provided legal representation to the three black persons arrested in August 1966 following the Gilmore shooting. 42 Although the Mississippi Supreme Court remanded for a recomputation of damages, its judgment is final for purposes of our jurisdiction. See Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 480, 95 S.Ct. 1029, 1038, 43 L.Ed.2d 328. 43 "Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances." U.S.Const., Amdt. 1. First Amendment freedoms are protected by the Fourteenth Amendment from invasion by the States. Edwards v. South Carolina, 372 U.S. 229, 235, 83 S.Ct. 680, 683, 9 L.Ed.2d 697. 44 Specifically, petitioner contended that respondent "aroused the fears of the local white residents that Negroes were coming into the area and then, exploiting the reactions and emotions so aroused, was able to secure listings and sell homes to Negroes." 402 U.S., at 416, 91 S.Ct., at 1576. 45 One of petitioner's officers testified at trial that he had hoped that respondent would be induced to sign the no-solicitation agreement by letting "his neighbors know what he was doing to us." Id., at 417, 91 S.Ct., at 1576. 46 See Watts v. United States, 394 U.S. 705, 708, 89 S.Ct. 1399, 1401, 22 L.Ed.2d 664 ("The language of the political arena, like the language used in labor disputes, see Linn v. United Plant Guard Workers of America, 383 U.S. 53, 58, 86 S.Ct. 657, 660, 15 L.Ed.2d 582 (1966), is often vituperative, abusive, and inexact"). See also Cohen v. California, 403 U.S. 15, 91 S.Ct. 1780, 29 L.Ed.2d 284; Farber, Commercial Speech and First Amendment Theory, 74 Nw.U.L.Rev. 372 (1979). 47 "To characterize the quality of the governmental interest which must appear, the Court has employed a variety of descriptive terms: compelling; substantial; subordinating; paramount; cogent; strong. Whatever imprecision inheres in these terms, we think it clear that a government regulation is sufficiently justified if it is within the constitutional power of the Government; if it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest." 391 U.S., at 376-377, 88 S.Ct., at 1678 (footnotes omitted). 48 In NAACP v. Alabama ex rel. Flowers, 377 U.S. 288, 84 S.Ct. 1302, 12 L.Ed.2d 325, the Court unanimously rejected Alabama's effort to oust the NAACP from that State. The State claimed, in part, that the NAACP was " 'engaged in organizing, supporting and financing an illegal boycott' " of Montgomery's bus system. Id., at 302, 84 S.Ct., at 1311. Writing for the Court, Justice Harlan described as "doubtful" the "assumption that an organized refusal to ride on Montgomery's buses in protest against a policy of racial segregation might, without more, in some circumstances violate a valid state law." Id., at 307, 84 S.Ct., at 1313. In Missouri v. National Organization for Women, Inc., 620 F.2d 1301, 1317 (CA8 1980), cert. denied, 449 U.S. 842, 101 S.Ct. 122, 66 L.Ed.2d 49, Judge Stephenson stated that "the right to petition is of such importance that it is not an improper interference [under state tort law] even when exercised by way of a boycott." 49 We need not decide in this case the extent to which a narrowly tailored statute designed to prohibit certain forms of anticompetitive conduct or certain types of secondary pressure may restrict protected First Amendment activity. No such statute is involved in this case. Nor are we presented with a boycott designed to secure aims that are themselves prohibited by a valid state law. See Hughes v. Superior Court, 339 U.S. 460, 70 S.Ct. 718, 94 L.Ed. 985. 50 "This Court's duty is not limited to the elaboration of constitutional principles; we must also in proper cases review the evidence to make certain that those principles have been constitutionally applied. This is such a case, particularly since the question is one of alleged trespass across 'the line between speech unconditionally guaranteed and speech which may legitimately be regulated.' Speiser v. Randall, 357 U.S. 513, 525, 78 S.Ct. 1332, 1341, 2 L.Ed.2d 1460. In cases where that line must be drawn, the rule is that we 'examine for ourselves the statements in issue and the circumstances under which they were made to see . . . whether they are of a character which the principles of the First Amendment, as adopted by the Due Process Clause of the Fourteenth Amendment, protect.' Pennekamp v. Florida, 328 U.S. 331, 335, 66 S.Ct. 1029, 1031, 90 L.Ed. 1295; see also One, Inc. v. Olesen, 355 U.S. 371, 78 S.Ct. 364, 2 L.Ed.2d 352; Sunshine Book Co. v. Summerfield, 355 U.S. 372, 78 S.Ct. 365, 2 L.Ed.2d 352. We must 'make an independent examination of the whole record,' Edwards v. South Carolina, 372 U.S. 229, 235, 83 S.Ct. 680, 683, 9 L.Ed.2d 697, so as to assure ourselves that the judgment does not constitute a forbidden intrusion on the field of free expression." New York Times Co. v. Sullivan, 376 U.S. 254, 285, 84 S.Ct. 710, 728, 11 L.Ed.2d 686. 51 Although this is a civil lawsuit between private parties, the application of state rules of law by the Mississippi state courts in a manner alleged to restrict First Amendment freedoms constitutes "state action" under the Fourteenth Amendment. New York Times Co. v. Sullivan, supra, at 265, 84 S.Ct., at 718. 52 See also Carroll v. Princess Anne, 393 U.S. 175, 184, 89 S.Ct. 347, 353, 21 L.Ed.2d 325; Keyishian v. Board of Regents, 385 U.S. 589, 604, 87 S.Ct. 675, 684, 17 L.Ed.2d 629. 53 See United States v. Robel, 389 U.S. 258, 88 S.Ct. 419, 19 L.Ed.2d 508; Elfbrandt v. Russell, 384 U.S. 11, 86 S.Ct. 1238, 16 L.Ed.2d 321; Aptheker v. Secretary of State, 378 U.S. 500, 84 S.Ct. 1659, 12 L.Ed.2d 992. 54 "Strictissimi juris." 367 U.S., at 299, 81 S.Ct., at 1521. 55 In Rizzo v. Goode, 423 U.S. 362, 96 S.Ct. 598, 46 L.Ed.2d 561, the Court vacated an injunction, directed against an entire police department, that had resulted from 20 specific incidents of police misconduct. The Court held that such collective responsibility should be limited to instances in which a concerted design existed to accomplish a wrongful objective. Id., at 373-376, 96 S.Ct., at 605. 56 Of course, the question whether an individual may be held liable for damages merely by reason of his association with others who committed unlawful acts is quite different from the question whether an individual may be held liable for unlawful conduct that he himself authorized or incited. See infra, at 925-926. 57 It is noteworthy that the portion of the chancellor's opinion discussing damages begins by referring expressly to the two theories of liability that the Mississippi Supreme Court rejected: "The complainants proved, in this record, that they suffered injury to their respective businesses as the direct and proximate result of the unlawful secondary boycott and the defendants' actions in restraining trade, all of which was accomplished by defendants through a conspiracy." App. to Pet. for Cert. 57b (footnote omitted). In a footnote, the chancellor added that "any kind of boycott is unlawful if executed with force or violence or threats." Id., at 57b, n. 21. 58 393 So.2d, at 1307. 59 Id., at 1302 (emphasis added). 60 Id., at 1300 (quoting trial court; see App. to Pet. for Cert. 39b). 61 393 So.2d, at 1300 (emphasis added). 62 The testimony of Julia Johnson—although itself only a small portion of a massive record—perhaps best illustrates this point: "Q. How did you observe the boycott? "A. I just stayed out of the stores, because I had my own personal reasons to stay out of the stores. There were some things I really wanted, and the things I wanted were the right to vote, the right to have a title—Mrs. or Mr. or whatever I am, and not uncle or aunt, boy or girl. So that's what I wanted. And if I wanted a job—a qualified job, I wanted to have the opportunity to be hired. Not hired because I'm black or white, but just hired. "Q. And this was your reason for observing the boycott? "A. Yes, it was. "Q. And you were in favor of the boycott? "A. Yes, I was in favor of the boycott. "Q. And it wasn't because somebody threatened you? "A. No, it wasn't because nobody threatened me. "Q. You weren't afraid? "A. Was I afraid? "Q. Yes. "A. No, I was not afraid." Record 15476. It is clear that losses were sustained because persons like Julia Johnson "wanted justice and equal opportunity." Id., at 6864 (testimony of Margaret Liggins). See id., at 6737, 12419, 13543-13544. 63 It is also noteworthy that virtually every victim of the acts of violence found by the chancellor testified that he or she continued to patronize the white merchants. See supra, at 904, and n. 37. 64 In Mine Workers v. Gibbs, 383 U.S. 715, 86 S.Ct. 1130, 16 L.Ed.2d 218, the Court stated that if "special facts" such as those presented in Meadowmoor "appeared in an action for damages after picketing marred by violence had occurred," they might "support the conclusion that all damages resulting from the picketing were proximately caused by its violent component or by the fear which that violence engendered." 383 U.S., at 731-732, 86 S.Ct., at 1141. 65 As described by the Court: "Witnesses testified to more than fifty instances of windowsmashing; explosive bombs caused substantial injury to the plants of Meadowmoor and another dairy using the vendor system and to five stores; stench bombs were dropped in five stores; three trucks of vendors were wrecked, seriously injuring one driver, and another was driven into a river; a store was set on fire and in large measure ruined; two trucks of vendors were burned; a storekeeper and a truck driver were severely beaten; workers at a dairy which, like Meadowmoor, used the vendor system were held with guns and severely beaten about the head while being told 'to join the union'; carloads of men followed vendors' trucks, threatening the drivers, and in one instance shot at the truck and driver." 312 U.S., at 291-292, 61 S.Ct., at 554. 66 See n. 62, supra. 67 For the same reasons, the permanent injunction entered by the chancellor must be dissolved. Since the boycott apparently has ended, the Mississippi Supreme Court may wish to vacate the entire injunction on the ground that it is no longer necessary; alternatively, the injunction must be modified to restrain only unlawful conduct and the persons responsible for conduct of that character. 68 See Record 1172. The strongest evidence of wrongdoing at the meetings was presented by petitioner Marjorie Brandon, who served at times as the local NAACP secretary. She testified that "in the meetings there were statements saying that you would be dealt with" if found trading in boycotted stores. Id., at 5637. She stated that she understood "dealt with" to mean "they would take care of you, do something to you, if you were caught going in." Ibid. Her testimony does not disclose who made the statements, how often they were made, or that they were in any way endorsed by others at the meetings. A massive damages judgment may not be sustained on the basis of this testimony; the fact that certain anonymous persons made such statements at some point during a 7-year period is insufficient to establish that the Association itself possessed unlawful aims or that any petitioner specifically intended to further an unlawful goal. 69 A legal duty to "repudiate"—to disassociate oneself from the acts of another—cannot arise unless, absent the repudiation, an individual could be found liable for those acts. As our decisions in Scales, Noto, and Healy make clear, see supra, at 920, civil liability may not be imposed merely because an individual belonged to a group, some members of which committed acts of violence. The chancellor in this case made no finding that the individuals who committed those acts of violence were "agents" or "servants" of those who attended the NAACP meetings; certainly such a relationship cannot be found simply because both shared certain goals. Cf. General Building Contractors Assn. v. Pennsylvania, 458 U.S. 375, 391-395, 102 S.Ct. 3141, 3150-3152, 73 L.Ed.2d 835. 70 "The most stringent protection of free speech would not protect a man in falsely shouting fire in a theatre and causing a panic." 249 U.S., at 52, 39 S.Ct., at 249. 71 In Watts v. United States, 394 U.S. 705, 89 S.Ct. 1399, 22 L.Ed.2d 664, the petitioner was convicted of willfully making a threat to take the life of the President. During a public rally at the Washington Monument, petitioner stated in a small discussion group: " 'They always holler at us to get an education. And now I have already received my draft classification as 1-A and I have got to report for my physical this Monday coming. I am not going. If they ever make me carry a rifle the first man I want to get in my sights is L.B.J.' " Id., at 706, 89 S.Ct., at 1400. This Court summarily reversed. The Court agreed with the petitioner that the statement, taken in context, was "a kind of very crude offensive method of stating a political opposition to the President." Id., at 708, 89 S.Ct., at 1401. 72 There is evidence that Evers occasionally served as a "store watcher," but there is no suggestion that anything improper occurred on those occasions. 73 See n. 69, supra. 74 Indeed it is noteworthy that Aaron Henry—who was president of the Mississippi State Conference of the NAACP, president of the Coahoma County Branch of the NAACP, and a member of the Board of Directors of the national NAACP—was the only defendant dismissed by the chancellor on the merits. 75 Cf. NAACP v. Alabama ex rel. Patterson, 357 U.S. 449, 78 S.Ct. 1163, 2 L.Ed.2d 1488; Bates v. Little Rock, 361 U.S. 516, 80 S.Ct. 412, 4 L.Ed.2d 480; Louisiana ex rel. Gremillion v. NAACP, 366 U.S. 293, 81 S.Ct. 1333, 6 L.Ed.2d 301; NAACP v. Button, 371 U.S. 415, 83 S.Ct. 328, 9 L.Ed.2d 405; Gibson v. Florida Legislative Investigation Committee, 372 U.S. 539, 83 S.Ct. 889, 9 L.Ed.2d 929; NAACP v. Alabama ex rel. Flowers, 377 U.S. 288, 84 S.Ct. 1302, 12 L.Ed.2d 325. 76 There is no question that Charles Evers—as its only paid representative in Mississippi—was an agent of the NAACP. 77 In a footnote to his discussion of the NAACP's liability, the chancellor wrote: "Aaron E. Henry, a prominent black leader in the State of Mississippi, who was president of the Mississippi State Conference of the NAACP, president of the Coahoma County Branch of the NAACP, and a member of the Board of Directors of the national NAACP, testified that the NAACP 'absolutely did not approve of the way the boycott was being conducted in Port Gibson.' There is also evidence in the record tending to show that Evers was called to account by the national NAACP because of the manner in which the boycott was conducted. However, the NAACP took no action whatever to curb Evers' activities in this connection." App. to Pet. for Cert. 42b, n. 9. Henry's testimony concerning Evers' having been "called to account by the National NAACP" concerned Evers' failure to make proper reports and Henry's understanding that there was a personality clash between Evers and an executive of the NAACP. Record 4905, 4907. We have found no evidence in the record that any representative of the national NAACP was advised of any facts concerning the manner in which the Port Gibson boycott was conducted. 78 The chancellor did find that the NAACP had posted bond and provided legal representation for arrested boycott violators. Since the NAACP regularly provides such assistance to indigent black persons throughout the country, this finding cannot support a determination that the national organization was aware of, and ratified, unauthorized violent conduct. Counsel for respondents does not contend otherwise. 79 In discussing the doctrine of criminal conspiracy, Justice Jackson noted: "The crime comes down to us wrapped in vague but unpleasant connotations. It sounds historical undertones of treachery, secret plotting and violence on a scale that menaces social stability and the security of the state itself. 'Privy conspiracy' ranks with sedition and rebellion in the Litany's prayer for deliverance. Conspiratorial movements do indeed lie back of the political assassination, the coup d'etat, the putsch, the revolution, and seizures of power in modern times, as they have in all history." Krulewitch v. United States, 336 U.S., at 448, 69 S.Ct., at 720 (concurring opinion). 80 "The most natural privilege of man, next to the right of acting for himself, is that of combining his exertions with those of his fellow creatures and of acting in common with them. The right of association therefore appears to me almost as inalienable in its nature as the right of personal liberty. No legislator can attack it without impairing the foundations of society." 1 A. de Tocqueville, Democracy in America 203 (P. Bradley ed. 1954).

23

458 U.S. 966 102 S.Ct. 3469 73 L.Ed.2d 1273 BOARD OF EDUCATION OF ROGERS, ARKANSAS, et al.v.Pete McCLUSKEY, by his next friend, Sally McCluskey. No. 81-1577. July 2, 1982. Rehearing Denied Sept. 9, 1982. See 458 U.S. 1132, 103 S.Ct. 16. PER CURIAM. 1 Respondent, a 10th-grade student in the Rogers, Ark. School District, left school on October 21, 1980, after the first period without permission, and, with four other students, consumed alcohol and became intoxicated. When he returned to school later that day to go on a band trip, he was notified that he was suspended from school. His parents were notified the next day that their son had been suspended pending a hearing before the Rogers School Board; a hearing was scheduled for October 29. At the hearing before the Board, none of the five students denied that they had been drinking, and the Board voted to expel all five for the remainder of the semester. 2 Respondent immediately sought injunctive relief under 42 U.S.C. § 1983 (1976 ed., Supp.IV), and the case was heard by the United States District Court for the Western District of Arkansas on December 4. The District Court decided that the School Board had violated respondent's right to substantive due process, and ordered that he be granted credit for the semester during which he was suspended and that all references to his suspension be expunged from his school records. 3 The District Court's action was based on its interpretation of the School Board's rules and its conclusions concerning which rules the Board invoked in suspending respondent. There is no doubt that the Board had the authority to suspend respondent under §§ 9 and 10 of its written Policies on Pupil Suspension. Section 9 provides that the Board may suspend or expel any student "for good cause." Section 10 defines "good cause," and provides that it includes "sale, use or possession of alcoholic beverages or illegal drugs." Thus it was clearly within the Board's discretion to suspend a student for becoming intoxicated. 4 The District Court decided that the Board had acted under § 11 of its rules, which provides for mandatory suspension when it applies. Section 11 provides: "For the protection of other pupils in the school grades 9-12, the school board shall expel for the remainder of the semester with loss of credit for the semester's work any pupil whenever it has been established to the satisfaction of the board, or the superintendent, or the principal, or his assistant in charge, that the pupil has on school premises or at school sponsored activities (including trips) used, sold, been under the influence of, or been in possession of narcotics or other hallucinogenics, drugs, or controlled substances classified as such by Act 590 of 1971, as amended." 5 There was conflicting testimony concerning which section the Board had invoked. The letters sent to respondent's parents informing them of the suspension and the hearing cited both § 10 and § 11. Adams, a Board member and a lawyer, testified that he based his motion to expel McCluskey on § 10 because he had doubts about the applicability of § 11. The Chairman of the Board testified that the Board had suspended students under § 11 for alcohol offenses for the past five years. 6 The District Court found as a matter of fact that the Board acted under § 11 when it suspended respondent. It then went on to decide that § 11 did not apply to alcohol.* Section 11 applies to "narcotics or other hallucinogenics, drugs, or controlled substances classified as such by Act 590 of 1971, as amended." Act 590, Ark.Stat. Ann. § 82-2602(e) (Supp.1981), specifically exempts alcohol from its coverage; therefore, alcohol is not a "controlled substance." Nor is it a "narcotic or other hallucinogenic." The District Court also concluded that alcohol is not a "drug." While technically alcohol is a drug, the District Court noted, it is not considered a drug in common parlance. For this reason, the District Court concluded, the Board had acted unreasonably by suspending respondent under § 11. It held that the Board violated substantive due process by suspending him under the mandatory terms of § 11, even though the Board had discretion to suspend him under § 10. 7 A divided Court of Appeals for the Eighth Circuit affirmed. 662 F.2d 1263 (1981). It reviewed the District Court's conclusion that the Board acted under § 11 rather than § 10 under the clearly-erroneous standard of Federal Rule of Civil Procedure 52(a), and held that the District Court's conclusion passed muster. It also affirmed the District Court's holding that § 11 cannot reasonably be interpreted to apply to alcohol because "the express terms of section 11 apply only to 'drugs' and expressly exempt alcohol." 662 F.2d, at 1267. For this reason, the Court of Appeals concluded, Wood v. Strickland, 420 U.S. 308, 95 S.Ct. 992, 43 L.Ed.2d 214 (1975), was distinguishable. There this Court had stated that "§ 1983 does not extend the right to relitigate in federal court evidentiary questions arising in school disciplinary proceedings or the proper construction of school regulations." Id., at 326, 95 S.Ct., at 1003. Although this Court had plainly stated that federal courts were not authorized to construe school regulations, the Court of Appeals concluded that Wood v. Strickland was distinguishable because the school board in that case had construed its regulations reasonably while here the Board had construed its regulations unreasonably. 662 F.2d, at 1267. Judge McMillian dissented because he concluded that Wood v. Strickland barred federal courts from construing the school regulations involved in this case differently than the Board had construed them. 8 Wood v. Strickland plainly requires that the Court of Appeals be reversed. There high school girls were expelled for "spiking" a punch served at a school meeting by adding two bottles of malt liquor. The malt liquor had an alcoholic content of 3.2% and the alcoholic content of the spiked punch was estimated at 0.91%. 420 U.S., at 326, 95 S.Ct., at 1003. The Court of Appeals had set aside the girls' expulsions because they had been expelled for adding an alcoholic beverage to the punch, but a state statute defined "intoxicating liquor" as a beverage with an alcoholic content exceeding 5%, and the court thought the 5% rule of the statute should apply to the school regulation. We held that the court erred in substituting its own notions for the school board's definition of its rules: 9 "[T]he Court of Appeals was ill advised to supplant the interpretation of the regulation of those officers who adopted it and are entrusted with its enforcement." Id., at 325, 95 S.Ct., at 1002. 10 The Court continued, as noted supra, by stating that 11 "§ 1983 does not extend the right to relitigate in federal court evidentiary questions arising in school disciplinary proceedings or the proper construction of school regulations." Id., at 326, 95 S.Ct., at 1003. (emphasis added). 12 The Court of Appeals and the District Court plainly erred in distinguishing Wood v. Strickland on the ground that the Board's interpretation of § 11 in this case was unreasonable while the school board's construction of "alcoholic beverage" in Wood v. Strickland was reasonable. A case may be hypothesized in which a school board's interpretation of its rules is so extreme as to be a violation of due process, but this is surely not that case. The Board's interpretation of § 11 is reasonable. Contrary to the Court of Appeals, alcohol is not expressly exempted from the coverage of § 11. Section 11 covers "controlled substances classified as such by Act 590," and Act 590 expressly exempts alcohol from its coverage. Therefore, alcohol is not a "controlled substance" under § 11. But § 11 also covers "drugs," and, as the District Court conceded, alcohol is a "drug." Moreover, § 11 mandates suspension of students under the influence of drugs while on school premises. Section 10, which gives the Board discretion to suspend students for drug use, is not limited in its application to drug use on school premises. It is reasonable to conclude that the regulations require suspension for any drug use, including use of alcohol, on school premises, while permitting suspension for drug use off school premises. 13 In any case, even if the District Court's and the Court of Appeals' views of § 11 struck us as clearly preferable to the Board's—which they do not—the Board's interpretation of its regulations controls under Wood v. Strickland. The Chairman of the Board testified that the Board had interpreted § 11 as requiring the suspension of students found intoxicated on school grounds for a number of years prior to respondent's suspension, and it is undisputed that the Board had the authority to suspend students for that reason. We conclude that the District Court and the Court of Appeals plainly erred in replacing the Board's construction of § 11 with their own notions under the facts of this case. Accordingly, the petition for certiorari is granted, and the judgment of the Court of Appeals is 14 Reversed. 15 Justice STEVENS, with whom Justice BRENNAN and Justice MARSHALL join, dissenting. 16 As Justice REHNQUIST has reminded us, in "our zeal to provide 'equal justice under law,' we must never forget that this Court is not a forum for the correction of errors." Boag v. MacDougall, 454 U.S. 364, 367-368, 102 S.Ct. 700, 702, 70 L.Ed.2d 551 (dissenting opinion). "To remain effective, the Supreme Court must continue to decide only those cases which present questions whose resolution will have immediate importance far beyond the particular facts and parties involved."1 This case illustrates how ineffectively the Court is supervising its discretionary docket. 17 The case is not of sufficient importance to warrant full briefing and argument. It is not worthy of an opinion signed by a Member of this Court. The disposition is explained by an anonymous author writing "per curiam"—that is to say, "for the Court." In ever-increasing numbers, appeals throughout the federal system are being decided in this anonymous fashion. It is not uncommon for courts of appeals to issue opinions that are not to be cited as authority in other cases.2 In one recent published case—which was sufficiently important to induce this Court to grant certiorari even before a conflict in the circuits had developed—the court purported to justify such an ad hoc adjudication by asserting that it lacked "precedential character."3 The threat to the quality of our work that is presented by the ever-increasing impersonalization and bureaucratization of the federal judicial system is far more serious than is generally recognized. Regrettably the example set by this Court in cases of this kind is not one of resistance, but rather of encouragement, to the rising administrative tide. 18 We are far too busy to correct every error that we perceive among the thousands of cases that litigants ask us to review. In recent years, when we have exercised our discretionary jurisdiction and issued per curiam rulings deciding cases summarily, we have most frequently come to the aid of a prosecutor or a warden who has been rebuffed by another court.4 Today we exercise our majestic power to enforce a School Board's suspension of a 10th-grade student who consumed too much alcohol on October 21, 1980. 19 If the student had been unjustly suspended, I wonder if the Court would consider the matter of sufficient national importance to require summary reversal. I doubt it. 20 I respectfully dissent. * The Board has since amended its regulations so as to remove all question that suspension for the remainder of the semester is mandatory if a student is intoxicated on school premises. 1 Address of Chief Justice Vinson before the American Bar Association, September 7, 1949 (quoted in R. Stern & E. Gressman, Supreme Court Practice 258 (5th ed. 1978)). 2 See Reynolds & Richman, The Non-Precedential Precedent Limited Publication and No-Citation Rules in the United States Courts of Appeals, 78 Colum.L.Rev. 1167 (1978); Note, Unreported Decisions in the United States Courts of Appeals, 63 Cornell L.Rev. 128 (1977). 3 Rowley v. Board of Education of Hendrick Hudson Central School District, 632 F.2d 945, 948 n.7 (CA2 1980), rev'd 458 U.S. 176, 102 S.Ct. 3034, 73 L.Ed.2d 690. 4 In this Term, see Duckworth v. Serrano, 454 U.S. 1, 102 S.Ct. 18, 70 L.Ed.2d 1 (1981); Jago v. Van Curen, 454 U.S. 14, 102 S.Ct. 31, 70 L.Ed.2d 13 (1981); Leeke v. Timmerman, 454 U.S. 83, 102 S.Ct. 69, 70 L.Ed.2d 65 (1981); California ex rel. Cooper v. Mitchell Brothers' Santa Ana Theater, 454 U.S. 90, 102 S.Ct. 172, 70 L.Ed.2d 262 (1981); Harris v. Rivera, 454 U.S. 339, 102 S.Ct. 460, 70 L.Ed.2d 530 (1981); Hutto v. Davis, 454 U.S. 370, 102 S.Ct. 703, 70 L.Ed.2d 556 (1982); Wainwright v. Torna, 455 U.S. 586, 102 S.Ct. 1300, 71 L.Ed.2d 475 (1982); Sumner v. Mata, 455 U.S. 591, 102 S.Ct. 1303, 71 L.Ed.2d 480 (1982); Fletcher v. Weir, 455 U.S. 603, 102 S.Ct. 1309, 71 L.Ed.2d 490 (1982); United States v. Hollywood Motor Car Co., 458 U.S. 263, 102 S.Ct. 3081, 73 L.Ed.2d 754; Michigan v. Thomas, 458 U.S. 259, 102 S.Ct. 3079, 73 L.Ed.2d 750. But see Boag v. MacDougall, 454 U.S. 364, 102 S.Ct. 700, 70 L.Ed.2d 551 (1982). It certainly cannot be said that egregious error is presented only in cases in which prosecutors and wardens seek review. See, e.g., McKinney v. Estelle, 657 F.2d 740 (CA5 1981), cert. denied 456 U.S. 937, 102 S.Ct. 1994, 72 L.Ed.2d 456; Tejeda-Mata v. INS, 626 F.2d 721 (CA9 1980), cert. denied 456 U.S. 994, 102 S.Ct. 2280, 73 L.Ed.2d 1291.

78

458 U.S. 782 102 S.Ct. 3368 73 L.Ed.2d 1140 Earl ENMUND, Petitionerv.FLORIDA. No. 81-5321. Argued March 23, 1982. Decided July 2, 1982. Syllabus Petitioner and a codefendant, at a jury trial in a Florida court, were convicted of first-degree murder and robbery of two elderly persons at their farmhouse, and were sentenced to death. The Florida Supreme Court affirmed. The court held that, although the record supported no more than the inference that petitioner was the person in a car parked by the side of the road near the farmhouse at the time of the killings waiting to help the robbers and killers (the codefendant and another) escape, this was enough under Florida law to make petitioner a constructive aider and abettor and hence a principal in first-degree murder upon whom the death penalty could be imposed. It was thus irrelevant to petitioner's challenge to the death sentence that he did not himself kill and was not present at the killings, or whether he intended that the victims be killed or anticipated that lethal force might be used to effectuate the robbery or escape. Held : The imposition of the death penalty upon petitioner is inconsistent with the Eighth and Fourteenth Amendments. Pp. 788-801. (a) The current judgments of legislatures, juries, and prosecutors weigh heavily on the side of rejecting capital punishment for the crime at issue. Only a small minority of States—eight—allow the death penalty to be imposed solely because the defendant somehow participated in the robbery in the course of which a murder was committed, but did not take or attempt or intend to take life, or intend that lethal force be employed. And the evidence is overwhelming that American juries have repudiated imposition of the death penalty for crimes such as petitioner's, the statistics demonstrating that juries—and perhaps prosecutors consider death a disproportionate penalty for those who fall within petitioner's category. Pp. 788-796. (b) While robbery is a serious crime deserving serious punishment, it is not a crime "so grievous an affront to humanity that the only adequate response may be the penalty of death." Gregg v. Georgia, 428 U.S. 153, 184, 96 S.Ct. 2909, 2930, 49 L.Ed.2d 859. The death penalty, which is "unique in its severity and irrevocability," id., at 187, 96 S.Ct., at 2931, is an excessive penalty for the robber, who, as such, does not take human life. Here, the focus must be on petitioner's culpability, not on those who committed the robbery and killings. He did not kill or intend to kill and thus his culpability is different from that of the robbers who killed, and it is impermissible for the State to treat them alike and attribute to petitioner the culpability of those who killed the victims. Pp. 797-798. (c) Neither deterrence of capital crimes nor retribution is a sufficient justification for executing petitioner. It is unlikely that the threat of the death penalty for murder will measurably deter one such as petitioner who does not kill or intend to kill. As to retribution, this depends on the degree of petitioner's culpability, which must be limited to his participation in the robbery. Putting him to death to avenge two killings that he did not commit or intend to commit or cause would not measurably contribute to the retribution end of ensuring that the criminal gets his just deserts. Pp. 798-801. 399 So.2d 1362, (Fla.) reversed and remanded. James S. Liebman, New York City, for petitioner, pro hac vice, by special leave of Court. Lawrence A. Kaden, Tallahassee, Fla., for respondent, pro hac vice, by special leave of Court. Justice WHITE delivered the opinion of the Court. 1 * The facts of this case, taken principally from the opinion of the Florida Supreme Court, are as follows. On April 1, 1975, at approximately 7:45 a. m., Thomas and Eunice Kersey, aged 86 and 74, were robbed and fatally shot at their farmhouse in central Florida. The evidence showed that Sampson and Jeanette Armstrong had gone to the back door of the Kersey house and asked for water for an overheated car. When Mr. Kersey came out of the house, Sampson Armstrong grabbed him, pointed a gun at him, and told Jeanette Armstrong to take his money. Mr. Kersey cried for help, and his wife came out of the house with a gun and shot Jeanette Armstrong, wounding her. Sampson Armstrong, and perhaps Jeanette Armstrong, then shot and killed both of the Kerseys, dragged them into the kitchen, and took their money and fled. 2 Two witnesses testified that they drove past the Kersey house between 7:30 and 7:40 a. m. and saw a large cream- or yellow-colored car parked beside the road about 200 yards from the house and that a man was sitting in the car. Another witness testified that at approximately 6:45 a. m. he saw Ida Jean Shaw, petitioner's common-law wife and Jeanette Armstrong's mother, driving a yellow Buick with a vinyl top which belonged to her and petitioner Earl Enmund. Enmund was a passenger in the car along with an unidentified woman. At about 8 a. m. the same witness saw the car return at a high rate of speed. Enmund was driving, Ida Jean Shaw was in the front seat, and one of the other two people in the car was lying down across the back seat. 3 Enmund, Sampson Armstrong, and Jeanette Armstrong were indicted for the first-degree murder and robbery of the Kerseys. Enmund and Sampson Armstrong were tried together.1 The prosecutor maintained in his closing argument that "Sampson Armstrong killed the old people." Record 1577. The judge instructed the jury that "[t]he killing of a human being while engaged in the perpetration of or in the attempt to perpetrate the offense of robbery is murder in the first degree even though there is no premeditated design or intent to kill." App. 6. He went on to instruct them that 4 "[i]n order to sustain a conviction of first degree murder while engaging in the perpetration of or in the attempted perpetration of the crime of robbery, the evidence must establish beyond a reasonable doubt that the defendant was actually present and was actively aiding and abetting the robbery or attempted robbery, and that the unlawful killing occurred in the perpetration of or in the attempted perpetration of the robbery." Id., at 9. 5 The jury found both Enmund and Sampson Armstrong guilty of two counts of first-degree murder and one count of robbery. A separate sentencing hearing was held and the jury recommended the death penalty for both defendants under the Florida procedure whereby the jury advises the trial judge whether to impose the death penalty. See Fla.Stat. § 921.141(2) (1981). The trial judge then sentenced Enmund to death on the two counts of first-degree murder. Enmund appealed, and the Florida Supreme Court remanded for written findings as required by Fla.Stat. § 921.141(3) (1981). The trial judge found four statutory aggravating circumstances: the capital felony was committed while Enmund was engaged in or was an accomplice in the commission of an armed robbery, Fla.Stat. § 921.141(5)(d) (1981); the capital felony was committed for pecuniary gain, § 921.141(5)(f); it was especially heinous, atrocious, or cruel, § 921.141(5)(h); and Enmund was previously convicted of a felony involving the use or threat of violence, § 921.141(5)(b). 399 So.2d 1362, 1371-1372 (Fla.1981). The court found that "none of the statutory mitigating circumstances applied" to Enmund and that the aggravating circumstances outweighed the mitigating circumstances. Id., at 1372. Enmund was therefore sentenced to death on each of the murder counts. 6 The Florida Supreme Court affirmed Enmund's conviction and sentences. It found that "[t]here was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered." Id., at 1370. However, it rejected petitioner's argument that at most he could be found guilty of second-degree murder under Florida's felony-murder rule. The court explained that the interaction of the " 'felony murder rule and the law of principals combine to make a felon generally responsible for the lethal acts of his co-felon.' " Id., at 1369, quoting Adams v. State, 341 So.2d 765, 768-769 (Fla.1976), cert. denied, 434 U.S. 878, 98 S.Ct. 232, 54 L.Ed.2d 158 (1977). Although petitioner could be convicted of second-degree murder only if he were an accessory before the fact rather than a principal, the Florida Supreme Court reasoned: 7 "[T]he only evidence of the degree of his participation is the jury's likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys' money. The evidence, therefore, was sufficient to find that the appellant was a principal of the second degree, constructively present aiding and abetting the commission of the crime of robbery. This conclusion supports the verdicts of murder in the first degree on the basis of the felony murder portion of section 782.04(1)(a)." 399 So.2d, at 1370.2 8 The State Supreme Court rejected two of the four statutory aggravating circumstances found by the trial court. It held that the findings that the murders were committed in the course of a robbery and that they were committed for pecuniary gain referred to the same aspect of petitioner's crime and must be treated as only one aggravating circumstance. Id., at 1373. In addition, the court held that "[t]he recited circumstance, that the murders were especially heinous, atrocious, and cruel, cannot be approved." Ibid., citing Armstrong v. State, 399 So.2d 953 (Fla.1981).3 However, because there were two aggravating circumstances and no mitigating circumstances, the death sentence was affirmed. In so doing, the court expressly rejected Enmund's submission that because the evidence did not establish that he intended to take life, the death penalty was barred by the Eighth Amendment of the United States Constitution. 399 So.2d, at 1371. 9 We granted Enmund's petition for certiorari, 454 U.S. 939, 102 S.Ct. 473, 70 L.Ed.2d 246 (1981), presenting the question whether death is a valid penalty under the Eighth and Fourteenth Amendments for one who neither took life, attempted to take life, nor intended to take life.4 II 10 As recounted above, the Florida Supreme Court held that the record supported no more than the inference that Enmund was the person in the car by the side of the road at the time of the killings, waiting to help the robbers escape. This was enough under Florida law to make Enmund a constructive aider and abettor and hence a principal in first-degree murder upon whom the death penalty could be imposed. It was thus irrelevant to Enmund's challenge to the death sentence that he did not himself kill and was not present at the killings; also beside the point was whether he intended that the Kerseys be killed or anticipated that lethal force would or might be used if necessary to effectuate the robbery or a safe escape. We have concluded that imposition of the death penalty in these circumstances is inconsistent with the Eighth and Fourteenth Amendments. A. 11 The Cruel and Unusual Punishments Clause of the Eighth Amendment is directed, in part, " 'against all punishments which by their excessive length or severity are greatly disproportioned to the offenses charged.' " Weems v. United States, 217 U.S. 349, 371, 30 S.Ct. 544, 551, 54 L.Ed. 793 (1910), quoting O'Neil v. Vermont, 144 U.S. 323, 339-340, 12 S.Ct. 693, 699-700, 36 L.Ed. 450 (1892) (Field, J., dissenting). This Court most recently held a punishment excessive in relation to the crime charged in Coker v. Georgia, 433 U.S. 584, 97 S.Ct. 2861, 53 L.Ed.2d 982 (1977). There the plurality opinion concluded that the imposition of the death penalty for the rape of an adult woman "is grossly disproportionate and excessive punishment for the crime of rape and is therefore forbidden by the Eighth Amendment as cruel and unusual punishment." Id., at 592, 97 S.Ct., at 2866. In reaching this conclusion, it was stressed that our judgment "should be informed by objective factors to the maximum possible extent." Ibid. Accordingly, the Court looked to the historical development of the punishment at issue, legislative judgments, international opinion, and the sentencing decisions juries have made before bringing its own judgment to bear on the matter. We proceed to analyze the punishment at issue in this case in a similar manner. B 12 The Coker plurality observed that "[a]t no time in the last 50 years have a majority of the States authorized death as a punishment for rape." Id., at 593, 97 S.Ct., at 2866. More importantly, in reenacting death penalty laws in order to satisfy the criteria established in Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972), only three States provided the death penalty for the rape of an adult woman in their revised statutes. 433 U.S., at 594, 97 S.Ct., at 2867. The plurality therefore concluded that "[t]he current judgment with respect to the death penalty for rape is not wholly unanimous among state legislatures, but it obviously weighs very heavily on the side of rejecting capital punishment as a suitable penalty for raping an adult woman." Id. 433 U.S., at 596, 97 S.Ct., at 2868 (footnote omitted). 13 Thirty-six state and federal jurisdictions presently authorize the death penalty. Of these, only eight jurisdictions authorize imposition of the death penalty solely for participation in a robbery in which another robber takes life.5 Of the remaining 28 jurisdictions, in 4 felony murder is not a capital crime.6 Eleven States require some culpable mental state with respect to the homicide as a prerequisite to conviction of a crime for which the death penalty is authorized. Of these 11 States, 8 make knowing, intentional, purposeful, or premeditated killing an element of capital murder.7 Three other States require proof of a culpable mental state short of intent, such as recklessness or extreme indifference to human life, before the death penalty may be imposed.8 In these 11 States, therefore, the actors in a felony murder are not subject to the death penalty without proof of their mental state, proof which was not required with respect to Enmund either under the trial court's instructions or under the law announced by the Florida Supreme Court. 14 Four additional jurisdictions do not permit a defendant such as Enmund to be put to death. Of these, one State flatly prohibits capital punishment in cases where the defendant did not actually commit murder.9 Two jurisdictions preclude the death penalty in cases such as this one where the defendant "was a principal in the offense, which was committed by another, but his participation was relatively minor, although not so minor as to constitute a defense to prosecution."10 One other State limits the death penalty in felony murders to narrow circumstances not involved here.11 15 Nine of the remaining States deal with the imposition of the death penalty for a vicarious felony murder in their capital sentencing statutes. In each of these States, a defendant may not be executed solely for participating in a felony in which a person was killed if the defendant did not actually cause the victim's death. For a defendant to be executed in these States, typically the statutory aggravating circumstances which are present must outweigh mitigating factors. To be sure, a vicarious felony murderer may be sentenced to death in these jurisdictions absent an intent to kill if sufficient aggravating circumstances are present. However, six of these nine States make it a statutory mitigating circumstance that the defendant was an accomplice in a capital felony committed by another person and his participation was relatively minor.12 By making minimal participation in a capital felony committed by another person a mitigating circumstance, these sentencing statutes reduce the likelihood that a person will be executed for vicarious felony murder. The remaining three jurisdictions exclude felony murder from their lists of aggravating circumstances that will support a death sentence.13 In each of these nine States, a nontriggerman guilty of felony murder cannot be sentenced to death for the felony murder absent aggravating circumstances above and beyond the felony murder itself. 16 Thus only a small minority of jurisdictions—eight—allow the death penalty to be imposed solely because the defendant somehow participated in a robbery in the course of which a murder was committed. Even if the nine States are included where such a defendant could be executed for an unintended felony murder if sufficient aggravating circumstances are present to outweigh mitigating circumstances—which often include the defendant's minimal participation in the murder—only about a third of American jurisdictions would ever permit a defendant who somehow participated in a robbery where a murder occurred to be sentenced to die. Moreover, of the eight States which have enacted new death penalty statutes since 1978, none authorize capital punishment in such circumstances.14 While the current legislative judgment with respect to imposition of the death penalty where a defendant did not take life, attempt to take it, or intend to take life is neither "wholly unanimous among state legislatures," Coker v. Georgia, 433 U.S., at 596, 97 S.Ct., at 2868, nor as compelling as the legislative judgments considered in Coker, it nevertheless weighs on the side of rejecting capital punishment for the crime at issue.15 C 17 Society's rejection of the death penalty for accomplice liability in felony murders is also indicated by the sentencing decisions that juries have made. As we have previously observed, " '[t]he jury . . . is a significant and reliable objective index of contemporary values because it is so directly involved.' " Coker v. Georgia, supra, at 596, 97 S.Ct., at 2868, quoting Gregg v. Georgia, 428 U.S. 153, 181, 96 S.Ct. 2909, 2929, 49 L.Ed.2d 859 (1976). The evidence is overwhelming that American juries have repudiated imposition of the death penalty for crimes such as petitioner's. First, according to the petitioner, a search of all reported appellate court decisions since 1954 in cases where a defendant was executed for homicide shows that of the 362 executions, in 339 the person executed personally committed a homicidal assault.16 In 2 cases the person executed had another person commit the homicide for him, and in 16 cases the facts were not reported in sufficient detail to determine whether the person executed committed the homicide.17 The survey revealed only 6 cases out of 362 where a nontriggerman felony murderer was executed. All six executions took place in 1955. By contrast, there were 72 executions for rape in this country between 1955 and this Court's decision in Coker v. Georgia in 1977.18 18 That juries have rejected the death penalty in cases such as this one where the defendant did not commit the homicide, was not present when the killing took place, and did not participate in a plot or scheme to murder is also shown by petitioner's survey of the Nation's death-row population.19 As of October 1, 1981, there were 796 inmates under sentences of death for homicide. Of the 739 for whom sufficient data are available, only 41 did not participate in the fatal assault on the victim. Of the 40 among the 41 for whom sufficient information was available, only 16 were not physically present when the fatal assault was committed. These 16 prisoners included only 3, including petitioner, who were sentenced to die absent a finding that they hired or solicited someone else to kill the victim or participated in a scheme designed to kill the victim. The figures for Florida are similar.20 Forty-five felony murderers are currently on death row. The Florida Supreme Court either found or affirmed a trial court or jury finding that the defendant intended life to be taken in 36 cases. In eight cases the courts made no finding with respect to intent, but the defendant was the triggerman in each case. In only one case—Enmund's—there was no finding of an intent to kill and the defendant was not the triggerman.21 The State does not challenge this analysis of the Florida cases. 19 The dissent criticizes these statistics on the ground that they do not reveal the percentage of homicides that were charged as felony murders or the percentage of cases where the State sought the death penalty for an accomplice guilty of felony murder. Post, at 818-819. We doubt whether it is possible to gather such information, and at any rate, it would be relevant if prosecutors rarely sought the death penalty for accomplice felony murder, for it would tend to indicate that prosecutors, who represent society's interest in punishing crime, consider the death penalty excessive for accomplice felony murder. The fact remains that we are not aware of a single person convicted of felony murder over the past quarter century who did not kill or attempt to kill, and did not intend the death of the victim, who has been executed, and that only three persons in that category are presently sentenced to die. Nor can these figures be discounted by attributing to petitioner the argument that "death is an unconstitutional penalty absent an intent to kill," post, at 819, and observing that the statistics are incomplete with respect to intent. Petitioner's argument is that because he did not kill, attempt to kill, and he did not intend to kill, the death penalty is disproportionate as applied to him, and the statistics he cites are adequately tailored to demonstrate that juries—and perhaps prosecutors as well—consider death a disproportionate penalty for those who fall within his category.22 III 20 Although the judgments of legislatures, juries, and prosecutors weigh heavily in the balance, it is for us ultimately to judge whether the Eighth Amendment permits imposition of the death penalty on one such as Enmund who aids and abets a felony in the course of which a murder is committed by others but who does not himself kill, attempt to kill, or intend that a killing take place or that lethal force will be employed. We have concluded, along with most legislatures and juries, that it does not. 21 We have no doubt that robbery is a serious crime deserving serious punishment. It is not, however, a crime "so grievous an affront to humanity that the only adequate response may be the penalty of death." Gregg v. Georgia, 428 U.S., at 184, 96 S.Ct., at 2930 (footnote omitted). "[I]t does not compare with murder, which does involve the unjustified taking of human life. Although it may be accompanied by another crime, [robbery] by definition does not include the death of or even the serious injury to another person. The murderer kills; the [robber], if no more than that, does not. Life is over for the victim of the murderer; for the [robbery] victim, life . . . is not over and normally is not beyond repair." Coker v. Georgia, 433 U.S., at 598, 97 S.Ct., at 2869 (footnote omitted). As was said of the crime of rape in Coker, we have the abiding conviction that the death penalty, which is "unique in its severity and irrevocability," Gregg v. Georgia, supra, 428 U.S., at 187, 96 S.Ct., at 2931, is an excessive penalty for the robber who, as such, does not take human life. 22 Here the robbers did commit murder; but they were subjected to the death penalty only because they killed as well as robbed. The question before us is not the disproportionality of death as a penalty for murder, but rather the validity of capital punishment for Enmund's own conduct. The focus must be on his culpability, not on that of those who committed the robbery and shot the victims, for we insist on "individualized consideration as a constitutional requirement in imposing the death sentence," Lockett v. Ohio, 438 U.S. 586, 605, 98 S.Ct. 2954, 2965, 57 L.Ed.2d 973 (1978) (footnote omitted), which means that we must focus on "relevant facets of the character and record of the individual offender." Woodson v. North Carolina, 428 U.S. 280, 304, 96 S.Ct. 2978, 2991, 49 L.Ed.2d 944 (1976). Enmund himself did not kill or attempt to kill; and, as construed by the Florida Supreme Court, the record before us does not warrant a finding that Enmund had any intention of participating in or facilitating a murder. Yet under Florida law death was an authorized penalty because Enmund aided and abetted a robbery in the course of which murder was committed. It is fundamental that "causing harm intentionally must be punished more severely than causing the same harm unintentionally." H. Hart, Punishment and Responsibility 162 (1968). Enmund did not kill or intend to kill and thus his culpability is plainly different from that of the robbers who killed; yet the State treated them alike and attributed to Enmund the culpability of those who killed the Kerseys. This was impermissible under the Eighth Amendment. 23 In Gregg v. Georgia the opinion announcing the judgment observed that "[t]he death penalty is said to serve two principal social purposes: retribution and deterrence of capital crimes by prospective offenders." 428 U.S., at 183, 96 S.Ct., at 2929-30 (footnote omitted). Unless the death penalty when applied to those in Enmund's position measurably contributes to one or both of these goals, it "is nothing more than the purposeless and needless imposition of pain and suffering," and hence an unconstitutional punishment. Coker v. Georgia, supra, 433 U.S., at 592, 97 S.Ct., at 2866. We are quite unconvinced, however, that the threat that the death penalty will be imposed for murder will measurably deter one who does not kill and has no intention or purpose that life will be taken. Instead, it seems likely that "capital punishment can serve as a deterrent only when murder is the result of premeditation and deliberation," Fisher v. United States, 328 U.S. 463, 484, 66 S.Ct. 1318, 1328, 90 L.Ed. 1382 (1946) (Frankfurter, J., dissenting), for if a person does not intend that life be taken or contemplate that lethal force will be employed by others, the possibility that the death penalty will be imposed for vicarious felony murder will not "enter into the cold calculus that precedes the decision to act." Gregg v. Georgia, supra, 428 U.S., at 186, 96 S.Ct., at 2931 (footnote omitted). 24 It would be very different if the likelihood of a killing in the course of a robbery were so substantial that one should share the blame for the killing if he somehow participated in the felony. But competent observers have concluded that there is no basis in experience for the notion that death so frequently occurs in the course of a felony for which killing is not an essential ingredient that the death penalty should be considered as a justifiable deterrent to the felony itself. Model Penal Code § 210.2, Comment, p. 38, and n. 96. This conclusion was based on three comparisons of robbery statistics, each of which showed that only about one-half of one percent of robberies resulted in homicide.23 The most recent national crime statistics strongly support this conclusion.24 In addition to the evidence that killings only rarely occur during robberies is the fact, already noted, that however often death occurs in the course of a felony such as robbery, the death penalty is rarely imposed on one only vicariously guilty of the murder, a fact which further attenuates its possible utility as an effective deterrence. 25 As for retribution as a justification for executing Enmund, we think this very much depends on the degree of Enmund's culpability—what Enmund's intentions, expectations, and actions were. American criminal law has long considered a defendant's intention—and therefore his moral guilt—to be critical to "the degree of [his] criminal culpability," Mullaney v. Wilbur, 421 U.S. 684, 698, 95 S.Ct. 1881, 1889, 44 L.Ed.2d 508 (1975), and the Court has found criminal penalties to be unconstitutionally excessive in the absence of intentional wrongdoing. In Robinson v. California, 370 U.S. 660, 667, 82 S.Ct. 1417, 1420, 8 L.Ed.2d 758 (1962), a statute making narcotics addiction a crime, even though such addiction "is apparently an illness which may be contracted innocently or involuntarily," was struck down under the Eighth Amendment. Similarly, in Weems v. United States, the Court invalidated a statute making it a crime for a public official to make a false entry in a public record but not requiring the offender to "injur[e] any one by his act or inten[d] to injure any one." 217 U.S., at 363, 30 S.Ct., at 547. The Court employed a similar approach in Godfrey v. Georgia, 446 U.S. 420, 433, 100 S.Ct. 1759, 1767, 64 L.Ed.2d 398 (1980), reversing a death sentence based on the existence of an aggravating circumstance because the defendant's crime did not reflect "a consciousness materially more 'depraved' than that of any person guilty of murder." 26 For purposes of imposing the death penalty, Enmund's criminal culpability must be limited to his participation in the robbery, and his punishment must be tailored to his personal responsibility and moral guilt. Putting Enmund to death to avenge two killings that he did not commit and had no intention of committing or causing does not measurably contribute to the retributive end of ensuring that the criminal gets his just deserts. This is the judgment of most of the legislatures that have recently addressed the matter, and we have no reason to disagree with that judgment for purposes of construing and applying the Eighth Amendment. IV 27 Because the Florida Supreme Court affirmed the death penalty in this case in the absence of proof that Enmund killed or attempted to kill, and regardless of whether Enmund intended or contemplated that life would be taken, we reverse the judgment upholding the death penalty and remand for further proceedings not inconsistent with this opinion. 28 So ordered. 29 Justice BRENNAN, concurring. 30 I join the Court's opinion. However, I adhere to my view that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. See Gregg v. Georgia, 428 U.S. 153, 227, 96 S.Ct. 2909, 2950, 49 L.Ed.2d 859 (1976) (dissenting opinion). 31 Justice O'CONNOR, with whom THE CHIEF JUSTICE, Justice POWELL, and Justice REHNQUIST join, dissenting. 32 Today the Court holds that the Eighth Amendment prohibits a State from executing a convicted felony murderer. I dissent from this holding not only because I believe that it is not supported by the analysis in our previous cases, but also because today's holding interferes with state criteria for assessing legal guilt by recasting intent as a matter of federal constitutional law. 33 * The evidence at trial showed that at approximately 7:30 a. m. on April 1, 1975, Sampson and Jeanette Armstrong approached the backdoor of Thomas and Eunice Kersey's farmhouse on the pretext of obtaining water for their overheated car.1 When Thomas Kersey retrieved a water jug to help the Armstrongs, Sampson Armstrong grabbed him, held a gun to him, and told Jeanette Armstrong to take his wallet. Hearing her husband's cries for help, Eunice Kersey came around the side of the house with a gun and shot Jeanette Armstrong. Sampson Armstrong, and perhaps Jeanette Armstrong, returned the fire, killing both of the Kerseys.2 The Armstrongs dragged the bodies into the kitchen, took Thomas Kersey's money, and fled to a nearby car, where the petitioner, Earl Enmund, was waiting to help the Armstrongs escape. Record 1348-1351.3 34 Ida Jean Shaw4 testified that on March 31 the petitioner and the two Armstrongs were staying at her house. When she awoke on April 1, the day of the murders, the petitioner, Jeanette, and Sampson, as well as Shaw's 1969 yellow Buick, were gone. Id., at 1185-1186. A little after eight o'clock, either the petitioner or Sampson Armstrong entered the house and told her that Jeanette had been shot. Id., at 1187-1188. After learning that Jeanette had been shot during a robbery, Shaw asked the petitioner "[W]hy he did it." Enmund answered that he had decided to rob Thomas Kersey after he had seen Kersey's money a few weeks earlier. Id., at 1205.5 At the same time, Sampson Armstrong volunteered that he had made sure that the Kerseys were dead. Id., at 1207-1208. 35 Ida Jean Shaw also testified that, pursuant to the petitioner's and Sampson Armstrong's instructions, she had disposed of a .22-caliber pistol that she normally kept in her car, as well as a .38-caliber pistol belonging to the Armstrongs. Id., at 1198-1202. The murder weapons were never recovered.6 36 In his closing argument, the prosecutor did not argue that Earl Enmund had killed the Kerseys. Instead, he maintained that the petitioner had initiated and planned the armed robbery, and was in the car during the killings. According to the prosecutor, "Sampson Armstrong killed the old people." Id., at 1577.7 37 After deliberating for four hours, the jury found Sampson Armstrong and the petitioner each guilty of two counts of first-degree murder8 and one count of robbery.9 The jury then heard evidence pertaining to the appropriate sentence for the two defendants, and recommended the death penalty for each defendant on each of the murder counts.10 38 In its sentencing findings,11 the trial court found four statutory aggravating circumstances regarding the petitioner's involvement in the murder: (1) the petitioner previously had been convicted of a felony involving the use of violence (an armed robbery in 1957), Fla.Stat. § 921.141(5)(b) (1981); (2) the murders were committed during the course of a robbery, § 921.141(5)(d); (3) the murders were committed for pecuniary gain, § 921.141(5)(f); and (4) the murders were especially heinous, atrocious, or cruel because the Kerseys had been shot in a prone position in an effort to eliminate them as witnesses, § 921.141(5)(h). App. 30-31; 399 So.2d 1362, 1371-1372 (Fla.1981).12 39 The trial court also found that "none of the statutory mitigating circumstances applied" to the petitioner. App. 32 (emphasis in original). Most notably, the court concluded that the evidence clearly showed that the petitioner was an accomplice to the capital felony and that his participation had not been "relatively minor," but had been major in that he "planned the capital felony and actively participated in an attempt to avoid detection by disposing of the murder weapons." Ibid.; 399 So.2d, at 1373. See Fla.Stat. § 921.141(6)(d) (1981).13 40 Considering these factors, the trial court concluded that the "aggravating circumstances of these capital felonies outweigh the mitigating circumstances," and imposed the death penalty for each count of murder. App. 32; 399 So.2d, at 1373. The court sentenced the petitioner to life imprisonment for the robbery. App. 28.14 41 On appeal, the Florida Supreme Court affirmed the petitioner's convictions and sentences.15 In challenging his convictions for first-degree murder, the petitioner claimed that there was no evidence that he had committed premeditated murder, or that he had been present aiding and abetting the robbery when the Kerseys were shot. He argued that since the jury properly could have concluded only that he was in the car on the highway when the murders were committed, he could be found guilty at most of second-degree murder under the State's felony-murder rule.16 42 The court rejected this argument. Quoting from an earlier case, the Florida Supreme Court held: 43 " '[A]n individual who personally kills another during the perpetration or attempt to perpetrate one of the enumerated felonies is guilty of first degree murder. . . . Moreover, the felon's liability for first degree murder extends to all of his co-felons who are personally present. As perpetrators of the underlying felony, they are principals in the homicide. In Florida, as in the majority of jurisdictions, the felony murder rule and the law of principals combine to make a felon generally responsible for the lethal acts of his co-felon. Only if the felon is an accessory before the fact and not personally present does liability attach under the second degree murder provision of the applicable statute in the instant case.' " 399 So.2d, at 1369 (quoting Adams v. State, 341 So.2d 765, 768-769 (Fla.1976) (footnote omitted), cert. denied, 434 U.S. 878, 98 S.Ct. 232, 54 L.Ed.2d 158 (1977)). 44 Consequently, the critical issue regarding liability was whether the petitioner's conduct would make him a principal or merely an accessory before the fact to the underlying robbery. Under Florida law at the time of the murders, "if the accused was present aiding and abetting the commission or attempt of one of the violent felonies listed in the first-degree murder statute, he is equally guilty, with the actual perpetrator of the underlying felony, of first-degree murder." 399 So.2d, at 1370. Moreover, 45 " 'the presence of the aider and abetter need not have been actual, but it is sufficient if he was constructively present, provided the aider, pursuant to a previous understanding, is sufficiently near and so situated as to abet or encourage, or to render assistance to, the actual perpetrator in committing the felonious act or in escaping after its commission.' " Ibid. (quoting Pope v. State, 84 Fla. 428, 446, 94 So. 865, 871 (1922)). 46 The court noted that there "was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered." 399 So.2d, at 1370.17 Instead, 47 "the only evidence of the degree of his participation is the jury's likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys' money." Ibid. 48 This evidence, the court concluded, was sufficient to find the petitioner to be a principal under state law, "constructively present aiding and abetting the commission of the crime of robbery," and thus guilty of first degree murder. Ibid. Turning to the trial court's written sentencing findings, the State Supreme Court rejected two of the four aggravating circumstances. First, the court held that two of the trial judge's findings—that the murders were committed both in the course of robbery and for pecuniary gain—referred to the same aspect of the petitioner's crime. Consequently, these facts supported only one aggravating circumstance. Second, citing Armstrong v. State, 399 So.2d 953 (Fla.1981), the court held that "[t]he recited circumstance, that the murders were especially heinous, atrocious, and cruel, cannot be approved." 399 So.2d, at 1373.18 The court affirmed the trial court's findings that none of the statutory mitigating circumstances applied. Ibid. Because one of those findings was that Enmund's participation in the capital felony was not minor, due to his role in planning the robbery, the State Supreme Court implicitly affirmed the finding that Enmund had planned the robbery. 49 Regarding the petitioner's claim that imposition of the death penalty, absent a showing that he intended to kill, would violate the Eighth Amendment's ban on cruel and unusual punishments, the court simply stated that the petitioner "offers us no binding legal authority that directly supports this proposition, and we therefore reject it." Id., at 1371. II 50 Earl Enmund's claim in this Court is that the death sentence imposed by the Florida trial court, and affirmed by the Florida Supreme Court, is unconstitutionally disproportionate to the role he played in the robbery and murders of the Kerseys.19 In particular, he contends that because he had no actual intent to kill the victims—in effect, because his behavior and intent were no more blameworthy than that of any robber—capital punishment is too extreme a penalty.20 51 In Gregg v. Georgia, 428 U.S. 153, 96 S.Ct. 2909, 49 L.Ed.2d 859 (1976), a majority of this Court concluded that the death penalty does not invariably violate the Cruel and Unusual Punishments Clause of the Eighth Amendment.21 See id., at 187, 96 S.Ct., at 2931 (opinion of Stewart, POWELL, and STEVENS, JJ.) ("[W]hen a life has been taken deliberately by the offender, we cannot say that the punishment is invariably disproportionate to the crime. It is an extreme sanction, suitable to the most extreme of crimes") (footnote omitted); id., at 226, 96 S.Ct., at 2949 (opinion of WHITE, J.) (rejecting the argument that "the death penalty, however imposed and for whatever crime, is cruel and unusual punishment"); id., at 227, 96 S.Ct., at 2950 (BLACKMUN, J., concurring in judgment). In no case since Gregg and its companion cases,22 has this Court retreated from that position.23 Recognizing the constitutionality of the death penalty, however, only marks the beginning of the inquiry, for Earl Enmund was not convicted of murder as it is ordinarily envisioned—a deliberate and premeditated, unlawful killing. Rather, through the doctrine of accessorial liability, the petitioner has been convicted of two murders that he did not specifically intend.24 Thus, it is necessary to examine the concept of proportionality as enunciated in this Court's cases to determine whether the penalty imposed on Earl Enmund is unconstitutionally disproportionate to his crimes. 52 The Eighth Amendment concept of proportionality was first fully expressed in Weems v. United States, 217 U.S. 349, 30 S.Ct. 544, 54 L.Ed. 793 (1910). In that case, defendant Weems was sentenced to 15 years at hard labor for falsifying a public document. After remarking that "it is a precept of justice that punishment for crime should be graduated and proportioned to offense," id., at 367, 30 S.Ct., at 549, and after comparing Weems' punishment to the punishments for other crimes, the Court concluded that the sentence was cruel and unusual. Id., at 381, 30 S.Ct., at 554-555. 53 Not until two-thirds of a century later, in Coker v. Georgia, 433 U.S. 584, 97 S.Ct. 2861, 53 L.Ed.2d 982 (1977), did the Court declare another punishment to be unconstitutionally disproportionate to the crime. Writing for himself and three other Members of the Court, Justice WHITE concluded that death is a disproportionate penalty for the crime of raping an adult woman. Id., at 597, 97 S.Ct., at 2868.25 In reaching this conclusion, the plurality was careful to inform its judgment "by objective factors to the maximum possible extent [by giving attention] to the public attitudes concerning a particular sentence—history and precedent, legislative attitudes, and the response of juries reflected in their sentencing decisions." Id., at 592, 97 S.Ct., at 2866. The plurality's resort to objective factors was no doubt an effort to derive "from the evolving standards of decency that mark the progress of a maturing society" the meaning of the requirement of proportionality contained within the Eighth Amendment. Trop v. Dulles, 356 U.S. 86, 101, 78 S.Ct. 590, 598, 2 L.Ed.2d 630 (1958) (opinion of Warren, C. J.). 54 The plurality noted that within the previous 50 years a majority of the States had never authorized death as a punishment for rape. More significantly to the plurality, only 3 of the 35 States that immediately reinstituted the death penalty following the Court's judgment in Furman v. Georgia, 408 U.S. 238, 92 S.Ct. 2726, 33 L.Ed.2d 346 (1972) (invalidating nearly all state capital punishment statutes), defined rape as a capital offense.26 The plurality also considered "the sentencing decisions that juries have made in the course of assessing whether capital punishment is an appropriate penalty for the crime being tried." 433 U.S., at 596, 97 S.Ct., at 2868. See Gregg v. Georgia, 428 U.S., at 181, 96 S.Ct., at 2928 (opinion of STEWART, POWELL, and STEVENS, JJ.) ("The jury also is a significant and reliable objective index of contemporary values because it is so directly involved"). From the available data, the plurality concluded that in at least 90% of the rape convictions since 1973, juries in Georgia had declined to impose the death penalty. 433 U.S., at 597, 97 S.Ct., at 2868. 55 Thus, the conclusion reached in Coker rested in part on the Court's observation that both legislatures and juries firmly rejected the penalty of death for the crime of rape. See Woodson v. North Carolina, 428 U.S. 280, 293, 96 S.Ct. 2978, 2986, 49 L.Ed.2d 944 (1976) (opinion of Stewart, POWELL, and STEVENS, JJ.) (concluding that the State's mandatory death penalty statute violates the Eighth Amendment because the "two crucial indicators of evolving standards of decency respecting the imposition of punishment in our society—jury determinations and legislative enactments—both point conclusively to the repudiation of automatic death sentences"). 56 In addition to ascertaining "contemporary standards," the plurality opinion also considered qualitative factors bearing on the question whether the death penalty was disproportionate, for "the Constitution contemplates that in the end our own judgment will be brought to bear on the question of the acceptability of the death penalty under the Eighth Amendment." 433 U.S., at 597, 97 S.Ct., at 2868. The plurality acknowledged that a rapist is almost as blameworthy as a murderer, describing the crime of rape as "highly reprehensible, both in a moral sense and in its almost total contempt for the personal integrity and autonomy of the female victim." Ibid. Despite the enormity of the crime of rape, however, the Court concluded that the death penalty was "grossly out of proportion to the severity of the crime," id., at 592, 97 S.Ct., at 2866, in part because the harm caused by a rape "does not compare with murder, which does involve the unjustified taking of human life." Id., at 598, 97 S.Ct., at 2869. 57 Coker teaches, therefore, that proportionality—at least as regards capital punishment—not only requires an inquiry into contemporary standards as expressed by legislators and jurors, but also involves the notion that the magnitude of the punishment imposed must be related to the degree of the harm inflicted on the victim, as well as to the degree of the defendant's blameworthiness.27 Moreover, because they turn on considerations unique to each defendant's case, these latter factors underlying the concept of proportionality are reflected in this Court's conclusion in Lockett v. Ohio, 438 U.S. 586, 605, 98 S.Ct. 2954, 2965, 57 L.Ed.2d 973 (1978), that "individualized consideration [is] a constitutional requirement in imposing the death sentence" (opinion of BURGER, C. J.) (footnote omitted). See id., at 613, 98 S.Ct., at 2969 (opinion of BLACKMUN, J.) ("the Ohio judgment in this case improperly provided the death sentence for a defendant who only aided and abetted a murder, without permitting any consideration by the sentencing authority of the extent of her involvement, or the degree of her mens rea, in the commission of the homicide"). 58 In sum, in considering the petitioner's challenge, the Court should decide not only whether the petitioner's sentence of death offends contemporary standards as reflected in the responses of legislatures and juries, but also whether it is disproportionate to the harm that the petitioner caused and to the petitioner's involvement in the crime, as well as whether the procedures under which the petitioner was sentenced satisfied the constitutional requirement of individualized consideration set forth in Lockett. B 59 Following the analysis set forth in Coker, the petitioner examines the historical development of the felony-murder rule, as well as contemporary legislation and jury verdicts in capital cases, in an effort to show that imposition of the death penalty on him would violate the Eighth Amendment. This effort fails, however, for the available data do not show that society has rejected conclusively the death penalty for felony murderers. 60 As the petitioner acknowledges, the felony-murder doctrine, and its corresponding capital penalty, originated hundreds of years ago,28 and was a fixture of English common law until 1957 when Parliament declared that an unintentional killing during a felony would be classified as manslaughter.29 The common-law rule was transplanted to the American Colonies, and its use continued largely unabated into the 20th century, although legislative reforms often restricted capital felony murder to enumerated violent felonies.30 61 The petitioner discounts the weight of this historical precedent by arguing that jurors and judges widely resisted the application of capital punishment by acquitting defendants in felony-murder cases or by convicting them of noncapital manslaughter.31 The force of the petitioner's argument is speculative at best, however, for it is unclear what fraction of the jury nullification in this country resulted from dissatisfaction with the capital felony-murder rule. Much of it, surely, was a reaction to the mandatory death penalty, and the failure of the common law and early state statutes to classify murder by degree. In fact, it was in response to juror attitudes toward capital punishment that most jurisdictions by the early part of this century replaced their mandatory death penalty statutes with statutes allowing juries the discretion to decide whether to impose or to recommend the death penalty. See Woodson v. North Carolina, 428 U.S., at 291-292, 96 S.Ct., at 2984-85 (opinion of Stewart, POWELL, and STEVENS, JJ.).32 Thus, it simply is not possible to conclude that historically this country conclusively has rejected capital punishment for homicides committed during the course of a felony. 62 The petitioner and the Court turn to jury verdicts in an effort to show that, by present standards at least, capital punishment is grossly out of proportion to the crimes that the petitioner committed. Surveying all reported appellate court opinions since 1954 involving executions, the petitioner has found that of the 362 individuals executed for homicide, 339 personally committed the homicidal assault, and two others each had another person commit the homicide on his behalf. Only six persons executed were "non-triggermen."33 A similar trend can be seen in the petitioner's survey of the current death row population.34 Of the 739 prisoners for whom sufficient data are available, only 40 did not participate in the homicidal assault, and of those, only 3 (including the petitioner) were sentenced to death absent a finding that they had collaborated with the killer in a specific plan to kill. Brief for Petitioner 35-36. See also App. to Reply Brief for Petitioner (showing that of the 45 felony murderers currently on death row in Florida, 36 were found by the State Supreme Court or a trial court to have had the intent to kill; in 8 cases, the state courts made no finding, but the defendant was the triggerman; and in 1, the petitioner's case, the defendant was not the triggerman, and there was no finding of intent to kill). 63 Impressive as these statistics are at first glance, they cannot be accepted uncritically. So stated, the data do not reveal the number or fraction of homicides that were charged as felony murders, or the number or fraction of cases in which the State sought the death penalty for an accomplice guilty of felony murder. Consequently, we cannot know the fraction of cases in which juries rejected the death penalty for accomplice felony murder. Moreover, as Justice BLACKMUN pointed out in his concurring opinion in Lockett v. Ohio, 438 U.S., at 615, n. 2, 98 S.Ct., at 2970, n. 2, many of these data classify defendants by whether they "personally committed a homicidal assault," and do not show the fraction of capital defendants who were shown to have an intent to kill. While the petitioner relies on the fact that he did not pull the trigger, his principal argument is, and must be, that death is an unconstitutional penalty absent an intent to kill, for otherwise defendants who hire others to kill would escape the death penalty. See n. 20, supra. Thus, the data he presents are not entirely relevant. Even accepting the petitioner's facts as meaningful, they may only reflect that sentencers are especially cautious in imposing the death penalty, and reserve that punishment for those defendants who are sufficiently involved in the homicide, whether or not there was specific intent to kill. 64 Finally, as the petitioner acknowledges, the jury verdict statistics cannot be viewed in isolation from state death penalty legislation. The petitioner and the Court therefore review recent legislation in order to support the conclusion that society has rejected capital felony murder. Of the 35 States that presently have a death penalty, however, fully 31 authorize a sentencer to impose a death sentence for a death that occurs during the course of a robbery.35 The States are not uniform in delimiting the circumstances under which the death penalty may be imposed for felony murder, but each state statute can be classified as one of three types. The first category, containing 20 statutes, includes those States that permit imposition of the death penalty for felony murder even though the defendant did not commit the homicidal act, and even though he had no actual intent to kill.36 Three additional States, while requiring some finding of intent, do not require the intent to kill that the petitioner believes is constitutionally mandated before the death sentence may be imposed.37 The second category, containing seven statutes, includes those States that authorize the death penalty only if the defendant had the specific intent (or some rough equivalent) to kill the victim.38 The third class of statutes, from only three States, restricts application of the death penalty to those felony murderers who actually commit the homicide.39 65 The Court's curious method of counting the States that authorize imposition of the death penalty for felony murder cannot hide the fact that 23 States permit a sentencer to impose the death penalty even though the felony murderer has neither killed nor intended to kill his victim. While the Court acknowledges that eight state statutes follow the Florida death penalty scheme, see ante, at 789, n. 5, it also concedes that 15 other statutes permit imposition of the death penalty where the defendant neither intended to kill or actually killed the victims. See ante, at 790, n. 8, (Arkansas, Delaware, and Kentucky); ante, at 793-794, n. 15 (New Mexico); ante, at 791, n. 10 (Colorado); ante, at 791, n. 11 (Vermont); ante, at 792, n. 12 (Arizona, Connecticut, Indiana, Montana, Nebraska, and North Carolina); ante, at 792, n. 13 (Idaho, Oklahoma, and South Dakota). Not all of the statutes list the same aggravating circumstances. Nevertheless, the question before the Court is not whether a particular species of death penalty statute is unconstitutional, but whether a scheme that permits imposition of the death penalty, absent a finding that the defendant either killed or intended to kill the victims, is unconstitutional. In short, the Court's peculiar statutory analysis cannot withstand closer scrutiny. 66 Thus, in nearly half of the States, and in two-thirds of the States that permit the death penalty for murder, a defendant who neither killed the victim nor specifically intended that the victim die may be sentenced to death for his participation in the robbery-murder. Far from "weigh[ing] very heavily on the side of rejecting capital punishment as a suitable penalty for" felony murder, Coker v. Georgia, 433 U.S., at 596, 97 S.Ct., at 2868, these legislative judgments indicate that our "evolving standards of decency" still embrace capital punishment for this crime. For this reason, I conclude that the petitioner has failed to meet the standards in Coker and Woodson that the "two crucial indicators of evolving standards of decency . . .—jury determinations and legislative enactments—both point conclusively to the repudiation" of capital punishment for felony murder. 428 U.S., at 293, 96 S.Ct., at 2986 (emphasis added). In short, the death penalty for felony murder does not fall short of our national "standards of decency." C 67 As I noted earlier, the Eighth Amendment concept of proportionality involves more than merely a measurement of contemporary standards of decency. It requires in addition that the penalty imposed in a capital case be proportional to the harm caused and the defendant's blameworthiness. Critical to the holding in Coker, for example, was that "in terms of moral depravity and of the injury to the person and to the public, [rape] does not compare with murder, which . . . involve[s] the unjustified taking of human life." 433 U.S., at 598, 97 S.Ct., at 2869. 68 Although the Court disingenuously seeks to characterize Enmund as only a "robber," ante, at 797, it cannot be disputed that he is responsible, along with Sampson and Jeanette Armstrong, for the murders of the Kerseys. There is no dispute that their lives were unjustifiably taken, and that the petitioner, as one who aided and abetted the armed robbery, is legally liable for their deaths.40 Quite unlike the defendant in Coker, the petitioner cannot claim that the penalty imposed is "grossly out of proportion" to the harm for which he admittedly is at least partly responsible. 69 The Court's holding today is especially disturbing because it makes intent a matter of federal constitutional law, requiring this Court both to review highly subjective definitional problems customarily left to state criminal law and to develop an Eighth Amendment meaning of intent. As Justice BLACKMUN pointed out in his concurring opinion in Lockett, the Court's holding substantially "interfere[s] with the States' individual statutory categories for assessing legal guilt." 438 U.S., at 616, 98 S.Ct., at 2970.41 See also id., at 635-636, 98 S.Ct., at 2976-77 (opinion of REHNQUIST, J.) (rejecting the idea that intent to kill must be proved before the State can impose the death penalty). Although the Court's opinion suggests that intent can be ascertained as if it were some historical fact, in fact it is a legal concept, not easily defined. Thus, while proportionality requires a nexus between the punishment imposed and the defendant's blameworthiness, the Court fails to explain why the Eighth Amendment concept of proportionality requires rejection of standards of blameworthiness based on other levels of intent, such as, for example, the intent to commit an armed robbery coupled with the knowledge that armed robberies involve substantial risk of death or serious injury to other persons. Moreover, the intent-to-kill requirement is crudely crafted; it fails to take into account the complex picture of the defendant's knowledge of his accomplice's intent and whether he was armed, the defendant's contribution to the planning and success of the crime, and the defendant's actual participation during the commission of the crime. Under the circumstances, the determination of the degree of blameworthiness is best left to the sentencer, who can sift through the facts unique to each case. Consequently, while the type of mens rea of the defendant must be considered carefully in assessing the proper penalty, it is not so critical a factor in determining blameworthiness as to require a finding of intent to kill in order to impose the death penalty for felony murder. 70 In sum, the petitioner and the Court have failed to show that contemporary standards, as reflected in both jury determinations and legislative enactments, preclude imposition of the death penalty for accomplice felony murder. Moreover, examination of the qualitative factors underlying the concept of proportionality do not show that the death penalty is disproportionate as applied to Earl Enmund. In contrast to the crime in Coker, the petitioner's crime involves the very type of harm that this Court has held justifies the death penalty. Finally, because of the unique and complex mixture of facts involving a defendant's actions, knowledge, motives, and participation during the commission of a felony murder, I believe that the factfinder is best able to assess the defendant's blameworthiness. Accordingly, I conclude that the death penalty is not disproportionate to the crime of felony murder, even though the defendant did not actually kill or intend to kill his victims.42 III 71 Although I conclude that the death penalty is not disproportionate to the crime of felony murder, I believe that, in light of the State Supreme Court's rejection of critical factual findings, our previous opinions require a remand for a new sentencing hearing.43 Repeatedly, this Court has emphasized that capital sentencing decisions must focus "on the circumstances of each individual homicide and individual defendant." Proffitt v. Florida, 428 U.S. 242, 258, 96 S.Ct. 2960, 2969, 49 L.Ed.2d 913 (1976) (opinion of Stewart, POWELL, and STEVENS, JJ.). In striking down the mandatory capital punishment statute in Woodson v. North Carolina, 428 U.S., at 304, 96 S.Ct., at 2991, a plurality of the Court wrote: 72 "A process that accords no significance to relevant facets of the character and record of the individual offender or the circumstances of the particular offense excludes from consideration in fixing the ultimate punishment of death the possibility of compassionate or mitigating factors stemming from the diverse frailties of humankind. It treats all persons convicted of a designated offense not as uniquely individual human beings, but as members of a faceless, undifferentiated mass to be subjected to the blind infliction of the penalty of death. 73 ". . . [W]e believe that in capital cases the fundamental respect for humanity underlying the Eighth Amendment requires consideration of the character and record of the individual offender and the circumstances of the particular offense as a constitutionally indispensable part of the process of inflicting the penalty of death." 74 In Lockett v. Ohio, 438 U.S., at 605, 98 S.Ct., at 2965, a plurality of this Court concluded: 75 "Given that the imposition of death by public authority is so profoundly different from all other penalties, we cannot avoid the conclusion that an individualized decision is essential in capital cases. The need for treating each defendant in a capital case with that degree of respect due the uniqueness of the individual is far more important than in noncapital cases. . . . The nonavailability of corrective or modifying mechanisms with respect to an executed capital sentence underscores the need for individualized consideration as a constitutional requirement in imposing the death sentence" (footnote omitted). 76 Accordingly, "the sentencer, in all but the rarest kind of capital case, [may] not be precluded from considering, as a mitigating factor, any aspect of the defendant's character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death" (footnotes omitted). Id., at 604, 98 S.Ct., at 2964. See id., at 613, 98 S.Ct., at 2969 (opinion of BLACKMUN, J.) (concluding that the Ohio capital sentencing statute is unconstitutional because it "provided the death sentence for a defendant who only aided and abetted a murder, without permitting any consideration by the sentencing authority of the extent of her involvement, or the degree of her mens rea, in the commission of the homicide"); Green v. Georgia, 442 U.S. 95, 97, 99 S.Ct. 2150, 2151, 60 L.Ed.2d 738 (1979) (per curiam) (holding that the exclusion of evidence, from the capital sentencing proceeding, that the petitioner was not present when the victim was killed violated due process because "[t]he excluded testimony was highly relevant to a critical issue in the punishment phase of the trial"); Eddings v. Oklahoma, 455 U.S. 104, 102 S.Ct. 869, 71 L.Ed.2d 1 (1982) (adopting the plurality's rule in Lockett). Thus, in deciding whether or not to impose capital punishment on a felony murderer, a sentencer must consider any relevant evidence or arguments that the death penalty is inappropriate for a particular defendant because of his relative lack of mens rea and his peripheral participation in the murder. Because of the peculiar circumstances of this case, I conclude that the trial court did not give sufficient consideration to the petitioner's role in the crimes, and thus did not consider the mitigating circumstances proffered by the defendant at his sentencing hearing.44 77 In sentencing the petitioner, the trial court found four statutory aggravating circumstances: the petitioner had been convicted previously of a violent felony; the murders had been committed during the course of a robbery; the murders had been committed for pecuniary gain; and the murders were especially heinous, atrocious, or cruel. In its factual findings, the trial court stated that the "armed robbery . . . was planned ahead of time by the defendant Enmund," App. 30, and that he had shot each of the victims while they lay prone in order to eliminate them as witnesses. Id., at 30-31. The court expressly found that "none of the statutory mitigating circumstances applied" to the petitioner. Id., at 32 (emphasis in original). Among other findings, the court rejected Enmund's claim that his participation in the murders had been "relatively minor," and found instead that "his participation in the capital felony was major. The defendant Enmund planned the capital felony and actively participated in an attempt to avoid detection by disposing of the murder weapons." Ibid. 78 The Florida Supreme Court rejected these findings in part. The court noted that there "was no direct evidence at trial that Earl Enmund was present at the back door of the Kersey home when the plan to rob the elderly couple led to their being murdered." 399 So.2d, at 1370. Rather, 79 "the only evidence of the degree of his participation is the jury's likely inference that he was the person in the car by the side of the road near the scene of the crimes. The jury could have concluded that he was there, a few hundred feet away, waiting to help the robbers escape with the Kerseys' money." Ibid. Consequently, the court expressly rejected the trial court's finding that Enmund personally had committed the homicides. Reviewing the aggravating circumstances, the Supreme Court consolidated two of them, and rejected the trial court's conclusion that the murders had been "heinous, atrocious, or cruel," since the evidence showed that the Armstrongs had killed the Kerseys in a gun battle arising from Mrs. Kersey's armed resistance, and not that the petitioner had killed them in an effort to eliminate them as witnesses. SeeArmstrong v. State, 399 So.2d, at 963. 80 Although the state statutory procedures did not prevent the trial judge from considering any mitigating circumstances,45 the trial judge's view of the facts, in part rejected by the State Supreme Court, effectively prevented such consideration. In his erroneous belief that the petitioner had shot both of the victims while they lay in a prone position in order to eliminate them as witnesses, the trial judge necessarily rejected the only argument offered in mitigation—that the petitioner's role in the capital felonies was minor, undeserving of the death penalty, because the petitioner was in the car when the fatal shots were fired. This fundamental misunderstanding of the petitioner's role in the crimes prevented the trial court from considering the "circumstances of the particular offense" in imposing sentence. Woodson v. North Carolina, 428 U.S., at 304, 96 S.Ct., at 2991. Moreover, this error was not so insignificant that we can be sure its effect on the sentencing judge's decision was negligible.46 Accordingly, I would vacate the decision below insofar as it affirms the death sentence, and remand the case for a new sentencing hearing. 1 Jeanette Armstrong's trial was severed and she was convicted of two counts of second-degree murder and one count of robbery and sentenced to three consecutive life sentences. 399 So.2d 1362, 1371 (Fla.1981). 2 The Florida Supreme Court's understanding of the evidence differed sharply from that of the trial court with respect to the degree of Enmund's participation. In its sentencing findings, the trial court concluded that Enmund was a major participant in the robbery because he planned the robbery in advance and himself shot the Kerseys. 399 So.2d, at 1372. Both of these findings, as we understand it, were rejected by the Florida Supreme Court's holding that the only supportable inference with respect to Enmund's participation was that he drove the getaway car. The dissent, while conceding that this holding negated the finding that Enmund was one of the triggermen, argues that the trial court's finding that Enmund planned the robbery was implicitly affirmed. Post, at 809. As we have said, we disagree with that view. In any event, the question is irrelevant to the constitutional issue before us, since the Florida Supreme Court held that driving the escape car was enough to warrant conviction and the death penalty, whether or not Enmund intended that life be taken or anticipated that lethal force would be used. 3 In Armstrong the Florida Supreme Court rejected the trial court's conclusion that the Kerseys had been killed in order to eliminate them as witnesses, and stated that according to the only direct account of the events, "the shootings were indeed spontaneous and were precipitated by the armed resistance of Mrs. Kersey." 399 So.2d, at 963. 4 The petitioner argues a second question: whether the degree of Enmund's participation in the killings was given the consideration required by the Eighth and Fourteenth Amendments. We need not deal with this question. 5 Cal.Penal Code Ann. §§ 189, 190.2(a)(17) (West Supp.1982); Fla.Stat. §§ 782.04(1)(a), 775.082(1), 921.141(5)(d) (1981); Ga.Code §§ 26-1101(b), (c), 27-2534.1(b)(2) (1978); Miss.Code Ann. §§ 97-3-19(2)(e), 99-19-101(5)(d) (Supp.1981); Nev.Rev.Stat. §§ 200.030(1)(b), 200.030(4), 200.033(4) (1981); S.C.Code §§ 16-3-10, 16-3-20(C)(a)(1) (1976 and Supp.1981); Tenn.Code Ann. §§ 39-2402(a), 39-2404(i)(7) (Supp.1981); Wyo.Stat. §§ 6-4-101, 6-4-102(h)(iv) (1977). 6 Mo.Rev.Stat. §§ 565.001, 565.003, 565.008(2) (1978) (death penalty may be imposed only for capital murder; felony murder is first-degree murder); N.H.Rev.Stat.Ann. §§ 630:1, 630:1(III), 630:1-a(I)(b)(2) (1974 and Supp.1981) (capital murder includes only killing a law enforcement officer, killing during a kidnaping, and murder for hire); 18 Pa.Cons.Stat. §§ 2502(a), (b), 1102 (1980) (death penalty may be imposed only for first-degree murder; felony murder is second-degree murder). Wash.Rev.Code §§ 9A.32.030, 10.95.020 (1981) (death penalty may be imposed only for premeditated killing). 7 Ala.Code §§ 13A-2-23, 13A-5-40(a)(2), 13A-6-2(a)(1) (1977 and Supp.1982) (to be found guilty of capital murder, accomplice must have had "intent to promote or assist the commission of the offense" and murder must be intentional); Ill.Rev.Stat., ch. 38, &Par; 9-1(a)(3), 9-1(b)(6) (1979) (capital crime only if defendant killed intentionally or with knowledge that his actions "created a strong probability of death or great bodily harm"); La.Rev.Stat.Ann. § 14:30(1) (West Supp.1982) ("specific intent to kill"); N.M.Stat.Ann. §§ 30-2-1(A)(2), 31-18-14(A), 31-20A-5 (Supp.1981) (felony murder is a capital crime but death penalty may not be imposed absent intent to kill unless victim was a peace officer); Ohio Rev.Code Ann. §§ 2903.01(B), (C), (D), 2929.02(A), 2929.04(A)(7) (1982) (accomplice not guilty of capital murder unless he intended to kill); Tex.Penal Code Ann. §§ 19.02(a), 19.03(a)(2) (1974) "intentionally commits the murder in the course of [a felony]"); Utah Code Ann. § 76-5-202(1) (1978) ("intentionally or knowingly causes the death of another"); Va.Code § 18.2-31(d) (1982) ("willful, deliberate and premeditated killing of any person in the commission of robbery while armed with a deadly weapon"). 8 Ark.Stat.Ann. § 41-1501(1)(a) (1977) ("extreme indifference to . . . life"); see also § 41-1501, Commentary ("an inadvertent killing in the course of a felony will not . . . support . . . a conviction entailing punishment by death"); Del.Code Ann., Tit. 11, §§ 636(a)(2), (6) (1979) ("recklessly" or "with criminal negligence" causes death during the commission of a felony); Ky.Rev.Stat. § 507.020(1)(b) (Supp.1980) (defendant must manifest "extreme indifference to human life" and "wantonly engag[e] in conduct which creates a grave risk of death . . . and thereby causes . . . death"); see also Commentary following Criminal Law of Kentucky Annotated, Penal Code § 507.020, p. 677 (1978) (each accomplice's "participation in [the] felony" must "constitut[e] wantonness manifesting extreme indifference to human life"). 9 Md.Code Ann., Art. 27, §§ 410, 412(b), 413(d)(10), 413(e)(1) (1982) (except in cases of murder for hire, only principal in the first degree subject to the death penalty). In addition, two jurisdictions already accounted for in n. 7, supra, also preclude the death penalty where the defendant did not commit the murder. Ill.Rev.Stat., ch. 38, &Par; 9-1(a)(3), 9-1(b)(6) (1979) (defendant must actually kill victim); Va.Code §§ 18.2-31(d), 18.2-10(a), 18.2-18 (1982) (except in cases of murder for hire, only principal in the first degree may be tried for capital murder). 10 Colo.Rev.Stat. § 16-11-103(5)(d)(1978); 49 U.S.C. § 1473(c)(6)(D) (same). 11 Vt.Stat.Ann., Tit. 13, §§ 2303(b), (c) (Supp.1981) (capital murder reserved for offenders who commit a second unrelated murder or murder of a correctional officer). 12 Ariz.Rev.Stat.Ann. § 13-703(G)(3) (Supp.1981-1982) ("relatively minor" participation); Conn.Gen.Stat. § 53a-46a(f)(4) (Supp.1982) (same); Ind.Code § 35-50-2-9(c)(4) (Supp.1981) (same); Mont.Code Ann. § 46-18-304(6) (1981) (same); Neb.Rev.Stat. § 29-2523(2)(e) (1979) (same); N.C.Gen.Stat. § 15A-2000(f)(4) (Supp.1981) (same). 13 Idaho Code § 19-2515(f) (1979); Okla.Stat., Tit. 21, § 701.12 (1981); S.D.Comp.Laws Ann. § 23A-27A-1 (Supp.1981). 14 See the Ala., Colo., Conn., Md., Ohio, Pa., S.D., and Wash. statutes cited in nn. 5-7, 9, 10, 12, and 13, supra. 15 The dissent characterizes the state statutes somewhat differently. It begins by noting that 31 States "authorize a sentencer to impose a death sentence for a death that occurs during the course of a robbery." Post, at 819. That is not relevant to this case, however. Rather, at issue is the number of States which authorize the death penalty where the defendant did not kill, attempt to kill, or intend to kill. The dissent divides the statutes into three categories. Its first category of 20 statutes include 8 about which there is no disagreement—Cal., Fla., Ga., Miss., Nev., S.C., Tenn., and Wyo. In 11 other States listed by the dissent—Ariz., Colo., Conn., Idaho, Ind., Mont., Neb., N.M., N.C., Okla., and S.D.—the dissent looks solely at the provisions defining the crime of capital murder. Colorado's capital sentencing statute makes a defendant's minimal participation in a murder an absolute defense to imposition of the death penalty. See n.10, supra. Contrary to the dissent's claim that this provision would have been of no help to petitioner, see post, at 820, n. 36, if the case is judged on the basis of the Florida Supreme Court's findings, see n.2, supra, Colorado law may well have barred imposition of the death penalty in this case. Similarly, the Ariz., Conn., Ind., Mont., Neb., and N.C. capital sentencing statutes do not permit capital punishment solely for vicarious felony murder and reduce the likelihood that the death penalty will be imposed on a vicarious felony murderer, even where aggravating circumstances are present, by making a defendant's minimal participation in the homicide a mitigating circumstance. See n.12, supra. Three other States—Idaho, Okla., and S.D.—allow a defendant who does not kill or actually kill to be executed only where other aggravating circumstances are present, and in those States the felony murder itself cannot serve as an aggravating circumstance. See n.13, supra. New Mexico's capital sentencing statute requires the jury to find at least one statutory aggravating circumstance before the death penalty may be imposed, and in addition aggravating circumstances must outweigh mitigating circumstances. N.M.Stat.Ann. §§ 31-20A-4(C)(1) and (2) (Supp.1981). The statute lists seven statutory aggravating circumstances, six of which require an intent to kill. §§ 31-20A-5(B)-(G). The only aggravating circumstance which does not include an intent element is not applicable here, for it requires that the victim must be "a peace officer who was acting in the lawful discharge of an official duty when he was murdered." § 31-20A-5(A). The remaining State, Vermont, limits the death penalty to narrow circumstances not present here. See n.11, supra. There is no disagreement that three States require a culpable mental state short of intent before a nontriggerman may be put to death, compare n. 8, supra, with post, at 821, n. 37, a mental state which Enmund was not proved to possess. Similarly, the dissent's second category of seven States which authorize the death penalty only if the defendant had specific intent to kill the victim differs from our group of specific-intent States only because we include New Mexico in that group. Compare n. 7, supra, with post, at 821-822, n. 38. Finally, there is no disagreement that three States restrict application of the death penalty to felony murderers who actually kill. Compare n. 9, supra, with post, at 822, n. 39. 16 See App. D to Brief for Petitioner. 17 There is no reason to believe that this group of 16 contains a higher proportion of nontriggermen than does the rest of the defendants studied. 18 See NAACP Legal Defense and Educational Fund, Inc., Death Row U.S.A. 1, n. * (Oct. 20, 1981). 19 See App. E to Brief for Petitioner; NAACP Legal Defense and Educational Fund, Inc., Death Row U.S.A. (Oct. 20, 1981). 20 See App. to Reply Brief for Petitioner A-1—A-7. 21 These statistics concerning the number of vicarious felony murderers who have been executed and the number of them on death row are consistent with the findings of a study of 111 cases in which the defendant was found guilty of a capital crime and hence could have received the death penalty. Kalven & Zeisel, The American Jury and the Death Penalty, 33 U.Chi.L.Rev. 769 (1966). The authors found that juries rebel "at imposing the death penalty for the vicarious criminal responsibility of the defendant," id., at 776, to the extent that felony murder and accomplice factors accounted for more jury decisions not to impose the death penalty when the trial judge decided to impose the death penalty than any other factor. Id., at 777. The authors had anticipated that "because of the rigidity of the felony murder rule, the jury's sense of equity would produce a broad area of disagreement." Id., at 776, n. 10. However, they found that "disagreement over the rule emerges only at the level of the death penalty." Ibid. 22 "[T]he climate of international opinion concerning the acceptability of a particular punishment" is an additional consideration which is "not irrelevant." Coker v. Georgia, 433 U.S. 584, 596, n. 10, 97 S.Ct. 2861, 2868, n. 10, 53 L.Ed.2d 982 (1977). It is thus worth noting that the doctrine of felony murder has been abolished in England and India, severely restricted in Canada and a number of other Commonwealth countries, and is unknown in continental Europe. ALI, Model Penal Code § 210.2, pp. 39-40 (Off. Draft and Revised Comments 1980) (hereafter Model Penal Code). It is also relevant that death sentences have not infrequently been commuted to terms of imprisonment on the grounds of the defendant's lack of premeditation and limited participation in the homicidal act. See Wolfgang, Kelly, & Nolde, Comparison of the Executed and Commuted Among Admissions to Death Row, 53 J.Crim.L.C. & P. S. 301, 310 (1962). 23 The statistics relied upon by the American Law Institute may be summarized as follows: Robberies Date & No. of Accompanied Location Robberies by Homicide % Cook County, Ill. 14,392 (est.) 71 .49 1926-1927 Philadelphia, Pa. 6,432 38 .59 1948-1952 New Jersey 16,273 66 .41 1975 Model Penal Code § 210.2, Comment, p. 38, n. 96. 24 An estimated total of 548,809 robberies occurred in the United States in 1980. U.S. Dept. of Justice, Federal Bureau of Investigation, Uniform Crime Reports 17 (1981). Approximately 2,361 persons were murdered in the United States in 1980 in connection with robberies, id., at 13, and thus only about 0.43% of robberies in the United States in 1980 resulted in homicide. See also Cook, The Effect of Gun Availability on Robbery and Robbery Murder, in 3 R. Haveman & B. Zellner, Policy Studies Review Annual 743, 747 (1980) (0.48% of all robberies result in murder). 1 Much of the evidence concerning these crimes came from J. B. Neal, to whom Sampson Armstrong made numerous admissions on the day of the murders. See Record 1344-1365. 2 J. B. Neal testified that Armstrong had told him that two guns were involved; Jeanette had one and Sampson had the other. Id., at 1354. 3 An autopsy revealed that Mr. Kersey had been shot twice, once with a .38-caliber bullet, and once with a .22-caliber bullet. Mrs. Kersey had been shot six times; of the bullets that could be identified, two were fired from a .38-caliber gun, and one from a .22-caliber gun. According to a firearms expert, the .22-caliber bullets were fired from the same gun, and the .38-caliber bullets were fired from the same gun. See 399 So.2d 1362, 1364 (Fla.1981). 4 Ida Jean Shaw was the petitioner's common-law wife and Jeanette Armstrong's mother. She was later given immunity from prosecution in return for her testimony. Record 1178-1179. 5 Thomas Kersey normally kept large sums of money in his wallet and indiscriminately showed the cash to people he dealt with. A few weeks before his murder, Kersey revealed the contents of his wallet to the petitioner and bragged that at any time he could "dig up $15,000, $16,000." 399 So.2d, at 1365. See Record 1205-1206. 6 Ida Jean Shaw's trial testimony contradicted her earlier statements to police. When police initially questioned her, she insisted that Jeanette had been shot by an unknown assailant while she and Jeanette had been traveling to a nearby town. Id., at 1191-1192. Later she gave investigators a statement implicating the petitioner and Sampson Armstrong in the murders. Id., at 1209-1210. Subsequently, she gave two more statements repudiating the statement implicating the petitioner. Id., at 1208-1209. In his closing argument, the prosecutor acknowledged the conflict between Ida Jean Shaw's testimony that she was not in the yellow Buick the morning of the murders, and the testimony of a witness who saw her in the car shortly before and after the murders. The prosecutor deemed the inconsistency irrelevant. Id., at 1571-1572. 7 At the sentencing hearing, the prosecutor theorized that the petitioner was not the "trigger man," but the "person who set it all up." Id., at 1679. The prosecutor admitted that he did not "know whether [the petitioner] set foot inside that house or not. But he drove them there. He set it up, planned it." Id., at 1679-1680. In this Court as well, the State acknowledges that the petitioner "was apparently not the triggerman in the two murders involved in his [sic ] case." Brief in Opposition 14. 8 In Florida at the time of the Kersey murders, first-degree murder was defined in Fla.Stat. § 782.04(1)(a) (1973) as "[t]he unlawful killing of a human being, when perpetrated from a premeditated design to effect the death of the person killed or any human being, or when committed by a person engaged in the perpetration of, or in the attempt to perpetrate, any . . . robbery. . . ." In instructing the jury on first-degree murder, the judge read the above provision verbatim. Record 1605-1606. He also added that "[t]he killing of a human being while engaged in the perpetration of or in the attempt to perpetrate the offense of robbery is murder in the first degree even though there is no premeditated design or intent to kill." Id., at 1606. Distinguishing first- and second-degree felony murder, the judge stated: "In order to sustain a conviction of first degree murder while engaging in the perpetration of or in the attempted perpetration of the crime of robbery, the evidence must establish beyond a reasonable doubt that the defendant was actually present and was actively aiding and abetting the robbery or attempted robbery, and that the unlawful killing occurred in the perpetration of or in the attempted perpetration of the robbery. "In order to sustain a conviction of second degree murder while engaged in the perpetration of or the attempted perpetration of robbery, the evidence must establish beyond a reasonable doubt that the unlawful killing was committed in the perpetration of or in the attempted perpetration of robbery, and that the defendant actually, although not physically present at the time of the commission of the offense, did, nonetheless, procure, counsel, command or aid another to commit the crime." Id., at 1609-1610. 9 On the motion of the petitioner and the prosecution, Jeanette Armstrong's trial had been severed from the trial of her co-defendants. Id., at 50, 57. Jeanette Armstrong was tried first and convicted of two counts of second-degree murder and one count of robbery. The trial judge sentenced her to three consecutive life sentences. 399 So.2d, at 1371. 10 Under Florida law, the "court shall conduct a separate sentencing proceeding to determine whether the defendant should be sentenced to death or life imprisonment." Fla.Stat. § 921.141(1) (1981). The jury renders only an "advisory sentence" based on the mitigating and aggravating circumstances. § 921.141(2). At the sentencing hearing, the petitioner presented no evidence, Record 1677, but his attorney argued that the death penalty was inappropriate because at most the evidence showed that the petitioner saw Thomas Kersey's money, suggested the robbery, and drove the Armstrongs to the Kersey house. Id., at 1683-1684. He also argued that death was an excessive penalty because the gunfight was spontaneous, and beyond the petitioner's control. Id., at 1684. 11 Initially, the trial court failed to make written findings as required by Fla.Stat. § 921.141(3) (1981). On the first state appeal, the Florida Supreme Court remanded the case for such findings. See App. 29. 12 Regarding the extent of the petitioner's involvement, the trial court reasoned that because two different guns had been used in the murders, and because Jeanette Armstrong had been seriously wounded by gunfire, the petitioner must have fired one of the guns. Moreover, since each of the Kerseys was injured by a bullet of each type, the petitioner must have shot each victim. Id., at 31; 399 So.2d, at 1372. 13 The court also rejected the other statutory mitigating circumstances. In particular, the petitioner did not have a record free of criminal convictions, Fla.Stat. § 921.141(6)(a) (1981); there was no evidence that he had acted under the influence of extreme mental or emotional disturbance, § 921.141(6)(b); there was no evidence that the victims were participants in or consented to the crimes, § 921.141(6)(c); there was no evidence that he acted under extreme duress or under the substantial domination of another person, § 921.141(6)(e); there was no evidence that the petitioner was incapable of appreciating the criminality of his conduct or conforming his conduct to the requirements of law, § 921.141(6)(f); and because he was 42 years old at the time of the offense, his age was not a mitigating factor, § 921.141(6)(g). App. 32; 399 So.2d, at 1372-1373. 14 The trial court made nearly identical findings for Sampson Armstrong. In particular, it found that the murders were committed during the course of a robbery, that they were committed for pecuniary gain, and that they were especially heinous, atrocious, or cruel. See Armstrong v. State, 399 So.2d 953, 960-961 (Fla.1981). The trial court considered the only possible mitigating circumstance to be Armstrong's age (23), but did not actually find that fact to be mitigating. See id., at 962 ("the factor of age was given no consideration"). Finding that the aggravating circumstances outweighed the mitigating circumstances, the trial judge imposed the death penalty for each murder conviction, and imposed a life sentence for the robbery. Id., at 955, 962. 15 The Florida Supreme Court also affirmed the convictions and sentences of Sampson Armstrong. See Armstrong v. State, supra, at 960. 16 Second-degree murder, based on felony murder, is defined in Fla.Stat. § 782.04(3) (1973): "[W]hen committed in the perpetration of, or in the attempt to perpetrate, any . . . robbery, . . . except as provided in subsection (1), it shall be murder in the second degree . . . punishable by imprisonment in the state prison for life or for such term of years as may be determined by the court." 17 The court also noted that Sampson Armstrong's admissions to J. B. Neal made no mention of the petitioner, and that the petitioner's admissions to Ida Jean Shaw indicated only "his complicity." 399 So.2d, at 1370. 18 In Armstrong, the Florida Supreme Court expressly had rejected the trial court's conclusion that the Kerseys were murdered in order to eliminate them as witnesses. "It simply cannot be said that there was proof that the robbers killed in order to assure that there would be no witnesses against them." 399 So.2d, at 963. On the contrary, "[t]he only direct account of what transpired is from the testimony of J. B. Neal about Armstrong's statement to him. By that account, the shootings were indeed spontaneous and were precipitated by the armed resistance of Mrs. Kersey." Ibid. In reaching this conclusion, the State Supreme Court also rejected the trial court's conclusions derived from the pathologist's testimony. Rather than indicating that the victims were prone when shot, the pathologist's testimony "as to the direction of fire and the positions of the victims when shot [was] equivocal at best." Ibid. 19 In this Court, the petitioner neither challenges his convictions for robbery and felony murder nor argues that the State has overstepped constitutional bounds in defining murder to include felony murder. The petitioner's sole challenge is to the penalty imposed for the murders. 20 Although the petitioner ostensibly relies on the fact that he was not the triggerman, the core of his argument is that the death penalty is disproportionate to his crime because he did not have the specific intent to kill the Kerseys. Pulling the trigger is only one factor, albeit a significant one, in determining intent. See Tr. of Oral Arg. 21-23 (counsel for petitioner asserting that so long as a defendant had the intent to kill, he need not actually have pulled the trigger in order to be subjected to capital punishment, and that even if he had pulled the trigger, he would not be subject to the death penalty absent a specific intent to kill). 21 The Eighth Amendment provides that "Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted." 22 See Roberts (Stanislaus) v. Louisiana, 428 U.S. 325, 96 S.Ct. 3001, 49 L.Ed.2d 974 (1976) (holding that Louisiana's mandatory death penalty statute violated the Eighth and Fourteenth Amendments); Woodson v. North Carolina, 428 U.S. 280, 96 S.Ct. 2978, 49 L.Ed.2d 944 (1976) (holding that the State's mandatory death penalty statute violated the Eighth and Fourteenth Amendments); Jurek v. Texas, 428 U.S. 262, 96 S.Ct. 2950, 49 L.Ed.2d 929 (1976) (upholding the Texas death penalty statute); Proffitt v. Florida, 428 U.S. 242, 96 S.Ct. 2960, 49 L.Ed.2d 913 (1976) (upholding Florida's death penalty statute). 23 In only one case since Gregg has this Court upheld a challenged death sentence. See Dobbert v. Florida, 432 U.S. 282, 97 S.Ct. 2290, 53 L.Ed.2d 344 (1977) (holding that changes in the death penalty statute between the time of the murder and the sentencing did not amount to an ex post facto violation). In five cases, the Court vacated the death sentence because the sentencer could not or did not consider all mitigating factors proffered by the defendant. See Roberts (Harry) v. Louisiana, 431 U.S. 633, 97 S.Ct. 1993, 52 L.Ed.2d 637 (1977) (per curiam); Lockett v. Ohio, 438 U.S. 586, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978) (plurality opinion); Bell v. Ohio, 438 U.S. 637, 98 S.Ct. 2977, 57 L.Ed.2d 1010 (1978) (plurality opinion); Green v. Georgia, 442 U.S. 95, 99 S.Ct. 2150, 60 L.Ed.2d 738 (1979) (per curiam); Eddings v. Oklahoma, 455 U.S. 104, 102 S.Ct. 869, 71 L.Ed.2d 1 (1982) (adopting the reasoning of the Lockett plurality as the holding of the Court). In two cases, the Court reversed the judgments affirming the death sentences because the jury had been selected in violation of Witherspoon v. Illinois, 391 U.S. 510, 88 S.Ct. 1770, 20 L.Ed.2d 776 (1968). See Adams v. Texas, 448 U.S. 38, 100 S.Ct. 2521, 65 L.Ed.2d 581 (1980); Davis v. Georgia, 429 U.S. 122, 97 S.Ct. 399, 50 L.Ed.2d 339 (1976) (per curiam). In five other cases, the Court vacated death sentences for a variety of reasons unrelated to the proportionality of the punishment to the crime. See Gardner v. Florida, 430 U.S. 349, 97 S.Ct. 1197, 51 L.Ed.2d 393 (1977) (plurality opinion) (due process violated when defendant had no chance to explain or deny information given to the sentencing judge); Godfrey v. Georgia, 446 U.S. 420, 100 S.Ct. 1759, 64 L.Ed.2d 398 (1980) (plurality opinion) (reversing the death sentence because the aggravating circumstance relied upon by jury was not so tailored as to avoid arbitrary and capricious infliction of death penalty); Beck v. Alabama, 447 U.S. 625, 100 S.Ct. 2382, 65 L.Ed.2d 392 (1980) (holding that death penalty may not be imposed where jury was precluded from considering lesser included noncapital offense, when evidence existed to support such a verdict); Bullington v. Missouri, 451 U.S. 430, 101 S.Ct. 1852, 68 L.Ed.2d 270 (1981) (holding that Double Jeopardy Clause prevented imposition of death sentence upon retrial when jury had imposed life imprisonment at the first trial); Estelle v. Smith, 451 U.S. 454, 101 S.Ct. 1866, 68 L.Ed.2d 359 (1981) (holding that admission of psychiatrist's testimony at the penalty phase of the capital trial violated the defendant's Fifth Amendment privilege against self-incrimination because he had not been told before his psychiatric examination that his statements could be used against him during the sentencing proceeding). In Coker v. Georgia, 433 U.S. 584, 97 S.Ct. 2861, 53 L.Ed.2d 982 (1977), the Court vacated a death sentence for a man who had been convicted of rape of an adult woman. Nevertheless, the Court made clear that the death penalty is not per se disproportionate to the crime of murder. See, e.g., id., at 591, 97 S.Ct., at 2865 (opinion of WHITE, J.) ("It is now settled that the death penalty is not invariably cruel and unusual punishment within the meaning of the Eighth Amendment; . . . neither is it always disproportionate to the crime for which it is imposed"); id., at 604, 97 S.Ct., at 2872 (opinion of BURGER, C. J.) (accepting "that the Eighth Amendment's concept of disproportionality bars the death penalty for minor crimes," but rejecting the argument that death is a disproportionate punishment for rape, much less murder). 24 Strictly speaking, this Court cannot state unequivocally whether the petitioner specifically intended either to kill the Kerseys or to have them killed because the trial court made no findings on these issues. The trial court, however, did make the finding, not rejected by the Florida Supreme Court, that the petitioner's participation was not minor, but "major" in that he "planned the capital felony and actively participated in an attempt to avoid detection by disposing of the murder weapons." App. 32. Accordingly, I proceed on the assumption that the petitioner's only intent was to commit an armed robbery with his accomplices, the Armstrongs. 25 Justice POWELL concurred in the plurality's reasoning in concluding that "ordinarily" death was disproportionate for such a crime, but stopped short of a per se rule. 433 U.S., at 601, 97 S.Ct., at 2870. Justice BRENNAN and Justice MARSHALL concurred in the judgment, adhering to their previously announced views that the death penalty is in all circumstances cruel and unusual punishment prohibited by the Eighth and Fourteenth Amendments. See id., at 600-601, 97 S.Ct., at 2870-71. 26 In fact, two of those States, Louisiana and North Carolina, did not define rape as a capital felony when they reenacted their death penalty statutes following their invalidation in Woodson v. North Carolina, 428 U.S. 280, 96 S.Ct. 2978, 49 L.Ed.2d 944 (1976), and Roberts v. Louisiana, 428 U.S. 325, 96 S.Ct. 3001, 49 L.Ed.2d 974 (1976). See 433 U.S., at 594, 97 S.Ct., at 2867. Consequently, at the time Coker was decided only Georgia authorized the death penalty for the rape of an adult woman. 27 The Court has conducted a less searching inquiry for punishments less than death. See Rummel v. Estelle, 445 U.S. 263, 100 S.Ct. 1133, 63 L.Ed.2d 382 (1980) (upholding, against an Eighth Amendment challenge, a life sentence imposed under a state recidivist statute); Hutto v. Davis, 454 U.S. 370, 102 S.Ct. 703, 70 L.Ed.2d 556 (1981) (per curiam ) (upholding, on the basis of Rummel, a 40-year sentence for two marijuana convictions). In Rummel, the Court expressly noted that, for purposes of Eighth Amendment analysis, those "decisions applying the prohibition of cruel and unusual punishments to capital cases are of limited assistance in deciding the constitutionality" of prison sentences. 445 U.S., at 272, 100 S.Ct., at 1138. 28 According to one source, at early common law most felonies were capital crimes, but attempts were punished as misdemeanors and accidental killings were not punishable at all. The felony-murder rule was an effort to create felony liability for accidental killings caused during the course of an attempted felony. See ALI, Model Penal Code § 210.2, Comment, p. 31, n. 74 (Off. Draft and Revised Comments 1980). 29 See English Homicide Act of 1957, 5 & 6 Eliz. 2, ch. 11. The English attitude toward capital punishment, as reflected in recent legislation, differs significantly from American attitudes as reflected in state legislation; in 1965, England abolished the death penalty for all murders. See Murder (Abolition of Death Penalty) Act of 1965, 8 Halsbury's Statutes of England 541 (3d ed. 1969). 30 See Comment, The Constitutionality of Imposing the Death Penalty for Felony Murder, 15 Hous.L.Rev. 356, 364-365 (1978); Alderstein, Felony-Murder in the New Criminal Codes, 4 Am.J.Crim.L. 249, 251-252 (1976). 31 See, e.g., Royal Commission on Capital Punishment 1949-1953, Report 31-33 (1953) (reporting that application of the felony-murder doctrine was limited to those cases in which the verdict could have been intentional murder); Law Revision Commission of the State of New York, 3d Annual Report 665, 668, and n. 444 (1937). It is significant that the New York Legislature rejected the Commission's recommendation of requiring some element of mens rea, and instead adopted a scheme giving jurors discretion to recommend life sentences. See 1937 N.Y.Laws, ch. 67. 32 The extent of jury nullification and the nearly complete repudiation of mandatory death penalty laws led a plurality of this Court to conclude that the "two crucial indicators of evolving standards of decency respecting the imposition of punishment in our society—jury determinations and legislative enactments—both point conclusively to the repudiation of automatic death sentences." Woodson v. North Carolina, 428 U.S., at 293, 96 S.Ct., at 2986 (opinion of Stewart, POWELL, and STEVENS, JJ.). These factors supported the Court's conclusion that North Carolina's mandatory death penalty law violated the Eighth Amendment. 33 See App. D to Brief for Petitioner. Moreover, the last nontriggerman was executed in 1955. By contrast, 72 rapists were executed between 1955 and this Court's 1977 decision in Coker. Brief for Petitioner 34-35. 34 See App. E to Brief for Petitioner; NAACP Legal Defense and Education Fund, Inc., Death Row U. S. A. (Oct. 20, 1981). 35 Only Missouri, New Hampshire, and Pennsylvania define felony murder as a crime distinct from capital murder. See Mo.Rev.Stat. §§ 565.001, 565.003, 565.008(2) (1978); N.H.Rev.Stat.Ann. §§ 630:1, 630:1-a(I)(b)(2), 630:1-a(III) (1974 and Supp.1981); 18 Pa.Cons.Stat. §§ 2502(a), (b), (d), 1102(b) (1980). One exception to the New Hampshire scheme is § 630:1(I)(b), which includes in the definition of capital murder a death caused "knowingly" in the course of a kidnaping. A fourth State, Washington, permits imposition of the death penalty if premeditated murder is aggravated by, inter alia, commission during a felony. Wash.Rev.Code §§ 9A.32.030(1)(a), 10.95.020(9) (1981). 36 See Ariz.Rev.Stat.Ann. §§ 13-1105(A)(2), (C) (Supp.1981-1982); Cal.Penal Code Ann. §§ 189, 190 (West Supp.1982); Colo.Rev.Stat. §§ 18-3-102(1)(b), 18-1-105(1)(a) (1978 and Supp.1981); Conn.Gen.Stat.Ann. §§ 53a-54b, 53a-54c, 53a-35a(1) (West Supp.1982); Fla.Stat. §§ 782.04(1)(a), 775.082(1) (1981); Ga.Code §§ 26-1101(b), (c) (1978); Idaho Code §§ 18-4003(d), 4004 (1979); Ind.Code §§ 35-42-1-1(2), 35-50-2-3(b) (Supp.1981); Miss.Code Ann. §§ 97-3-19(2)(e), 97-3-21 (Supp.1981); Mont.Code Ann. §§ 45-5-102(1)(b), (2) (1981); Neb.Rev.Stat. §§ 28-303(2), 28-105(1) (1979); Nev.Rev.Stat. §§ 200.030(1)(b), 200.030(4)(a) (1981); N.M.Stat.Ann. §§ 30-2-1(A)(2), 31-18-14(A), 31-20A-5 (Supp.1981); N.C.Gen.Stat. § 14-17 (1981); Okla.Stat., Tit. 21, §§ 701.7(B), 701.9(A) (1981); S.C.Code §§ 16-3-10, 16-3-20(C)(a)(1) (1976 and Supp.1981); S.D.Codified Laws §§ 22-16-4, 22-16-12, 22-6-1(1), 22-3-3 (1979 and Supp.1981); Tenn.Code Ann. §§ 39-2402(a), (b) (Supp.1981); Vt.Stat.Ann., Tit. 13, §§ 2301, 2303(b), (c) (1974 and Supp.1981); and Wyo.Stat. §§ 6-4-101(a), (b) (1977). Two of these States, Colorado and Connecticut, provide that it is an affirmative defense to the capital crime if the accomplice did not "in any way solicit, request, command, importune, cause or aid the commission" of the homicidal act; was not armed with a deadly weapon and had no reason to believe that his cofelons were so armed; and did not engage or intend to engage, and had no reason to believe that his cofelons would engage, in conduct "likely to result in death or serious bodily injury." See Colo.Rev.Stat. § 18-3-102(2) (1978); Conn.Gen.Stat. § 53a-54c (Supp.1982). Colorado also prevents imposition of the death penalty if the defendant's role, though sufficient to establish guilt, was "relatively minor." Colo.Rev.Stat. § 16-11-103(5)(d) (1978). Even if they were available under the Florida statute, these provisions would have been of no help to the petitioner since the trial court found that there were no mitigating circumstances, in part because Enmund's role in the capital felony was not minor. See Fla.Stat. § 921.141(6)(d) (1981). The State Supreme Court expressly affirmed the trial court's finding of no mitigating circumstances, and therefore the finding that the petitioner's role was not minor. 399 So.2d, at 1373. Of course, not all of the statutes listed above are identical. Several of them provide that robbery murder is a capital felony, but require proof of additional aggravating circumstances, e.g., the defendant had been convicted previously of a violent felony, or the victim was a correctional officer, before the death penalty can be imposed. See, e.g., Okla.Stat., Tit. 21, § 701.12 (1981); N.M.Stat.Ann. §§ 30-2-1(A)(2), 31-18-14(A), 31-20A-5 (Supp.1981). Others, like the Florida statute, define robbery murder as a capital offense and use the robbery as an aggravating circumstance. The common thread in all of these statutes, however, is that the defendant need not have the intent to kill in order to be subject to the death penalty. The Court's additional subdivision of this group of statutes, see ante, at 791-793, and nn. 10-13, serves only to obscure the point that 20 States permit imposition of the death penalty even though the defendant did not actually kill, and had no intent to kill. 37 See Ark.Stat.Ann. §§ 41-1501(1)(a), (2), (3) (1977) (a capital crime if death occurs during commission of the felony "under circumstances manifesting extreme indifference to the value of human life"); Del.Code Ann., Tit. 11, §§ 636(a)(6), 636(b), 4209(a) (1979) (a capital crime only if the death is caused "with criminal negligence"); Ky.Rev.Stat. § 507.020(1)(b), (2) (Supp.1980) (defendant must "caus[e] the death of another person" under "circumstances manifesting extreme indifference to human life [and while] wantonly engag[ing] in conduct which creates a grave risk of death to another person"). It is an affirmative defense to capital felony murder in Arkansas if the "defendant did not commit the homicide act or in any way solicit, command, induce, procure, counsel, or aid its commission." Ark.Stat.Ann. § 41-1501(2) (1977). At oral argument, counsel for petitioner stated that "the determining factor is the intent to take life, conscious purpose to take life." Tr. of Oral Arg. 18. Under the petitioner's proposed standard, these statutes would be unconstitutional. 38 See Ala.Code §§ 13A-2-23, 13A-5-40(a)(2), (b), (c), (d), 13A-6-2(a)(1) (1977 and Supp.1982) (the accomplice is not guilty of capital murder unless the killing is intentional, and the accomplice had "intent to promote or assist the commission" of the murder); Ill.Rev.Stat., ch. 38, &Par; 9-1(a)(3), 9-1(b)(6) (1979) (a capital crime only if the defendant killed intentionally or with knowledge that his actions "created a strong probability of death or great bodily harm"); La.Rev.Stat.Ann. § 14.30(1) (West Supp.1982) (defendant is guilty of capital murder only if he had "specific intent to kill or to inflict great bodily harm"); Ohio Rev.Code Ann. §§ 2903.01(B), (C), (D), 2929.02(A), 2929.04(A)(7) (1982) (accomplice is not guilty of the capital crime unless he "purposely cause[d]" the death and was "specifically found to have intended to cause the death of another"; if defendant is not the "principal offender," the death penalty is precluded unless he "committed the aggravated murder with prior calculation and design"); Tex.Penal Code Ann. §§ 12.31, 19.03(a)(2), 19.02(a)(1) (1974) (defendant is guilty of capital murder only if he "intentionally or knowingly" caused death during the course of the robbery); Utah Code Ann. §§ 76-5-202(1)(d), (2), 76-3-206(1) (1978) (defendant is guilty of capital murder only if he "intentionally or knowingly" caused the death during the course of the robbery); and Va.Code §§ 18.2-31(d), 18.2-10(a) (1982) (capital murder only if killing is "willful, deliberate and premeditated"). 39 See Ill.Rev.Stat., ch. 38, &Par; 9-1(a)(3), 9-1(b)(6) (1979) (a capital crime only if the defendant actually killed the victim and the defendant killed intentionally or with knowledge that his actions "created a strong probability of death or great bodily harm"); Md.Ann.Code, Art. 27, §§ 410, 412(b), 413(d)(10), (e)(1) (1982) (except in cases of murder for hire, only principal in the first degree subject to the death penalty); Va.Code §§ 18.2-31(d), 18.2-10(a), 18.2-18 (1982) (except in cases of murder for hire, only the immediate perpetrator of the homicide, and not accomplice before the fact or principal in the second degree, may be tried for capital murder). Note that Illinois and Virginia also require an intent to kill. See n. 38, supra. 40 The Court's attempt to downplay the significance of Enmund's role in the murders, see ante, at 786-787, n. 2, does not square with the facts of this case. The trial court expressly found that because Enmund had planned the robbery, his role was not minor, and that therefore no statutory mitigating circumstances applied. The Florida Supreme Court affirmed the finding of no mitigating circumstances, thereby affirming the underlying factual predicate—Enmund had planned the armed robbery. Moreover, even Enmund's trial counsel conceded at the sentencing hearing that Enmund initiated the armed robbery and drove the getaway car. See n. 10, supra. The Court misreads the opinion below in suggesting that the State Supreme Court deduced from the sentencing hearing that Enmund's only participation was as the getaway driver. In fact, the court made that statement with respect to the guilt phase of the trial. As I mentioned above, Enmund's counsel conceded at the sentencing hearing that Enmund had initiated the armed robbery. 41 It is not true, as the petitioner suggests, that an intent-to-kill requirement would not interfere with the State's substantive categories of murder. Prohibiting the death penalty for accomplice felony murder would create a category of murder between capital murder, for which the death penalty is permitted, and the next statutory degree, for which some term of years (typically less than life imprisonment) is imposed. 42 The petitioner and the Court also contend that capital punishment for felony murder violates the Eighth Amendment because it "makes no measurable contribution to acceptable goals of punishment." Coker v. Georgia, 433 U.S., at 592, 97 S.Ct., at 2866. In brief, the petitioner and the Court reason that since he did not specifically intend to kill the Kerseys, since the probability of death during an armed robbery is so low, see ALI, Model Penal Code, supra n. 28, § 210.2, Comment, p. 38, n. 96 (concluding from several studies that a homicide occurs in about one-half of one percent of all robberies), and since the death penalty is so rarely imposed on nontriggermen, capital punishment could not have deterred him or anyone else from participating in the armed robbery. The petitioner and the Court also reject the notion that the goal of retribution might be served because his "moral guilt" is too insignificant. At their core, these conclusions are legislative judgments regarding the efficacy of capital punishment as a tool in achieving retributive justice and deterring violent crime. Surely, neither the petitioner nor the Court has shown that capital punishment is ineffective as a deterrent for his crime; the most the Court can do is speculate as to its effect on other felony murders and rely on "competent observers" rather than legislative judgments. See ante, at 799-800. Moreover, the decision of whether or not a particular punishment serves the admittedly legitimate goal of retribution seems uniquely suited to legislative resolution. Because an armed robber takes a serious risk that someone will die during the course of his crime, and because of the obviousness of that risk, we cannot conclude that the death penalty "makes no measurable contribution to acceptable goals of punishment." 43 Apparently, the Court also intends that the case be remanded for a new death sentence hearing, consistent, of course, with its holding today. 44 Although the petitioner challenges the constitutionality of his sentencing hearing, he does not challenge the constitutionality of the statutory capital sentencing procedures. See Proffitt v. Florida, 428 U.S. 242, 96 S.Ct. 2960, 49 L.Ed.2d 913 (1976) (upholding the Florida scheme). 45 See Songer v. State, 365 So.2d 696, 700 (Fla.1978) (holding that Fla.Stat. § 921.141(6) (1981), which lists mitigating circumstances, does not restrict the sentencer's consideration of mitigating circumstances to those expressly listed in the statute); Shriner v. State, 386 So.2d 525, 533 (Fla.1980), cert. denied, 449 U.S. 1103, 101 S.Ct. 899, 66 L.Ed.2d 829 (1981); 399 So.2d, at 1371. As noted above, the petitioner offered no additional evidence at the sentencing hearing in mitigation of his crime. See Record 1677. His counsel argued, however, that the petitioner did not deserve the death penalty because his role in the crime was relatively minor. Id., at 1683-1685. 46 The Florida Supreme Court's opinion fails to correct this error either by remanding for new sentencing or by evaluating the impact of the trial court's fundamental misperception of the petitioner's role in the killings. Rather, the court simply repeats three times, without any discussion of the evidence, that there are "no mitigating circumstances." 399 So.2d, at 1373. In light of the court's dramatically different factual findings, this review is inadequate to satisfy the Lockett principle.

01

458 U.S. 832 102 S.Ct. 3394 73 L.Ed.2d 1174 RAMAH NAVAJO SCHOOL BOARD, INC., et al., Appellantsv.BUREAU OF REVENUE OF NEW MEXICO. No. 80-2162. Argued April 28, 1982. Decided July 2, 1982. Syllabus Held: Federal law pre-empts New Mexico's tax imposed on the gross receipts that appellant non-Indian construction company received from appellant tribal school board for the construction of a school for Indian children on the reservation. White Mountain Apache Tribe v. Bracker, 448 U.S. 136, 100 S.Ct. 2636, 65 L.Ed.2d 742, controlling. Pp. 836-847. (a) In view of the federal and tribal interests arising from Congress' broad power to regulate tribal affairs under the Indian Commerce Clause, Art. I, § 8, cl. 3, and from the semi-autonomous status of Indian tribes, the exercise of state authority over commercial activity on an Indian reservation may be pre-empted by federal law, or it may interfere with the tribe's ability to exercise its sovereign functions. Traditional notions of tribal sovereignty, and the recognition and encouragement of such sovereignty in congressional Acts promoting tribal independence and economic development, inform the pre-emption analysis. Ambiguities in federal law should be construed generously, and federal preemption is not limited to those situations where Congress has explicitly announced an intention to pre-empt state activity. Pp. 837-839. (b) Federal statutes (particularly the Indian Self-Determination and Education Assistance Act) reflect the federal policy of encouraging the development of Indian-controlled institutions on the reservation, and under detailed regulations governing school construction the Bureau of Indian Affairs has wide-ranging authority to monitor and review subcontracting agreements between the Indian organization, which is viewed as the general contractor, and the non-Indian firm that actually constructs the facilities. The direction and supervision provided by the comprehensive federal regulatory scheme for the construction of Indian schools leave no room for the additional burden sought to be imposed by New Mexico. There is no merit to the contention that the state tax is not pre-empted merely because the federal statutes and regulations do not specifically express the intention to pre-empt this exercise of state authority. The interest asserted by the State relating to its providing services to the non-Indian contractor for its activities off the reservation is not a legitimate justification for a tax whose ultimate burden falls on the tribal organization. Nor is the State's purpose in imposing the tax pursuant to a general desire to increase revenues sufficient to justify the additional burdens thereby imposed on the comprehensive federal scheme regulating the creation and maintenance of educational opportunities for Indian children and on the express federal policy of encouraging Indian self-sufficiency in the area of education. Pp. 839-845. (c) Pre-emption analysis in this area need not be modified by applying a new approach relying on the Indian Commerce Clause. Existing pre-emption analysis governing this type of case provides sufficient guidance to state courts and also allows for more flexible consideration of the federal, state, and tribal interests at issue. Pp.845-846 95 N.M. 708, 625 P.2d 1225, reversed and remanded. Michael P. Gross, Santa Fe, N. M., for appellants. Louis F. Claiborne, Washington, D. C., for the United States as amicus curiae, by special leave of Court. Jan Unna, Santa Fe, N. M., for appellees. Justice MARSHALL delivered the opinion of the Court. 1 In this case, we address the question whether federal law pre-empts a state tax imposed on the gross receipts that a non-Indian construction company receives from a tribal school board for the construction of a school for Indian children on the reservation. The New Mexico Court of Appeals held that the gross receipts tax imposed by the State of New Mexico was permissible. Because the decision below is inconsistent with White Mountain Apache Tribe v. Bracker, 448 U.S. 136, 100 S.Ct. 2578, 65 L.Ed.2d 665 (1980) (White Mountain), we reverse. 2 * Approximately 2,000 members of the Ramah Navajo Chapter of the Navajo Indian Tribe live on tribal trust and allotment lands located in west central New Mexico. Ramah Navajo children attended a small public high school near the reservation until the State closed this facility in 1968. Because there were no other public high schools reasonably close to the reservation, the Ramah Navajo children were forced either to abandon their high school education or to attend federal Indian boarding schools far from the reservation. In 1970, the Ramah Navajo Chapter exercised its authority under Navajo Tribal Code, Title 10, § 51 (1969), and established its own school board in order to remedy this situation. Appellant Ramah Navajo School Board, Inc. (the Board), was organized as a nonprofit corporation to be operated exclusively by members of the Ramah Navajo Chapter. The Board is a Navajo "tribal organization" within the meaning of 25 U.S.C. § 450b(c), 88 Stat. 2204. With funds provided by the federal Bureau of Indian Affairs (BIA) and the Navajo Indian Tribe, the Board operated a school in the abandoned public school facility, thus creating the first independent Indian school in modern times.1 3 In 1972, the Board successfully solicited from Congress funds for the design of new school facilities. Pub. L. 92-369, 86 Stat. 510. The Board then contracted with the BIA for the design of the new school and hired an architect. In 1974, the Board contracted with the BIA for the actual construction of the new school to be built on reservation land. Funding for the construction of this facility was provided by a series of congressional appropriations specifically earmarked for this purpose.2 The contract specified that the Board was the design and building contractor for the project, but that the Board could subcontract the actual construction work to third parties. The contract further provided that any subcontracting agreement would have to include certain clauses governing pricing, wages, bonding, and the like, and that it must be approved by the BIA. 4 The Board then solicited bids from area building contractors for the construction of the school, and received bids from two non-Indian firms. Each firm included the state gross receipts tax as a cost of construction in their bids, although the tax was not itemized separately. Appellant Lembke Construction Co. (Lembke) was the low bidder and was awarded the contract. The contract between the Board and Lembke provides that Lembke is to pay all "taxes required by law." Lembke began construction of the school facilities in 1974 and continued this work for over five years. During that time, Lembke paid the gross receipts tax and, pursuant to standard industry practice, was reimbursed by the Board for the full amount paid. Before the second contract between Lembke and the Board was executed in 1977, a clause was inserted into the contract recognizing that the Board could litigate the validity of this tax and was entitled to any refund. 5 Both Lembke and the Board protested the imposition of the gross receipts tax. In 1978, after exhausting administrative remedies, they filed this refund action against appellee New Mexico Bureau of Revenue in the New Mexico District Court. At the time of trial, the parties stipulated that the Board had reimbursed Lembke for tax payments of $232,264.38 and that the Board would receive any refund that might be awarded. 6 The trial court entered judgment for the State Bureau of Revenue. After noting that the "legal incidence" of the tax fell on the non-Indian construction firm, the court rejected appellants' arguments that the tax was pre-empted by comprehensive federal regulation and that it imposed an impermissible burden on tribal sovereignty. The Court of Appeals for the State of New Mexico affirmed. 95 N.M. 708, 625 P.2d 1225 (1980). Although acknowledging that the economic burden of the tax fell on the Board, the Court of Appeals concluded that the tax was not preempted by federal law and that it did not unlawfully burden tribal sovereignty. The Board filed a petition for rehearing in light of this Court's intervening decisions in White Mountain, supra, and Central Machinery Co. v. Arizona State Tax Comm'n, 448 U.S. 160, 100 S.Ct. 2592, 65 L.Ed.2d 684 (1980). The Court of Appeals denied the petition, stating only that this case did not involve either "a comprehensive or pervasive scheme of federal regulation" or "federal regulation similar to the Indian trader statutes." App. to Juris. Statement 36. After initially granting discretionary review, the New Mexico Supreme Court quashed the writ as improvidently granted. 96 N.M. 17, 627 P.2d 412 (1981). We noted probable jurisdiction. 454 U.S. 1079, 102 S.Ct. 631, 70 L.Ed.2d 612 (1981). II 7 In recent years, this Court has often confronted the difficult problem of reconciling "the plenary power of the States over residents within their borders with the semi-autonomous status of Indians living on tribal reservations." McClanahan v. Arizona State Tax Comm'n, 411 U.S. 164, 165, 93 S.Ct. 1257, 1258, 36 L.Ed.2d 129 (1973). Although there is no definitive formula for resolving the question whether a State may exercise its authority over tribal members or reservation activities, we have recently identified the relevant federal, tribal, and state interests to be considered in determining whether a particular exercise of state authority violates federal law. See White Mountain, 448 U.S., at 141-145, 100 S.Ct., at 2582-2584. A. 8 In White Mountain, we recognized that the federal and tribal interests arise from the broad power of Congress to regulate tribal affairs under the Indian Commerce Clause, Art. I, § 8, cl. 3, and from the semi-autonomous status of Indian tribes. 448 U.S., at 142, 100 S.Ct., at 2583. These interests tend to erect two "independent but related" barriers to the exercise of state authority over commercial activity on an Indian reservation: state authority may be pre-empted by federal law, or it may interfere with the tribe's ability to exercise its sovereign functions. Ibid. (citing, inter alia, Warren Trading Post Co. v. Arizona Tax Comm'n, 380 U.S. 685, 85 S.Ct. 1242, 14 L.Ed.2d 165 (1965); McClanahan v. Arizona State Tax Comm'n, supra; and Williams v. Lee, 358 U.S. 217, 79 S.Ct. 269, 3 L.Ed.2d 251 (1959)). As we explained in White Mountain : 9 "The two barriers are independent because either, standing alone, can be a sufficient basis for holding state law inapplicable to activity undertaken on the reservation or by tribal members. They are related, however, in two important ways. The right of tribal self-government is ultimately dependent on and subject to the broad power of Congress. Even so, traditional notions of Indian self-government are so deeply engrained in our jurisprudence that they have provided an important 'back-drop,' . . . against which vague or ambiguous federal enactments must always be measured." 448 U.S., at 143, 100 S.Ct., at 2583 (quoting McClanahan v. Arizona State Tax Comm'n, supra, 358 U.S. at 172, 93 S.Ct., at 1262). 10 The State's interest in exercising its regulatory authority over the activity in question must be examined and given appropriate weight. Pre-emption analysis in this area is not controlled by "mechanical or absolute conceptions of state or tribal sovereignty"; it requires a particularized examination of the relevant state, federal, and tribal interests. 448 U.S., at 145, 100 S.Ct., at 2584. The question whether federal law, which reflects the related federal and tribal interests, pre-empts the State's exercise of its regulatory authority is not controlled by standards of pre-emption developed in other areas. Id., at 143-144, 100 S.Ct., at 2583-2584. Instead, the traditional notions of tribal sovereignty, and the recognition and encouragement of this sovereignty in congressional Acts promoting tribal independence and economic development, inform the pre-emption analysis that governs this inquiry. See id., at 143, and n. 10, 100 S.Ct., at 2583, and n. 10. Relevant federal statutes and treaties must be examined in light of "the broad policies that underlie them and the notions of sovereignty that have developed from historical traditions of tribal independence." Id., at 144-145, 100 S.Ct., at 2584-2585. As a result, ambiguities in federal law should be construed generously, and federal pre-emption is not limited to those situations where Congress has explicitly announced an intention to pre-empt state activity. Id., at 143-144, 150-151, 100 S.Ct., at 2583-2584, 2587-2588. 11 In White Mountain, we applied these principles and held that federal law pre-empted application of the state motor carrier license and use fuel taxes to a non-Indian logging company's activity on tribal land. We found the federal regulatory scheme for harvesting Indian timber to be so pervasive that it precluded the imposition of additional burdens by the relevant state taxes. Id., at 148, 100 S.Ct., at 2586. The Secretary of the Interior (Secretary) had promulgated detailed regulations for developing " 'Indian forests by the Indian people for the purpose of promoting self-sustaining communities.' " Id., at 147, 100 S.Ct., at 2585 (quoting 25 CFR § 141.3(a)(3) (1979)). Under these regulations, the BIA was involved in virtually every aspect of the production and marketing of Indian timber. 448 U.S., at 145-148, 100 S.Ct., at 2585-2586. In particular, the Secretary and the BIA extensively regulated the contractual relationship between the Indians and the non-Indians working on the reservation: they established the bidding procedure, set mandatory terms to be included in every contract, and required that all contracts be approved by the Secretary. Id., at 147, 100 S.Ct., at 2585. 12 We found that the state taxes in question would "threaten the overriding federal objective of guaranteeing Indians that they will 'receive . . . the benefit of whatever profit [the forest] is capable of yielding. . . .' " Id., at 149, 100 S.Ct., at 2586 (quoting 25 CFR § 141.3(a)(3) (1979)). We concluded that the imposition of state taxes would also undermine the Secretary's ability to carry out his obligations to set fees and rates for the harvesting and sale of the timber, and it would impede the "Tribe's ability to comply with the sustained-yield management policies imposed by federal law." 448 U.S., at 149-150, 100 S.Ct., at 2586-2587. Balanced against this intrusion into the federal scheme, the State asserted only "a general desire to raise revenue" as its justification for imposing the taxes. Id., at 150, 100 S.Ct., at 2587. In this context, this interest is insufficient to justify the State's intrusion into a sphere so heavily regulated by the Federal Government. Ibid. B 13 This case is indistinguishable in all relevant respects from White Mountain. Federal regulation of the construction and financing of Indian educational institutions is both comprehensive and pervasive. The Federal Government's concern with the education of Indian children can be traced back to the first treaties between the United States and the Navajo Tribe.3 Since that time, Congress has enacted numerous statutes empowering the BIA to provide for Indian education both on and off the reservation. See, e.g., Snyder Act, 42 Stat. 208 (1921), 25 U.S.C. § 13; Johnson-O'Malley Act, 48 Stat. 596 (1934), 25 U.S.C. § 452 et seq.; Navajo-Hopi Rehabilitation Act, 64 Stat. 44 (1950), 25 U.S.C. § 631 et seq.; Indian Self-Determination and Education Assistance Act, 88 Stat. 2203 (1975), 25 U.S.C. § 450 et seq. (Self-Determination Act). Although the early focus of the federal efforts in this area concentrated on providing federal or state educational facilities for Indian children, in the early 1970's the federal policy shifted toward encouraging the development of Indian-controlled institutions on the reservation. See 6 Weekly Comp. of Pres. Doc. 894, 899-900 (1970) (Message of President Nixon). 14 This federal policy has been codified in the Indian Financing Act of 1974, 88 Stat. 77, 25 U.S.C. § 1451 et seq., and most notably in the Self-Determination Act. The Self-Determination Act declares that a "major national goal of the United States is to provide the quantity and quality of educational services and opportunities which will permit Indian children to compete and excel in the life areas of their choice, and to achieve the measure of self-determination essential to their social and economic well-being." 88 Stat. 2203, as set forth in 25 U.S.C. § 450a(c). In achieving this goal, Congress expressly recognized that "parental and community control of the educational process is of crucial importance to the Indian people." 88 Stat. 2203, as set forth in 25 U.S.C. § 450(b)(3). 15 Section 450k empowers the Secretary to promulgate regulations to accomplish the purposes of the Act. 88 Stat. 2212, 25 U.S.C. § 450k. Pursuant to this authority, the Secretary has promulgated detailed and comprehensive regulations respecting "school construction for previously private schools now controlled and operated by tribes or tribally approved Indian organizations." 25 CFR § 274.1 (1981). Under these regulations, the BIA has wide-ranging authority to monitor and review the subcontracting agreements between the Indian organization, which is viewed as the general contractor, and the non-Indian firm that actually constructs the facilities. See 25 CFR § 274.2 (1981).4 Specifically, the BIA must conduct preliminary on-site inspections, and prepare cost estimates for the project in cooperation with the tribal organization. 25 CFR § 274.22 (1981). The Board must approve any architectural or engineering agreements executed in connection with the project. 25 CFR § 274.32(c) (1981). In addition, the regulations empower the BIA to require that all subcontracting agreements contain certain terms, ranging from clauses relating to bonding and pay scales, 41 CFR § 14H-70.632 (1981), to preferential treatment for Indian workers. 25 CFR § 274.38 (1981). Finally, to ensure that the Tribe is fulfilling its statutory obligations, the regulations require the tribal organization to maintain records for the Secretary's inspection. 25 CFR § 274.41 (1981). 16 This detailed regulatory scheme governing the construction of autonomous Indian educational facilities is at least as comprehensive as the federal scheme found to be pre-emptive in White Mountain.5 The direction and supervision provided by the Federal Government for the construction of Indian schools leave no room for the additional burden sought to be imposed by the State through its taxation of the gross receipts paid to Lembke by the Board. This burden, although nominally falling on the non-Indian contractor, necessarily impedes the clearly expressed federal interest in promoting the "quality and quantity" of educational opportunities for Indians by depleting the funds available for the construction of Indian schools.6 17 The Bureau of Revenue argues that imposition of the state tax is not pre-empted because the federal statutes and regulations do not specifically express the intention to pre-empt this exercise of state authority. This argument is clearly foreclosed by our precedents. In White Mountain we flatly rejected a similar argument. 448 U.S., at 150-151, 100 S.Ct., at 2586-2587 (citing Warren Trading Post Co. v. Arizona Tax Comm'n, 380 U.S. 685, 85 S.Ct. 1242, 14 L.Ed.2d 165 (1965); Williams v. Lee, 358 U.S. 217, 79 S.Ct. 269, 3 L.Ed.2d 251 (1959); and Kennerly v. District Court of Montana, 400 U.S. 423, 91 S.Ct. 480, 27 L.Ed.2d 507 (1971)). There is nothing unique in the nature of a gross receipts tax or in the federal laws governing the development of tribal self-sufficiency in the area of education that requires a different analysis. 18 In this case, the State does not seek to assess its tax in return for the governmental functions it provides to those who must bear the burden of paying this tax. Having declined to take any responsibility for the education of these Indian children, the State is precluded from imposing an additional burden on the comprehensive federal scheme intended to provide this education—a scheme which has "left the State with no duties or responsibilities." Warren Trading Post Co. v. Arizona Tax Comm'n, supra, at 691, 85 S.Ct., at 1246.7 Nor has the State asserted any specific, legitimate regulatory interest to justify the imposition of its gross receipts tax. The only arguably specific interest advanced by the State is that it provides services to Lembke for its activities off the reservation. This interest, however, is not a legitimate justification for a tax whose ultimate burden falls on the tribal organization.8 Furthermore, although the State may confer substantial benefits on Lembke as a state contractor, we fail to see how these benefits can justify a tax imposed on the construction of school facilities on tribal lands pursuant to a contract between the tribal organization and the non-Indian contracting firm.9 The New Mexico gross receipts tax is intended to compensate the State for granting "the privilege of engaging in business." N.M.Stat.Ann. §§ 7-9-3(F) and 7-9-4(A) (1980). New Mexico has not explained the source of its power to levy such a tax in this case where the "privilege of doing business" on an Indian reservation is exclusively bestowed by the Federal Government. 19 The State's ultimate justification for imposing this tax amounts to nothing more than a general desire to increase revenues. This purpose, as we held in White Mountain, 448 U.S., at 150, 100 S.Ct., at 2586, is insufficient to justify the additional burdens imposed by the tax on the comprehensive federal scheme regulating the creation and maintenance of educational opportunities for Indian children and on the express federal policy of encouraging Indian self-sufficiency in the area of education.10 This regulatory scheme precludes any state tax that "stands as an obstacle to the accomplishment of the full purposes and objectives of Congress." Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 404, 85 L.Ed. 581 (1941). C 20 The Solicitor General, in an amicus brief filed on behalf of the United States, suggests that we modify our pre-emption analysis and rely on the dormant Indian Commerce Clause, Art. I, § 8, cl. 3, to hold that on-reservation activities involving a resident tribe are presumptively beyond the reach of state law even in the absence of comprehensive federal regulation, thus placing the burden on the State to demonstrate that its intrusion is either condoned by Congress or justified by a compelling need to protect legitimate, specified state interests other than the generalized desire to collect revenue. He argues that adopting this approach is preferable for several reasons: it would provide guidance to the state courts addressing these issues, thus reducing the need for our case-by-case review of these decisions; it would avoid the tension created by focusing on the pervasiveness of federal regulation as a principal barrier to state assertions of authority when the primary federal goal is to encourage tribal self-determination and self-government; and it would place a higher burden on the State to articulate clearly its particularized interests in taxing the transaction and to demonstrate the services it provides in assisting the taxed transaction. 21 We do not believe it necessary to adopt this new approach—the existing pre-emption analysis governing these cases is sufficiently sensitive to many of the concerns expressed by the Solicitor General. Although clearer rules and presumptions promote the interest in simplifying litigation, our precedents announcing the scope of pre-emption analysis in this area provide sufficient guidance to state courts and also allow for more flexible consideration of the federal, state, and tribal interests at issue. We have consistently admonished that federal statutes and regulations relating to tribes and tribal activities must be "construed generously in order to comport with . . . traditional notions of [Indian] sovereignty and with the federal policy of encouraging tribal independence." White Mountain, supra, at 144, 100 S.Ct., at 2584; see also McClanahan v. Arizona State Tax Comm'n, 411 U.S., at 174-175, and n. 13, 93 S.Ct., at 1263; Warren Trading Post Co. v. Arizona Tax Comm'n, 380 U.S., at 690-691, 85 S.Ct., at 1245. This guiding principle helps relieve the tension between emphasizing the pervasiveness of federal regulation and the federal policy of encouraging Indian self-determination. Although we must admit our disappointment that the courts below apparently gave short shrift to this principle and to our precedents in this area, we cannot and do not presume that state courts will not follow both the letter and the spirit of our decisions in the future. III 22 In sum, the comprehensive federal regulatory scheme and the express federal policy of encouraging tribal self-sufficiency in the area of education preclude the imposition of the state gross receipts tax in this case. Accordingly, the judgment of the New Mexico Court of Appeals is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. 23 It is so ordered. 24 Justice REHNQUIST, with whom Justice WHITE and Justice STEVENS join, dissenting. 25 The Court today reproves the New Mexico Court of Appeals for failing to heed our precedents, much as a disappointed parent would rebuke a wayward child.1 I do not think the Court of Appeals deserves the rebuke; it seems to me that the state court applied our precedents at least as faithfully, and coherently, as the Court itself. In its desire to reach a result that it evidently finds quite salutary as a matter of policy, the Court finds "indistinguishable" a case that is considerably off the mark, and it finds "pervasively regulated" an activity that is largely free of federal regulation. It ultimately accords a dependent Indian tribal organization greater tax immunity than it accorded the sovereignty of the United States a short three months ago in a case involving the precise state taxes at issue here. 26 * The general question presented by this case has occupied the Court many times in the recent past, and seems destined to demand its attention over and over again until the Court sees fit to articulate, and follow, a consistent and predictable rule of law. This insistent question concerns the extent to which the States can tax economic activity on Indian reservations within their borders. I believe the dominant trend of our cases is toward treating the scope of reservation immunity from nondiscriminatory state taxation as a question of pre-emption, ultimately dependent on congressional intent. In such a framework, the tradition of Indian sovereignty stands as an independent barrier to discriminatory taxes, and otherwise serves only as a guide to the ascertainment of the congressional will. 27 The principles announced in White Mountain Apache Tribe v. Bracker, 448 U.S. 136, 100 S.Ct. 2578, 65 L.Ed.2d 665 (1980), are consistent with this trend.2 Thus, the Court in White Mountain recognized federal pre-emption as a principal barrier to the assertion of state regulatory authority over tribal reservations and members, id., at 142, 100 S.Ct., at 2583, and specifically invalidated the challenged assertion of taxing authority on that basis, id., at 148, 151, n. 15, 100 S.Ct., at 2586, 2587, n. 15. The Court also recognized that in some instances a state law may be invalid because it infringes " 'the right of reservation Indians to make their own laws and be ruled by them.' " Id., at 142, 100 S.Ct., at 2583 (quoting Williams v. Lee, 358 U.S. 217, 220, 79 S.Ct. 269, 3 L.Ed.2d 251 (1959)). But apart from those rare instances in which the State attempts to interfere with the residual sovereignty of a tribe to govern its own members, the "tradition of tribal sovereignty" merely provides a "backdrop" against which the pre-emptive effect of federal statutes or treaties must be assessed. See 448 U.S., at 143, 100 S.Ct., at 2583. 28 The Court today pays homage to these principles but then promptly bestows its favors on a new analytical framework in which the extent of economic burden on the tribe, and not the pre-emptive effect of federal regulations, appears to be the paramount consideration. Such a shift is necessary, for the Court's purported reliance on White Mountain will not withstand even superficial scrutiny. II 29 The Court declares that "[t]his case is indistinguishable in all relevant respects from White Mountain." Ante, at 839. This statement is quite inaccurate. White Mountain involved an attempt by the State of Arizona to apply its motor carrier license and use fuel taxes to the logging operations of a non-Indian company doing business exclusively on the reservation. The Court concluded that application of the State's taxes was inconsistent with the pervasive federal regulation of the very activity subject to taxation. The Court repeatedly emphasized the comprehensiveness of the regulations on which it relied. 30 "Under these regulations, the Bureau of Indian Affairs exercises literally daily supervision over the harvesting and management of tribal timber. In the present case, contracts between [the tribal organization] and [the non-Indian contractor] must be approved by the Bureau; indeed, the record shows that some of those contracts were drafted by employees of the Federal Government. Bureau employees regulate the cutting, hauling, and marking of timber by [the tribal organization and the contractor]. The Bureau decides such matters as how much timber will be cut, which trees will be felled, which roads are to be used, which hauling equipment [the contractor] should employ, the speeds at which logging equipment may travel, and the width, length, height, and weight of loads. 31 "The Secretary has also promulgated detailed regulations governing the roads developed by the Bureau of Indian Affairs. . . . On the Fort Apache Reservation the Forestry Department of the Bureau has required [the tribal organization] and its contractors . . . to repair and maintain existing Bureau and tribal roads and in some cases to construct new logging roads. . . . A high percentage of [the contractor's] receipts are expended for those purposes, and it has maintained separate personnel and equipment to carry out a variety of tasks relating to road maintenance." 448 U.S., at 147-148, 100 S.Ct., at 2586. 32 But the Court in White Mountain did not merely review the comprehensiveness of the regulations and conclude, ipso facto, that state taxes on the logging operations were preempted. It found, with considerable attention to specifics, that "the assessment of state taxes would obstruct federal policies." Id., at 148, 100 S.Ct., at 2586. 33 "At the most general level, the taxes would threaten the overriding federal objective of guaranteeing Indians that they will 'receive . . . the benefit of whatever profit [the forest] is capable of yielding. . . .' 25 CFR § 141.3(a)(3) (1979). Underlying the federal regulatory program rests a policy of assuring that the profits derived from timber sales will inure to the benefit of the Tribe subject only to administrative expenses incurred by the Federal Government. . . . 34 "In addition, the taxes would undermine the Secretary's ability to make the wide range of determinations committed to his authority concerning the setting of fees and rates with respect to the harvesting and sale of tribal timber. The Secretary reviews and approves the terms of the Tribe's agreements with its contractors, sets fees for services rendered to the Tribe by the Federal Government, and determines stumpage rates for timber to be paid to the Tribe. Most notably in reviewing or writing the terms of the contracts between [the tribal organization] and its contractors, federal agents must predict the amount and determine the proper allocation of all business expenses, including fuel costs. The assessment of state taxes would throw additional factors into the federal calculus, reducing tribal revenues and diminishing the profitability of the enterprise for potential contractors. 35 "Finally, the imposition of state taxes would adversely affect the Tribe's ability to comply with the sustained-yield management policies imposed by federal law." Id., at 149-150, 100 S.Ct., at 2586-2587. 36 As noted, the Court thinks that this case is "indistinguishable in all relevant respects from White Mountain." Ante, at 839. It finds that "[f]ederal regulation of the construction and financing of Indian educational institutions is both comprehensive and pervasive." Ibid. But the regulations on which the Court relies do not regulate school construction, which is the activity taxed. They merely detail procedures by which tribes may apply for federal funds in order to carry out school construction. 37 The purpose of the regulations, which the Court quotes only in part, ante, at 840-841, "is to give the application and approval process for obtaining a contract or services from the Bureau for school construction for previously private schools now controlled and operated by tribes or tribally approved Indian organizations. . . ." 25 CFR § 274.1 (1981) (emphasis added). The regulations that follow explain the procedures by which tribes may obtain, complete, and file application forms for federal funding or services. §§ 274.12-274.18. As the Court observes, ante, at 841, the regulations also authorize the BIA to approve or disapprove plans and specifications for construction as well as construction contracts let by the tribe, which are treated as subcontracts of the funding contract between the tribe and the BIA. The contracts are required to contain a clause establishing a hiring preference for Indians. § 274.38. And the BIA is given access to the tribe's records for auditing purposes. § 274.41. That is the extent of the regulations. 38 In this case the BIA "contracted" with the School Board in order to convey federal funds for the construction project. It also approved the Board's construction "subcontract" with the construction contractor. It played no role in the selection of the contractor and it played no role in regulating or supervising the actual construction of the school. The Court concludes that this scheme, which is little more than a grant application process, "is at least as comprehensive as the federal scheme found to be pre-emptive in White Mountain." Ante, at 841. I simply cannot agree. 39 More important, the Court concludes in the very next sentence that "[t]he direction and supervision provided by the Federal Government for the construction of Indian schools leaves no room for the additional burden sought to be imposed by the State through its taxation of the gross receipts paid to Lembke by the Board." Ante, at 841-842. This statement constitutes the sum total of the Court's pre-emption analysis in this case. In White Mountain the Court engaged in a detailed examination of the extent to which state taxes would interfere both with the Secretary's ability to carry out his congressional mandate and with the tribe's ability to carry out federal policy. In the place of such careful analysis, the Court today relies on ipse dixit. It does so because there is no realistic basis for concluding that the State's taxes would interfere with a "pervasive" regulatory scheme. The BIA simply does not regulate the construction activity which the State seeks to tax. It provides federal money to eligible tribes and tribal organizations and it establishes a contract-approval and auditing mechanism as a means of attempting to ensure that the money is put to the use for which it is earmarked.3 III 40 A careful reading of the Court's opinion demonstrates that the single, determinative factor in its judgment is the fact that the challenged state taxes have increased the financial burden of constructing a tribal school. Whether the federal regulations are detailed and comprehensive or largely a matter of bookkeeping is an irrelevancy, for the Court concludes that the tax burden "impedes the clearly expressed federal interest in promoting the 'quality and quantity' of educational opportunities for Indians by depleting the funds available for the construction of Indian schools." Ante, at 842 (emphasis added). The Court recognizes that the legal incidence of the tax is on the non-Indian contractor, but asserts that "in White Mountain . . . we found it significant that the economic burden of the asserted taxes would ultimately fall on the Tribe, even though the legal incidence of the tax was on the non-Indian logging company." Ante, at 844, n. 8. 41 The Court in White Mountain did indeed note that "the economic burden of the asserted taxes will ultimately fall on the Tribe." 448 U.S., at 151, 100 S.Ct., at 2587. But in a footnote immediately following that sentence, which is today ignored, the Court declared: 42 "Of course, the fact that the economic burden of the tax falls on the Tribe does not by itself mean that the tax is pre-empted, as Moe v. Salish & Kootenai Tribes, 425 U.S. 463 [96 S.Ct. 1634, 48 L.Ed.2d 96] (1976), makes clear. Our decision today is based on the pre-emptive effect of the comprehensive federal regulatory scheme, which . . . leaves no room for the additional burdens sought to be imposed by state law." Id., at 151, n. 15, 100 S.Ct., at 2587. 43 Despite its references to the supposed "comprehensive and pervasive" regulatory scheme in this case, the Court clearly has chosen to bar the State from taxing Lembke's gross receipts principally because the tax imposes an indirect economic burden on the tribal organization. As the Court in White Mountain recognized, our precedents undeniably view that as an insufficient basis for the recognition of an Indian tax immunity. See Washington v. Confederated Tribes of Colville Indian Reservation, 447 U.S. 134, 156, 100 S.Ct. 2069, 2082, 65 L.Ed.2d 10 (1980) ("Washington does not infringe the right of reservation Indians to 'make their own laws and be ruled by them,' . . . merely because the result of imposing its taxes will be to deprive the Tribes of revenues which they currently are receiving"); Moe v. Salish & Kootenai Tribes, 425 U.S. 463, 481-482, 96 S.Ct. 1634, 1645, 48 L.Ed.2d 96 (1976) (upholding tax on cigarette sales from Indians to non-Indians because the legal incidence of the tax was on the consumer); Mescalero Apache Tribe v. Jones, 411 U.S. 145, 156-157, 93 S.Ct. 1267, 1274-1275, 36 L.Ed.2d 114 (1973) (refusing to imply tax immunity despite economic burden on tribal enterprise).4 Even under the modified form of pre-emption doctrine applicable to state regulation of reservation activities, there must be some affirmative indication that Congress did not intend the State to exercise the sovereign power challenged in the suit. Until today, the mere fact that the asserted power will impose an economic burden on a tribal endeavor has not provided that affirmative indication. 44 I do not disagree with the Court's judgment that congressional enactments such as the Indian Financing Act and the Indian Self-Determination and Education Assistance Act embody a federal policy encouraging the development of Indian-controlled educational institutions. But it is a considerable leap to infer from that policy the independent principle that all state laws which might increase the cost of such an endeavor are to be considered null and void. It is perfectly conceivable that Congress favored Indian education, but also contemplated that all costs of obtaining that end would be paid in a normal fashion. State taxes are as much a normal cost of school construction as the cost of cement and labor. The cost of taxes was included in the bids submitted to the Board by the construction contractors, and it apparently was also included in the funding requests submitted by the Board to Congress. The Board cannot be faulted for attempting to stretch its federal construction funds as far as possible, but that is a woefully inadequate basis for interfering with the sovereign prerogatives of the State of New Mexico. IV 45 A short three months ago, this Court considered whether the State of New Mexico could impose its gross receipts and compensating use taxes on private contractors that conduct business with the Federal Government. We concluded that tax immunity was appropriate in only one circumstance: "when the levy falls on the United States itself, or on an agency or instrumentality so closely connected to the Government that the two cannot realistically be viewed as separate entities, at least insofar as the activity being taxed is concerned." United States v. New Mexico, 455 U.S. 720, 735, 102 S.Ct. 1373, 1382, 71 L.Ed. 580 (1982). In reaching this conclusion, we held that "immunity may not be conferred simply because the tax has an effect on the United States, or even because the Federal Government shoulders the entire economic burden of the levy." Id., at 734, 102 S.Ct., at 1382. If the legal incidence of the tax is on the contractor, it is to be considered valid, absent specific congressional action, as long as "the contractors can realistically be considered entities independent of the United States." Id., at 738, 102 S.Ct., at 1384.5 46 In this case, as in United States v. New Mexico, the legal incidence of the New Mexico tax is on the private contractor, not on the entity whose status might be the source of a tax immunity. And, as in United States v. New Mexico, it is evident that Lembke is a separate taxable entity completely independent of the tribal school board. Were the tax immunity of Tribe no greater than that of the United States, it seems plain that New Mexico's tax would have to be upheld as applied to the gross receipts of the non-Indian contractor. But the Court reaches a different conclusion because it finds that the tax imposes an economic burden on the Tribe's effort to build a school with federal funds. Thus, the Court accords an Indian Tribe, whose sovereignty "exists only at the sufferance of Congress and is subject to complete defeasance," United States v. Wheeler, 435 U.S. 313, 323, 98 S.Ct. 1079, 1086, 55 L.Ed.2d 303 (1978), greater immunity from state taxes than is enjoyed by the sovereignty of the United States on whom it is dependent.6 47 For these reasons, I dissent from the Court's judgment. 1 On July 8, 1970, in his Message to the Congress on Indian Affairs, President Nixon referred specifically to these efforts of the Board to assume responsibility for the education of tribal children abandoned by the State as a "notable exampl[e]" of Indian self-determination. 6 Weekly Comp. of Pres. Doc. 894, 899 (1970). 2 See Pub. L. 93-245, 87 Stat. 1073 (1973) (amending Pub. L. 93-120, 87 Stat. 431 (1973) to specifically earmark funds appropriated there for the construction of the Ramah school facility); Pub. L. 93-404, 88 Stat. 810 (1974); Pub. L. 94-165, 89 Stat. 985 (1975); Pub. L. 95-74, 91 Stat. 293 (1977). 3 Article VI of the 1868 Treaty between the United States and the Navajo Tribe, 15 Stat. 669, provides that "[i]n order to insure the civilization of the Indians entering into this treaty, the necessity of education is admitted." 4 Although these regulations did not become effective until several months after the BIA and the Board had executed the initial contracts, the Secretary and the BIA had applied similar requirements under the authority of the Johnson-O'Malley Act, 48 Stat. 596, 25 U.S.C. § 452 et seq. In any event, the two subsequent agreements between the BIA, the Board and Lembke, accounting for two-thirds of the total construction, were signed after the effective date of these regulations, which clearly authorize the BIA to monitor these construction agreements. 5 Justice REHNQUIST asserts that the comprehensive federal regulatory scheme outlined above "do[es] not regulate school construction, which is the activity taxed." Post, at 851. The dissent fails to explain, however, how this fact distinguishes this case from White Mountain. In that case, we struck down Arizona's use fuel tax and motor carrier license tax, not because of any federal interest in gasoline, licenses, or highways, but because the imposition of these state taxes on a non-Indian contractor doing work on the reservation was pre-empted by the "comprehensive regulation of the harvesting and sale of tribal timber." 448 U.S., at 151, 100 S.Ct., at 2587. We find that New Mexico is similarly precluded from impeding the federal interest in the construction of autonomous Indian educational institutions by imposing its gross receipts tax on Lembke. Justice REHNQUIST's contention that the New Mexico tax is somehow compatible with this federal interest because such taxes "are as much a normal cost of school construction as the cost of cement and labor," post, at 855, is also foreclosed by White Mountain. Surely, state use fuel and motor carrier license taxes are considered part of the cost of harvesting and marketing timber. Yet in White Mountain, we concluded that these taxes impeded the federal interest in "guaranteeing Indians that they will 'receive . . . the benefit of whatever profit [the forest] is capable of yielding,' " 448 U.S., at 149, 100 S.Ct., at 2586, despite the dissent's argument that the taxes amounted to less than 1% of the annual profits produced by the logging operation. Here, as in White Mountain, Justice REHNQUIST continues to press this argument. 6 Appellee would have us impute congressional awareness and approval of the state gross receipts tax from appropriations bills which earmarked funds for the construction of these facilities, see n. 2, supra. Brief for Appellee 21-22. Appellee strains to find this awareness and approval by arguing that the same architects who prepared the cost estimates and requests that the Board submitted to Congress also prepared the bid specifications pursuant to which Lembke submitted its bid. However, as we have indicated, the bid specifications only required prospective bidders to include "all taxes required by law," and the submitted bids did not specify the gross receipts tax as a separate line item. Supra, at 835. Therefore, it is by no means clear, and the Board disputes the contention, that the Board ever intended to have these state taxes included in the construction costs of its school facilities. Furthermore, there is absolutely no indication that Congress was even made aware of the existence of these taxes when it appropriated funds for the construction of the Ramah Navajo school. In any event, as we have noted in a related context, courts should be wary of inferring congressional intent to alter the force of existing law from an appropriations Act. Cf. TVA v. Hill, 437 U.S. 153, 189-191, 98 S.Ct. 2279, 2299-2300, 57 L.Ed.2d 117 (1978). 7 Of course, these statutes and regulations do not prevent the States from providing for the education of Indian children within their boundaries. Indeed, the Self-Determination Act specifically authorizes the Secretary to enter into contracts with any State willing to construct educational institutions for Indian children on or near the reservation. 88 Stat. 2214, 25 U.S.C. § 458. This case would be different if the State were actively seeking tax revenues for the purpose of constructing, or assisting in the effort to provide, adequate educational facilities for Ramah Navajo children. 8 The Bureau of Revenue invites us to adopt the "legal incidence" test, under which the legal incidence and not the actual burden of the tax would control the pre-emption inquiry. Of course, in some contexts, the fact that the legal incidence of the tax falls on a non-Indian is significant. See Washington v. Confederated Tribes of Colville Indian Reservation, 447 U.S. 134, 150-151, 100 S.Ct. 2069, 2079-2080, 65 L.Ed.2d 10 (1980); Moe v. Salish & Kootenai Tribes, 425 U.S. 463, 96 S.Ct. 1634, 48 L.Ed.2d 96 (1976). However, in White Mountain, 448 U.S., at 151, 100 S.Ct., at 2587, we found it significant that the economic burden of the asserted taxes would ultimately fall on the Tribe, even though the legal incidence of the tax was on the non-Indian logging company. Given the comprehensive federal regulatory scheme at issue here, we decline to allow the State to impose additional burdens on the significant federal interest in fostering Indian-run educational institutions, even if those burdens are imposed indirectly through a tax on a non-Indian contractor for work done on the reservation. 9 In Central Machinery Co. v. Arizona State Tax Comm'n, 448 U.S. 160, 100 S.Ct. 2592, 65 L.Ed.2d 684 (1980), we held that the Indian trader statutes, 19 Stat. 200, 25 U.S.C. § 261 et seq., pre-empted the State's jurisdiction to tax the sale of farm machinery to the Indian Tribe, notwithstanding the substantial services that the State undoubtedly provided to the off-reservation activities of the non-Indian seller. Presumably, the state tax revenues derived from Lembke's off-reservation business activities are adequate to reimburse the State for the services it provides to Lembke. 10 We are similarly unpersuaded by the State's argument that the significant services it provides to the Ramah Navajo Indians justify the imposition of this tax. The State does not suggest that these benefits are in any way related to the construction of schools on Indian land. Furthermore, the evidence introduced below by the State on this issue is far from clear. Although the State does provide services to the Ramah Navajo Indians, it receives federal funds for providing some of these services, and the State conceded at trial that it saves approximately $380,000 by not having to provide education for the Ramah Navajo children. App. 95, 105-106, 108. 1 "Although we must admit our disappointment that the courts below apparently gave short shrift to this principle and to our precedents in this area, we cannot and do not presume that state courts will not follow both the letter and the spirit of our decisions in the future." Ante, at 846. 2 Nevertheless, the Solicitor General has again suggested that on-reservation activities affecting resident tribes be considered presumptively beyond the reach of state law by operation of the "principle of tribal sovereignty." See Brief for United States as Amicus Curiae 17-24. The same suggestion was urged, and rejected, in White Mountain. It has proved no more appealing in this case. 3 The Court ignores other distinctions between this case and White Mountain. For example, the logging contractor in the latter case, although a non-Indian corporation, operated exclusively to harvest timber on the reservation; it conducted no off-reservation activities whatsoever. See 448 U.S., at 139, 100 S.Ct., at 2581. The contractor in this case is a general building contractor doing business throughout the State of New Mexico, and enjoying state services to the same extent as any other commercial enterprise in New Mexico. The Court dismisses this factor with the statement that "[p]resumably, the state tax revenues derived from Lembke's off-reservation business activities are adequate to reimburse the State for the services it provides to Lembke." Ante, at 844, n. 9. The Court's "presumptions," however, are no substitute for the considered judgment of the state taxing authority. Indeed, in assessing the validity of a state tax, the Court has previously recognized that the State's interests are strongest when the taxpayer is the recipient of state services. See Washington v. Confederated Tribes of Colville Indian Reservation, 447 U.S. 134, 157, 100 S.Ct. 2069, 2083, 65 L.Ed.2d 10 (1980). To the extent presumptions are relevant, the Court has inverted the one that ought to apply. Another distinction is also relevant. The activity taxed in White Mountain was the exploitation of natural resources located on the reservation and devoted to the beneficial use and enjoyment of reservation Indians. Indeed, over 90% of the total profits generated by tribal enterprises were derived from the Tribe's logging operations. 448 U.S., at 138, 100 S.Ct., at 2581. In this case, the state taxes diminish, not the income generated by the Tribe for its own preservation and welfare, but federal funds appropriated by Congress for the purpose of school construction. No tribal funds are devoted to this endeavor, and congressional appropriations were based on funding requests that included the gross receipts tax as part of the estimated construction cost. 4 In other areas of tax immunity, the Court has steadfastly refused to assess the validity of a tax by reference to the economic burdens it imposes if those burdens are nondiscriminatory and comport with due process. See United States v. New Mexico, 455 U.S. 720, 102 S.Ct. 1373, 71 L.Ed.2d 580 (1982) (state taxation of federal contractors); United States v. County of Fresno, 429 U.S. 452, 97 S.Ct. 699, 50 L.Ed.2d 683 (1977) (state taxation of Federal Government); New York v. United States, 326 U.S. 572, 66 S.Ct. 310, 90 L.Ed. 326 (1946) (federal taxation of state government); Michelin Tire Corp. v. Wages, 423 U.S. 276, 96 S.Ct. 535, 46 L.Ed.2d 495 (1976) (state taxation of imports and exports). 5 We recognized one possible exception to this general rule: "In the case of a sales tax . . . it is arguable that an entity serving as a federal procurement agent can be so closely associated with the Government, and so lack an independent role in the purchase, as to make the sale—in both a real and a symbolic sense—a sale to the United States, even though the purchasing agent has not otherwise been incorporated into the Government structure." 455 U.S., at 742, 102 S.Ct., at 1386. In this case, there is no basis for arguing that Lembke has acted merely as a purchasing agent for the Board or the BIA. 6 Of course, the Court purports to rest its decision on the pre-emptive effect of federal law. But the immunity of federal contractors from state taxes is also dependent on "generalized notions of federal supremacy." United States v. New Mexico, supra, at 730, 102 S.Ct., at 1380. The critical question, both in United States v. New Mexico and in this case, is what factors will the Court examine to determine whether the State has exceeded limits imposed by the Supremacy Clause and by Congress. I think it is evident that in the area of federal tax immunity, the Court has required evidence of more than mere economic burdens before it will invalidate a state tax as applied. As this case demonstrates, tribal tax immunity may be invoked on no greater showing than the fact of economic burdens on a federally supported tribal endeavor. Since both immunities derive from precisely the same source—the supremacy of federal law—I find the Court's decision today inexplicable. "With the abandonment of the notion that the economic—as opposed to the legal—incidence of the tax is relevant, it becomes difficult to maintain that federal tax immunity is designed to insulate federal operations from the effects of state taxation." United States v. New Mexico, supra, at 735, n. 11, 102 S.Ct., at 1383, n. 11.

12

458 U.S. 747 102 S.Ct. 3348 73 L.Ed.2d 1113 NEW YORK, Petitionerv.Paul Ira FERBER. No. 81-55. Argued April 27, 1982. Decided July 2, 1982. Syllabus A New York statute prohibits persons from knowingly promoting a sexual performance by a child under the age of 16 by distributing material which depicts such a performance. The statute defines "sexual performance" as any performance that includes sexual conduct by such a child, and "sexual conduct" is in turn defined as actual or simulated sexual intercourse, deviate sexual intercourse, sexual bestiality, masturbation, sado-masochistic abuse, or lewd exhibition of the genitals. Respondent bookstore proprietor was convicted under the statute for selling films depicting young boys masturbating, and the Appellate Division of the New York Supreme Court affirmed. The New York Court of Appeals reversed, holding that the statute violated the First Amendment as being both underinclusive and overbroad. The court reasoned that in light of the explicit inclusion of an obscenity standard in a companion statute banning the knowing dissemination of similarly defined material, the statute in question could not be construed to include an obscenity standard, and therefore would prohibit the promotion of materials traditionally entitled to protection under the First Amendment. Held: As applied to respondent and others who distribute similar material, the statute in question does not violate the First Amendment as applied to the States through the Fourteenth Amendment. Pp. 753-774. (a) The States are entitled to greater leeway in the regulation of pornographic depictions of children for the following reasons: (1) the legislative judgment that the use of children as subjects of pornographic materials is harmful to the physiological, emotional, and mental health of the child, easily passes muster under the First Amendment; (2) the standard of Miller v. California, 413 U.S. 15, 93 S.Ct. 2607, 37 L.Ed.2d 419, for determining what is legally obscene is not a satisfactory solution to the child pornography problem; (3) the advertising and selling of child pornography provide an economic motive for and are thus an integral part of the production of such materials, an activity illegal throughout the Nation; (4) the value of permitting live performances and photographic reproductions of children engaged in lewd exhibitions is exceedingly modest, if not de minimis ; and (5) recognizing and classifying child pornography as a category of material outside the First Amendment's protection is not incompatible with this Court's decisions dealing with what speech is unprotected. When a definable class of material, such as that covered by the New York statute, bears so heavily and pervasively on the welfare of children engaged in its production, the balance of competing interests is clearly struck, and it is permissible to consider these materials as without the First Amendment's protection. Pp. 756-764. (b) The New York statute describes a category of material the production and distribution of which is not entitled to First Amendment protection. Accordingly, there is nothing unconstitutionally "underinclusive" about the statute, and the State is not barred by the First Amendment from prohibiting the distribution of such unprotected materials produced outside the State. Pp. 764-766. (c) Nor is the New York statute unconstitutionally overbroad as forbidding the distribution of material with serious literary, scientific, or educational value. The substantial overbreath rule of Broadrick v. Oklahoma, 413 U.S. 601, 93 S.Ct. 2908, 37 L.Ed.2d 830, applies. This is the paradigmatic case of a state statute whose legitimate reach dwarfs its arguably impermissible applications. "[W]hatever overbreadth may exist should be cured through case-by-case analysis of the fact situations to which [the statute's] sanctions, assertedly, may not be applied." Broadrick v. Oklahoma, supra, at 615-616, 93 S.Ct., at 2917-2918. Pp. 766-774. 52 N.Y.2d 674, 439 N.Y.S.2d 863, 422 N.E.2d 523, reversed and remanded. Robert M. Pitler, New York City, for petitioner. Herald Price Fahringer, New York City, for respondent. Justice WHITE delivered the opinion of the Court. 1 At issue in this case is the constitutionality of a New York criminal statute which prohibits persons from knowingly promoting sexual performances by children under the age of 16 by distributing material which depicts such performances. 2 * In recent years, the exploitive use of children in the production of pornography has become a serious national problem.1 The Federal Government and 47 States have sought to combat the problem with statutes specifically directed at the production of child pornography. At least half of such statutes do not require that the materials produced be legally obscene. Thirty-five States and the United States Congress have also passed legislation prohibiting the distribution of such materials; 20 States prohibit the distribution of material depicting children engaged in sexual conduct without requiring that the material be legally obscene.2 3 New York is one of the 20. In 1977, the New York Legislature enacted Article 263 of its Penal Law. N.Y. Penal Law, Art. 263 (McKinney 1980). Section 263.05 criminalizes as a class C felony the use of a child in a sexual performance: 4 "A person is guilty of the use of a child in a sexual performance if knowing the character and content thereof he employs, authorizes or induces a child less than sixteen years of age to engage in a sexual performance or being a parent, legal guardian or custodian of such child, he consents to the participation by such child in a sexual performance." 5 A "[s]exual performance" is defined as "any performance or part thereof which includes sexual conduct by a child less than sixteen years of age." § 263.00(1). "Sexual conduct" is in turn defined in § 263.00(3): 6 " 'Sexual conduct' means actual or simulated sexual intercourse, deviate sexual intercourse, sexual bestiality, masturbation, sado-masochistic abuse, or lewd exhibition of the genitals." 7 A performance is defined as "any play, motion picture, photograph or dance" or "any other visual representation exhibited before an audience." § 263.00(4). 8 At issue in this case is § 263.15, defining a class D felony:3 9 "A person is guilty of promoting a sexual performance by a child when, knowing the character and content thereof, he produces, directs or promotes any performance which includes sexual conduct by a child less than sixteen years of age." To "promote" is also defined: 10 " 'Promote' means to procure, manufacture, issue, sell, give, provide, lend, mail, deliver, transfer, transmute, publish, distribute, circulate, disseminate, present, exhibit or advertise, or to offer or agree to do the same." § 263.00(5). 11 A companion provision bans only the knowing dissemination of obscene material. § 263.10. 12 This case arose when Paul Ferber, the proprietor of a Manhattan bookstore specializing in sexually oriented products, sold two films to an undercover police officer. The films are devoted almost exclusively to depicting young boys masturbating. Ferber was indicted on two counts of violating § 263.10 and two counts of violating § 263.15, the two New York laws controlling dissemination of child pornography.4 After a jury trial, Ferber was acquitted of the two counts of promoting an obscene sexual performance, but found guilty of the two counts under § 263.15, which did not require proof that the films were obscene. Ferber's convictions were affirmed without opinion by the Appellate Division of the New York State Supreme Court. 74 A.D.2d 558, 424 N.Y.S.2d 967 (1980). 13 The New York Court of Appeals reversed, holding that § 263.15 violated the First Amendment. 52 N.Y.2d 674, 439 N.Y.S.2d 863, 422 N.E.2d 523 (1981). The court began by noting that in light of § 263.10's explicit inclusion of an obscenity standard, § 263.15 could not be construed to include such a standard. Therefore, "the statute would . . . prohibit the promotion of materials which are traditionally entitled to constitutional protection from government interference under the First Amendment." 52 N.Y.2d, at 678, 439 N.Y.S.2d, at 865, 422 N.E.2d, at 525. Although the court recognized the State's "legitimate interest in protecting the welfare of minors" and noted that this "interest may transcend First Amendment concerns," id., at 679, 439 N.Y.S.2d, at 866, 422 N.E.2d, at 525-526, it nevertheless found two fatal defects in the New York statute. Section 263.15 was underinclusive because it discriminated against visual portrayals of children engaged in sexual activity by not also prohibiting the distribution of films of other dangerous activity. It was also overbroad because it prohibited the distribution of materials produced outside the State, as well as materials, such as medical books and educational sources, which "deal with adolescent sex in a realistic but nonobscene manner." 52 N.Y.2d, at 681, 439 N.Y.S.2d, at 866, 422 N.E.2d, at 526. Two judges dissented. We granted the State's petition for certiorari, 454 U.S. 1052, 102 S.Ct. 595, 70 L.Ed.2d 587 (1981), presenting the single question: 14 "To prevent the abuse of children who are made to engage in sexual conduct for commercial purposes, could the New York State Legislature, consistent with the First Amendment, prohibit the dissemination of material which shows children engaged in sexual conduct, regardless of whether such material is obscene?" II 15 The Court of Appeals proceeded on the assumption that the standard of obscenity incorporated in § 263.10, which follows the guidelines enunciated in Miller v. California, 413 U.S. 15, 93 S.Ct. 2607, 37 L.Ed.2d 419 (1973),5 constitutes the appropriate line dividing protected from unprotected expression by which to measure a regulation directed at child pornography. It was on the premise that "nonobscene adolescent sex" could not be singled out for special treatment that the court found § 263.15 "strikingly underinclusive." Moreover, the assumption that the constitutionally permissible regulation of pornography could not be more extensive with respect to the distribution of material depicting children may also have led the court to conclude that a narrowing construction of § 263.15 was unavailable. 16 The Court of Appeals' assumption was not unreasonable in light of our decisions. This case, however, constitutes our first examination of a statute directed at and limited to depictions of sexual activity involving children. We believe our inquiry should begin with the question of whether a State has somewhat more freedom in proscribing works which portray sexual acts or lewd exhibitions of genitalia by children. 17 * In Chaplinsky v. New Hampshire, 315 U.S. 568, 62 S.Ct. 766, 86 L.Ed. 1031 (1942), the Court laid the foundation for the excision of obscenity from the realm of constitutionally protected expression: 18 "There are certain well-defined and narrowly limited classes of speech, the prevention and punishment of which have never been thought to raise any Constitutional problem. These include the lewd and obscene . . .. It has been well observed that such utterances are no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality." Id., at 571-572, 62 S.Ct., at 769 (footnotes omitted). 19 Embracing this judgment, the Court squarely held in Roth v. United States, 354 U.S. 476, 77 S.Ct. 1304, 1 L.Ed.2d 1498 (1957), that "obscenity is not within the area of constitutionally protected speech or press." Id., at 485, 77 S.Ct., at 1309. The Court recognized that "rejection of obscenity as utterly without redeeming social importance" was implicit in the history of the First Amendment: The original States provided for the prosecution of libel, blasphemy, and profanity, and the "universal judgment that obscenity should be restrained [is] reflected in the international agreement of over 50 nations, in the obscenity laws of all of the 48 states, and in the 20 obscenity laws enacted by Congress from 1842 to 1956." Id., at 484-485, 77 S.Ct., at 1309 (footnotes omitted). 20 Roth was followed by 15 years during which this Court struggled with "the intractable obscenity problem." Interstate Circuit, Inc. v. Dallas, 390 U.S. 676, 704, 88 S.Ct. 1298, 1313, 20 L.Ed.2d 225 (1968) (opinion of Harlan, J.). See, e.g., Redrup v. New York, 386 U.S. 767, 87 S.Ct. 1414, 18 L.Ed.2d 515 (1967). Despite considerable vacillation over the proper definition of obscenity, a majority of the Members of the Court remained firm in the position that "the States have a legitimate interest in prohibiting dissemination or exhibition of obscene material when the mode of dissemination carries with it a significant danger of offending the sensibilities of unwilling recipients or of exposure to juveniles." Miller v. California, supra, at 18-19, 93 S.Ct., at 2612 (footnote omitted); Stanley v. Georgia, 394 U.S. 557, 567, 89 S.Ct. 1243, 1249, 22 L.Ed.2d 542 (1969); Ginsberg v. New York, 390 U.S. 629, 637-643, 88 S.Ct. 1274, 1279-1282, 20 L.Ed.2d 195 (1968); Interstate Circuit, Inc. v. Dallas, supra, 390 U.S., at 690, 88 S.Ct., at 1306; Redrup v. New York, supra, 386 U.S., at 769, 87 S.Ct., at 1415; Jacobellis v. Ohio, 378 U.S. 184, 195, 84 S.Ct. 1676, 1682, 12 L.Ed.2d 793 (1964). 21 Throughout this period, we recognized "the inherent dangers of undertaking to regulate any form of expression." Miller v. California, supra, at 23, 93 S.Ct., at 2614. Consequently, our difficulty was not only to assure that statutes designed to regulate obscene materials sufficiently defined what was prohibited, but also to devise substantive limits on what fell within the permissible scope of regulation. In Miller v. California, supra, a majority of the Court agreed that a "state offense must also be limited to works which, taken as a whole, appeal to the prurient interest in sex, which portray sexual conduct in a patently offensive way, and which, taken as a whole, do not have serious literary, artistic, political, or scientific value." Id., at 24, 93 S.Ct., at 2615. Over the past decade, we have adhered to the guidelines expressed in Miller,6 which subsequently has been followed in the regulatory schemes of most States.7 B 22 The Miller standard, like its predecessors, was an accommodation between the State's interests in protecting the "sensibilities of unwilling recipients" from exposure to pornographic material and the dangers of censorship inherent in unabashedly content-based laws. Like obscenity statutes, laws directed at the dissemination of child pornography run the risk of suppressing protected expression by allowing the hand of the censor to become unduly heavy. For the following reasons, however, we are persuaded that the States are entitled to greater leeway in the regulation of pornographic depictions of children. 23 First. It is evident beyond the need for elaboration that a State's interest in "safeguarding the physical and psychological well-being of a minor" is "compelling." Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 607, 102 S.Ct. 2613, 2620, 73 L.Ed.2d 248 (1982). "A democratic society rests, for its continuance, upon the healthy, well-rounded growth of young people into full maturity as citizens." Prince v. Massachusetts, 321 U.S. 158, 168, 64 S.Ct. 438, 443, 88 L.Ed. 645 (1944). Accordingly, we have sustained legislation aimed at protecting the physical and emotional well-being of youth even when the laws have operated in the sensitive area of constitutionally protected rights. In Prince v. Massachusetts, supra, the Court held that a statute prohibiting use of a child to distribute literature on the street was valid notwithstanding the statute's effect on a First Amendment activity. In Ginsberg v. New York, supra, we sustained a New York law protecting children from exposure to nonobscene literature. Most recently, we held that the Government's interest in the "well-being of its youth" justified special treatment of indecent broadcasting received by adults as well as children. FCC v. Pacifica Foundation, 438 U.S. 726, 98 S.Ct. 3026, 57 L.Ed.2d 1073 (1978). 24 The prevention of sexual exploitation and abuse of children constitutes a government objective of surpassing importance. The legislative findings accompanying passage of the New York laws reflect this concern: 25 "[T]here has been a proliferation of exploitation of children as subjects in sexual performances. The care of children is a sacred trust and should not be abused by those who seek to profit through a commercial network based upon the exploitation of children. The public policy of the state demands the protection of children from exploitation through sexual performances." 1977 N.Y.Laws, ch. 910, § 1.8 26 We shall not second-guess this legislative judgment. Respondent has not intimated that we do so. Suffice it to say that virtually all of the States and the United States have passed legislation proscribing the production of or otherwise combating "child pornography." The legislative judgment, as well as the judgment found in the relevant literature, is that the use of children as subjects of pornographic materials is harmful to the physiological, emotional, and mental health of the child.9 That judgment, we think, easily passes muster under the First Amendment. 27 Second. The distribution of photographs and films depicting sexual activity by juveniles is intrinsically related to the sexual abuse of children in at least two ways. First, the materials produced are a permanent record of the children's participation and the harm to the child is exacerbated by their circulation.10 Second, the distribution network for child pornography must be closed if the production of material which requires the sexual exploitation of children is to be effectively controlled. Indeed, there is no serious contention that the legislature was unjustified in believing that it is difficult, if not impossible, to halt the exploitation of children by pursuing only those who produce the photographs and movies. While the production of pornographic materials is a low-profile, clandestine industry, the need to market the resulting products requires a visible apparatus of distribution. The most expeditious if not the only practical method of law enforcement may be to dry up the market for this material by imposing severe criminal penalties on persons selling, advertising, or otherwise promoting the product. Thirty-five States and Congress have concluded that restraints on the distribution of pornographic materials are required in order to effectively combat the problem, and there is a body of literature and testimony to support these legislative conclusions.11 Cf. United States v. Darby, 312 U.S. 100, 61 S.Ct. 451, 85 L.Ed. 609 (1941) (upholding federal restrictions on sale of goods manufactured in violation of Fair Labor Standards Act). 28 Respondent does not contend that the State is unjustified in pursuing those who distribute child pornography. Rather, he argues that it is enough for the State to prohibit the distribution of materials that are legally obscene under the Miller test. While some States may find that this approach properly accommodates its interests, it does not follow that the First Amendment prohibits a State from going further. The Miller standard, like all general definitions of what may be banned as obscene, does not reflect the State's particular and more compelling interest in prosecuting those who promote the sexual exploitation of children. Thus, the question under the Miller test of whether a work, taken as a whole, appeals to the prurient interest of the average person bears no connection to the issue of whether a child has been physically or psychologically harmed in the production of the work. Similarly, a sexually explicit depiction need not be "patently offensive" in order to have required the sexual exploitation of a child for its production. In addition, a work which, taken on the whole, contains serious literary, artistic, political, or scientific value may nevertheless embody the hardest core of child pornography. "It is irrelevant to the child [who has been abused] whether or not the material . . . has a literary, artistic, political or social value." Memorandum of Assemblyman Lasher in Support of § 263.15. We therefore cannot conclude that the Miller standard is a satisfactory solution to the child pornography problem.12 29 Third. The advertising and selling of child pornography provide an economic motive for and are thus an integral part of the production of such materials, an activity illegal throughout the Nation.13 "It rarely has been suggested that the constitutional freedom for speech and press extends its immunity to speech or writing used as an integral part of conduct in violation of a valid criminal statute." Giboney v. Empire Storage & Ice Co., 336 U.S. 490, 498, 69 S.Ct. 684, 688, 93 L.Ed. 834 (1949).14 We note that were the statutes outlawing the employment of children in these films and photographs fully effective, and the constitutionality of these laws has not been questioned, the First Amendment implications would be no greater than that presented by laws against distribution: enforceable production laws would leave no child pornography to be marketed.15 30 Fourth. The value of permitting live performances and photographic reproductions of children engaged in lewd sexual conduct is exceedingly modest, if not de minimis. We consider it unlikely that visual depictions of children performing sexual acts or lewdly exhibiting their genitals would often constitute an important and necessary part of a literary performance or scientific or educational work. As a state judge in this case observed, if it were necessary for literary or artistic value, a person over the statutory age who perhaps looked younger could be utilized.16 Simulation outside of the prohibition of the statute could provide another alternative. Nor is there any question here of censoring a particular literary theme or portrayal of sexual activity. The First Amendment interest is limited to that of rendering the portrayal somewhat more "realistic" by utilizing or photographing children. 31 Fifth. Recognizing and classifying child pornography as a category of material outside the protection of the First Amendment is not, incompatible with our earlier decisions. "The question whether speech is, or is not protected by the First Amendment often depends on the content of the speech." Young v. American Mini Theatres, Inc., 427 U.S. 50, 66, 96 S.Ct. 2440, 2450, 49 L.Ed.2d 310 (1976) (opinion of STEVENS, J., joined by BURGER, C.J., and WHITE and REHNQUIST, JJ.). See also FCC v. Pacifica Foundation, 438 U.S. 726, 742-748, 98 S.Ct. 3026, 3036-3039, 57 L.Ed.2d 1073 (1978) (opinion of STEVENS, J., joined by BURGER, C.J., and REHNQUIST, J.). "[I]t is the content of [an] utterance that determines whether it is a protected epithet or an unprotected 'fighting comment.' " Young v. American Mini Theatres, Inc., supra, 427 U.S., at 66, 96 S.Ct., at 2450. See Chaplinsky v. New Hampshire, 315 U.S. 568, 62 S.Ct. 766, 86 L.Ed. 1031 (1942). Leaving aside the special considerations when public officials are the target, New York Times Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 95 L.Ed. 1412 (1964), a libelous publication is not protected by the Constitution. Beauharnais v. Illinois, 343 U.S. 250, 72 S.Ct. 725, 96 L.Ed. 919 (1952). Thus, it is not rare that a content-based classification of speech has been accepted because it may be appropriately generalized that within the confines of the given classification, the evil to be restricted so overwhelmingly outweighs the expressive interests, if any, at stake, that no process of case-by-case adjudication is required. When a definable class of material, such as that covered by § 263.15, bears so heavily and pervasively on the welfare of children engaged in its production, we think the balance of competing interests is clearly struck and that it is permissible to consider these materials as without the protection of the First Amendment. C 32 There are, of course, limits on the category of child pornography which, like obscenity, is unprotected by the First Amendment. As with all legislation in this sensitive area, the conduct to be prohibited must be adequately defined by the applicable state law, as written or authoritatively construed. Here the nature of the harm to be combated requires that the state offense be limited to works that visually depict sexual conduct by children below a specified age.17 The category of "sexual conduct" proscribed must also be suitably limited and described. 33 The test for child pornography is separate from the obscenity standard enunciated in Miller, but may be compared to it for the purpose of clarity. The Miller formulation is adjusted in the following respects: A trier of fact need not find that the material appeals to the prurient interest of the average person; it is not required that sexual conduct portrayed be done so in a patently offensive manner; and the material at issue need not be considered as a whole. We note that the distribution of descriptions or other depictions of sexual conduct, not otherwise obscene, which do not involve live performance or photographic or other visual reproduction of live performances, retains First Amendment protection. As with obscenity laws, criminal responsibility may not be imposed without some element of scienter on the part of the defendant. Smith v. California, 361 U.S. 147, 80 S.Ct. 215, 4 L.Ed.2d 205 (1959); Hamling v. United States, 418 U.S. 87, 94 S.Ct. 2887, 41 L.Ed.2d 590 (1974). D 34 Section 263.15's prohibition incorporates a definition of sexual conduct that comports with the above-stated principles. The forbidden acts to be depicted are listed with sufficient precision and represent the kind of conduct that, if it were the theme of a work, could render it legally obscene: "actual or simulated sexual intercourse, deviate sexual intercourse, sexual bestiality, masturbation, sado-masochistic abuse, or lewd exhibition of the genitals." § 263.00(3). The term "lewd exhibition of the genitals" is not unknown in this area and, indeed, was given in Miller as an example of a permissible regulation. 413 U.S., at 25, 93 S.Ct., at 2615. A performance is defined only to include live or visual depictions: "any play, motion picture, photograph or dance . . . [or] other visual representation exhibited before an audience." § 263.00(4). Section 263.15 expressly includes a scienter requirement. 35 We hold that § 263.15 sufficiently describes a category of material the production and distribution of which is not entitled to First Amendment protection. It is therefore clear that there is nothing unconstitutionally "underinclusive" about a statute that singles out this category of material for proscription.18 It also follows that the State is not barred by the First Amendment from prohibiting the distribution of unprotected materials produced outside the State.19 III 36 It remains to address the claim that the New York statute is unconstitutionally overbroad because it would forbid the distribution of material with serious literary, scientific, or educational value or material which does not threaten the harms sought to be combated by the State. Respondent prevailed on that ground below, and it is to that issue that we now turn. 37 The New York Court of Appeals recognized that overbreadth scrutiny has been limited with respect to conduct-related regulation, Broadrick v. Oklahoma, 413 U.S. 601, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973), but it did not apply the test enunciated in Broadrick because the challenged statute, in its view, was directed at "pure speech." The court went on to find that § 263.15 was fatally overbroad: "[T]he statute would prohibit the showing of any play or movie in which a child portrays a defined sexual act, real or simulated, in a nonobscene manner. It would also prohibit the sale, showing, or distributing of medical or educational materials containing photographs of such acts. Indeed, by its terms, the statute would prohibit those who oppose such portrayals from providing illustrations of what they oppose." 52 N.Y.2d, at 678, 439 N.Y.S.2d at 865, 422 N.E.2d, at 525. 38 While the construction that a state court gives a state statute is not a matter subject to our review, Wainwright v. Stone, 414 U.S. 21, 22-23, 94 S.Ct. 190, 192, 38 L.Ed.2d 179 (1973); Gooding v. Wilson, 405 U.S. 518, 520, 92 S.Ct. 1103, 1105, 31 L.Ed.2d 408 (1972), this Court is the final arbiter of whether the Federal Constitution necessitated the invalidation of a state law. It is only through this process of review that we may correct erroneous applications of the Constitution that err on the side of an overly broad reading of our doctrines and precedents, as well as state-court decisions giving the Constitution too little shrift. A state court is not free to avoid a proper facial attack on federal constitutional grounds. Bigelow v. Virginia, 421 U.S. 809, 817, 95 S.Ct. 2222, 2230, 44 L.Ed.2d 600 (1975). By the same token, it should not be compelled to entertain an overbreadth attack when not required to do so by the Constitution. A. 39 The traditional rule is that a person to whom a statute may constitutionally be applied may not challenge that statute on the ground that it may conceivably be applied unconstitutionally to others in situations not before the Court. Broadrick v. Oklahoma, supra, at 610, 93 S.Ct., at 2914; United States v. Raines, 362 U.S. 17, 21, 80 S.Ct. 519, 522, 4 L.Ed.2d 524 (1960); Carmichael v. Southern Coal & Coke Co., 301 U.S. 495, 513, 57 S.Ct. 868, 874, 81 L.Ed. 1245 (1937); Yazoo & M. V. R. Co. v. Jackson Vinegar Co., 226 U.S. 217, 219-220, 33 S.Ct. 40-41, 57 L.Ed. 193 (1912). In Broadrick, we recognized that this rule reflects two cardinal principles of our constitutional order: the personal nature of constitutional rights, McGowan v. Maryland, 366 U.S. 420, 429, 81 S.Ct. 1101, 1106, 6 L.Ed.2d 393 (1961), and prudential limitations on constitutional adjudication.20 In United States v. Raines, supra, 362 U.S., at 21, 80 S.Ct., at 522, we noted the "incontrovertible proposition" that it " 'would indeed be undesirable for this Court to consider every conceivable situation which might possibly arise in the application of complex and comprehensive legislation,' " (quoting Barrows v. Jackson, 346 U.S. 249, 256, 73 S.Ct. 1031, 1035, 97 L.Ed. 1586 (1953)). By focusing on the factual situation before us, and similar cases necessary for development of a constitutional rule,21 we face "flesh-and-blood"22 legal problems with data "relevant and adequate to an informed judgment."23 This practice also fulfills a valuable institutional purpose: it allows state courts the opportunity to construe a law to avoid constitutional infirmities. 40 What has come to be known as the First Amendment overbreadth doctrine is one of the few exceptions to this principle and must be justified by "weighty countervailing policies." United States v. Raines, supra, 362 U.S., at 22-23, 80 S.Ct., at 523. The doctrine is predicated on the sensitive nature of protected expression: "persons whose expression is constitutionally protected may well refrain from exercising their rights for fear of criminal sanctions by a statute susceptible of application to protected expression." Village of Schaumburg v. Citizens for a Better Environment, 444 U.S. 620, 634, 100 S.Ct. 826, 834, 63 L.Ed.2d 73 (1980); Gooding v. Wilson, supra, at 521, 92 S.Ct., at 1105. It is for this reason that we have allowed persons to attack overly broad statutes even though the conduct of the person making the attack is clearly unprotected and could be proscribed by a law drawn with the requisite specificity. Dombrowski v. Pfister, 380 U.S. 479, 486, 85 S.Ct. 1116, 1120, 14 L.Ed.2d 22 (1965); Thornhill v. Alabama, 310 U.S. 88, 97-98, 60 S.Ct. 736, 741-742, 84 L.Ed. 1093 (1940); United States v. Raines, supra, at 21-22, 80 S.Ct., at 522-523; Gooding v. Wilson, supra, at 521, 92 S.Ct., at 1105. 41 The scope of the First Amendment overbreadth doctrine, like most exceptions to established principles, must be carefully tied to the circumstances in which facial invalidation of a statute is truly warranted. Because of the wide-reaching effects of striking down a statute on its face at the request of one whose own conduct may be punished despite the First Amendment, we have recognized that the overbreadth doctrine is "strong medicine" and have employed it with hesitation, and then "only as a last resort." Broadrick, 413 U.S., at 613, 93 S.Ct., at 2916. We have, in consequence, insisted that the overbreadth involved be "substantial" before the statute involved will be invalidated on its face.24 42 In Broadrick, we explained the basis for this requirement: 43 "[T]he plain import of our cases is, at the very least, that facial overbreadth adjudication is an exception to our traditional rules of practice and that its function, a limited one at the outset, attenuates as the otherwise unprotected behavior that it forbids the State to sanction moves from 'pure speech' toward conduct and that conduct—even if expressive—falls within the scope of otherwise valid criminal laws that reflect legitimate state interests in maintaining comprehensive controls over harmful, constitutionally unprotected conduct. Although such laws, if too broadly worded, may deter protected speech to some unknown extent, there comes a point where that effect—at best a prediction—cannot, with confidence, justify invalidating a statute on its face and so prohibiting a State from enforcing the statute against conduct that is admittedly within its power to proscribe. Cf. Alderman v. United States, 394 U.S. 165, 174-175, 89 S.Ct. 961, 966-967, 22 L.Ed.2d 176 (1969)." Id., at 615, 93 S.Ct., at 2917. 44 We accordingly held that "particularly where conduct and not merely speech is involved, we believe that the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute's plainly legitimate sweep." Ibid.25 45 Broadrick examined a regulation involving restrictions on political campaign activity, an area not considered "pure speech," and thus it was unnecessary to consider the proper overbreadth test when a law arguably reaches traditional forms of expression such as books and films. As we intimated in Broadrick, the requirement of substantial overbreadth extended "at the very least" to cases involving conduct plus speech. This case, which poses the question squarely, convinces us that the rationale of Broadrick is sound and should be applied in the present context involving the harmful employment of children to make sexually explicit materials for distribution. 46 The premise that a law should not be invalidated for overbreadth unless it reaches a substantial number of impermissible applications is hardly novel. On most occasions involving facial invalidation, the Court has stressed the embracing sweep of the statute over protected expression.26 Indeed, Justice BRENNAN observed in his dissenting opinion in Broadrick : 47 "We have never held that a statute should be held invalid on its face merely because it is possible to conceive of a single impermissible application, and in that sense a requirement of substantial overbreadth is already implicit in the doctrine." Id., at 630, 93 S.Ct., at 2925. 48 The requirement of substantial overbreadth is directly derived from the purpose and nature of the doctrine. While a sweeping statute, or one incapable of limitation, has the potential to repeatedly chill the exercise of expressive activity by many individuals, the extent of deterrence of protected speech can be expected to decrease with the declining reach of the regulation.27 This observation appears equally applicable to the publication of books and films as it is to activities, such as picketing or participation in election campaigns, which have previously been categorized as involving conduct plus speech. We see no appreciable difference between the position of a publisher or bookseller in doubt as to the reach of New York's child pornography law and the situation faced by the Oklahoma state employees with respect to that State's restriction on partisan political activity. Indeed, it could reasonably be argued that the bookseller, with an economic incentive to sell materials that may fall within the statute's scope, may be less likely to be deterred than the employee who wishes to engage in political campaign activity. Cf. Bates v. State Bar of Arizona, 433 U.S. 350, 380-381, 97 S.Ct. 2691, 2707, 53 L.Ed.2d 810 (1977) (overbreadth analysis inapplicable to commercial speech). 49 This requirement of substantial overbreadth may justifiably be applied to statutory challenges which arise in defense of a criminal prosecution as well as civil enforcement or actions seeking a declaratory judgment. Cf. Parker v. Levy, 417 U.S. 733, 760, 94 S.Ct. 2547, 2563, 41 L.Ed.2d 439 (1974). Indeed, the Court's practice when confronted with ordinary criminal laws that are sought to be applied against protected conduct is not to invalidate the law in toto, but rather to reverse the particular conviction. Cantwell v. Connecticut, 310 U.S. 296, 60 S.Ct. 900, 84 L.Ed. 1213 (1940); Edwards v. South Carolina, 372 U.S. 229, 83 S.Ct. 680, 9 L.Ed.2d 697 (1973). We recognize, however, that the penalty to be imposed is relevant in determining whether demonstrable overbreadth is substantial. We simply hold that the fact that a criminal prohibition is involved does not obviate the need for the inquiry or a priori warrant a finding of substantial overbreadth. B 50 Applying these principles, we hold that § 263.15 is not substantially overbroad. We consider this the paradigmatic case of a state statute whose legitimate reach dwarfs its arguably impermissible applications. New York, as we have held, may constitutionally prohibit dissemination of material specified in § 263.15. While the reach of the statute is directed at the hard core of child pornography, the Court of Appeals was understandably concerned that some protected expression, ranging from medical textbooks to pictorials in the National Geographic would fall prey to the statute. How often, if ever, it may be necessary to employ children to engage in conduct clearly within the reach of § 263.15 in order to produce educational, medical, or artistic works cannot be known with certainty. Yet we seriously doubt, and it has not been suggested, that these arguably impermissible applications of the statute amount to more than a tiny fraction of the materials within the statute's reach. Nor will we assume that the New York courts will widen the possibly invalid reach of the statute by giving an expansive construction to the proscription on "lewd exhibition[s] of the genitals." Under these circumstances, § 263.15 is "not substantially overbroad and . . . Swhatever overbreadth may exist should be cured through case-by-case analysis of the fact situations to which its sanctions, assertedly, may not be applied." Broadrick v. Oklahoma, 413 U.S., at 615-616, 93 S.Ct., at 2917-2918. IV 51 Because § 263.15 is not substantially overbroad, it is unnecessary to consider its application to material that does not depict sexual conduct of a type that New York may restrict consistent with the First Amendment. As applied to Paul Ferber and to others who distribute similar material, the statute does not violate the First Amendment as applied to the States through the Fourteenth.28 The judgment of the New York Court of Appeals is reversed, and the case is remanded to that court for further proceedings not inconsistent with this opinion. 52 So ordered. 53 Justice BLACKMUN concurs in the result. 54 Justice O'CONNOR, concurring. 55 Although I join the Court's opinion, I write separately to stress that the Court does not hold that New York must except "material with serious literary, scientific, or educational value," ante, at 766, from its statute. The Court merely holds that, even if the First Amendment shelters such material, New York's current statute is not sufficiently overbroad to support respondent's facial attack. The compelling interests identified in today's opinion, see ante, at 756-764, suggest that the Constitution might in fact permit New York to ban knowing distribution of works depicting minors engaged in explicit sexual conduct, regardless of the social value of the depictions. For example, a 12-year-old child photographed while masturbating surely suffers the same psychological harm whether the community labels the photograph "edifying" or "tasteless." The audience's appreciation of the depiction is simply irrelevant to New York's asserted interest in protecting children from psychological, emotional, and mental harm. 56 An exception for depictions of serious social value, moreover, would actually increase opportunities for the content-based censorship disfavored by the First Amendment. As drafted, New York's statute does not attempt to suppress the communication of particular ideas. The statute permits discussion of child sexuality, forbidding only attempts to render the "portrayal[s] somewhat more 'realistic' by utilizing or photographing children." Ante, at 763. Thus, the statute attempts to protect minors from abuse without attempting to restrict the expression of ideas by those who might use children as live models. 57 On the other hand, it is quite possible that New York's statute is overbroad because it bans depictions that do not actually threaten the harms identified by the Court. For example, clinical pictures of adolescent sexuality, such as those that might appear in medical textbooks, might not involve the type of sexual exploitation and abuse targeted by New York's statute. Nor might such depictions feed the poisonous "kiddie porn" market that New York and other States have attempted to regulate. Similarly, pictures of children engaged in rites widely approved by their cultures, such as those that might appear in issues of the National Geographic, might not trigger the compelling interests identified by the Court. It is not necessary to address these possibilities further today, however, because this potential overbreadth is not sufficiently substantial to warrant facial invalidation of New York's statute. 58 Justice BRENNAN, with whom Justice MARSHALL joins, concurring in the judgment. 59 I agree with much of what is said in the Court's opinion. As I made clear in the opinion I delivered for the Court in Ginsburg v. New York, 390 U.S. 629, 88 S.Ct. 1274, 20 L.Ed.2d 195 (1968), the State has a special interest in protecting the well-being of its youth. Id., at 638-641, 88 S.Ct., at 1279-1281. See also Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 607, 102 S.Ct. 2613, 2621, 73 L.Ed.2d 248 (1982). This special and compelling interest, and the particular vulnerability of children, afford the State the leeway to regulate pornographic material, the promotion of which is harmful to children, even though the State does not have such leeway when it seeks only to protect consenting adults from exposure to such material. Ginsburg v. New York, supra, at 637, 638, n. 6, 642-643, n. 10, 88 S.Ct., at 1279, 1280, n. 6, 1282, n. 10. See also Jacobellis v. Ohio, 378 U.S. 184, 195, 84 S.Ct. 1676, 1682, 12 L.Ed.2d 793 (1964) (opinion of BRENNAN, J.). I also agree with the Court that the "tiny fraction," ante, at 773, of material of serious artistic, scientific, or educational value that could conceivably fall within the reach of the statute is insufficient to justify striking the statute on the grounds of overbreadth. See Broadrick v. Oklahoma, 413 U.S. 601, 630, 93 S.Ct. 2908, 2925, 37 L.Ed.2d 830 (1973) (BRENNAN, J., dissenting). 60 But in my view application of § 263.15 or any similar statute to depictions of children that in themselves do have serious literary, artistic, scientific, or medical value, would violate the First Amendment. As the Court recognizes, the limited classes of speech, the suppression of which does not raise serious First Amendment concerns, have two attributes. They are of exceedingly "slight social value," and the State has a compelling interest in their regulation. See Chaplinsky v. New Hampshire, 315 U.S. 568, 571-572, 62 S.Ct. 766, 768-769, 86 L.Ed. 1031 (1942). The First Amendment value of depictions of children that are in themselves serious contributions to art, literature, or science, is, by definition, simply not "de minimis." See ante, at 761. At the same time, the State's interest in suppression of such materials is likely to be far less compelling. For the Court's assumption of harm to the child resulting from the "permanent record" and "circulation" of the child's "participation," ante, at 759, lacks much of its force where the depiction is a serious contribution to art or science. The production of materials of serious value is not the "lowprofile, clandestine industry" that according to the Court produces purely pornographic materials. See ante, at 760. In short, it is inconceivable how a depiction of a child that is itself a serious contribution to the world of art or literature or science can be deemed "material outside the protection of the First Amendment." See ante, at 763. 61 I, of course, adhere to my view that, in the absence of exposure, or particular harm, to juveniles or unconsenting adults, the State lacks power to suppress sexually oriented materials. See, e.g., Paris Adult Theatre I v. Slaton, 413 U.S. 49, 73, 93 S.Ct. 2628, 2642, 37 L.Ed.2d 446 (1973) (BRENNAN, J., dissenting). With this understanding, I concur in the Court's judgment in this case. 62 Justice STEVENS, concurring in the judgment. 63 Two propositions seem perfectly clear to me. First, the specific conduct that gave rise to this criminal prosecution is not protected by the Federal Constitution; second, the state statute that respondent violated prohibits some conduct that is protected by the First Amendment. The critical question, then, is whether this respondent, to whom the statute may be applied without violating the Constitution, may challenge the statute on the ground that it conceivably may be applied unconstitutionally to others in situations not before the Court. I agree with the Court's answer to this question but not with its method of analyzing the issue. 64 Before addressing that issue, I shall explain why respondent's conviction does not violate the Constitution. The two films that respondent sold contained nothing more than lewd exhibition; there is no claim that the films included any material that had literary, artistic, scientific, or educational value.1 Respondent was a willing participant in a commercial market that the State of New York has a legitimate interest in suppressing. The character of the State's interest in protecting children from sexual abuse justifies the imposition of criminal sanctions against those who profit, directly or indirectly, from the promotion of such films. In this respect my evaluation of this case is different from the opinion I have expressed concerning the imposition of criminal sanctions for the promotion of obscenity in other contexts.2 65 A holding that respondent may be punished for selling these two films does not require us to conclude that other users of these very films, or that other motion pictures containing similar scenes, are beyond the pale of constitutional protection. Thus, the exhibition of these films before a legislative committee studying a proposed amendment to a state law, or before a group of research scientists studying human behavior, could not, in my opinion, be made a crime. Moreover, it is at least conceivable that a serious work of art, a documentary on behavioral problems, or a medical or psychiatric teaching device, might include a scene from one of these films and, when viewed as a whole in a proper setting, be entitled to constitutional protection. The question whether a specific act of communication is protected by the First Amendment always requires some consideration of both its content and its context. 66 The Court's holding that this respondent may not challenge New York's statute as overbroad follows its discussion of the contours of the category of nonobscene child pornography that New York may legitimately prohibit. Having defined that category in an abstract setting,3 the Court makes the empirical judgment that the arguably impermissible application of the New York statute amounts to only a "tiny fraction of the materials within the statute's reach." Ante, at 773. Even assuming that the Court's empirical analysis is sound,4 I believe a more conservative approach to the issue would adequately vindicate the State's interest in protecting its children and cause less harm to the federal interest in free expression. 67 A hypothetical example will illustrate my concern. Assume that the operator of a New York motion picture theater specializing in the exhibition of foreign feature films is offered a full-length movie containing one scene that is plainly lewd if viewed in isolation but that nevertheless is part of a serious work of art. If the child actor resided abroad, New York's interest in protecting its young from sexual exploitation would be far less compelling than in the case before us. The federal interest in free expression would, however, be just as strong as if an adult actor had been used. There are at least three different ways to deal with the statute's potential application to that sort of case. 68 First, at one extreme and as the Court appears to hold, the First Amendment inquiry might be limited to determining whether the offensive scene, viewed in isolation, is lewd. When the constitutional protection is narrowed in this drastic fashion, the Court is probably safe in concluding that only a tiny fraction of the materials covered by the New York statute is protected. And with respect to my hypothetical exhibitor of foreign films, he need have no uncertainty about the permissible application of the statute; for the one lewd scene would deprive the entire film of any constitutional protection. 69 Second, at the other extreme and as the New York Court of Appeals correctly perceived, the application of this Court's cases requiring that an obscenity determination be based on the artistic value of a production taken as a whole would afford the exhibitor constitutional protection and result in a holding that the statute is invalid because of its overbreadth. Under that approach, the rationale for invalidating the entire statute is premised on the concern that the exhibitor's understanding about its potential reach could cause him to engage in self-censorship. This Court's approach today substitutes broad, unambiguous, state-imposed censorship for the self-censorship that an overbroad statute might produce. 70 Third, as an intermediate position, I would refuse to apply overbreadth analysis for reasons unrelated to any prediction concerning the relative number of protected communications that the statute may prohibit. Specifically, I would postpone decision of my hypothetical case until it actually arises. Advocates of a liberal use of overbreadth analysis could object to such postponement on the ground that it creates the risk that the exhibitor's uncertainty may produce self-censorship. But that risk obviously interferes less with the interest in free expression than does an abstract, advance ruling that the film is simply unprotected whenever it contains a lewd scene, no matter how brief. 71 My reasons for avoiding overbreadth analysis in this case are more qualitative than quantitative. When we follow our traditional practice of adjudicating difficult and novel constitutional questions only in concrete factual situations, the adjudications tend to be crafted with greater wisdom. Hypothetical rulings are inherently treacherous and prone to lead us into unforeseen errors; they are qualitatively less reliable than the products of case-by-case adjudication. 72 Moreover, it is probably safe to assume that the category of speech that is covered by the New York statute generally is of a lower quality than most other types of communication. On a number of occasions, I have expressed the view that the First Amendment affords some forms of speech more protection from governmental regulation than other forms of speech.5 Today the Court accepts this view, putting the category of speech described in the New York statute in its rightful place near the bottom of this hierarchy. Ante, at 761-763. Although I disagree with the Court's position that such speech is totally without First Amendment protection, I agree that generally marginal speech does not warrant the extraordinary protection afforded by the overbreadth doctrine.6 73 Because I have no difficulty with the statute's application in this case, I concur in the Court's judgment. 1 "[C]hild pornography and child prostitution have become highly organized, multimillion dollar industries that operate on a nationwide scale." S.Rep.No.95-438, p. 5 (1977), U.S.Code Cong. & Admin.News 1978, pp. 40, 42. One researcher has documented the existence of over 260 different magazines which depict children engaging in sexually explicit conduct. Ibid. "Such magazines depict children, some as young as three to five years of age . . .. The activities featured range from lewd poses to intercourse, fellatio, cunnilingus, masturbation, rape, incest and sado-masochism." Id., at 6. In Los Angeles alone, police reported that 30,000 children have been sexually exploited. Sexual Exploitation of Children, Hearings before the Subcommittee on Select Education of the House Committee on Education and Labor, 95th Cong., 1st Sess., 41-42 (1977). 2 In addition to New York, 19 States have prohibited the dissemination of material depicting children engaged in sexual conduct regardless of whether the material is obscene. Ariz.Rev.Stat.Ann. § 13-3553 (Supp.1981-1982); Colo.Rev.Stat. § 18-6-403 (Supp.1981); Del.Code Ann., Tit. 11, §§ 1108, 1109 (1979); Fla.Stat. § 847.014 (1981); Haw.Rev.Stat. § 707-751 (Supp.1981); Ky.Rev.Stat. §§ 531.320, 531.340-531.360 (Supp.1980); La.Rev.Stat.Ann. § 14:81.1(A)(3) (West Supp.1982); Mass.Gen.Laws Ann., ch. 272, § 29A (West Supp.1982-1983); Mich.Comp.Laws Ann. § 750.145c(3) (1982-1983); Miss.Code Ann. § 97-5-33(4) (Supp.1981); Mont.Code Ann. § 45-5-625 (1981); N.J.Stat.Ann. § 2C:24-4(b)(5) (West 1981); Okla.Stat., Tit. 21, § 1021.2 (1981); Pa.Stat.Ann., Tit. 18, § 6312(c) (Purdon 1982-1983); R.I.Gen.Laws § 11-9-1.1 (1981); Tex. Penal Code Ann. § 43.25 (1982); Utah Code Ann. § 76-10-1206.5(3) (Supp.1981); W.Va.Code § 61-8C-3 (Supp.1981); Wis.Stat. § 940.203(4) (1979-1980). Fifteen States prohibit the dissemination of such material only if it is obscene. Ala.Code §§ 13-7-231, 13-7-232 (Supp.1981); Ark.Stat.Ann. § 41-4204 (Supp.1981); Cal.Penal Code Ann. § 311.2(b) (West Supp.1982) (general obscenity statute); Ill.Rev.Stat., ch. 38, ¶ 11-20a(b)(1) (1979); Ind.Code § 35-30-10.1-2 (1979); Me.Rev.Stat.Ann., Tit. 17, § 2923(1) (Supp.1981-1982); Minn.Stat. §§ 617.246(3) and (4) (1980); Neb.Rev.Stat. § 28-1463(2) (1979); N.H.Rev.Stat.Ann. § 650:2(II) (Supp.1981); N.D.Cent.Code § 12.1-27.1-01 (1976) (general obscenity statute); Ohio Rev.Code Ann. § 2907.321(A) (1982); Ore.Rev.Stat. § 163.485 (1981); S.D. Codified Laws §§ 22-22-24, 22-22-25 (1979); Tenn.Code Ann. § 39-1020 (Supp.1981); Wash.Rev.Code § 9.68A.030 (1981). The federal statute also prohibits dissemination only if the material is obscene. 18 U.S.C. § 2252(a) (1976 ed., Supp. IV). Two States prohibit dissemination only if the material is obscene as to minors. Conn.Gen.Stat. § 53a-196b (1981); Va.Code § 18.2-374.1 (1982). Twelve States prohibit only the use of minors in the production of the material. Alaska Stat.Ann. § 11.41.455 (1978); Ga.Code § 26-9943a(b) (1978); Idaho Code § 44-1306 (1977); Iowa Code § 728.12 (1981); Kan.Stat.Ann. § 21-3516 (1981); Md.Ann.Code, Art. 27, § 419A (Supp.1981); Mo.Rev.Stat. § 568.060(1)(b) (1978); Nev.Rev.Stat. § 200.509 (1981); N.M.Stat.Ann. § 30-6-1 (Supp.1982); N.C.Gen.Stat. § 14-190.6 (1981); S.C.Code § 16-15-380 (Supp.1981); Wyo.Stat. § 14-3-102(a)(v)(E) (1978). 3 Class D felonies carry a maximum punishment of imprisonment for up to seven years as to individuals, and as to corporations a fine of up to $10,000. N.Y.Penal Law §§ 70.00, 80.10 (McKinney 1975). Respondent Ferber was sentenced to 45 days in prison. 4 A state judge rejected Ferber's First Amendment attack on the two sections in denying a motion to dismiss the indictment. 96 Misc.2d 669, 409 N.Y.S.2d 632 (1978). 5 N.Y. Penal Law § 235.00(1) (McKinney 1980); People v. Illardo, 48 N.Y.2d 408, 415, and n. 3, 423 N.Y.S.2d 470, 472-473, and n. 3, 399 N.E.2d 59, 62-63, and n. 3 (1979). 6 Hamling v. United States, 418 U.S. 87, 94 S.Ct. 2887, 41 L.Ed.2d 590 (1974); Jenkins v. Georgia, 418 U.S. 153, 94 S.Ct. 2750, 41 L.Ed.2d 642 (1974); Ward v. Illinois, 431 U.S. 767, 97 S.Ct. 2085, 52 L.Ed.2d 738 (1977); Marks v. United States, 430 U.S. 188, 97 S.Ct. 990, 51 L.Ed.2d 260 (1977); Pinkus v. United States, 436 U.S. 293, 98 S.Ct. 1808, 56 L.Ed.2d 293 (1978). 7 Thirty-seven States and the District of Columbia have either legislatively adopted or judicially incorporated the Miller test for obscenity. Ala.Code § 13A-12-150 (Supp.1981); Ariz.Rev.Stat.Ann. § 13-3501(2) (1978); Ark.Stat.Ann. § 41-3502(6) (Supp.1981); Colo.Rev.Stat. § 18-7-101(2) (Supp.1981); Del.Code Ann., Tit. 11, § 1364 (1979); Lakin v. United States, 363 A.2d 990 (D.C.1976); Ga.Code § 26-2101(b) (1978); Haw.Rev.Stat. § 712-1210(6) (Supp.1981); Idaho Code § 18-4101(A) (1979); Iowa Code § 728.4 (1981) (only child pornography covered); Ind.Code § 35-30-10.1-1(c) (1979); Kan.Stat.Ann. § 21-4301(2)(a) (1981); Ky.Rev.Stat. § 531.010(3) (1975); La.Rev.Stat.Ann. §§ 14:106(A)(2) and (A)(3) (West Supp.1982); Ebert v. Maryland State Bd. of Censors, 19 Md.App. 300, 313 A.2d 536 (1973); Mass.Gen.Laws Ann., ch. 272, § 31 (West Supp.1982-1983); People v. Neumayer, 405 Mich. 341, 275 N.W.2d 230 (1979); State v. Welke, 298 Minn. 402, 216 N.W.2d 641 (1974); Mo.Rev.Stat. § 573.010(1) (1978); Neb.Rev.Stat. § 28-807(9) (1979); Nev.Rev.Stat. § 201.235 (1981); N.H.Rev.Stat.Ann. § 650:1(IV) (Supp.1981); N.J.Stat.Ann. § 2C:34-2 (West 1981); N.Y.Penal Law § 235.00(1) (McKinney 1980); N.C.Gen.Stat. § 14-190.1(b) (1981); N.D.Cent.Code § 12.1-27.1-01(4) (1976); State v. Burgun, 56 Ohio St.2d 354, 384 N.E.2d 255 (1978); McCrary v. State, 533 P.2d 629 (Okl.Cr.App.1974); Ore.Rev.Stat. § 167.087(2) (1981); 18 Pa.Stat.Ann., Tit. 18, § 5903(b) (Purdon Supp.1982-1983); R.I.Gen.Laws § 11-31-1 (1981); S.C.Code § 16-15-260(a) (Supp.1981); S.D. Codified Laws § 22-24-27(10) (1979); Tenn.Code Ann. § 39-3001(1) (Supp.1981); Tex. Penal Code Ann. § 43.21(a) (1982); Utah Code Ann. § 76-10-1203(1) (1978); Va.Code § 18.2-372 (1982); 1982 Wash.Laws., ch. 184, § 1(2). Four States continue to follow the test approved in Memoirs v. Massachusetts, 383 U.S. 413, 86 S.Ct. 975, 16 L.Ed.2d 1 (1966). Cal.Penal Code Ann. § 311(a) (West Supp.1982); Conn.Gen.Stat. § 53a-193 (1981); Fla.Stat. § 847.07 (1981); Ill.Rev.Stat., ch. 38, ¶ 11-20(b) (1979). Five States regulate only the distribution of pornographic material to minors. Me.Rev.Stat.Ann., Tit. 17, § 2911 (Supp.1981-1982); Mont.Code Ann. § 45-8-201 (1981); N.M.Stat.Ann. § 30-37-2 (Supp.1982); Vt.Stat.Ann., Tit. 13, § 2802 (1974); W.Va.Code § 61-8A-2 (1977). Three state obscenity laws do not fall into any of the above categories. Miss.Code Ann. § 97-29-33 (1973), declared invalid in ABC Interstate Theatres, Inc. v. State, 325 So.2d 123 (Miss.1976); Wis.Stat. § 944.21(1)(a) (1979-1980), declared invalid in State v. Princess Cinema of Milwaukee, Inc., 96 Wis.2d 646, 292 N.W.2d 807 (1980); Wyo.Stat. § 6-5-303 (1977). Alaska has no current state obscenity law. A number of States employ a different obscenity standard with respect to material distributed to children. See, e.g., Fla.Stat. § 847.0125 (1981). 8 In addition, the legislature found: "the sale of these movies, magazines and photographs depicting the sexual conduct of children to be so abhorrent to the fabric of our society that it urge[d] law enforcement officers to aggressively seek out and prosecute . . . the peddlers . . . of this filth by vigorously applying the sanctions contained in this act." 1977 N.Y.Laws, ch. 910, § 1. 9 "[T]he use of children as . . . subjects of pornographic materials is very harmful to both the children and the society as a whole." S.Rep.No. 95-438, p. 5 (1977), U.S.Code Cong. & Admin.News 1978, p. 42. It has been found that sexually exploited children are unable to develop healthy affectionate relationships in later life, have sexual dysfunctions, and have a tendency to become sexual abusers as adults. Schoettle, Child Exploitation: A Study of Child Pornography, 19 J.Am.Acad.Child Psychiatry 289, 296 (1980) (hereafter cited as Child Exploitation); Schoettle, Treatment of the Child Pornography Patient, 137 Am.J.Psychiatry 1109, 1110 (1980); Densen-Gerner, Child Prostitution and Child Pornography: Medical, Legal, and Societal Aspects of the Commercial Exploitation of Children, reprinted in U. S. Dept. of Health and Human Services, Sexual Abuse of Children: Selected Readings 77, 80 (1980) (hereafter cited as Commercial Exploitation) (sexually exploited children predisposed to self-destructive behavior such as drug and alcohol abuse or prostitution). See generally Burgess & Holmstrom, Accessory-to-Sex: Pressure, Sex, and Secrecy, in A. Burgess, A. Groth, L. Holmstrom, & S. Sgroi, Sexual Assault of Children and Adolescents 85, 94 (1978); V. De Francis, Protecting the Child Victim of Sex Crimes Committed by Adults 169 (1969); Ellerstein & Canavan, Sexual Abuse of Boys, 134 Am.J. Diseases of Children 255, 256-257 (1980); Finch, Adult Seduction of the Child: Effects on the Child, Medical Aspects of Human Sexuality 170, 185 (Mar.1973); Groth, Sexual Trauma in the Life Histories of Rapists and Child Molesters, 4 Victimology 10 (1979). Sexual molestation by adults is often involved in the production of child sexual performances. Sexual Exploitation of Children, A Report to the Illinois General Assembly by the Illinois Legislative Investigating Commission 30-31 (1980). When such performances are recorded and distributed, the child's privacy interests are also invaded. See n. 10, infra. 10 As one authority has explained: "[P]ornography poses an even greater threat to the child victim than does sexual abuse or prostitution. Because the child's actions are reduced to a recording, the pornography may haunt him in future years, long after the original misdeed took place. A child who has posed for a camera must go through life knowing that the recording is circulating within the mass distribution system for child pornography." Shouvlin, Preventing the Sexual Exploitation of Children: A Model Act, 17 Wake Forest L.Rev. 535, 545 (1981). See also Child Exploitation 292 ("[I]t is the fear of exposure and the tension of keeping the act secret that seem to have the most profound emotional repercussions"); Note, Protection of Children from Use in Pornography: Toward Constitutional and Enforceable Legislation, 12 U.Mich.J. Law Reform 295, 301 (1979) (hereafter cited as Use in Pornography) (interview with child psychiatrist) ("The victim's knowledge of publication of the visual material increases the emotional and psychic harm suffered by the child"). Thus, distribution of the material violates "the individual interest in avoiding disclosure of personal matters." Whalen v. Roe, 429 U.S. 589, 599, 97 S.Ct. 869, 876, 51 L.Ed.2d 64 (1977). Respondent cannot undermine the force of the privacy interests involved here by looking to Cox Broadcasting Corp. v. Cohn, 420 U.S. 469, 95 S.Ct. 1029, 43 L.Ed.2d 328 (1975), and Smith v. Daily Mail Publishing Co., 443 U.S. 97, 99 S.Ct. 2667, 61 L.Ed.2d 399 (1979), cases protecting the right of newspapers to publish, respectively, the identity of a rape victim and a youth charged as a juvenile offender. Those cases only stand for the proposition that "if a newspaper lawfully obtains truthful information about a matter of public significance then state officials may not constitutionally punish publication of the information, absent a need . . . of the highest order." Id., at 103, 99 S.Ct., at 2671. 11 See Sexual Exploitation of Children, Hearings before the Subcommittee on Crime of the House Judiciary Committee, 95th Cong., 1st Sess., 34 (1977) (statement of Charles Rembar) ("It is an impossible prosecutorial job to try to get at the acts themselves"); id., at 11 (statement of Frank Osanka, Professor of Social Justice and Sociology) ("[W]e have to be very careful . . . that we don't take comfort in the existence of statutes that are on the books in the connection with the use of children in pornography. . . . There are usually no witnesses to these acts of producing pornography"); id., at 69 (statement of Investigator Lloyd Martin, Los Angeles Police Department) (producers of child pornography use false names making difficult the tracing of material back from distributor). See also L. Tribe, American Constitutional Law 666, n. 62 (1978); Note, Child Pornography: A New Role for the Obscenity Doctrine, 1978 U.Ill. Law Forum 711, 716, n. 29; Use in Pornography 315 ("passage of criminal laws aimed at producers without similar regulation of distributors will arguably shift the production process further underground"). 12 In addition, legal obscenity under Miller is a function of "contemporary community standards." 413 U.S., at 24, 93 S.Ct., at 2615. "It is neither realistic nor constitutionally sound to read the First Amendment as requiring that the people of Maine or Mississippi accept public depiction of conduct found tolerable in Las Vegas, or New York City." Id., at 32, 93 S.Ct., at 2619. It would be equally unrealistic to equate a community's toleration for sexually oriented material with the permissible scope of legislation aimed at protecting children from sexual exploitation. Furthermore, a number of States rely on stricter obscenity tests, see n. 7, supra, under which successful prosecution for child pornography may be even more difficult. 13 One state committee studying the problem declared: "The act of selling these materials is guaranteeing that there will be additional abuse of children." Texas House Select Committee on Child Pornography: Its Related Causes and Control 32 (1978). See also Commercial Exploitation 80 ("Printed materials cannot be isolated or removed from the process involved in developing them"). 14 In Giboney, a unanimous Court held that labor unions could be restrained from picketing a firm in support of a secondary boycott which a State had validly outlawed. In Pittsburgh Press Co. v. Pittsburgh Comm'n on Human Relations, 413 U.S. 376, 93 S.Ct. 2553, 37 L.Ed.2d 669 (1973), the Court allowed an injunction against a newspaper's furtherance of illegal sex discrimination by placing of job advertisements in gender-designated columns. The Court stated: "Any First Amendment interest which might be served by advertising an ordinary commercial proposal and which might arguably outweigh the governmental interest supporting the regulation is altogether absent when the commercial activity itself is illegal and the restriction on advertising is incidental to a valid limitation on economic activity." Id., at 389, 93 S.Ct., at 2561. 15 In this connection we note that 18 U.S.C. § 2251 (1976 ed., Supp.IV), making it a federal offense for anyone to use children under the age of 16 in the production of pornographic materials, embraces all "sexually explicit conduct" without imposing an obscenity test. In addition, half of the state laws imposing criminal liability on the producer do not require the visual material to be legally obscene. Use in Pornography 307-308. 16 96 Misc.2d, at 676, 409 N.Y.S.2d, at 637. This is not merely a hypothetical possibility. See Brief for Petitioner 25 and examples cited therein. 17 Sixteen States define a child as a person under age 18. Four States define a child as under 17 years old. The federal law and 16 States, including New York, define a child as under 16. Illinois and Nebraska define a child as a person under age 16 or who appears as a prepubescent. Ill.Rev.Stat., ch. 38, ¶ 11-20a(a)(1)(A) (1979); Neb.Rev.Stat. § 28-1463 (1979). Indiana defines a child as one who is or appears to be under 16. Ind.Code. §§ 35-30-10.1-2, 35-30-10.1-3 (1979). Kentucky provides for two age classifications (16 and 18) and varies punishment according to the victim's age. Ky.Rev.Stat. §§ 531.300-531.370 (Supp.1980). See Use in Pornography 307, n. 71 (collecting statutes). 18 Erznoznik v. City of Jacksonville, 422 U.S. 205, 95 S.Ct. 2268, 45 L.Ed.2d 125 (1975), relied upon by the Court of Appeals, struck down a law against drive-in theaters showing nude scenes if movies could be seen from a public place. Since nudity, without more is protected expression, id., at 213, 95 S.Ct., at 2274, we proceeded to consider the underinclusiveness of the ordinance. The Jacksonville ordinance impermissibly singled out movies with nudity for special treatment while failing to regulate other protected speech which created the same alleged risk to traffic. Today, we hold that child pornography as defined in § 263.15 is unprotected speech subject to content-based regulation. Hence, it cannot be underinclusive or unconstitutional for a State to do precisely that. 19 It is often impossible to determine where such material is produced. The Senate Report accompanying federal child pornography legislation stressed that "it is quite common for photographs or films made in the United States to be sent to foreign countries to be reproduced and then returned to this country in order to give the impression of foreign origin." S.Rep.No.95-438, p. 6 (1977), U.S.Code Cong. & Admin.News 1978, p. 44. In addition, States have not limited their distribution laws to material produced within their own borders because the maintenance of the market itself "leaves open the financial conduit by which the production of such material is funded and materially increases the risk that [local] children will be injured." 52 N.Y.2d 674, 688, 439 N.Y.S.2d 863, 871, 422 N.E.2d 523, 531 (1981) (Jasen, J., dissenting). 20 In addition to prudential restraints, the traditional rule is grounded in Art. III limits on the jurisdiction of federal courts to actual cases and controversies. "This Court, as is the case with all federal courts, 'has no jurisdiction to pronounce any statute, either of a State or of the United States, void, because irreconcilable with the Constitution, except as it is called upon to adjudge the legal rights of litigants in actual controversies. In the exercise of that jurisdiction, it is bound by two rules, to which it has rigidly adhered, one, never to anticipate a question of constitutional law in advance of the necessity of deciding it; the other never to formulate a rule of constitutional law broader than is required by the precise facts to which it is to be applied.' Liverpool, New York & Philadelphia S.S. Co. v. Commissioners of Emigration, 113 U.S. 33, 39, 5 S.Ct. 352, 355, 28 L.Ed. 899." United States v. Raines, 362 U.S. 17, 21, 80 S.Ct. 519, 522, 4 L.Ed.2d 524 (1960). 21 Overbreadth challenges are only one type of facial attack. A person whose activity may be constitutionally regulated nevertheless may argue that the statute under which he is convicted or regulated is invalid on its face. See, e.g., Terminiello v. City of Chicago, 337 U.S. 1, 5, 69 S.Ct. 894, 896, 93 L.Ed. 1131 (1949). See generally Monaghan, Overbreadth, 1981 S.Ct. Rev. 1, 10-14. 22 A. Bickel, The Least Dangerous Branch 115-116 (1962). 23 Frankfurter & Hart, The Business of the Supreme Court at October Term, 1934, 49 Harv.L.Rev. 68, 95-96 (1935). 24 When a federal court is dealing with a federal statute challenged as overbroad, it should, of course, construe the statute to avoid constitutional problems, if the statute is subject to such a limiting construction. Crowell v. Benson, 285 U.S. 22, 62, 52 S.Ct. 285, 296, 76 L.Ed. 598 (1932). Accord, e.g., Haynes v. United States, 390 U.S. 85, 92, 88 S.Ct. 722, 727, 19 L.Ed.2d 923 (1968) (dictum); Schneider v. Smith, 390 U.S. 17, 27, 88 S.Ct. 682, 687, 19 L.Ed.2d 799 (1968); United States v. Rumely, 345 U.S. 41, 45, 73 S.Ct. 543, 545, 97 L.Ed. 770 (1953); Ashwander v. TVA, 297 U.S. 288, 348, 56 S.Ct. 466, 483, 80 L.Ed. 688 (1936) (Brandeis, J. concurring). Furthermore, if the federal statute is not subject to a narrowing construction and is impermissibly overbroad, it nevertheless should not be stricken down on its face; if it is severable, only the unconstitutional portion is to be invalidated. United States v. Thirty-seven Photographs, 402 U.S. 363, 91 S.Ct. 1400, 28 L.Ed.2d 822 (1971). A state court is also free to deal with a state statute in the same way. If the invalid reach of the law is cured, there is no longer reason for proscribing the statute's application to unprotected conduct. Here, of course, we are dealing with a state statute on direct review of a state-court decision that has construed the statute. Such a construction is binding on us. 25 Parker v. Levy, 417 U.S. 733, 760, 94 S.Ct. 2547, 2563, 41 L.Ed.2d 439 (1974) ("This Court has . . . repeatedly expressed its reluctance to strike down a statute on its face where there were a substantial number of situations to which it might be validly applied. Thus, even if there are marginal applications in which a statute would infringe on First Amendment values, facial invalidation is inappropriate if the 'remainder of the statute . . . covers a whole range of easily identifiable and constitutionally proscribable . . . conduct. . . .' CSC v. Letter Carriers, 413 U.S. 548, 580-581 [93 S.Ct. 2880, 2897-2898, 37 L.Ed.2d 796] (1973)"). See Bogen, First Amendment Ancillary Doctrines, 37 Md.L.Rev. 679, 712-714 (1978); Note, The First Amendment Overbreadth Doctrine, 83 Harv.L.Rev. 844, 860-861 (1970). 26 In Gooding v. Wilson, 405 U.S. 518, 519, 527, 92 S.Ct. 1103, 1104, 1108, 31 L.Ed.2d 408 (1972), the Court's invalidation of a Georgia statute making it a misdemeanor to use " 'opprobrious words or abusive language, tending to cause a breach of the peace' " followed from state judicial decisions indicating that "merely to speak words offensive to some who hear them" could constitute a "breach of the peace." Cases invalidating laws requiring members of a "subversive organization" to take a loyalty oath, Baggett v. Bullitt, 377 U.S. 360, 84 S.Ct. 1316, 12 L.Ed.2d 377 (1964), or register with the government, Dombrowski v. Pfister, 380 U.S. 479, 85 S.Ct. 1116, 14 L.Ed.2d 22 (1965), can be explained on the basis that the laws involved, unlike § 263.15, defined no central core of constitutionally regulable conduct; the entire scope of the laws was subject to the uncertainties and vagaries of prosecutorial discretion. See also Bigelow v. Virginia, 421 U.S. 809, 817, 95 S.Ct. 2222, 2230, 44 L.Ed.2d 600 (1975) ("the facts of this case well illustrate 'the statute's potential for sweeping and improper applications' ") (citation omitted); NAACP v. Button, 371 U.S. 415, 433, 83 S.Ct. 328, 338, 9 L.Ed.2d 405 (1963) ("We read the decree of the Virginia Supreme Court of Appeals . . . as proscribing any arrangement by which prospective litigants are advised to seek the assistance of particular attorneys"); Thornhill v. Alabama, 310 U.S. 88, 97, 60 S.Ct. 736, 741, 84 L.Ed. 1093 (1940) (the statute "does not aim specifically at evils within the allowable area of state control but, on the contrary, sweeps within its ambit other activities that in ordinary circumstances constitute an exercise of freedom of speech or of the press"). 27 "A substantial overbreadth rule is implicit in the chilling effect rationale . . . the presumption must be that only substantially overbroad laws set up the kind and degree of chill that is judicially cognizable." Moreover, "[w]ithout a substantial overbreadth limitation, review for overbreadth would be draconian indeed. It is difficult to think of a law that is utterly devoid of potential for unconstitutionality in some conceivable application." Note, 83 Harv.L.Rev., supra n. 25, at 859, and n. 61. 28 There is no argument that the films sold by respondent do not fall squarely within the category of activity we have defined as unprotected. Therefore, no independent examination of the material is necessary to assure ourselves that the judgment here "does not constitute a forbidden intrusion on the field of free expression." New York Times Co. v. Sullivan, 376 U.S. 254, 285, 84 S.Ct. 710, 728, 11 L.Ed.2d 686 (1964). 1 Respondent's counsel conceded at oral argument that a finding that the films are obscene would have been consistent with the Miller definition. Tr. of Oral Arg. 41. 2 See Burch v. Louisiana, 441 U.S. 130, 139, 99 S.Ct. 1623, 1628, 60 L.Ed.2d 96 (STEVENS, J., concurring); Pinkus v. United States, 436 U.S. 293, 305, 98 S.Ct. 1808, 1815, 56 L.Ed.2d 293 (STEVENS, J., concurring); Ballew v. Georgia, 435 U.S. 223, 245, 98 S.Ct. 1029, 1041, 55 L.Ed.2d 234 (STEVENS, J., concurring); Smith v. United States, 431 U.S. 291, 311-321, 97 S.Ct. 1756, 1769-1774, 52 L.Ed.2d 324 (STEVENS, J., dissenting); Marks v. United States, 430 U.S. 188, 198, 97 S.Ct. 990, 996, 51 L.Ed.2d 260 (STEVENS, J., concurring in part and dissenting in part); see also Schad v. Borough of Mount Ephraim, 452 U.S. 61, 84, 101 S.Ct. 2176, 2190, 68 L.Ed.2d 671 (STEVENS, J., concurring in judgment); FCC v. Pacifica Foundation, 438 U.S. 726, 750, 98 S.Ct. 3026, 3040, 57 L.Ed.2d 1073 (opinion of STEVENS, J.). 3 "The test for child pornography is separate from the obscenity standard enunciated in Miller, but may be compared to it for the purpose of clarity. The Miller formulation is adjusted in the following respects: A trier of fact need not find that the material appeals to the prurient interest of the average person; it is not required that sexual conduct portrayed be done so in a patently offensive manner; and the material at issue need not be considered as a whole." Ante, at 764. 4 The Court's analysis is directed entirely at the permissibility of the statute's coverage of nonobscene material. Its empirical evidence, however, is drawn substantially from congressional Committee Reports that ultimately reached the conclusion that a prohibition against obscene child pornography coupled with sufficiently stiff sanctions—is an adequate response to this social problem. The Senate Committee on the Judiciary concluded that "virtually all of the materials that are normally considered child pornography are obscene under the current standards," and that "[i]n comparison with this blatant pornography, non-obscene materials that depict children are very few and very inconsequential." S.Rep.No. 95-438, p. 13 (1977), U.S.Code Cong. & Admin.News 1978, p. 50; see also H.R.Rep.No. 95-696, pp. 7-8 (1977). The coverage of the federal statute is limited to obscene material. See 18 U.S.C. § 2252(a) (1976 ed., Supp.IV). 5 See, e.g., Schad v. Borough of Mount Ephraim, 452 U.S., at 80, 83, 101 S.Ct., at 2189, 2190 (STEVENS, J., concurring in judgment); Consolidated Edison Co. v. Public Service Comm'n, 447 U.S. 530, 544-548, 100 S.Ct. 2326, 2337-2339, 65 L.Ed.2d 319 (STEVENS, J., concurring in judgment); FCC v. Pacifica Foundation, 438 U.S., at 744-748, 98 S.Ct., at 3037-3039 (opinion of STEVENS, J.); Carey v. Population Services International, 431 U.S. 678, 716-717, 97 S.Ct. 2010, 2032, 52 L.Ed.2d 675 (STEVENS, J., concurring in part and concurring in judgment); Smith v. United States, 431 U.S., at 317-319, 97 S.Ct., at 1772-1773 (STEVENS, J., dissenting); Young v. American Mini Theatres, Inc., 427 U.S. 50, 66-71, 96 S.Ct. 2440, 2450-2452, 49 L.Ed.2d 310 (opinion of STEVENS, J.). 6 See FCC v. Pacifica Foundation, supra, at 742-743, 98 S.Ct., at 3036-3037 (opinion of STEVENS, J.); Young v. American Mini Theatres, Inc., supra, at 59-61, 96 S.Ct., at 2446-2447; see also Metromedia, Inc. v. City of San Diego, 453 U.S. 490, 544-548, 101 S.Ct. 2882, 2911-2913, 69 L.Ed.2d 800 (STEVENS, J., dissenting in part); Schad v. Borough of Mount Ephraim, supra, at 85, 101 S.Ct., at 2191 (STEVENS, J., concurring in judgment).

23

459 U.S. 1 103 S.Ct. 250 74 L.Ed.2d 1 State of CALIFORNIA, ex rel. STATE LANDS COMMISSION, plaintiff,v.UNITED STATES of America No. 89 Supreme Court of the United States October 18, 1982 On cross-motions for entry of decree DECREE 1 For the purpose of giving effect to the decision of this Court herein, announced June 18, 1982, rehearing having been denied on September 9, 1982, IT IS ORDERED, ADJUDGED AND DECREED: 2 1. As against the State of California and all those claiming under it, the United States holds all right, title and interest in the parcel of land described in the Complaint filed herein and in Appendix A hereto, the seaward boundary of which shall be the line of mean high water of the Pacific Ocean, as heretofore or hereafter modified by accretion, erosion or reliction, whether attributable to natural or artificial causes; 3 2. The State of California has no right, title or interest in the said parcel of land, and the State, its agencies, political subdivisions, officers and agents, and all those claiming under them or in privity with them, are enjoined from interfering in any way with the right, title, and interest of the United States in the said parcel; 4 3. Each party shall bear its own costs. EXHIBIT A 5 All that certain real property in the State of California, County of Humboldt situated in Townships 4 and 5 North, Range 1 West, Humboldt Base & Meridian ("HB&M") and particularly described as follows: 6 COMMENCING at the east 1/4 corner of Section 31, Township 5 North, Range 1 West HB&M, thence from said point of commencement; N 88o 01'20 W, 1981.14 feet along the north line of U. S. Lot 3 of said Section 31, as said lot is shown on the official United States Government Township Plat, to the United States Meander Line of the Pacific Ocean as surveyed by J. S. Murray under contract dated October 18, 1854, and the TRUE POINT OF BEGINNING: thence from said true point of beginning southerly along the shore of the Pacific Ocean with the meander lines of said Section 31 the following (3) courses: 7 1. § 14o38'54 W, 395.44 feet; 2. § 03o38'54 W, 1863.84 feet; and 8 3. § 10o21'06 E, 400.10 feet; to the United States Meander Corner on the Township line common to said Townships 4 and 5 North Range 1 West; thence southerly along the shore of the Pacific Ocean with the meander lines of Section 6 of Township 4 North, Range 1 West as surveyed by J. H. Miller under contract dated October 19, 1854, the following (3) courses: 9 1. § 08o24'17 W, 968.24 feet; 2. § 01o24'17 W, 869.50 feet; and 10 3. § 11o35'43 E, 646.26 feet more or less to the centerline of the North Jetty at the entrance to Humboldt Bay; thence westerly along said centerline the following (6) courses: 1. N 75o15'58 W, 307.31 feet; 11 2. N 65o00'58 W, 431.97 feet; 12 3. N 52o05'24 W, 442.91 feet; 13 4. N 53o15'24 W, 408.72 feet; 14 5. N 50o02'05 W, 400.00 feet; 15 6. N 46o08'24 W, 1427 feet more or less to the line of mean high water of the Pacific Ocean; thence northerly along said line of mean high water to a point which bears N 88o 01' 20 W, from the true point of beginning; thence § 88o 01' 20 E, along the north line of U. S. Lot 3, of Section 31 of Township 5 North, Range 1 West produced, to the true point of beginning. 16 Bearings, and distances are based on the State of California Coordinate System (Lambert Conformal Projection), Zone 1, derived locally from that certain map entitled "Record of Survey, Surplus Property," recorded in Book 29 of Surveys at Page 137, Humboldt County Records as surveyed by the United States Coast Guard; 12th District.

910

74 L.Ed.2d 3 103 S.Ct. 276 459 U.S. 4 Charles E. ANDERSON, Wardenv.Jack E. HARLESS. No. 81-2066. Nov. 1, 1982. PER CURIAM. 1 Respondent was convicted of two counts of first degree murder and was sentenced to life imprisonment. The Michigan Court of Appeals affirmed respondent's conviction, People v. Harless, 78 Mich.App. 745, 261 N.W.2d 41 (1977), and the Michigan Supreme Court, on review of the record, denied respondent's request for relief. Petn for Cert 30a. 2 Respondent then filed a petition for writ of habeas corpus, pursuant to 28 U.S.C. s 2254, in the United States District Court for the Eastern District of Michigan. He alleged, inter alia, that the trial court's instruction on "malice"-a crucial element in distinguishing between second degree murder and manslaughter under Michigan law-was unconstitutional.1 In particular, respondent focused on the following language from the trial court's lengthy charge: 3 "Malice is implied from the nature of the act which caused the death. Malice can be implied from using the weapon on another person. You are not obligated to reach the conclusion, but you must imply malice if you find death was implied (sic) by the use of a gun against another." Petn for Cert 59a. 4 Relying primarily on Sandstrom v. Montana, 442 U.S. 510, 99 S.Ct. 2450, 61 L.Ed.2d 39 (1979), the District Court held that this instruction unconstitutionally shifted the burden of proof to respondent and was inconsistent with the presumption of innocence. Harless v. Anderson, 504 F.Supp. 1135 (1981). The court also held that respondent had exhausted available state-court remedies, as required by 28 U.S.C. §§ 2254(b) and (c), since his conviction had been reviewed by both the Michigan Court of Appeals and the Michigan Supreme Court. The District Court ordered that the application for writ of habeas corpus be granted unless respondent was retried within 90 days. 5 The United States Court of Appeals for the Sixth Circuit affirmed. Harless v. Anderson, 664 F.2d 610 (1982). The court held that respondent's claim had been properly exhausted in the state courts, because respondent had presented to the Michigan Court of Appeals the facts on which he based his federal claim and had argued that the malice instruction was "reversible error." See People v. Harless, 78 Mich.App., at 748, 261 N.W.2d, at 43. The court also emphasized that respondent, in his brief to the Michigan Court of Appeals, had cited People v. Martin, 392 Mich. 553, 221 N.W.2d 336 (1974)-a decision predicated solely on state law in which no federal issues were decided, but in which the defendant had argued broadly that failure to properly instruct a jury violates the Sixth and Fourteenth Amendments. In the view of the United States Court of Appeals, respondent's assertion before the Michigan Court of Appeals that the trial court's malice instruction was erroneous, coupled with his citation of People v. Martin, supra, provided the Michigan courts with sufficient opportunity to consider the issue encompassed by respondent's subsequent federal habeas petition. 6 We reverse. In Picard v. Connor, 404 U.S. 270, 92 S.Ct. 509, 30 L.Ed.2d 438 (1971), we made clear that 28 U.S.C. § 2254 requires a federal habeas petitioner to provide the state courts with a "fair opportunity" to apply controlling legal principles to the facts bearing upon his constitutional claim. Id., at 276-277, 92 S.Ct., at 512-513. It is not enough that all the facts necessary to support the federal claim were before the state courts, id., at 277, 92 S.Ct., at 513, or that a somewhat similar state-law claim was made. See, e.g., Gayle v. LeFevre, 613 F.2d 21 (CA2 1980); Paullet v. Howard, 634 F.2d 117, 119-120 (CA3 1980); Wilks v. Israel, 627 F.2d 32, 37-38 (CA7), cert. denied, 449 U.S. 1086, 101 S.Ct. 874, 66 L.Ed.2d 811 (1980); Connor v. Auger, 595 F.2d 407, 413 (CA8), cert. denied, 444 U.S. 851, 100 S.Ct. 104, 62 L.Ed.2d 67 (1979). In addition, the habeas petitioner must have "fairly presented" to the state courts the "substance" of his federal habeas corpus claim. Picard, supra, 404 U.S. at 275, 277-278, 92 S.Ct. at 512, 513-514. See also, Rose v. Lundy, —- U.S. ——, ——, 102 S.Ct. 1198, 1204, 71 L.Ed.2d 379 (1982). 7 In this case respondent argued on appeal that the trial court's instruction on the element of malice was "erroneous." He offered no support for this conclusion other than a citation to, and three excerpts from, People v. Martin, supra-a case which held that, under Michigan law, malice should not be implied from the fact that a weapon is used. See Petn for Cert 47a-49a, 51a-53a.2 Not surprisingly, the Michigan Court of Appeals interpreted respondent's claim as being predicated on the state-law rule of Martin, and analyzed it accordingly. 78 Mich.App., at 748-750, 261 N.W.2d, at 43. 8 The United States Court of Appeals concluded that "the due process ramifications" of respondent's argument to the Michigan court "were self-evident," and that respondent's "reliance on Martin was sufficient to present the state courts with the substance of his due process challenge to the malice instruction for habeas exhaustion purposes." 664 F.2d, at 612. We disagree. The District Court based its grant of habeas relief in this case on the doctrine that certain sorts of "mandatory presumptions" may undermine the prosecution's burden to prove guilt beyond a reasonable doubt and thus deprive a criminal defendant of due process. See Sandstrom, supra; In re Winship, 397 U.S. 358, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970). The Court of Appeals affirmed on the same rationale. However, it is plain from the record that this constitutional argument was never presented to, or considered by, the Michigan courts. Nor is this claim even the same as the constitutional claim advanced in Martin-the defendant there asserted a broad federal due process right to jury instructions that "properly explain" state law, 392 Mich., at 558, 221 N.W.2d, at 339, and did not rely on the more particular analysis developed in cases such as Sandstrom, supra.3 9 Since it appears that respondent is still free to present his Sandstrom claim to Michigan Court of Appeals, see People v. Berry, 10 Mich.App. 469, 474-475, 157 N.W.2d 310, 312-313 (1968), we conclude that he has not exhausted his available state-court remedies as required by 28 U.S.C. § 2254. Accordingly, the petition for certiorari and respondent's motion for leave to proceed in forma pauperis are granted, the judgment of the United States Court of Appeals for the Sixth Circuit is reversed, and the case is remanded to that court for further proceedings consistent with this opinion. 10 It is so ordered. 11 Justice STEVENS, with whom Justice BRENNAN, and Justice MARSHALL join, dissenting. 12 Few issues consume as much of the scarce time of federal judges as the question whether a state prisoner adequately exhausted his state remedies before filing a petition for a federal writ of habeas corpus. Distressingly, the Court seems oblivious of this fact and takes action in this case that can only exacerbate that problem. 13 On the merits the question presented by this case is whether a somewhat garbled jury instruction contained a mandatory presumption that required a finding of malice or merely a permissive inference that allowed the jury to make such a finding.1 The parties seem to agree that if the instruction is considered mandatory, the respondent's conviction must be set aside under principles that are well settled in Michigan2 and in the federal courts.3 14 The Michigan Court of Appeals rejected respondent's construction of the jury instruction and therefore affirmed his conviction. After the Supreme Court of Michigan denied leave to appeal and after the state trial court denied a subsequent motion for a new trial, respondent commenced this federal habeas corpus proceeding. 15 The Federal District Court carefully analyzed the difference between a permissive inference and a mandatory presumption and concluded that the Michigan Court of Appeals' construction of the jury instruction was simply untenable. Harless v. Anderson, 504 F.Supp. 1135, 1138 (1981). It also considered and rejected the argument that the respondent had not exhausted his state remedies. Id., at 1139. 16 On appeal, the Court of Appeals for the Sixth Circuit affirmed, 664 F.2d 610 (1982). That Court first carefully considered the warden's contention that respondent's state remedies had not been exhausted because his federal claim had not been fairly presented to the state courts. After explaining in some detail why the federal claim necessarily presented the very question that the state court had already resolved against the respondent, the Court of Appeals concluded "that the Michigan courts had a fair opportunity to consider the issue encompassed by Harless' habeas corpus petition." Id., at 611. Thereafter, the Court of Appeals reviewed and upheld the District Court's holding on the merits. 17 I agree with the sensible approach to the exhaustion issue that was followed by the District Court and the Court of Appeals.4 I also believe that approach was entirely faithful to Picard v. Connor, 404 U.S. 270, 92 S.Ct. 509, 30 L.Ed.2d 438, which requires only that the "substance" of the federal claim (not the form) be "fairly presented" to the state courts.5 In this case the only arguable justification for dismissing the petition for failure to exhaust is a possibility that the state court might decide the instruction issue differently if phrased in terms of Sandstrom v. United States rather than in terms of People v. Martin. See note 4, supra. That possibility is virtually nonexistent. The Court apparently perceives this case as a simple application of Picard; I think it can only be explained as an expansion of Picard. Such an expansion should be accompanied by a more careful analysis than the Court provides in this case, and it should not be undertaken without full briefing and argument. 18 But even if I shared the Court's analysis of the exhaustion question in this particular case, I would nevertheless take issue with its decision to grant certiorari for the sole purpose of correcting what it considers to be a technical, procedural error. It is not appropriate for this Court to expend its scarce resources crafting opinions that correct technical errors in cases of only local importance where the correction in no way promotes the development of the law.6 19 If the Court of Appeals was correct in its appraisal of the merits, the respondent should be given a prompt retrial. If that court was incorrect on the merits, nothing will be gained by requiring the respondent to present his claim to three more sets of Michigan judges and two more sets of federal judges before this Court decides whether the substantive error merits our review. Cf. Rose v. Lundy, —- U.S. ——, ——, 102 S.Ct. 1198, 1217, 71 L.Ed.2d 379 (1982) (STEVENS, J., dissenting). 20 I respectfully dissent. 1 Respondent was convicted of first degree murder, which requires not only proof of "malice" but also of premeditation. For this reason, petitioner argues that any error in the trial court's definition of malice was harmless. In light of our disposition, we do not reach the issue. 2 Respondent was represented by counsel on appeal. 3 We doubt that a defendant's citation to a state-court decision predicated solely on state law ordinarily will be sufficient to fairly apprise a reviewing court of a potential federal claim merely because the defendant in the cited case advanced a federal claim. However, it is clear that such a citation is insufficient when, as here, the federal claim asserted in the cited case is not even the same as the federal claim on which federal habeas relief is sought. See Picard, supra, 404 U.S. at 276, 92 S.Ct. at 512. 1 The instruction stated: "Members of the jury, the term malice is a technical term which has to do with the doing of a cruel act against another human being without excuse or justification. The doing of a cruel act against another human being without excuse or justification. Malice is implied from the nature of the act which caused the death. Malice can be implied from using the weapon on another person. You are not obligated to reach the conclusion, but you must imply malice if you find death was implied by the use of a gun against another." Pet. for Cert., at 59a. 2 See People v. Martin, 392 Mich. 553, 221 N.W.2d 336 (1974). 3 Compare Ulster County Court v. Allen, 442 U.S. 140, 99 S.Ct. 2213, 60 L.Ed.2d 777 with Sandstrom v. Montana, 442 U.S. 510, 99 S.Ct. 2450, 61 L.Ed.2d 39. 4 Because this Court's description of the Court of Appeals' treatment of the exhaustion issue is so abbreviated it seems appropriate to quote in full the relevant portion of that opinion: "In our view, Harless adequately exhausted available state remedies for purposes of 28 U.S.C. §§ 2254(b) and (c). The respondent concedes that Harless presented to the state appellate courts all the facts on which he based his federal constitutional claim. Respondent contends, however, that the state courts had no opportunity to correct the constitutional error, because Harless did not explicitly complain to the state courts that the malice instruction denied him due process. "Although we do not have before us Harless' state appellate papers, we learn from the Michigan Court of Appeals opinion the substance of Harless' contention before that court: 'The trial court committed reversible error by instructing the jury incorrectly on the implication (sic) of malice that might be drawn from defendant's use of a deadly weapon, the effect of which was to remove the possible finding of manslaughter." People v. Harless, 78 Mich.App. 745, 261 N.W.2d 41, 43 (1977) (emphasis added). At Harless' trial the court gave the following definition of malice to the jury: "Malice is implied from the nature of the act which caused the death. Malice can be implied from using the weapon on another person. You are not obligated to reach the conclusion, but you must imply malice if you find death was implied (sic) by the use of a gun against another. (emphasis added). "Harless claimed on appeal that this instruction was reversible error under People v. Martin, 392 Mich. 553, 221 N.W.2d 336 (1974), a case holding that the law does not imply malice from the use of a deadly weapon. In Martin, appellant challenged his murder conviction on numerous state and constitutional grounds. The gist of Martin's appeal was that he was denied a fair trial because certain instructions failed to provide the jury with sufficient understanding of the elements of the crimes charged, to enable them to perceive the crucial distinctions between first and second degree murder, and manslaughter. In particular, Martin challenged a malice instruction in which the jury was informed that the law implies malice from the use of a deadly weapon. The Court of Appeals ultimately decided as a matter of state law that the malice instruction 'erroneously categorized (the issue of malice) as a presumption of law rather than a permissible inference.' 221 N.W.2d, at 341. However, Martin's constitutional argument was broadly phrased: failure to explain the law properly to a jury through adequate instructions abridges the due process right to a fair trial. The Court of Appeals neither rejected nor refined this constitutional argument. Rather, it seems to have accepted it tacitly, for in response to defendant's argument that his Sixth and Fourteenth Amendment rights to a fair trial had been abridged by an inadequate manslaughter instruction, the court stated that 'an erroneous or misleading charge denies the defendant the right to have a properly instructed jury pass upon the evidence.' Id., at 341. Although it did not specifically label this a federal right, the court clearly felt that a properly instructed jury is a prerequisite to a fair trial. "In our view, Harless' reliance on Martin was sufficient to present the state courts with the substance of his due process challenge to the malice instruction for habeas exhaustion purposes. The substance of Harless' state appeal, although unartfully phrased, sufficiently asked the state court to consider that the incorrect malice instruction denied Harless a fair jury trial by effectively eliminating the possibility of a manslaughter verdict from the jury's consideration. In our view, the due process ramifications were self-evident. Under Picard v. Connor, 404 U.S. 270, 92 S.Ct. 509, 30 L.Ed.2d 438 (1971), a habeas petitioner need not label his state claim as federal or constitutional. Given the elasticity of the due process concept, we are convinced that the Michigan courts had a fair opportunity to consider the issue encompassed by Harless' habeas corpus petition." 664 F.2d, at 611-612. 5 In Picard, the habeas petitioner had argued in the Massachusetts courts that his indictment had been contrary to Massachusetts law and that, if the procedures used in his case had been in accord with Massachusetts law, then those procedures could not be approved without reference to whether the Fifth Amendment's requirement of a grand jury indictment applied to the States through incorporation in the Fourteenth Amendment. 404 U.S., at 277, 92 S.Ct., at 513. This Court held that such an argument did not give the state a fair opportunity to consider the proposition that its treatment of the petitioner denied him equal protection, in view of the Commonwealth's otherwise universal commitment to grand jury indictments in felony cases. In this case, the federal courts ordered habeas relief on the theory that the "operative effect" of the instruction quoted above was to cause the jury to use an unconstitutional mandatory presumption of malice. It held that the instruction was therefore inconsistent with Sandstrom, supra. 664 F.2d, at 612. Accord, majority opinion, supra, at 3. In state court, the defendant had argued that the instruction was inconsistent with People v. Martin, supra, because Martin had struck down an instruction that caused the jury to use a mandatory presumption of malice. See People v. Harless, 78 Mich.App. 745, 749, 261 N.W.2d 41 (1977). The substance of the argument-that the instruction deprived him of fair jury consideration because it created a mandatory presumption of malice-is the same. 6 In many respects this case is merely a sequel to Board of Education of Rogers, Arkansas v. McCluskey, —- U.S. ——, 102 S.Ct. 3469, 73 L.Ed.2d 1273 (1982). In my dissent in that case, I observed: "As Justice REHNQUIST has reminded us, in 'our zeal to provide "equal justice under law," we must never forget that this Court is not a forum for the correction of errors.' Boag v. MacDougall, —- U.S. ——, ——, 102 S.Ct. 700, 702, 70 L.Ed.2d 551 (dissenting opinion). 'To remain effective, the Supreme Court must continue to decide only those cases which present questions whose resolution will have immediate importance far beyond the particular facts and parties involved.' This case illustrates how ineffectively the Court is supervising its discretionary docket." Id., —- U.S. at ——, 102 S.Ct., at 3472 (STEVENS, J., dissenting) (footnote omitted).

01

74 L.Ed.2d 12 103 S.Ct. 281 459 U.S. 14 IMMIGRATION AND NATURALIZATION SERVICEv.Horacio Ramos MIRANDA. No. 82-29. Nov. 8, 1982. PER CURIAM. 1 Respondent Horacio Miranda, a citizen of the Philippines, entered the United States in 1971 on a temporary visitor's visa. After his visa expired, he stayed in this country, eventually marrying Linda Milligan, a citizen of the United States, on May 26, 1976. Shortly thereafter, Milligan filed a visa petition with the Immigration and Naturalization Service (INS) on respondent's behalf. She requested that he be granted an immigrant visa as her spouse.1 Respondent simultaneously filed an application requesting the INS to adjust his status to that of a permanent resident alien. Section 245(a) of the Immigration and Nationality Act conditions the granting of permanent resident status to an alien on the immediate availability of an immigrant visa.2 Milligan's petition, if approved, would have satisfied this condition. 2 The INS did not act on either Milligan's petition or respondent's application for eighteen months. Following the breakup of her marriage with respondent, Milligan withdrew her petition in December 1977. At that point, the INS denied respondent's application for permanent residence because he had not shown that an immigrant visa was immediately available to him. The INS also issued an order to show cause why he should not be deported. 3 At a deportation hearing, respondent conceded his deportability but renewed his application for permanent resident status because of his marriage to Milligan. Although the marriage had ended, he claimed that a previous marriage was sufficient to support his application. The immigration judge rejected this claim, concluding that the immediate availability of an immigrant visa was a necessary condition to respondent's application. Since Milligan had withdrawn her petition for an immigrant visa before the INS had acted on it, respondent was ineligible for permanent resident status. 4 Respondent appealed the decision to the Board of Immigration Appeals. For the first time, he raised the claim that the INS was estopped from denying his application because of its "unreasonable delay." He argued that the "failure to act was not only unreasonable, unfair and unjust but also an abuse of governmental process if the delay was deliberate." Record 44. The Board rejected respondent's claim. It found "no evidence of any 'affirmative misconduct' " and no basis for an equitable estoppel. Record 4. 5 Respondent sought review of the Board's decision in the Court of Appeals for the Ninth Circuit. The Court of Appeals reversed, holding that "(t)he unexplained failure of the INS to act on the visa petition for an eighteen-month period prior to the petitioner's withdrawal ... was affirmative misconduct by the INS." Miranda v. INS, 638 F.2d 83, 84 (CA9 1980). We granted certiorari, vacated the judgment of the Court of Appeals, and remanded the case for further consideration in light of Schweiker v. Hansen, 450 U.S. 785, 101 S.Ct. 1468, 67 L.Ed.2d 685 (1981). See 454 U.S. 808, 102 S.Ct. 81, 70 L.Ed.2d 77 (1981). 6 On remand, the Court of Appeals adhered to its earlier decision. 673 F.2d 1105 (CA9 1982) (per curiam). It found Hansen inapplicable for three reasons. First, the Government's conduct in Hansen had not risen to the level of affirmative misconduct. In this case, however, affirmative misconduct was established by the INS's unexplained delay in processing respondent's application. Second, although the private party in Hansen subsequently had been able to correct the Government's error, the INS's error here inflicted irrevocable harm on respondent. Finally, unlike the private party in Hansen who sought to recover from the public treasury, respondent was seeking only to become a permanent resident-a result that would entail no burden on the public fisc. The Court of Appeals determined "that the Supreme Court's conclusion that the government was not estopped in Hansen neither compels nor suggests the same conclusion here." Id., at 1106. 7 In Hansen, we did not consider whether estoppel will lie against the Government when there is evidence of affirmative misconduct. We found that a government official's misstatement to an applicant for federal insurance benefits, conceded to be less than affirmative misconduct, did not justify allowing the applicant to collect retroactive benefits from the public treasury. See 450 U.S., at 788-789, 101 S.Ct., at 1470-1471. Although Hansen involved estoppel in the context of a claim against the public treasury, we observed that "(i)n two cases involving denial of citizenship, the Court has declined to decide whether even 'affirmative misconduct' would estop the Government from denying citizenship, for in neither case was 'affirmative misconduct' involved." Id., at 788, 101 S.Ct., at 1470. 8 The Court of Appeals thus correctly considered whether, as an initial matter, there was a showing of affirmative misconduct. See INS v. Hibi, 414 U.S. 5, 8-9, 94 S.Ct. 19, 21-22, 38 L.Ed.2d 7 (1973) (per curiam); Montana v. Kennedy, 366 U.S. 308, 314-315, 81 S.Ct. 1336, 1340-1341, 6 L.Ed.2d 313 (1961). Hibi and Montana indicate, however, that the Court of Appeals erred in determining that the evidence in this case established affirmative misconduct. In Montana, a government official had incorrectly informed the petitioner's mother that she was unable to return to the United States because she was pregnant. The Court found that the official's misstatement "falls far short of misconduct such as might prevent the United States from relying on petitioner's foreign birth" as a basis for denying him citizenship. Id., at 314-315, 81 S.Ct., at 1340-1341. In Hibi, Congress had exempted aliens serving in the United States Armed Forces from certain requirements normally imposed on persons seeking naturalization. We found that neither the Government's failure to publicize fully the rights accorded by Congress nor its failure to make an authorized naturalization representative available to aliens serving outside of the United States estopped the Government from rejecting respondent's untimely application for naturalization. See 414 U.S., at 8-9, 94 S.Ct., at 21-22. 9 Unlike Montana and Hibi, where the Government's error was clear, the evidence that the Government failed to fulfill its duty in this case is at best questionable. The only indication of negligence is the length of time that the INS took to process respondent's application. Although the time was indeed long, we cannot say in the absence of evidence to the contrary that the delay was unwarranted.3 Cf. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 415, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971) (presumption of regularity supports official act of public officer); United States v. Chemical Foundation, 272 U.S. 1, 14-15, 47 S.Ct. 1, 6, 71 L.Ed. 131 (1926) (same). Both the number of the applications received by the INS and the need to investigate their validity may make it difficult for the agency to process an application as promptly as may be desirable.4 Even if the INS arguably was negligent in not acting more expeditiously, its conduct was not significantly different from that in Montana and Hibi. Nor is the harm to respondent different. Montana and Hibi make clear that neither the Government's conduct nor the harm to the respondent is sufficient to estop the Government from enforcing the conditions imposed by Congress for residency in this country. 10 The final distinction drawn by the Court of Appeals between this case and Hansen is unpersuasive. It is true that Hansen relied on a line of cases involving claims against the public treasury. But there was no indication that the Government would be estopped in the absence of the potential burden on the fisc. An increasingly important interest, implicating matters of broad public concern, is involved in cases of this kind. Enforcing the immigration laws, and the conditions for residency in this country, is becoming more difficult. See note 4, supra. Moreover, the INS is the agency primarily charged by Congress to implement the public policy underlying these laws. See, e.g., INS v. Wang, 450 U.S. 139, 144-145, 101 S.Ct. 1027, 1031, 67 L.Ed.2d 123 (1981) (per curiam); Hibi, supra, 414 U.S., at 8, 94 S.Ct., at 21. Appropriate deference must be accorded its decisions. 11 This case does not require us to reach the question we reserved in Hibi, whether affirmative misconduct in a particular case would estop the Government from enforcing the immigration laws. Proof only that the Government failed to process promptly an application falls far short of establishing such conduct. Accordingly, we grant the petition for certiorari and reverse the judgment of the Court of Appeals. 12 It is so ordered. 13 Justice MARSHALL, dissenting. 14 I dissent from the Court's summary reversal of the Court of Appeals. The Court concedes that the INS's 18-month delay in processing respondent's application "was indeed long," but concludes that it "cannot say in the absence of evidence to the contrary that the delay was unwarranted." Ante, at 18. The Court relies on a presumption of regularity which it says attends the official acts of public officers. Ibid. In view of the unusual delay in the processing of respondent's application, I do not agree that this case should be summarily disposed of on the basis of this convenient presumption. If the Court believes, as I do not, that this case raises an issue of sufficient importance to justify the exercise of our certiorari jurisdiction, and if the Court also believes that oral argument should be dispensed with, I would at least notify the parties that the Court is considering a summary disposition, so that they may have an opportunity to submit briefs on the merits. 1 Section 201(b) of the Immigration and Nationality Act of 1952, as amended, provides for the admission of immigrants who are immediate relatives of United States citizens. See 8 U.S.C. § 1151(b). 2 Section 245(a) provides that the status of an alien who was admitted into the United States "may be adjusted by the Attorney General, in his discretion and under such regulations as he may prescribe, to that of an alien lawfully admitted for permanent residence if (1) the alien makes an application for such adjustment, (2) the alien is eligible to receive an immigrant visa and is admissible to the United States for permanent residence, and (3) an immigrant visa is immediately available to him at the time his application is filed." See 8 U.S.C. § 1255(a). 3 The INS has maintained consistently that the eighteen-month delay was reasonable because of the need to investigate the validity of respondent's marriage. Because the issue of estoppel was raised initially on appeal, the parties were unable to develop any factual record on the issue. 4 In 1976, the year in which Milligan filed her petition on behalf of respondent, some 206,319 immediate relative petitions were filed. See 1976 Annual Report: Immigration and Naturalization Service 11. The Service has noted that "(i)n dealing with these petitions, an inordinate amount of fraud, particularly in relation to claimed marriages, has been uncovered.... For a fee, partners are provided and marriages contracted to establish eligibility under the statutes for visa issuance benefits." Ibid. We cannot discount the need for careful investigation by the INS that these petitions demand.

12

459 U.S. 56 103 S.Ct. 400 74 L.Ed.2d 225 Robert C. GRIGGS and Jacqueline M. Griggsv.PROVIDENT CONSUMER DISCOUNT COMPANY. No. 82-5082. Nov. 29, 1982. PER CURIAM. 1 The petition for certiorari questions the validity of a notice of appeal filed after the entry of the District Court's judgment but while the appellant's motion to alter or amend that judgment remained pending in the District Court. 2 The petitioners brought this civil action in the United States District Court for the Eastern District of Pennsylvania, seeking statutory damages for an alleged violation of the Truth in Lending Act, 15 U.S.C. § 1601 et seq., and Regulation Z of the Federal Reserve Board, 12 CFR § 226.1 et seq. On December 24, 1980, the District Court granted the petitioners' motion for summary judgment, finding that the respondent's disclosure of its security interests in after-acquired property had been inaccurate and misleading. Griggs v. Provident Consumer Discount Co., 503 F.Supp. 246 (E.D.Pa.1980). On November 5, 1981, the District Court entered an order pursuant to Fed.Rule Civ.Proc. 54(b) directing that a final judgment be entered. On November 12, the respondent filed a timely motion to alter or amend the judgment, pursuant to Fed.Rule Civ.Proc. 59. On November 19, while that motion was still pending, the respondent filed a notice of appeal. On November 23, the District Court denied the motion to alter or amend the judgment. Neither the opinion below nor the response to the petition for a writ of certiorari indicates that any further notice of appeal was filed. 3 The United States Court of Appeals for the Third Circuit accepted jurisdiction of the appeal and reversed the District Court's judgment. Griggs v. Provident Consumer Discount Co., 680 F.2d 927 (CA3 1982). The Court of Appeals explained its decision to take jurisdiction as follows: 4 "The Griggses urge that this matter is not appealable because Rule 4(a)(4) of the Federal Rules of Appellate Procedure provides that '[a] notice of appeal filed before the disposition of any of the above motions shall have no effect.' Appellant did fail to satisfy Rule 4(a)(4) but though a premature notice of appeal is subject to dismissal, we have generally allowed appellant to proceed unless the appellee can show prejudice resulting from the premature filing of the notice. Tose v. First Pennsylvania Bank, N.A., 648 F.2d 879, 882 n. 2 (3d Cir.), cert. denied 454 U.S. 893, 102 S.Ct. 390, 70 L.Ed.2d 208 (1981); Hodge v. Hodge, 507 F.2d 87, 89 (3d Cir.1975); accord Williams v. Town of Okoboji, 599 F.2d 238 (8th Cir.1979). See also 9 Moore's Federal Practice ¶ 204.14 (2d ed. 1982). In our case, the Griggses have shown no prejudice by the premature filing of a notice of appeal." Id., at 929 n. 2. 5 Because this analysis of Rule 4(a)(4) conflicts with the decisions of other Courts of Appeals1 and is contrary to the language and purposes of the 1979 Amendments to the Federal Rules of Appellate Procedure, we grant the petitioners' request for leave to proceed in forma pauperis and their petition for a writ of certiorari, and we reverse. 6 Even before 1979, it was generally understood that a federal district court and a federal court of appeals should not attempt to assert jurisdiction over a case simultaneously. The filing of a notice of appeal is an event of jurisdictional significance—it confers jurisdiction on the court of appeals and divests the district court of its control over those aspects of the case involved in the appeal. See, e.g., United States v. Hitchmon, 587 F.2d 1357 (CA5 1979). Cf. Ruby v. Secretary of United States Navy, 365 F.2d 385, 389 (CA9 1966) (en banc) (notice of appeal from unappealable order does not divest district court of jurisdiction), cert. denied, 386 U.S. 1011, 87 S.Ct. 1357, 18 L.Ed.2d 440 (1967). Under pre-1979 procedures, a district court lacked jurisdiction to entertain a motion to vacate, alter, or amend a judgment after a notice of appeal was filed. See Hattersley v. Bollt, 512 F.2d 209 (CA3 1975); Edmond v. Moore-McCormack Lines, 253 F.2d 143 (CA2 1958). However, if the timing was reversed—if the notice of appeal was filed after the motion to vacate, alter, or amend the judgment—two seemingly inconsistent conclusions were generally held to follow: the district court retained jurisdiction to decide the motion, but the notice of appeal was nonetheless considered adequate for purposes of beginning the appeals process. E.g., Yaretsky v. Blum, 592 F.2d 65, 66 (CA2 1979), cert. denied, 450 U.S. 925, 101 S.Ct. 1379, 67 L.Ed.2d 355 (1981); Williams v. Town of Okoboji, 599 F.2d 238 (CA8 1979); Alexander v. Aero Lodge No. 735, 565 F.2d 1364, 1371 (CA6 1977), cert. denied, 436 U.S. 946, 98 S.Ct. 2849, 56 L.Ed.2d 787 (1978); Dougherty v. Harper's Magazine Co., 537 F.2d 758, 762 (CA3 1976); Stokes v. Peyton's Inc., 508 F.2d 1287 (CA5 1975); Song Jook Suh v. Rosenberg, 437 F.2d 1098 (CA9 1971). Cf. Foman v. Davis, 371 U.S. 178, 83 S.Ct. 227, 9 L.Ed.2d 222 (1962). But see Century Laminating, Ltd. v. Montgomery, 595 F.2d 563 (CA10), cert. granted, 444 U.S. 897, 100 S.Ct. 204, 62 L.Ed.2d 132, cert. dismissed, 444 U.S. 987, 100 S.Ct. 516, 62 L.Ed.2d 417 (1979). The reason this theoretical inconsistency was tolerable in practice was that the district courts did not automatically inform the courts of appeals when a notice of appeal had been filed, and there was therefore little danger a district court and a court of appeals would be simultaneously analyzing the same judgment. 7 In 1979, the Rules were amended to clarify both the litigants' timetable and the courts' respective jurisdictions. The new requirement that a district court "transmit forthwith" any valid notice of appeal to the court of appeals advanced the time when that court could begin processing an appeal. Fed.Rule App.Proc. 3(d). At the same time, in order to prevent unnecessary appellate review, the district court was given express authority to entertain a timely motion to alter or amend the judgment under Rule 59, even after a notice of appeal had been filed. Fed.Rule App.Proc. 4(a)(4). If these had been the only changes, the theoretical inconsistency noted above would have suddenly taken on practical significance. A broad class of situations would have been created in which district courts and courts of appeals would both have had the power to modify the same judgment. The 1979 Amendments avoided that potential conflict by depriving the courts of appeals of jurisdiction in such situations. New Rule 4(a)(4) states:2 8 "If a timely motion under the Federal Rules of Civil Procedure is filed in the District Court by any party . . . under Rule 59 . . ., the time for appeal for all parties shall run from the entry of the order denying . . . such motion. A notice of appeal filed before the disposition of [such motion] shall have no effect. A new notice of appeal must be filed within the prescribed time measured from the entry of the order disposing of the motion as provided above. No additional fees shall be required for such filing." Professor Moore has aptly described the post-1979 effect of a Rule 59 motion on a previously filed notice of appeal: "The appeal simply self-destructs." 9 J. Moore, B. Ward & J. Lucas, Moore's Federal Practice ¶ 204.12[1], at 4-65 n. 17 (2d ed. 1982). Moreover, a subsequent notice of appeal is also ineffective if it is filed while a timely Rule 59 motion is still pending. See 16 C. Wright, A. Miller, E. Cooper & E. Gressman, Federal Practice and Procedure § 3950 (1982 Supp.). 9 The United States Court of Appeals for the Third Circuit has taken the position that, notwithstanding the 1979 Amendments, it retains discretion under Fed.Rule App.Proc. 2 to waive the conceded defects in a premature notice of appeal. Tose v. First Pennsylvania Bank, N.A., 648 F.2d 879, 882 n. 2. (CA3), cert. denied, 454 U.S. 893, 102 S.Ct. 390, 70 L.Ed.2d 208 (1981). We disagree. The notice of appeal filed in this case on November 19, 1980, was not merely defective; it was a nullity. Under the plain language of the current rule, a premature notice of appeal "shall have no effect"; a new notice of appeal "must be filed." In short, it is as if no notice of appeal were filed at all. And if no notice of appeal is filed at all, the Court of Appeals lacks jurisdiction to act. It is well settled that the requirement of a timely notice of appeal is "mandatory and jurisdictional." Browder v. Director, Illinois Dept. of Corrections, 434 U.S. 257, 264, 98 S.Ct. 556, 560, 54 L.Ed.2d 521 (1978).3 10 The motion of petitioners for leave to proceed in forma pauperis and the petition for a writ of certiorari are granted. The judgment is vacated and the case is remanded for further proceedings consistent with this opinion. 11 Justice MARSHALL, dissenting. 12 Without the benefit of briefing or argument on the merits, the majority—in a conclusory footnote—decides that a Court of Appeals cannot invoke Rule 2 of the Federal Rules of Appellate Procedure to waive a defect in a notice of appeal. The Court's exercise of its majestic power to decide this question is inappropriate in this case because an alternative ground for the lower court's disposition exists: respondent in fact filed an effective notice of appeal following the denial of its motion to amend the District Court's judgment.1 In any event, the majority's interpretation of Rule 2 is inconsistent with the language of the Rule and with prior Court decisions, and the decision may have grave consequences for pro se litigants. At a minimum, the Court should allow the parties an opportunity to address these issues in a brief on the merits. I respectfully dissent. 13 * While the majority describes respondent's filing of a premature notice of appeal, it fails to mention the subsequent actions taken by respondent in the Court of Appeals following the District Court's denial of the Fed.Rule Civ.Proc. 59 motion on November 23, 1981. Respondent's actions within 30 days of November 23 amply satisfied the content requirements of Fed.Rule App.Proc. 3(c). 14 On December 4, the Court of Appeals docketed the appeal and the record from the District Court was filed. That same day, the clerk for the Court of Appeals sent a letter to respondent's counsel with a copy to petitioners' counsel notifying them that the case had been docketed and the record filed. The clerk's letter noted that a brief on the merits of the appeal had already been filed by respondent, due to a prior misunderstanding.2 The clerk asked respondent's counsel to advise the Court "in writing if it is your intention to rely on the briefs previously filed." See Petn., Appendix C. 15 On December 12, respondent sent two letters to the Court of Appeals, both of which were received on December 15.3 The first letter stated that respondent intended to file a new brief in the docketed case but would rely on the same appendix that had previously been filed. The letter also included a disclosure statement in order to comply with a local Third Circuit rule. The second letter provided, in accordance with Fed.Rule App.Proc. 30(b), a statement of the issues which respondent intended to present for review to the Court of Appeals and also a designation of the portions of the appendix on which respondent would rely. Copies of both of these letters were served on counsel for petitioners. 16 Under the circumstances of this case, viewed in their entirety, respondent clearly filed a timely notice of appeal as defined by Rule 3(c). That rule was amended in 1979 to provide that "[a]n appeal shall not be dismissed for informality of form or title of the notice of appeal." The Advisory Committee Notes explain the significance of the amendment as follows: 17 "[I]t is important that the right to appeal not be lost by mistakes of mere form. In a number of decided cases it has been held that so long as the function of notice is met by the filing of a paper indicating an intention to appeal, the substance of the rule has been complied with. See, e.g., Cobb v. Lewis (CA5th, 1974) 488 F.2d 41; Holley v. Capps (CA5th, 1972) 468 F.2d 1366. The proposed amendment would give recognition to this practice." (emphasis added) 18 The Cobb case cited by the Advisory Committee is particularly instructive. There, the Court of Appeals concluded that "the notice of appeal requirement may be satisfied by any statement, made either to the District Court or to the Court of Appeals, that clearly evinces the party's intent to appeal." 488 F.2d, at 45 (emphasis added). The court reasoned that such a statement "accomplishes the two basic objectives of the Rule 3 notice requirement: (1) to notify the Court of the taking of an appeal; and (2) to notify the opposing party of the taking of an appeal." Ibid. 19 The actions undertaken by respondent during the 30 days after November 23 amply satisfied the Rule's requirement of notice to the Court of Appeals4 and to the opposing party. Within 30 days after November 23, 1981, the Court of Appeals had before it the record of the case, respondent's previously filed brief on the merits, a letter from respondent indicating its intention to file a new brief on the merits and also containing a disclosure statement, and a letter from respondent stating precisely those issues which were to be raised on appeal and also providing designations of the portions of the previously filed appendix upon which respondent would rely. Similarly, petitioners had received a notice from the Court of Appeals that the case had been docketed and the record filed, and they had received from respondent the letters sent to the Court of Appeals, which included a Rule 30(b) statement of the issues to be presented. 20 The specific actions taken by respondent after November 23 provided adequate notice of its intent to appeal. Any other conclusion would exalt empty form and ritual over common sense. As the court stated in Cobb v. Lewis, supra, a decision upon which the Advisory Committee relied in amending Rule 3(c), " 'it would we think be a harking back to formalistic rigorism of an earlier and outmoded time, as well as a travesty upon justice, to hold the extremely simple procedure required by the Rule is itself a kind of Mumbo Jumbo, and that the failure to comply formalistically with it defeats substantial rights.' " 488 F.2d, at 45, quoting Crump v. Hill, 104 F.2d 36, 38 (CA5 1939). 21 Because respondent filed an effective notice of appeal, the Court of Appeals was compelled to reach the merits of the appeal. The lower court's interpretation of its discretionary authority under Rule 2 of the Federal Rules of Appellate Procedure was thus unnecessary to the proper disposition of respondent's appeal. Consequently, I do not think this case is an appropriate vehicle for making new procedural law. II 22 Even if this case warranted review, I would decline to join the majority in summarily rejecting the basis provided by the Court of Appeals for its decision to reach the merits of respondent's appeal. The court relied on Rule 2 of the Federal Rules of Appellate Procedure, which provides that for good cause "a court of appeals may, except as otherwise provided in Rule 26(b), suspend the requirements or provisions of any of these rules in a particular case . . . on its own motion . . . ." According to the Advisory Committee notes, the Rule "contains a general authorization to the courts to relieve litigants of the consequences of default where manifest injustice would otherwise result." 23 Invoking its discretionary authority under Rule 2, the Third Circuit declines to dismiss appeals based on Rule 4(a)(4) defaults in the absence of a showing of prejudice to the appellee. See Tose v. First Pennsylvania Bank, N.A., 648 F.2d 879, 882 n. 2 (CA3), cert. denied, 454 U.S. 893, 102 S.Ct. 390, 70 L.Ed.2d 208 (1981); Hodge v. Hodge, 507 F.2d 87, 89 (CA3 1979), cited in Griggs v. Provident Consumer Discount Co., 680 F.2d 927, 929 n. 2 (CA3 1982). On this ground, the Court of Appeals exercised its discretion in this case after concluding that petitioners had failed to show any prejudice.5 24 In a two-sentence footnote rejecting the lower court's interpretation of Rule 2, the majority notes only that the discretion granted in Rule 2 is explicitly limited by Fed.Rule App.Proc. 26(b), which states that a court of appeals "may not enlarge the time for filing a notice of appeal." The majority does not explain the relevance of Rule 26(b) to this case. The common-sense meaning of the Rule is that a court may not recognize a late notice of appeal. See United States v. Robinson, 361 U.S. 220, 224, 80 S.Ct. 282, 285, 4 L.Ed.2d 259 (1960). Rule 26 by its very title deals with an extension of time; in the words of the Advisory Committee Notes to Rule 2, "Rule 26(b) prohibits a court of appeals from extending the time for taking appeal or seeking review" (emphasis added). In similar fashion, the provisions of Fed.Rule Civ.Proc. 6, on which Rule 26 is based,6 discuss enlargement in terms of extending the expiration date of a period. In short, there is little question that a court of appeals may not—consistent with the mandate of Rule 26(b)—give effect to a late notice of appeal. But it is certainly debatable whether Rule 26(b) prohibits the recognition of a premature notice of appeal. Only Rule 4(a)(4) explicitly bars such recognition, but Rule 4(a)(4) does not serve as an express limitation on Rule 2. 25 The Court concludes that, because of respondent's failure to refile the same notice of appeal filed four days prematurely, the Court of Appeals was absolutely barred from addressing the merits of its appeal. This conclusion flies in the face of our previous declaration that it is "too late in the day and entirely contrary to the spirit of the Federal Rules of Civil Procedure for decisions on the merits to be avoided on the basis of such mere technicalities." Foman v. Davis, 371 U.S. 178, 181, 83 S.Ct. 227, 229, 9 L.Ed.2d 222 (1962) (discussing a notice of appeal under Fed.Rule Civ.Proc. 73(a), the predecessor of Fed.Rule App.Proc. (4). See also Bankers Trust Co. v. Mallis, 435 U.S. 381, 387, 98 S.Ct. 1117, 1121, 55 L.Ed.2d 357 (1978) (per curiam ) ("the technical requirements [imposed by the Rules of Appellate Procedure] for a notice of appeal were not mandatory where the notice 'did not mislead or prejudice' "). 26 The Court's interpretation of Rule 4(a)(4) also creates new and serious pitfalls for pro se and other unsophisticated litigants. The reports are filled with cases in which litigants filed post-judgment motions to "reconsider," to "vacate," to "set aside," or to "reargue" adverse judgments. The lower courts have almost without exception treated these as Rule 59 motions, regardless of their label.7 Indeed, even motions captioned under Rule 60(b), but filed within 10 days of judgment, are normally deemed Rule 59 motions.8 According to the majority, a notice of appeal becomes a "nullity" if it is filed while a Rule 59 motion is pending. Thus, under the majority's approach, litigants could unwittingly file invalid notices of appeal simply because they had previously filed a motion questioning a District Court judgment which, unbeknownst to them, is a Rule 59 motion. The mere failure to appreciate the distinction between a Rule 59 motion and a Rule 60(b) motion, when combined with the draconian application of Rule 4(a)(4) adopted by the majority, would require the dismissal of an appeal. See, e.g., Apel v. Wainwright, 677 F.2d 116 (CA11, 1982) (on petition for rehearing), petition for cert. pending, No. 82-5503. III 27 If the Court believes, as I do not, that it is necessary in this case to examine the Court of Appeals' interpretation of Rule 2, I would at least notify the parties that the Court is considering a summary disposition, so that they may have an opportunity to submit briefs on the merits. Without such briefing, the risk of error necessarily increases. I therefore dissent. 1 See United States v. Valdosta-Lowndes County Hospital Authority, 668 F.2d 1177, 1178 n. 2 (CA11 1982); Beam v. Youens, 664 F.2d 1275 (CA5 1982); Williams v. Bolger, 633 F.2d 410 (CA5 1980); Century Laminating, Ltd. v. Montgomery, 595 F.2d 563 (CA10), cert. granted, 444 U.S. 897, cert. dismissed, 444 U.S. 987, 100 S.Ct. 516, 62 L.Ed.2d 417 (1979). Cf. United States v. Jones, 669 F.2d 559, 561 (CA8 1982) (dictum); Calhoun v. United States, 647 F.2d 6, 10 (CA9 1981); United States v. Moore, 616 F.2d 1030, 1032 n. 2 (CA7 1980) (dictum), cert. denied, 446 U.S. 987, 100 S.Ct. 2972, 64 L.Ed.2d 844 (1980). But cf. Laser Alignment, Inc. v. Warlick, 32 Fed.R.Serv.2d 776 (CA4 1981). 2 The Advisory Committee on Appellate Rules explained the modification as follows: "The proposed amendment would make it clear that after the filing of the specified post trial motions, a notice of appeal should await disposition of the motion. . . . [I]t would be undesirable to proceed with the appeal while the district court has before it a motion the granting of which would vacate or alter the judgment appealed from. . . . Under the present rule, since docketing may not take place until the record is transmitted, premature filing is much less likely to involve waste effort. See, e.g. Stokes v. Peyton's Inc., 508 F.2d 1287 (5th Cir.1975). Further, since a notice of appeal filed before the disposition of a post trial motion, even if it were treated as valid for purposes of jurisdiction, would not embrace objections to the denial of the motion, it is obviously preferable to postpone the notice of appeal until after the motion is disposed of. "The present rule [pre-1979], since it provides for the 'termination' of the 'running' of the appeal time, is ambiguous in its application to a notice of appeal filed prior to a post trial motion filed within the 10-day limit. The amendment would make it clear that in such circumstances the appellant should not proceed with the appeal during the pendency of the motion but should file a new notice of appeal after the motion is disposed of." Note of Advisory Committee on Appellate Rules, 28 U.S.C.A. following Rule 4. 3 Rule 2 does not purport to vest unlimited discretion in the Court of Appeals. That Rule explicitly states that the discretion it authorizes is limited by Rule 26(b), which prohibits Courts of Appeals from enlarging the time for filing a notice of appeal. 1 Presumably, the majority's remand for "further proceedings" will allow the Court of Appeals to consider whether respondent filed an effective notice of appeal. Cf. United States v. Hollywood Motor Car Company, Inc., --- U.S. ----, 102 S.Ct. 3081, 73 L.Ed.2d 754 (1982) (per curiam ) (where the lower court lacks jurisdiction, Court reverses and remands with instructions to the Court of Appeals to dismiss the appeal); Browder v. Director, Department of Corrections of Illinois, 434 U.S. 257, 98 S.Ct. 556, 54 L.Ed.2d 521 (1978) (Court simply reverses where Court of Appeals lacked jurisdiction due to untimely notice of appeal). 2 Respondent filed a brief on appeal in early 1981 in the mistaken belief that a final summary judgment had been entered. On October 2, 1981, the Court of Appeals remanded the case to the District Court, which subsequently entered an order directing entry of final judgment pursuant to Fed.Rule Civ.Proc. 54(b). 3 The Clerk's Office for the Court of Appeals for the Third Circuit enters only pleadings on its docket sheet. It maintains a separate file with all correspondence relating to a docketed case. The letters sent by respondent are in the Court of Appeals correspondence file for case No. 81-2989, the Court of Appeals docket number for this case. The docket sheet for the Court of Appeals in No. 81-2989 states that on December 4, 1981, a notice of appeal by respondent's counsel was filed. This is apparently a reference to a certified copy of the premature notice of appeal, which the District Court transmitted along with the record. 4 The papers filed by respondent after November 23 were transmitted to the Court of Appeals rather than to the District Court. Cobb v. Lewis, 488 F.2d 41, 45 (CA5 1974), makes clear that the notice requirement may be satisfied by a statement made either to the District Court or to the Court of Appeals. In a similar vein, Fed.Rule App.Proc. 4(a)(1) states that if a notice of appeal "is mistakenly filed in the court of appeals," the clerk of that court should note the date of the notice, and the notice "shall be deemed filed in the district court on the date so noted." Thus, a mistaken filing in the Court of Appeals is clearly not a fatal defect under the Rules. In this case, respondent appears to have filed a notice of appeal as defined by Rule 3(c) with the Court of Appeals on December 15. By that date, the District Court had already transmitted the record and a certified copy of the premature notice of appeal to the Court of Appeals, and the appellate court had docketed the appeal. Under these circumstances, the Court of Appeals would have been the sensible place in which to file a new notice. Respondent should not have been expected to return to District Court after December 4, when that court no longer had the record. 5 The majority apparently does not dispute the Court of Appeals' conclusion that the dismissal of an appeal based on an appellant's failure to comply with the technical requirements of Rule 4(a)(4) would be a manifest injustice in the absence of prejudice to the appellee. 6 See 1967 Advisory Committee Note to Fed.Rule App.Proc. 26. 7 See 9 J. Moore, B. Ward & J. Lucas, Moore's Federal Practice ¶ 204.12 at 4-67 & n. 26 (2d ed. 1982). In the Third Circuit alone, see, e.g., Richerson v. Jones, 572 F.2d 89, 93 (CA3 1978) (motion to reconsider judgment); Sonneblick-Goldman Corp. v. Norwalk, 420 F.2d 858, 859 (CA3 1970) (motion to vacate judgment); Gainey v. Brotherhood of Railway & Steamship Clerks, 303 F.2d 716, 718 (CA3 1962) (motion for rehearing or reconsideration). Sometimes the characterization has resulted in the dismissal of an appeal. 8 E.g., Dove v. Codesco, 569 F.2d 807 (CA4 1978); Alley v. Dodge Hotel, 551 F.2d 442 (CADC 1977); Sea Ranch Ass'n v. California Coastal Zone Conservation Commissions, 537 F.2d 1058 (CA9 1976); Woodham v. American Cystoscope Co., 335 F.2d 551 (CA5 1964).

89

459 U.S. 42 103 S.Ct. 394 74 L.Ed.2d 214 Donald W. WYRICK, Wardenv.Edward FIELDS. No. 82-158. Decided Nov. 29, 1982. PER CURIAM. 1 In this case, the United States Court of Appeals for the Eighth Circuit, over a dissent by Judge Ross, directed that respondent Edward Fields' petition for a writ of habeas corpus be granted; it did so on the ground that Fields had been convicted with evidence obtained in violation of his Fifth Amendment right to have counsel present at an interrogation. 682 F.2d 154 (1982). We have concluded that the Court of Appeals' majority misconstrued this Court's recent decision in Edwards v. Arizona, 451 U.S. 477, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981), and imposed a new and unjustified limit on police questioning of a suspect who voluntarily, knowingly, and intelligently waives his right to have counsel present. 2 * Respondent, a soldier then stationed at Fort Leonard Wood, Mo., was charged with raping an 81 year-old woman-on September 21, 1974. After his arrest on September 25, Fields was released on his own recognizance. He retained private defense counsel. After discussing the matter with his counsel and with a military attorney provided him by the Army, Fields requested a polygraph examination. This request was granted and the examination was conducted on December 4 by an agent of the Army's Criminal Investigation Division (CID) at the fort. 3 Prior to undergoing the polygraph examination, Fields was given a written consent document, which he signed, informing him of his rights, as required by Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966), and of his rights under the Uniform Code of Military Justice and the Eighth Amendment. In addition, the CID agent read to Fields the following detailed statement: 4 "Before I ask you any questions, you must understand your rights. You do not have to answer my questions or say anything. Anything you say or do can be used against you in a criminal trial. You have a right to talk to a lawyer before questioning or have a lawyer present with you during the questioning. This lawyer can be a civilian lawyer of your own choice, or a military lawyer, detailed for you at no expense to you. Also, you may ask for a military lawyer of your choice by name and he will be detailed for you if superiors determine he's reasonably available. If you are now going to discuss the offense under investigation, which is rape, with or without a lawyer present, you have a right to stop answering questions at any time or speak to a lawyer before answering further, even if you sign a waiver certificate. Do you want a lawyer at this time?" See State v. Fields, 538 S.W.2d 348, 350 n. 1 (Mo.App.1976) (emphasis added). 5 Fields answered: "No." 6 At the conclusion of the polygraph examination, which took less than two hours, the CID agent told Fields that there had been some deceit, and asked him if he could explain why his answers were bothering him. Fields then admitted having intercourse with the victim on September 21, but said that she had instigated and consented to it. The agent asked Fields if he wished to discuss the matter further with another CID agent and with the Waynesville, Mo., Chief of Police. Fields said that he did. Then, in his turn, the police chief read Fields his Miranda warnings once again before questioning him. Fields repeated that he had had sexual contact with the victim, but that it had been consensual. 7 Respondent was tried before a jury in Circuit Court, Pulaski County, Mo. He sought to suppress the testimony of the two CID agents and the police chief regarding his "confessions" to voluntary intercourse. The trial court denied the motion, ruling that Fields had waived his rights. The testimony was admitted. Fields was convicted, and was sentenced to 25 years in prison. The Missouri Court of Appeals affirmed the judgment on the ground that Fields "had been repeatedly and amply advised of his rights and . . . voluntarily, knowingly and intelligently waived his rights." 538 S.W.2d, at 350. 8 Eventually, Fields sought a writ of habeas corpus in the United States District Court for the Eastern District of Missouri. The District Court, agreeing with the Missouri Court of Appeals that Fields had voluntarily, knowingly, and intelligently waived his right to counsel, denied respondent's petition. On appeal, however, the Eighth Circuit reversed and remanded the case with directions to order the State either to release Fields or to afford him a new trial. 682 F.2d, at 162. II 9 The Court of Appeals found that the police conduct in question contravened the "clear import" of this Court's decision in Edwards v. Arizona: "a defendant's right to have counsel present at custodial interrogations must be zealously guarded." 682 F.2d, at 158. In Edwards, this Court had held that once a suspect invokes his right to counsel, he may not be subjected to further interrogation until counsel is provided unless the suspect himself initiates dialogue with the authorities. 451 U.S., at 484-487, 101 S.Ct., at 1884-1886. The Eighth Circuit recognized that what it called the "per se rule" of Edwards "does not resolve the issue present here." 682 F.2d, at 158. Fields and his counsel had agreed that Fields should take the polygraph examination, and Fields appeared voluntarily and stated that he did not want counsel present during the interrogation. Thus, the Court of Appeals conceded that "Fields thereby 'initiated' further dialogue with the authorities after his right to counsel had been invoked." Ibid. 10 When the suspect has initiated the dialogue, Edwards makes clear that the right to have a lawyer present can be waived: 11 "If, as frequently would occur in the course of a meeting initiated by the accused, the conversation is not wholly one-sided, it is likely that the officers will say or do something that clearly would be 'interrogation.' In that event, the question would be whether a valid waiver of the right to counsel and the right to silence had occurred, that is, whether the purported waiver was knowing and intelligent and found to be so under the totality of the circumstances, including the necessary fact that the accused, not the police, reopened the dialogue with the authorities." 451 U.S., at 486, n. 9, 101 S.Ct., at 1885, n. 9. 12 Citing this language, the Eighth Circuit acknowledged—as it had to—that "[t]here is no question that Fields waived his right to have counsel present while the [polygraph] examination itself was being conducted." 682 F.2d, at 160. Yet that court found that the State had failed to satisfy its burden of proving that "Fields knowingly and intelligently waived his right to have counsel present at the post-test interrogation." The court suggested that had the CID agent merely "paus[ed] to remind the defendant" of his rights, thus providing "meaningfully timed Miranda warnings" (emphasis in original), there would have been no violation. Ibid. III 13 In reaching this result, the Court of Appeals did not examine the "totality of the circumstances," as Edwards requires. Fields did not merely initiate a "meeting." By requesting a polygraph examination, he initiated interrogation. That is, Fields waived not only his right to be free of contact with the authorities in the absence of an attorney, but also his right to be free of interrogation about the crime of which he was suspected. Fields validly waived his right to have counsel present at "post-test" questioning, unless the circumstances changed so seriously that his answers no longer were voluntary, or unless he no longer was making a "knowing and intelligent relinquishment or abandonment" of his rights. 451 U.S., at 482, 101 S.Ct., at 1883. 14 The Court of Appeals relied on two facts indicating the need for a new set of warnings: the polygraph examination had been discontinued, and Fields was asked if he could explain the test's unfavorable results. To require new warnings because of these two facts is unreasonable. Disconnecting the polygraph equipment effectuated no significant change in the character of the interrogation. The CID agent could have informed Fields during the examination that his answers indicated deceit; asking Fields, after the equipment was disconnected, why the answers were bothering him was not any more coercive. The Court of Appeals stated that there was no indication that Fields or his lawyer anticipated that Fields would be asked questions after the examination. But it would have been unreasonable for Fields and his attorneys to assume that Fields would not be informed of the polygraph readings and asked to explain any unfavorable result. Moreover, Fields had been informed that he could stop the questioning at any time, and could request at any time that his lawyer join him. Merely disconnecting the polygraph equipment could not remove this knowledge from Fields' mind.* 15 The only plausible explanation for the court's holding is that, encouraged by what it regarded as a per se rule established in Edwards, it fashioned another rule of its own: that, notwithstanding a voluntary, knowing, and intelligent waiver of the right to have counsel present at a polygraph examination, and notwithstanding clear evidence that the suspect understood that right and was aware of his power to stop questioning at any time or to speak to an attorney at any time, the police again must advise the suspect of his rights before questioning him at the same interrogation about the results of the polygraph. The court indicated that this rule was needed because it thought that the use of polygraph "results" in questioning, although it does not necessarily render a response involuntary, is inherently coercive. But Courts of Appeals, including a different panel of the Eighth Circuit itself, and state courts, have rejected such a rule. See, e.g., United States v. Little Bear, 583 F.2d 411, 414 (CA8 1978); Keiper v. Cupp, 509 F.2d 238, 241-242 (CA9 1975); People v. Barreto, 256 Cal.App.2d 392, 64 Cal.Rptr. 211 (1967); State v. Henry, 352 So.2d 643 (La.1977). Cf. Henry v. Dees, 658 F.2d 406 (CA5 1981) (waiver not voluntary, knowing, and intelligent in the total circumstances of the case, including mental retardation of suspect). The Eighth Circuit's rule certainly finds no support in Edwards, which emphasizes that the totality of the circumstances, including the fact that the suspect initiated the questioning, is controlling. Nor is the rule logical; the questions put to Fields after the examination would not have caused him to forget the rights of which he had been advised and which he had understood moments before. The rule is simply an unjustifiable restriction on reasonable police questioning. IV 16 According to the dissent, a substantial question as to the admissibility of Fields' statements may be raised under the Sixth Amendment. Post, at 52-55. The Sixth Amendment issues raised by the dissent, however, are not before us. The Court of Appeals rested its judgment exclusively on the Fifth Amendment "right to have counsel present during a custodial interrogation" and on its interpretation of this Court's decision in Edwards. 682 F.2d, at 158. That interpretation was flawed and the judgment of the Court of Appeals must be reversed. We express no view as to whether any constitutional safeguards not mentioned by the Court of Appeals bear on this case. 17 Because the Court of Appeals misapplied Edwards and created an unjustified per se rule, the petition for a writ of certiorari is granted and that court's judgment is reversed and the case is remanded. 18 It is so ordered. 19 Justice STEVENS, concurring. 20 There is much force to what Justice MARSHALL has written in dissent. I share his concern about the Court's practice of deciding cases summarily, partly because there is a special risk of error in summary dispositions and partly because the practice represents an unwise use of the Court's scarce resources. I do not, however, agree with Justice MARSHALL's suggestion that we should invite the parties to submit briefs on the merits before a case is summarily decided. I fear that the institution of such a practice would tend to regularize and expand the number of our summary dispositions. 21 In this case I believe the correct procedure for the Court to have followed would have been simply to deny the petition for a writ of certiorari. No conflict has yet developed on the precise question presented and, as Justice MARSHALL demonstrates, the Court of Appeals' conclusion is not without reasoned support. The Court, however, has granted the petition. Although I voted against that action, I am now persuaded that the Court's resolution of the merits is correct and therefore join its disposition. 22 Justice MARSHALL, dissenting. 23 A summary reversal is an exceptional disposition. It should be reserved for situations in which the applicable law is settled and stable, the facts are not disputed, and the decision below is clearly in error.1 Because I do not believe that this is such a case, I dissent. 24 * I do not agree that respondent's consent to the polygraph examination necessarily constituted a waiver of his Fifth Amendment rights with respect to the post-examination interrogation. In my view, this case is not controlled by the footnote in Edwards v. Arizona, 451 U.S. 477, 486 n. 9, 101 S.Ct. 1880, 1885 n. 9, 68 L.Ed.2d 378 (1981), on which the Court relies. That footnote dealt with the hypothetical case, not before the Court in Edwards, of a suspect who initiates a meeting with the police. The Court indicated that, even if the police said or did something in the meeting that constituted interrogation, incriminating statements thereby elicited would be admissible if a knowing and intelligent waiver of the suspect's Fifth Amendment rights could be inferred in light of "the totality of the circumstances, including the necessary fact that the accused, not the police, reopened the dialogue with the authorities." Ibid. (Emphasis supplied.) In this case, "the totality of the circumstances" includes the undisputed facts that respondent agreed only to submit to a polygraph examination, and that he was never told he would be subjected to a post-examination interrogation. Moreover, an agreement to submit to a polygraph examination differs in an important respect from the initiation of an ordinary conversation with the authorities. When a suspect commences a conversation with a policeman, he has reason to expect that, as in any conversation, there will be a give-and-take extending beyond the subject matter of his original remarks. It may therefore be appropriate to conclude that the suspect's waiver of his Fifth Amendment rights extends to the entire conversation. By contrast, a polygraph examination is a discrete test. It has a readily identifiable beginning and end. An individual who submits to such an examination does not necessarily have any reason whatsoever to expect that he will be subjected to a post-examination interrogation.2 While in some cases the prosecution may be able to prove that a suspect knew there would be questioning after the test, here there is "no evidence that Fields or his lawyer anticipated that the CID officer would attempt to elicit incriminating statements from Fields after the examination was run." 682 F.2d 154, 160 (CA8 1982). 25 In any event, I do not believe that this substantial constitutional question should be disposed of summarily. I recognize, of course, that this Court's expanding docket has increased the pressure to accelerate the disposition process. I cannot agree, however, that summary reversal is proper in a case that involves a significant issue not settled by our prior decisions. If the Court concludes that there are "special and important reasons," Sup.Ct.Rule 17.1, for granting certiorari but also concludes that this case should not be set for oral argument, the Court should at least give the parties notice that it is considering a summary disposition, so that they may have an opportunity to submit briefs on the merits. See Brown, Foreword: Process of Law, 72 Harv.L.Rev. 77, 94-95 (1958). II 26 Today's decision holds only that the post-examination interrogation did not violate respondent's Fifth Amendment privilege against self-incrimination. The Court's ruling does not preclude the Court of Appeals from considering on remand whether the interrogation nevertheless violated his Sixth Amendment right to counsel.3 See Massiah v. United States, 377 U.S. 201, 84 S.Ct. 1199, 12 L.Ed.2d 246 (1964). Because "the policies underlying the two constitutional protections are quite distinct," Rhode Island v. Innis, 446 U.S. 291, 300 n. 4, 100 S.Ct. 1682, 1689 n. 4, 64 L.Ed.2d 297 (1980), a suspect may waive his Fifth Amendment right to remain silent without waiving his Sixth Amendment right to counsel. 27 Where only the Fifth Amendment applies, the ultimate question is whether the conduct alleged to constitute a waiver demonstrates that, despite "the compulsion inherent in custodial surroundings," Miranda v. Arizona, 384 U.S. 436, 458, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966), the suspect's statements were given voluntarily. To make the Fifth Amendment protection against compelled self-incrimination effective, this Court has held that a suspect has a right to have counsel present at any custodial interrogation. Id., at 469-472, 86 S.Ct., at 1625-1626. Once a suspect in custody asks to speak to a lawyer, he "is not subject to further interrogation by the authorities until counsel has been made available to him, unless the accused himself initiates further communication, exchanges, or conversations with the police." Edwards v. Arizona, 451 U.S. 477, 484-485, 101 S.Ct. 1880, 1884-1885, 68 L.Ed.2d 378 (1981). When a suspect has indicated that he needs legal advice before deciding whether to talk further, any subsequent statements made at the authorities' insistence without counsel being present are unlikely to be voluntary. See Michigan v. Mosley, 423 U.S. 96, 110 n. 2, 96 S.Ct. 321, n. 2, 46 L.Ed.2d 313 (1975) (WHITE, J., concurring in the result). If, on the other hand, the subsequent statements are made in a conversation initiated by the accused, they may well be voluntary. See Edwards v. Arizona, 451 U.S., at 486 n. 9, 101 S.Ct., at 1885 n. 9. Since the underlying purpose of the privilege against self-incrimination is to prevent the State from coercing an individual to give evidence against himself, it makes sense to find a waiver of the privilege where a suspect's conduct provides assurance that his statements were made voluntarily. 28 The determination of whether there has been a valid waiver of the Sixth Amendment right to counsel has a different focus, for the values underlying that right are different. The purpose of the Sixth Amendment right to counsel is to provide the defendant with legal assistance during the critical stages of the criminal process. See, e.g., Brewer v. Williams, 430 U.S. 387, 398, 97 S.Ct. 1232, 1239, 51 L.Ed.2d 424 (1977); Powell v. Alabama, 287 U.S. 45, 57, 53 S.Ct. 55, 59, 77 L.Ed. 158 (1932). To give effect to this protection, this Court has insisted that the State deal with a defendant through his attorney. Once the State has commenced adversary criminal proceedings against an individual, as Missouri did in this case more than two months before the polygraph examination was held, the Sixth Amendment forbids all efforts to elicit information from him in the absence of counsel, regardless of whether he is in custody, see United States v. Henry, 447 U.S. 264, 273-274 n. 11, 100 S.Ct. 2183, 2188-2189 n. 11, 65 L.Ed.2d 115 (1980); Massiah v. United States, supra, and regardless of whether the technique used to extract information is in any way coercive, see McLeod v. Ohio, 381 U.S. 356, 85 S.Ct. 1556, 14 L.Ed.2d 682 (1965). 29 To establish a waiver of the Sixth Amendment right to counsel, it is therefore not enough for the State to point to conduct—such as the initiation of a conversation—that demonstrates that the defendant's statements were made voluntarily. Since a Sixth Amendment violation does not depend upon coercion, the protection of the Sixth Amendment is not waived by conduct that shows only that a defendant's statements were not coerced. The State must show that the defendant intelligently and knowingly relinquished his right not to be questioned in the absence of counsel. The State can establish a waiver only by proving " 'an intentional relinquishment or abandonment' " of the right to have counsel present. Brewer v. Williams, supra, 430 U.S., at 404, 97 S.Ct., at 1242, quoting Johnson v. Zerbst, 304 U.S. 458, 464, 58 S.Ct. 1019, 1023, 82 L.Ed. 1461 (1938).4 30 Given the different policies underlying the Fifth and Sixth Amendments, it is not surprising that a number of courts have held that " '[w]arnings by law enforcement officers and subsequent action by the accused that might suffice to comply with Fifth Amendment strictures against testimonial compulsion [do] not necessarily meet . . . the higher standard with respect to waiver of the right to counsel that applies when the Sixth Amendment right to counsel has attached.' " United States v. Mohabir, 624 F.2d 1140, 1147 (CA2 1980), quotingUnited States v. Massimo, 432 F.2d 324, 327 (CA2 1970) (Friendly, J., dissenting) (majority did not reach the issue), cert. denied, 400 U.S. 1022, 91 S.Ct. 586, 27 L.Ed.2d 633 (1971).5 Today's decision therefore does not foreclose the Court of Appeals from considering on remand whether the post-examination interrogation violated the Sixth Amendment. * The dissent suggests that, because the results of polygraph examinations are inadmissible in Missouri, Fields might reasonably have expected that he would not be subjected "to additional questioning that can produce admissible evidence." Post, at 51, n. 2. Although the results of the polygraph examination might not have been admissible evidence, the statements Fields made in response to questioning during the course of the polygraph examination surely would have been. 1 See generally Brown, Foreword: Process of Law, 72 Harv.L.Rev. 77 (1958). 2 Certainly no one would argue that a suspect who consented to a blood test, a lineup, or fingerprinting thereby consented to be questioned about the results of those procedures. In this case, it is particularly inappropriate to assume that Fields must have realized that the CID agent would conduct a post-examination interrogation. The results of polygraph examinations are inadmissible in Missouri. See State v. Biddle, 599 S.W.2d 182, 191 (Mo.1980) (en banc); State v. Weindorf, 361 S.W.2d 806, 811 (Mo.1962). When a defendant, after consultation with his attorney, agrees to submit to an examination the results of which are inadmissible, the authorities have no justification for inferring that the defendant has also agreed to submit to additional questioning that can produce admissible evidence. 3 I do not share the majority's certainty that the Court of Appeals relied "exclusively on the Fifth Amendment." Ante, at 49. Although the opinion below does discuss Edwards v. Arizona, 451 U.S. 477, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981), at considerable length, the court phrased its holding in terms of the "right to counsel" without referring specifically to the Fifth Amendment or the Sixth Amendment. See 682 F.2d, at 157 ("we conclude that Fields did not knowingly and intelligently waive his right to have counsel present at the interrogation"); id., at 161 ("The government has simply introduced no evidence from which we can conclude that when Fields was confronted with the accusatory statement that the 'lie-detector' showed he was lying, he waived his right to the protection of counsel in this coercive situation."). See also id., at 161 n. 12 (relying on Brewer v. Williams, 430 U.S. 387, 97 S.Ct. 1232, 51 L.Ed.2d 424 (1977), a Sixth Amendment case). It is noteworthy that the Magistrate, whose report the District Court adopted, pointed to the Sixth Amendment problem by observing that it is "a somewhat empty gesture to appoint an attorney for an accused . . . and then pursue an interrogation . . . without his attorney." In addition, the petition for certiorari asserts that the decision below "expands the rights guaranteed an accused during interrogation under the Fifth and Sixth Amendments." Petition for Certiorari at 7 (emphasis supplied). In any event, since the Court today construes the Court of Appeals' opinion as resting solely on the Fifth Amendment, the Sixth Amendment issue remains open on remand. 4 Edwards v. Arizona, 451 U.S. 477, 101 S.Ct. 1880, 68 L.Ed.2d 378 (1981), addressed only the standard governing waiver of the Fifth Amendment privilege against self-incrimination. Since the Court concluded that Edwards had been interrogated in violation of the Fifth Amendment, it had no occasion to consider whether the Sixth Amendment applied or whether, if so, Edwards had waived its protection. See id., at 480 n. 7, 101 S.Ct., at 1882 n. 7. 5 See United States ex rel. O'Connor v. New Jersey, 405 F.2d 632, 636 (CA7), cert. denied, 395 U.S. 923, 89 S.Ct. 1770, 23 L.Ed.2d 240 (1969); Hancock v. White, 378 F.2d 479, 482 (CA1 1967). See also United States v. Springer, 460 F.2d 1344, 1354-1355 (CA7) (Stevens, J., dissenting), cert. denied, 409 U.S. 873, 93 S.Ct. 205, 34 L.Ed.2d 125 (1972); People v. Arthur, 22 N.Y.2d 325, 292 N.Y.S.2d 663, 239 N.E.2d 537, 539 (1968). See generally Note, 82 Colum.L.Rev. 363 (1982).

01

459 U.S. 70 103 S.Ct. 407 74 L.Ed.2d 235 UNITED STATES, Appellant,v.SECURITY INDUSTRIAL BANK et al. No. 81-184. Argued Oct. 6, 1982. Decided Nov. 30, 1982. Syllabus A provision of the Bankruptcy Reform Act of 1978, 11 U.S.C. § 522(f)(2) (1976 ed., Supp. V), permits individual debtors in bankruptcy proceedings to avoid nonpossessory, nonpurchase-money liens on certain property, including household furnishings and appliances. Appellees loaned individual debtors money and obtained and perfected such liens on the debtors' household furnishings and appliances before the 1978 Act was enacted. Subsequently, these debtors instituted separate bankruptcy proceedings under the 1978 Act. Sections 522(b) and (d) exempt household items from the property included within debtors' estates. The debtors claimed these exemptions, relying on § 522(f)(2) to avoid the liens. The Bankruptcy Courts refused to apply § 522(f)(2) retroactively to abrogate the liens. The Court of Appeals in consolidated appeals affirmed, holding that, although the 1978 Act was intended to apply retrospectively and thus was designed to invalidate liens acquired before the enactment date, such an application violates the Takings Clause of the Fifth Amendment. Held: Section 522(f)(2) was not intended to be applied retrospectively to destroy pre-enactment property rights. Pp. 74-82. (a) Where there is substantial doubt whether retroactive destruction of appellees' liens would comport with the Fifth Amendment, the cardinal principle that this Court will first determine whether a construction of a statute is fairly possible by which the constitutional question may be avoided warrants a consideration of whether, as a matter of statutory construction, § 522(f)(2) must necessarily be applied retroactively. Pp. 74-78. (b) No bankruptcy law shall be construed to eliminate property rights that existed before the law was enacted in the absence of an explicit command from Congress. In light of this principle, in the absence of a clear expression of Congress' intent to apply § 522(f)(2) to property rights established before the enactment date, the statute will not be construed in a manner that could call upon this Court to resolve difficult and sensitive questions arising out of the guarantees of the Takings Clause. P. 414. 642 F.2d 1193, (10th Cir.) affirmed. Alan I. Horowitz, Washington, D.C., for appellant. Henry F. Field, Chicago, Ill., for appellees. Justice REHNQUIST delivered the opinion of the Court. 1 This case concerns the effect of 11 U.S.C. § 522(f)(2), which permits individual debtors in bankruptcy proceedings to avoid liens on certain property. The Court of Appeals consolidated seven appeals from the Bankruptcy Courts for the Districts of Kansas and Colorado. In each case the debtor was an individual who instituted bankruptcy proceedings after the Bankruptcy Reform Act of 1978, Pub.L. No. 95-598, 92 Stat. 2549 ("the 1978 Act"), became effective on October 1, 1979. In each case one of the appellees had loaned the debtor money and obtained and perfected a lien on the debtor's household furnishings and appliances before the 1978 Act was enacted on November 6, 1978. None of these liens was possessory, and none secured purchase-money obligations. 2 Included within the personal property subject to the appellees' liens were household items that are exempt from the property included within the debtors' estates by virtue of subsections (b) and (d) of § 522.1 The debtors claimed these exemptions in their respective bankruptcy proceedings, relying on § 522(f)(2) to avoid the liens. That section provides: 3 "Notwithstanding any waiver of exemptions, the debtor may avoid the fixing of a lien on an interest of the debtor in property to the extent such lien impairs an exemption to which the debtor would have been entitled under subsection (b) of this section, if such lien is— 4 (2) a nonpossessory, nonpurchase-money security interest in any— 5 (A) household furnishings, household goods, wearing apparel, appliances, books, animals, crops, musical instruments, or jewelry that are held primarily for the personal, family, or household use of the debtor or a dependent of the debtor; 6 (B) implements, professional books, or tools, of the trade of the debtor or of a dependent of the debtor; or 7 (C) professionally prescribed health aids for the debtor or a dependent of the debtor." 8 The appellees asserted that application of § 522(f)(2) to liens acquired before the enactment date would violate the Fifth Amendment. The United States intervened in each case to defend the constitutionality of the federal statute,2 but the bankruptcy courts in each case refused to apply § 522(f)(2) to abrogate liens acquired before the enactment date.3 9 The Court of Appeals consolidated the cases and affirmed the judgments of the bankruptcy courts. 642 F.2d 1193, 1195 (CA10 1981). It held that the 1978 Act was intended to apply retrospectively, and thus was designed to invalidate liens acquired before the enactment date. It also held, however, that such an application violates the Fifth Amendment. The court stated that § 522(f)(2) effects a "complete taking of the secured creditors' property interest," and is thus invalid under Louisville Joint Stock Land Bank v. Radford, 295 U.S. 555, 55 S.Ct. 854, 79 L.Ed.2d 1593 (1935).4 The United States appealed, and we noted probable jurisdiction. --- U.S. ----, 102 S.Ct. 969, 71 L.Ed.2d 108 (1982). 10 The appellees, of course, defend the judgment of the Court of Appeals.5 The government argues at some length that retrospective application of § 522(f)(2) to these liens would not violate the Fifth Amendment. It contends that the enactment is a "rational" exercise of Congress' bankruptcy power, that for "bankruptcy purposes" property interests are all but indistinguishable from contractual interests, that these particular interests were "insubstantial" and therefore their destruction does not amount to a "taking" of property requiring compensation. We do not decide the constitutional question reached by the Court of Appeals. We address it only to determine whether the attack on the retrospective application of the statute raises substantial enough constitutional doubts to warrant the employment of the canon of statutory construction referred to post, 8-10. 11 It may be readily agreed that § 522(f)(2) is a rational exercise of Congress' authority under Article I, Section 8, Clause 4, and that this authority has been regularly construed to authorize the retrospective impairment of contractual obligations. Hanover National Bank v. Moyses, 186 U.S. 181, 188, 22 S.Ct. 857, 860, 46 L.Ed. 1113 (1902). Such agreement does not, however, obviate the additional difficulty that arises when that power is sought to be used to defeat traditional property interests. The bankruptcy power is subject to the Fifth Amendment's prohibition against taking private property without compensation. Louisville Joint Stock Land Bank v. Radford, 295 U.S. 555, 55 S.Ct. 854, 79 L.Ed. 1593 (1935). Thus, however "rational" the exercise of the bankruptcy power may be, that inquiry is quite separate from the question whether the enactment takes property within the prohibition of the Fifth Amendment. 12 The government apparently contends (Brief for the United States, at 30-32) that because cases such as Arnett v. Kennedy, 416 U.S. 134, 94 S.Ct. 1633, 40 L.Ed.2d 15 (1974) and Goldberg v. Kelly, 397 U.S. 254, 90 S.Ct. 1011, 25 L.Ed.2d 287 (1970) defined "property" for purposes of the Due Process Clause sufficiently broadly to include rights which at common law would have been deemed contractual, traditional property rights are entitled to no greater protection under the takings clause than traditional contract rights. It argues that "bankruptcy principles do not support a sharp distinction between the rights of secured and unsecured creditors." Brief for the United States, at 31. However "bankruptcy principles" may speak to this question, our cases recognize, as did the common law, that the contractual right of a secured creditor to obtain repayment of his debt may be quite different in legal contemplation from the property right of the same creditor in the collateral. Compare Hanover National Bank v. Moyses, supra, with Louisville Joint Stock Land Bank v. Radford, supra, and Kaiser-Aetna v. United States, 444 U.S. 164, 100 S.Ct. 383, 62 L.Ed.2d 332 (1979). 13 Since the governmental action here would result in a complete destruction of the property right of the secured party, the case fits but awkwardly into the analytic framework employed in Penn Central Transportation Co. v. New York City, 438 U.S. 104, 98 S.Ct. 2646, 57 L.Ed.2d 631 (1978) and PruneYard Shopping Center v. Robins, 447 U.S. 74, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980), where governmental action affected some but not all of the "bundle of rights" which comprise the "property" in question. The government argues that the interest of a secured party such as was involved here is "insubstantial," apparently in part because it is a nonpurchase-money, non-possessory interest in personal property. The "bundle of rights" which accrues to a secured party is obviously smaller than that which accrues to an owner in fee simple, but the government cites no cases supporting the proposition that differences such as these relegate the secured party's interest to something less than property.6 And our decisions in Radford, supra, and Armstrong v. United States, 364 U.S. 40, 80 S.Ct. 1563, 4 L.Ed.2d 1554 (1961), militate against such a proposition. 14 In Radford, we held that the Frazier-Lemke Act, June 28, 1934, c. 869, 48 Stat. 1289, violated the takings clause. The bank held a nonpurchase-money mortgage on Radford's farm. Radford defaulted and instituted bankruptcy proceedings. The Frazier-Lemke Act, which by its terms applied only retrospectively, permitted the debtor to purchase the property for less than its fair market value.7 We held the statute was void because it effected a "taking of substantive rights in specific property acquired by the Bank prior to" its enactment. 295 U.S., at 590, 55 S.Ct. at 863. In his opinion for the Court, Justice Brandeis stated: 15 "[T]he Fifth Amendment commands that, however great the Nation's need, private property shall not be thus taken even for a wholly public use without just compensation. If the public interest requires, and permits, the taking of property of individual mortgagees in order to relieve the necessities of individual mortgagors, resort must be had to proceedings by eminent domain; so that, through taxation, the burden of the relief afforded in the public interest may be borne by the public." Id., at 602, 55 S.Ct., at 869. 16 In Armstrong, materialmen delivered materials to a prime contractor for use in constructing navy personnel boats. Under state law, they obtained liens in the vessels.8 The prime contractor defaulted on his obligations to the United States, and the government took title to and possession of the uncompleted hulls and unused materials, thus making it impossible for the materialmen to enforce their liens. We held that this constituted a taking: 17 "The total destruction by the government of all compensable value of these liens, which constitute compensable property, has every possible element of a Fifth Amendment 'taking' and is not a mere 'consequential incidence' of a valid regulatory measure." 364 U.S., at 48, 80 S.Ct., at 1568-69. 18 The government seeks to distinguish Armstrong on the ground that it was a classical "taking" in the sense that the government acquired for itself the property in question, while in the instant case the government has simply imposed a general economic regulation which in effect transfers the property interest from a private creditor to a private debtor. While the classical taking is of the sort that the government describes, our cases show that takings analysis is not necessarily limited to outright acquisitions by the government for itself. See Loretto v. Teleprompter Manhattan CATV Corp., --- U.S. ----, 102 S.Ct. 3164, 73 L.Ed.2d 868 (1982); PruneYard Shopping Center v. Robins, supra; Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 415, 43 S.Ct. 158, 160, 67 L.Ed. 322 (1922). 19 The government finally contends that because the resale value of household goods is generally low, and because creditors therefore view the principal value of their security as a lever to negotiate for reaffirmation of the debt rather than as a vehicle for foreclosure, the property interests involved here do not merit protection under the takings clause. While this contention cannot be dismissed out of hand, it seems to run counter to the state's characterization of the interest as property, see note 5, supra, to our reliance in other "takings" cases on state law characterizations, see, e.g., Kaiser-Aetna v. United States, supra, 444 U.S. 164, 179, 100 S.Ct. 383, 392, 62 L.Ed.2d 332 and also to at least some of the implications of Radford, supra, and Armstrong, supra. 20 The foregoing discussion satisfies us that there is substantial doubt whether the retroactive destruction of the appellees' liens in these cases comports with the Fifth Amendment. We now consider whether, as a matter of statutory construction, § 522(f)(2) must necessarily be applied in that manner. We consider the statutory question because of the " 'cardinal principle that this Court will first ascertain whether a construction of the statute is fairly possible by which the constitutional question may be avoided.' " Lorillard v. Pons, 434 U.S. 575, 577, 98 S.Ct. 866, 868, 55 L.Ed.2d 40 (1978), quoting Crowell v. Benson, 285 U.S. 22, 62, 52 S.Ct. 285, 296, 76 L.Ed. 598 (1932). 21 The Court of Appeals thought § 522(f)(2) must apply retroactively, that is, to liens which attached before the enactment date, because "there would be no bankruptcy law applicable to cases involving such liens if it did not." 642 F.2d, at 1197. The court apparently thought that if § 522(f)(2) does not apply to liens which came into existence before the enactment date, then no part of the 1978 Act could apply to cases involving such liens. This is not necessarily the case. The liens, of course, exist under state law independently of the 1978 Act. Although the 1978 Act, in general, is effective for all cases commenced after its effective date, Congress might have intended that provisions that destroy previously vested property rights apply only to interests that came into effect after the enactment date. If § 522(f)(2) is such a provision, the remainder of the 1978 Act would not affect the enforceability of these liens, but would still apply to these liens and these cases. We think that the analysis of the Court of Appeals did not adequately dispose of the question as to the retrospective effect of § 522(f), and we therefore pursue the inquiry further. 22 The principle that statutes operate only prospectively, while judicial decisions operate retrospectively, is familiar to every law student. Compare Sands, Sutherland's Statutory Construction § 106 with Linkletter v. Walker, 381 U.S. 618, 622-625, 85 S.Ct. 1731, 1733-35, 14 L.Ed.2d 601 (1965). This Court has often pointed out that 23 the first rule of construction is that legislation must be considered as addressed to the future, not to the past. . . . The rule has been expressed in varying degrees of strength but always of one import, that a retrospective operation will not be given to a statute which interferes with antecedent rights . . . unless such be "the unequivocal and inflexible import of the terms, and the manifest intention of the legislature." 24 Union Pacific R. Co. v. Laramie Stock Yards Co., 231 U.S. 190, 199, 34 S.Ct. 101, 102, 58 L.Ed. 179 (1913) (citations omitted). See, e.g., United States Fidelity & Guaranty Co. v. Struthers Wells Co., 209 U.S. 306, 314, 28 S.Ct. 537, 539, 52 L.Ed. 804 (1908) ("The presumption is very strong that a statute was not meant to act retrospectively, and it ought never to receive such a construction if it is susceptible of any other."); United States v. The Peggy, 5 U.S. (1 Cranch) 103, 110, 2 L.Ed. 49 (1801). 25 This principle has been repeatedly applied to bankruptcy statutes affecting property rights. In Holt v. Henley, 232 U.S. 637, 34 S.Ct. 459, 58 L.Ed. 767 (1914), the Court had before it a new statute granting bankruptcy trustees the position of a lienholder with priority over sellers on conditional sales contracts. Act of June 25, 1910, c. 412, § 8, 26 Stat. 838, 840. This provision, like § 522(f)(2), could be read literally to divest property interests which had been created before it was enacted. The 1910 statute, like the 1978 Act, applied to all bankruptcy cases instituted after it became effective.9 Nonetheless, the Court followed the lead of the lower courts in refusing to infer retroactivity absent an explicitly "expressed intent of Congress." Arctic Ice Machine Co. v. Armstrong County Trust Co., 192 F. 114, 116 (CCA 3 1911). See also In re Schneider, 203 F. 589, 590 (E.D.Pa.1913). In his opinion for the unanimous Court, Justice Holmes stated "that the reasonable and usual interpretation of [bankruptcy] statutes is to confine their effect, so far as it may be, to property rights established after they were passed." 232 U.S., at 639, 34 S.Ct., at 460. See Auffm'ordt v. Rasin, 102 U.S. 620, 622, 26 L.Ed. 262 (1881). 26 The government nonetheless contends that bankruptcy statutes are usually construed to apply to preexisting rights. This statement is unobjectionable in the context of traditional contract rights, Hanover National Bank v. Moyses, supra, 186 U.S., at 188, 22 S.Ct., at 860-61, but none of the cases cited by the government extend it to property rights such as those involved here.10 27 Neither these cases, nor any other that has come to our attention, casts doubt on the principle of statutory construction deducible from Holt and Auffm'ordt: No bankruptcy law shall be construed to eliminate property rights which existed before the law was enacted in the absence of an explicit command from Congress. In light of this principle, the legislative history of the 1978 Act suggests that Congress may not have intended that § 522(f) operate to destroy pre-enactment property rights. 28 An early version of the 1978 Act contained an explicit requirement that all its provisions "shall apply in all cases or proceedings instituted after its effective date, regardless of the occurrence of any of the operative facts determining legal rights, duties or liabilities hereunder." H.R. 31, 94th Cong., 1st Sess., § 10-103(a) (1975), reprinted in Bankruptcy Act Revision: Hearings on H.R. 31 and H.R. 32 before the Subcomm. on Civil and Constitutional Right of the House Comm. on the Judiciary, 94th Cong., 1st Sess. (1975), n. 14 app., at 320-321. This provision may or may not have been deleted directly in response to the comments of witness William Plumb to the effect that retroactive invalidation of liens may be an unconstitutional taking. Id., at 2066-2067. Nonetheless, Congress's elimination of an explicit command is some evidence that it did not intend to depart from the usual principle of construction. See Bradley v. Richmond School Board, 416 U.S. 696, 716, n. 23, 94 S.Ct. 2006, 2018, n. 23, 40 L.Ed.2d 476 (1974) ("We are reluctant to read into the statute the very . . . limitation that Congress eliminated."). 29 "Accordingly, in the absence of a clear expression of Congress' intent to" apply § 522(f)(2) to property rights established before the enactment date,11 "we decline to construe the Act in a manner that could in turn call upon the Court to resolve difficult and sensitive questions arising out of the guarantees of the" takings clause. N.L.R.B. v. Catholic Bishop of Chicago, 440 U.S. 490, 507, 99 S.Ct. 1313, 1322, 59 L.Ed.2d 533 (1979).12 The judgment of the Court of Appeals must therefore be 30 Affirmed. 31 Justice BLACKMUN, with whom Justice BRENNAN and Justice MARSHALL join, concurring in the judgment. 32 This case concerns the Bankruptcy Act of 1978, 11 U.S.C. § 101 et seq. (1976 ed., Supp. V), and, in particular, the exemption provisions of § 522 of that Act. Specifically at issue is the effect of certain of these exemption provisions upon nonpossessory, nonpurchase-money obligations given by debtors to small loan companies before the enactment of the Act. The purported liens apply generally, not specifically, to property of the kind described and, as a practicable matter, there is nothing to prevent the debtor's selling the property and replacing it or not replacing it, just as he chooses. 33 Section 522, for the first time, established a set of federal exemptions for individual debtors. Concededly, the section, as all similar statutes, was enacted to protect the debtor's essential needs and to enable him to have a fresh start economically. Section 522(f)(2) permits the debtor to "avoid the fixing" of a nonpossessory, nonpurchase-money security interest in certain property, but the subsection does not extend to all property otherwise exempt under § 522(d). It is limited to certain personal items, such as household furnishings, wearing apparel, jewelry, tools of the debtor's trade, and professionally prescribed health aids. 34 The Court naturally struggles with the question of the application of the new exemption provisions to obligations created before the new Act. It notes its concern with constitutional problems and it also greets with obvious relief the possibility of construing the Act as being only prospective in its operation. It then quickly pursues the latter route in order to avoid any constitutional issue. 35 I understand and can sympathize with the Court's desire thus to resolve the case. It is usually much easier to construe a statute so as to avoid a constitutional issue than it is to resolve the constitutional issue itself. And, of course, the Court's cases have announced that, where feasible, this is the preferred method. See, e.g., Lorillard v. Pons, 434 U.S. 575, 577, 98 S.Ct. 866, 868, 55 L.Ed.2d 40 (1978). 36 Were we writing on a "clean slate," however, I would not pursue, in this case, that principle of construction-preference, for I think that the case would deserve consideration in greater depth. I see nothing in the statute with which we are concerned that speaks or hints of only prospective applicability, or that compels it, and I would find it necessary to reach the constitutional issue. I would then resolve that issue in favor of the debtor and against the small loan company creditor. I would do so because the exemptions in question are limited as to kinds of property and as to values; because the amount loaned has little or no relationship to the value of the property; because these asserted lien interests come close to being contracts of adhesion; because repossessions by small loan companies in this kind of situation are rare; because the purpose of the statute is salutary and is to give the debtor a fresh start with a minimum for necessities; because there has been creditor abuse; because Congress merely has adjusted priorities, and has not taken for the Government's use or for public use; because the exemption provisions in question affect the remedy and not the debt; because the security interest seems to have little direct value and weight in its own right and appears useful mainly as a convenient tool with which to threaten the debtor to reaffirm the underlying obligation; because the statute is essentially economic regulation and insubstantial at that; and because there is an element of precedent favorable to the debtor to be found in such cases as Penn Central Transp. Co. v. New York City, 438 U.S. 104, 98 S.Ct. 2646, 57 L.Ed.2d 631 (1978), and PruneYard Shopping Center v. Robins, 447 U.S. 74, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980). 37 But we are not writing on a clean slate. It seems to me that the case of Holt v. Henley, 232 U.S. 637, 34 S.Ct. 459, 58 L.Ed. 767 (1914), is precisely in point and, unless the Court chooses to overrule it, must control the present case. There, Holt and the eventual bankrupt signed an agreement in 1909 for the installation of an automatic sprinkler system on the property of the eventual bankrupt. The agreement specified that the system was to remain Holt's property until paid for and that he was to have a right to enter and remove it upon failure to pay as agreed. Thereafter, but also in 1909, a mortgage deed was executed covering the plant and what was "acquired and placed upon the said premises during the continuance of this trust." Id., at 639, 34 S.Ct., at 459. Section 8 of the Act of June 25, 1910, ch. 412, 36 Stat. 840, amended § 47a(2) of the then Bankruptcy Act to give the trustee in bankruptcy, as to property coming into the custody of the bankruptcy court, the rights of a creditor holding a lien. Upon Holt's debtor's bankruptcy, the mortgagees claimed the sprinkler system. 38 Justice Holmes, writing for a unanimous Court, observed that before the amendment "Holt had a better title than the trustees would have got" and that the Court was of the opinion "that the Act should not be construed to impair it." 232 U.S., at 639, 34 S.Ct., at 460. He went on: 39 "We do not need to consider whether or how far in any event the constitutional power of Congress would have been limited. It is enough that the reasonable and usual interpretation of such statutes is to confine their effect, so far as may be, to property rights established after they were passed. . . . That is a familiar and natural mode of interpretation. . . . We are of opinion that [Holt's title] was not affected by the enactment of later date than the conditional sale. The opposite construction would not simply extend a remedy but would impute to the Act of Congress an intent to take away rights lawfully retained, and unimpeachable at the moment when they took their start." Id., at 639-640, 34 S.Ct., at 460. 40 The Court then ruled against the claim of the mortgagees because they had made no advance on the faith of the sprinkler system and were not purchasers for value as against Holt, and because removal "would not affect the integrity of the structure on which the mortgagees advanced." Id., at 641, 34 S.Ct., at 460. 41 Holt v. Henley thus also involved a pre-existing agreement, a subsequent change in the then Bankruptcy Act, and the Court's preservation of the pre-existing right. I see no way to distinguish that case from this one, and I would affirm the judgment of the Court of Appeals simply on the compelling authority of Holt v. Henley. See also Auffm'ordt v. Rasin, 102 U.S. 620, 622, 26 L.Ed. 262 (1881). I would much prefer to avoid in this way the dicta the Court enunciates with respect to "takings." 1 The exemptions were designed to permit individual debtors to retain exempt property so that they will be able to enjoy a "fresh start" after bankruptcy. Subsections (b) and (d) of § 522 provide in pertinent part: (b) [A]n individual debtor may exempt from the property of the estate . . . (1) property that is specified under subsection (d) of this section. (d) The following property may be exempted under subsection (b)(1) of this section: (3) The debtor's interest, not to exceed $200 in value in any particular item, in household furnishings, household goods, wearing apparel, appliances, books, animals, crops, or musical instruments, that are held primarily for the personal, family or household use of the debtor or a dependent of the debtor. (4) The debtor's aggregate interest, not to exceed $500 in value, in jewelry held primarily for the personal, family, or household use of the debtor or the dependent of the debtor. (6) The debtor's aggregate interest, not to exceed $750 in value, in any implements, professional books, or tools, of the trade of the debtor or the trade of a dependent of the debtor. (9) Professionally described health aids for the debtor or a dependent of the debtor. 2 See 28 U.S.C. § 2403(a). 3 In Schulte v. Beneficial Finance of Kansas, Inc., and Hunter v. Beneficial Finance of Kansas, Inc., 8 B.R. 12, the Bankruptcy Court for the District of Kansas noted that retrospective application of § 522(f)(2) creates constitutional problems and held that it should be applied only prospectively. In Jackson v. Security Industrial Bank, Stevens v. Liberty Loan Corp., 4 B.R. 293, Rodrock v. Security Industrial Bank, and Knezel v. Security Industrial Bank, 3 B.R. 629, the Bankruptcy Court for the District of Colorado concluded that § 522(f)(2), as applied retrospectively, violates the Due Process Clause of the Fifth Amendment. In Hoops v. Freedom Finance, 3 B.R. 635, the Bankruptcy Court for the District of Colorado concluded that § 522(f)(2), as applied retrospectively, violates "substantive due process." 4 In re Gifford, 688 F.2d 447, (CA7 1982) (en banc), holds that § 522(f)(2) constitutionally applies to liens created before the enactment date. In re Webber, 674 F.2d 796 (CA9 1982), holds that § 522(f)(2) constitutionally applies to liens created before the Act became effective but after the enactment date. In re Ashe, 669 F.2d 105 (CA3 1982), holds that § 522(f)(1), which permits avoidance of certain judicial liens, constitutionally applies to a cognovit note created before the enactment date. 5 Appellee Beneficial Finance of Kansas, Inc. asserts that the judgments should be affirmed because the Act violates Article III of the Constitution by granting judicial power to non-Article III bankruptcy judges. See Northern Pipeline Construction Co. v. Marathon Pipe Line Co., --- U.S. ----, 102 S.Ct. 2858, 73 L.Ed.2d 598 (1982) (plurality opinion of Brennan, J.); id., at ----, 102 S.Ct., at 2882 (Rehnquist, J., concurring in the judgment). Because our decision in Northern Pipeline is prospective only, id., at ----, 102 S.Ct. 288, and because we have stayed the issuance of our mandate in that case to December 24, 1982, --- U.S. ----, 103 S.Ct. 99, 73 L.Ed.2d ----, that decision does not affect the judgment in this case. 6 At oral argument the government conceded that the liens at issue in this case are treated as property under state law. Tr. Oral Arg., at 21. Both Kansas and Colorado have adopted the Uniform Commercial Code. Although under the Code the priority among secured parties is often affected by the purchase money or possessory character of security interests, see, e.g., § 9-312, these characterizations do not affect the nature of the security interest. See §§ 9-107 (defining "purchase money security interest"), 9-305 (providing for perfection of security interests by possession). Section 101(28) of the 1978 Act defines a lien as a "charge against or interest in property to secure payment of a debt or performance of an obligation." It does not make distinctions based on the purchase-money or possessory nature of a lien. 7 The Frazier-Lemke Act permitted the farmer, if the mortgagee assented, to purchase the property at its then appraised value on a deferred payment plan. If the mortgagee refused to assent, the court was required to stay all proceedings for 5 years, during which time the farmer could retain possession by paying a reasonable rent. After 5 years the property could be reappraised, but the farmer still had the right to purchase it free and clear for the appraised value regardless of the amount of the lien. See Radford, supra, 295 U.S., at 557-558, 55 S.Ct., at 855. Given the interest rate of 1%, the present value of the deferred payments was much less than the value of the property. Id., at 591-593, 55 S.Ct., at 864-65. 8 Under the Uniform Commercial Code definition, these statutory liens would be nonpossessory, nonpurchase-money liens in personal property. See note 5, supra. 9 The transition provisions of the 1910 statute, id., § 14, 36 Stat. at 842, are, in substance, the same as those of the 1978 Act. Pub.L. No. 95-598, Title IV, §§ 402, 403(a), Nov. 6, 1978, 92 Stat. 2682, 2683. 10 Claridge Apartments Co. v. Commissioner, 323 U.S. 141, 65 S.Ct. 172, 89 L.Ed. 139 (1944), involved rights to certain tax benefits, not to property rights. Dickinson Industrial Site, Inc. v. Cowan, 309 U.S. 382, 383, 60 S.Ct. 595, 596, 84 L.Ed. 819 (1940), dealt with the application of new procedural rules to a bankruptcy proceeding that was pending when the new statute was enacted. Allowing an appeal to the Circuit Court rather than the District Court in that case did not eliminate any property rights. Carpenter v. Wabash Ry., 309 U.S. 23, 60 S.Ct. 416, 84 L.Ed. 558 (1940), involved a provision giving personal injury judgments the status of operating expenses and thus priority over mortgages in ongoing railroad reorganizations. Although that statute may have disadvantaged the mortgagees by reducing the amount of cash available to pay their notes, it did not affect their property right in the collateral securing the mortgages. McFaddin v. Evans-Snider-Buel Co., 185 U.S. 505, 22 S.Ct. 758, 46 L.Ed. 1012 (1902), considered a curative statute providing the methods by which valid mortgages could be created in the Indian Territory. The Legal Tender Cases, 79 U.S. (12 Wall.) 457, 549-550, 20 L.Ed. 287 (1870), decided only that debts could be paid in legal tender as defined by Congress at the time of payment without impairing the obligation of contracts. 11 Because all of the liens at issue in this case were established before the enactment date we have no occasion to consider whether § 522(f)(2) should be applied to liens established after Congress passed the Act, but before it became effective. 12 "When aid to construction of the meaning of words, as used in the statute, is available, there certainly can be no 'rule of law' which forbids its use, however clear the words may appear on 'superficial examination.' . . . Obviously there is danger that the courts' conclusion as to legislative purpose will be unconsciously influenced by the judges' own views or by factors not considered by the enacting body. A lively appreciation of the danger is the best assurance of escape from its threat but hardly justifies an acceptance of a literal interpretation dogma which withholds from the courts available information for reaching a correct conclusion. A few words of general connotation appearing in the text of statutes should not be given a wide meaning, contrary to a settled policy, 'excepting as a different purpose is plainly shown.' " United States v. American Trucking Association, 310 U.S. 534, 543-544, 60 S.Ct. 1059, 1063-64, 84 L.Ed. 1345 (1940) (footnotes omitted).

78

459 U.S. 86 103 S.Ct. 484 74 L.Ed.2d 249 The GILLETTE COMPANY, petitioner,v.Steven MINER No. 81-1493 Supreme Court of the United States December 6, 1982 On writ of certiorari to the Supreme Court of Illinois. PER CURIAM. 1 There being no final judgment, the writ of certiorari is dismissed for want of jurisdiction.

89

459 U.S. 87 103 S.Ct. 416 74 L.Ed.2d 250 Ted W. BROWN et al., Appellantsv.SOCIALIST WORKERS '74 CAMPAIGN COMMITTEE (OHIO) et al. No. 81-776. Argued Oct. 4, 1982. Decided Dec. 8, 1982. Syllabus Held: The disclosure provisions of the Ohio Campaign Expense Reporting Law requiring every candidate for political office to report the names and addresses of campaign contributors and recipients of campaign disbursements, cannot be constitutionally applied to appellee Socialist Workers Party (SWP), a minor political party that historically has been the object of harassment by government officials and private parties. Pp. 91-102. (a) The First Amendment prohibits a State from compelling disclosures by a minor political party that will subject those persons identified to the reasonable probability of threats, harassment, or reprisals. Buckley v. Valeo, 424 U.S. 1, 74, 96 S.Ct. 612, 661, 46 L.Ed.2d 659. Moreover, minor parties must be allowed sufficient flexibility in the proof of injury. Ibid. These principles for safeguarding the First Amendment interests of minor parties and their members and supporters apply not only to the compelled disclosure of campaign contributors but also to the compelled disclosure of recipients of campaign disbursements. Pp. 91-98. (b) Here, the District Court, in upholding appellees' challenge to the constitutionality of the Ohio disclosure provisions, properly concluded that the evidence of private and government hostility toward the SWP and its members establishes a reasonable probability that disclosing the names of contributors and recipients will subject them to threats, harassment, and reprisals. Pp. 98-101. Affirmed. Gary Elson Brown, Columbus, Ohio, for appellants. Thomas D. Buckley, Jr., Cleveland, Ohio, for appellees. Justice MARSHALL delivered the opinion of the Court. 1 This case presents the question whether certain disclosure requirements of the Ohio Campaign Expense Reporting Law, Ohio Rev.Code Ann. § 3517.01 et seq., can be constitutionally applied to the Socialist Workers Party, a minor political party which historically has been the object of harassment by government officials and private parties. The Ohio statute requires every political party to report the names and addresses of campaign contributors and recipients of campaign disbursements. In Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), this Court held that the First Amendment prohibits the government from compelling disclosures by a minor political party that can show a "reasonable probability" that the compelled disclosures will subject those identified to "threats, harassment, or reprisals." Id., at 74, 96 S.Ct., at 661. Employing this test, a three-judge District Court for the Southern District of Ohio held that the Ohio statute is unconstitutional as applied to the Socialist Workers Party. We affirm. 2 * The Socialist Workers Party ("SWP") is a small political party with approximately sixty members in the State of Ohio. The party states in its constitution that its aim is "the abolition of capitalism and the establishment of a workers' government to achieve socialism." As the District Court found, the SWP does not advocate the use of violence. It seeks instead to achieve social change through the political process, and its members regularly run for public office. The SWP's candidates have had little success at the polls. In 1980, for example, the Ohio SWP's candidate for the United States Senate received fewer than 77,000 votes, less than 1.9% of the total vote. Campaign contributions and expenditures in Ohio have averaged about $15,000 annually since 1974. 3 In 1974 appellees instituted a class action1 in the District Court for the Northern District of Ohio challenging the constitutionality of the disclosure provisions of the Ohio Campaign Expense Reporting Law. Ohio Rev.Code Ann. § 3517.01 et seq. The Ohio statute requires every candidate for political office to file a statement identifying each contributor and each recipient of a disbursement of campaign funds. § 3517.10.2 The "object or purpose"3 of each disbursement must also be disclosed. The lists of names and addresses of contributors and recipients are open to public inspection for at least six years. Violations of the disclosure requirements are punishable by fines of up to $1,000 for each day of violation. § 3517.99. 4 On November 6, 1974, the District Court for the Northern District of Ohio entered a temporary restraining order barring the enforcement of the disclosure requirements against the class pending a determination of the merits.4 The case was then transferred to the District Court for the Southern District of Ohio, which entered an identical temporary restraining order in February 1975.5 Accordingly, since 1974 appellees have not disclosed the names of contributors and recipients but have otherwise complied with the statute. A three-judge district court was convened pursuant to 28 U.S.C. § 2281 (1976). Following extensive discovery, the trial was held in February 1981. After reviewing the "substantial evidence of both governmental and private hostility toward and harassment of SWP members and supporters," the three-judge Court concluded that under Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), the Ohio disclosure requirements are unconstitutional as applied to appellees.6 II 5 The Constitution protects against the compelled disclosure of political associations and beliefs. Such disclosures "can seriously infringe on privacy of association and belief guaranteed by the First Amendment." Buckley v. Valeo, supra, 424 U.S., at 64, 96 S.Ct., at 656, citing Gibson v. Florida Legislative Comm., 372 U.S. 539, 83 S.Ct. 889, 9 L.Ed.2d 929 (1963); NAACP v. Button, 371 U.S. 415, 83 S.Ct. 328, 9 L.Ed.2d 405 (1963); Shelton v. Tucker, 364 U.S. 479, 81 S.Ct. 247, 5 L.Ed.2d 231 (1960); Bates v. Little Rock, 361 U.S. 516, 80 S.Ct. 412, 4 L.Ed.2d 480 (1960); NAACP v. Alabama, 357 U.S. 449, 78 S.Ct. 1163, 2 L.Ed.2d 1488 (1958). "Inviolability of privacy in group association may in many circumstances be indispensable to preservation of freedom of association, particularly where a group espouses dissident beliefs." NAACP v. Alabama, supra, 357 U.S., at 462, 78 S.Ct., at 1172. The right to privacy in one's political associations and beliefs will yield only to a "subordinating interest of the State [that is] compelling," NAACP v. Alabama, supra, 357 U.S., at 463, 78 S.Ct., at 1172 (quoting Sweezy v. New Hampshire, 354 U.S. 234, 265, 77 S.Ct. 1203, 1219, 1 L.Ed.2d 1311 (1957) (concurring opinion)), and then only if there is a "substantial relation between the information sought and [an] overriding and compelling state interest." Gibson v. Florida Legislative Comm., supra, 372 U.S., at 546, 83 S.Ct., at 893. 6 In Buckley v. Valeo this Court upheld against a First Amendment challenge the reporting and disclosure requirements imposed on political parties by the Federal Election Campaign Act of 1971. 2 U.S.C. § 431 et seq. (1976). 424 U.S., at 60-74, 96 S.Ct., at 654-661. The Court found three government interests sufficient in general to justify requiring disclosure of information concerning campaign contributions and expenditures:7 enhancement of voters' knowledge about a candidate's possible allegiances and interests, deterrence of corruption, and the enforcement of contribution limitations.8 The Court stressed, however, that in certain circumstances the balance of interests requires exempting minor political parties from compelled disclosures. The government's interests in compelling disclosures are "diminished" in the case of minor parties. Id., at 70, 96 S.Ct., at 659. Minor party candidates "usually represent definite and publicized viewpoints" well known to the public, and the improbability of their winning reduces the dangers of corruption and vote-buying. Ibid. At the same time, the potential for impairing First Amendment interests is substantially greater: "We are not unmindful that the damage done by disclosure to the associational interests of the minor parties and their members and to supporters of independents could be significant. These movements are less likely to have a sound financial base and thus are more vulnerable to falloffs in contributions. In some instances fears of reprisals may deter contributions to the point where the movement cannot survive. The public interest also suffers if that result comes to pass, for there is a consequent reduction in the free circulation of ideas both within and without the political arena." Id., at 71, 96 S.Ct., at 659 (footnotes omitted). 7 We concluded that in some circumstances the diminished government interests furthered by compelling disclosures by minor parties does not justify the greater threat to First Amendment values. 8 Buckley v. Valeo set forth the following test for determining when the First Amendment requires exempting minor parties from compelled disclosures: 9 "The evidence offered need show only a reasonable probability that the compelled disclosure of a party's contributors' names will subject them to threats, harassment, or reprisals from either Government officials or private parties." Id., at 74, 96 S.Ct., at 661. 10 The Court acknowledged that "unduly strict requirements of proof could impose a heavy burden" on minor parties. Ibid. Accordingly, the Court emphasized that "[m]inor parties must be allowed sufficient flexibility in the proof of injury." Ibid. 11 "The proof may include, for example, specific evidence of past or present harassment of members due to their associational ties, or of harassment directed against the organization itself. A pattern of threats or specific manifestations of public hostility may be sufficient. New parties that have no history upon which to draw may be able to offer evidence of reprisals and threats directed against individuals or organizations holding similar views." Ibid. 12 Appellants concede that the Buckley test for exempting minor parties governs the disclosure of the names of contributors, but they contend that the test has no application to the compelled disclosure of names of recipients of campaign disbursements.9 Appellants assert that the state has a substantial interest in preventing the misuse of campaign funds.10 They also argue that the disclosure of the names of recipients of campaign funds will have no significant impact on First Amendment rights, because, unlike a contribution, the mere receipt of money for commercial services does not affirmatively express political support. 13 We reject appellants' unduly narrow view of the minor party exemption recognized in Buckley. Appellants' attempt to limit the exemption to laws requiring disclosure of contributors is inconsistent with the rationale for the exemption stated in Buckley. The Court concluded that the government interests supporting disclosure are weaker in the case of minor parties, while the threat to First Amendment values is greater. Both of these considerations apply not only to the disclosure of campaign contributors but also to the disclosure of recipients of campaign disbursements. 14 Although appellants contend that requiring disclosure of recipients of disbursements is necessary to prevent corruption, this Court recognized in Buckley that this concededly legitimate government interest has less force in the context of minor parties. The federal law considered in Buckley, like the Ohio law at issue here, required campaign committees to identify both campaign contributors and recipients of campaign disbursements. 2 U.S.C. §§ 432(c) and (d), and 434(a) and (b). We stated that "by exposing large contributions and expenditures to the light of publicity," disclosure requirements "ten[d] to 'prevent the corrupt use of money to affect elections.' " Id., at 67, 96 S.Ct., at 657 (emphasis added), quoting Burroughs v. United States, 290 U.S. 534, 548, 54 S.Ct. 287, 291, 78 L.Ed. 484 (1934). We concluded, however, that because minor party candidates are unlikely to win elections, the government's general interest in "deterring the 'buying' of elections" is "reduced" in the case of minor parties. 424 U.S., at 70, 96 S.Ct., at 659.11 15 Moreover, appellants seriously understate the threat to First Amendment rights that would result from requiring minor parties to disclose the recipients of campaign disbursements. Expenditures by a political party often consist of reimbursements, advances, or wages paid to party members, campaign workers, and supporters, whose activities lie at the very core of the First Amendment.12 Disbursements may also go to persons who choose to express their support for an unpopular cause by providing services rendered scarce by public hostility and suspicion.13 Should their involvement be publicized, these persons would be as vulnerable to threats, harassment, and reprisals as are contributors whose connection with the party is solely financial.14 Even individuals who receive disbursements for "merely" commercial transactions may be deterred by the public enmity attending publicity, and those seeking to harass may disrupt commercial activities on the basis of expenditure information.15 Because an individual who enters into a transaction with a minor party purely for commercial reasons lacks any ideological commitment to the party, such an individual may well be deterred from providing services by even a small risk of harassment.16 Compelled disclosure of the names of such recipients of expenditures could therefore cripple a minor party's ability to operate effectively and thereby reduce "the free circulation of ideas both within and without the political arena." Buckley, supra, 424 U.S., at 71, 96 S.Ct., at 659 (footnotes omitted). See Sweezy v. New Hampshire, 354 U.S. 234, 250-251, 77 S.Ct. 1203, 1211-1212, 1 L.Ed.2d 1311 (1957) (plurality opinion) ("Any interference with the freedom of a party is simultaneously an interference with the freedom of its adherents."). 16 We hold, therefore, that the test announced in Buckley for safeguarding the First Amendment interests of minor parties and their members and supporters applies not only to the compelled disclosure of campaign contributors but also to the compelled disclosure of recipients of campaign disbursements. III 17 The District Court properly applied the Buckley test to the facts of this case. The District Court found "substantial evidence of both governmental and private hostility toward and harassment of SWP members and supporters." Appellees introduced proof of specific incidents of private and government hostility toward the SWP and its members within the four years preceding the trial. These incidents, many of which occurred in Ohio and neighboring states, included threatening phone calls and hate mail, the burning of SWP literature, the destruction of SWP members' property, police harassment of a party candidate, and the firing of shots at an SWP office. There was also evidence that in the 12-month period before trial 22 SWP members, including four in Ohio, were fired because of their party membership. Although appellants contend that two of the Ohio firings were not politically motivated, the evidence amply supports the District Court's conclusion that "private hostility and harassment toward SWP members make it difficult for them to maintain employment." 18 The District Court also found a past history of government harassment of the SWP. FBI surveillance of the SWP was "massive" and continued until at least 1976. The FBI also conducted a counterintelligence program against the SWP and the Young Socialist Alliance (YSA), the SWP's youth organization. One of the aims of the "SWP Disruption Program" was the dissemination of information designed to impair the ability of the SWP and YSA to function. This program included "disclosing to the press the criminal records of SWP candidates, and sending anonymous letters to SWP members, supporters, spouses, and employers."17 Until at least 1976, the FBI employed various covert techniques to obtain information about the SWP, including information concerning the sources of its funds and the nature of its expenditurers. The District Court specifically found that the FBI had conducted surveillance of the Ohio SWP and had interfered with its activities within the State.18 Government surveillance was not limited to the FBI. The United States Civil Service Commission also gathered information on the SWP, the YSA, and their supporters, and the FBI routinely distributed its reports to Army, Navy and Air Force Intelligence, the United States Secret Service, and the Immigration and Naturalization Service. 19 The District Court properly concluded that the evidence of private and government hostility toward the SWP and its members establishes a reasonable probability that disclosing the names of contributors and recipients will subject them to threats, harassment, and reprisals.19 There were numerous instances of recent harassment of the SWP both in Ohio and in other states.20 There was also considerable evidence of past government harassment. Appellants challenge the relevance of this evidence of government harassment in light of recent efforts to curb official misconduct. Notwithstanding these efforts, the evidence suggests that hostility toward the SWP is ingrained and likely to continue. All this evidence was properly relied on by the District Court. Buckley, supra, 424 U.S., at 74, 96 S.Ct., at 661. IV 20 The First Amendment prohibits a state from compelling disclosures by a minor party that will subject those persons identified to the reasonable probability of threats, harassment or reprisals. Such disclosures would infringe the First Amendment rights of the party and its members and supporters. In light of the substantial evidence of past and present hostility from private persons and government officials against the SWP, Ohio's campaign disclosure requirements cannot be constitutionally applied to the Ohio SWP. 21 The judgment of the three-judge District Court for the Southern District of Ohio is affirmed. 22 Justice BLACKMUN, concurring in part and concurring in the judgment. 23 I join Parts I, III, and IV of the Court's opinion and agree with much of what is said in Part II. But I cannot agree, with the Court or with the partial dissent, that we should reach the issue whether a standard of proof different from that applied to disclosure of campaign contributions should be applied to disclosure of campaign disbursements. See ante, at 94, n. 9; post, at 112-113, n. 7.1 Appellants did not suggest in the District Court that different standards might apply. Nor was the issue raised in appellants' jurisdictional statement or in their brief on the merits in this Court. Consequently, I would merely assume for purposes of our present decision—as appellants apparently have assumed throughout this litigation and as the District Court clearly assumed—that the flexible proof rule of Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), applies equally to forced disclosure of contributions and to forced disclosure of expenditures. I would leave for another day, when the issue is squarely presented, considered by the courts below, and adequately briefed here, the significant question that now divides the Court. 24 This Court's Rule 15.1(a) states: "Only the questions set forth in the jurisdictional statement or fairly included therein will be considered by the Court." Appellants' jurisdictional statement presented a single question: 25 "Whether, under the standards set forth by this Court in Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), the provisions of Sections 3517.10 and 3517.11 of the Ohio Revised Code, which require that the campaign committee of a candidate for public office file a report disclosing the full names and addresses of persons making contributions to or receiving expenditures from such committee, are consistent with the right of privacy of association guaranteed by the First and Fourteenth Amendments of the Constitution of the United States when applied to the committees of candidates of a minority party which can establish only isolated instances of harassment directed toward the organization or its members within Ohio during recent years." Juris. Statement i. 26 The question assumes the applicability of Buckley to the entire case, and asks this Court to decide only whether the evidence presented to and facts found by the District Court were sufficient to support that court's conclusion that the Buckley test was satisfied. 27 Absent extraordinary circumstances, this Court does not decide issues beyond those it has agreed to review. Mayor v. Educational Equality League, 415 U.S. 605, 623, 94 S.Ct. 1323, 1334, 39 L.Ed.2d 630 (1974); United States v. Bass, 404 U.S. 336, 339, n. 4, 92 S.Ct. 515, 518, n. 4, 30 L.Ed.2d 488 (1971); General Talking Pictures Co. v. Western Electric Co., 304 U.S. 175, 178-179, 58 S.Ct. 849, 851-852, 82 L.Ed. 1273 (1938). According to the Court, however, the issue whether the flexible standard of proof established in Buckley applies to recipients of expenditures is " 'fairly included' in the question presented." Ante, at 94, n. 9. But appellants' failure to present the issue was not a mere oversight in phrasing that question. That appellants did not invoke this Court's jurisdiction to review specifically the proper standard for disclosure of campaign expenditures is also apparent from appellants' arguments in their jurisdictional statement and their brief on the merits. In their jurisdictional statement, under the heading "The Question is Substantial," appellants stated: 28 "The standards governing the resolution of actions involving challenges to reporting requirements by minority parties were set forth by this Court in the case of Buckley v. Valeo, 424 U.S. 1 [96 S.Ct. 612, 46 L.Ed.2d 659] (1976). In Buckley the Court held that in order to receive relief from reporting requirements such as those at issue in this action a minority party must establish '. . . a reasonable probability that the compelled disclosure of a party's contributors' names will subject them to threats, harassment or reprisals from either Government officials or private parties.' 424 U.S., at 74 [96 S.Ct., at 661]." Juris. Statement 10. 29 Appellants went on to state that the flexible standard of proof of injury established in Buckley applied to "disclosure requirements." Juris. Statement 12-13. Similar assertions are found in appellants' brief on the merits. See Brief for Appellants 12 ("Summary of Argument"); id., at 18 ("While refusing to grant minority parties a blanket exemption from financial disclosure requirements, the Court in Buckley established a standard under which they may obtain relief . . . .") 30 Thus, appellants' exclusive theme in the initial presentation of their case here was that the District Court erred in finding that the Buckley standard was satisfied. They did not suggest that the standard was inapplicable, or applied differently, to campaign expenditure requirements. It was not until their reply brief, submitted eight years after this suit was instituted and at a time when appellees had no opportunity to respond in writing, that appellants sought to inject this new issue into the case. See Irvine v. California, 347 U.S. 128, 129, 74 S.Ct. 381, 98 L.Ed. 561 (1954) (plurality opinion of Jackson, J.). In my view, it simply cannot be said that it was "fairly included" in the jurisdictional statement. 31 Moreover, "[w]here issues are neither raised before nor considered [by the court below], this Court will not ordinarily consider them." Adickes v. S.H. Kress & Co., 398 U.S. 144, 147, n. 2, 90 S.Ct. 1598, 1603 n. 2, 26 L.Ed.2d 142 (1970); Lawn v. United States, 355 U.S. 339, 362-363, n. 16, 78 S.Ct. 311, 324-325, n. 16, 2 L.Ed.2d 321 (1958). The District Court did not address the question whether some standard other than that developed in Buckley should apply to disclosure of campaign expenditures. The reason for this was that appellants conceded in the District Court, as they concede here, that the "flexibility in proof of injury" applicable to disclosure of contributors governed the entire case. In their post-trial memorandum, for example, appellants did not even hint that a different standard should govern disclosure of the identities of recipients of expenditures. Instead, they quoted the Buckley test and granted that "evidence of past harassment may be presented by plaintiffs in cases such as the instant one." Defendants' Post-Trial Memorandum 4-5. 32 This case presents no extraordinary circumstances justifying deviation from this Court's Rule 15.1(a) and its long established practice respecting issues not presented below. We have deviated from the rule when jurisdictional issues have been omitted by the parties and lower courts, see, e.g., United States v. Storer Broadcasting Co., 351 U.S. 192, 197, 76 S.Ct. 763, 767, 100 L.Ed. 1081 (1956), or when the Court has noticed "plain error" not assigned, see Brotherhood of Carpenters v. United States, 330 U.S. 395, 412, 67 S.Ct. 775, 784, 91 L.Ed.2d 973 (1947). Obviously, the issue that divides the Court from the partial dissent is not jurisdictional. Nor, as the Court's opinion persuasively demonstrates, is application of the Buckley test to disclosure of campaign disbursements "plain error." Indeed, I consider it quite possible that, after full consideration, the Court would adopt the Buckley standard in this context for the reasons stated by the Court. I also consider it quite possible that, after full consideration, the Court might wish to revise the Buckley standard as applied to campaign disbursements—perhaps to take account of the different types of expenditures covered and their differing impacts on associational rights, or perhaps along the lines suggested in the partial dissent. But this significant constitutional decision should not be made until the question is properly presented so that the record includes data and arguments adequate to inform the Court's judgment. 33 The Court's apparent reliance on Procunier v. Navarette, 434 U.S. 555, 560, n. 6, 98 S.Ct. 855, 859, n. 6, 55 L.Ed.2d 24 (1978), does not provide a rationale for deciding this issue at this time. The petitioner there had included in his petition for certiorari all the questions we eventually decided. Notwithstanding the fact that the Court limited its grant of the petition to a single question, the parties fully briefed the questions on which review had been denied. Deciding those questions, therefore, was neither unwise nor unfair. In this case, in contrast, appellants affirmatively excluded the point at issue in their jurisdictional statement and in their brief on the merits. By failing to raise it until their reply brief, appellants prevented appellees from responding to the argument in writing. There can be no question that, as the Court observes, "our power to decide is not limited by the precise terms of the question presented." Ante, at 421, n. 9 (quoting Procunier v. Navarette, 434 U.S., at 560, n. 6, 98 S.Ct., at 859, n. 6) (emphasis supplied). But Rule 15.1(a) is designed, as a prudential matter, to prevent the possibility that such tactics will result in ill-considered decisions. It is cases like this one that show the wisdom of the Rule. 34 Thus, for purposes of this case, I would assume, as appellants' jurisdictional statement and brief on the merits assume, that the Buckley standard applies to campaign expenditures just as it applies to contributions.2 Appellees presented "specific evidence of past or present harassment of members due to their associational ties, or of harassment directed against the organization itself," sufficient under the rule in Buckley to establish a "reasonable probability" that the Ohio law would trigger "threats, harassment, or reprisals" against contributors. 424 U.S., at 74, 96 S.Ct., at 661. On this basis, I would affirm the judgment of the District Court in its entirety. 35 Justice O'CONNOR, with whom Justice REHNQUIST and Justice STEVENS join, concurring in part and dissenting in part. 36 I concur in the judgment that the Socialist Workers Party ("SWP") has sufficiently demonstrated a reasonable probability that disclosure of contributors will subject those persons to threats, harassment, or reprisals, and thus under Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), the state of Ohio cannot constitutionally compel the disclosure. Further, I agree that the broad concerns of Buckley apply to the required disclosure of recipients of campaign expenditures. But, as I view the record presented here, the SWP has failed to carry its burden of showing that there is a reasonable probability that disclosure of recipients of expenditures will subject the recipients themselves or the SWP to threats, harassment, or reprisals. Moreover, the strong public interest in fair and honest elections outweighs any damage done to the associational rights of the party and its members by application of the state's expenditure-disclosure law. 37 * Buckley upheld the validity of the Federal Election Campaign Act of 1971, which requires the disclosure of names of both contributors to a campaign and recipients of expenditures from the campaign. Buckley recognized three major governmental interests in disclosure requirements: deterrence of corruption; enhancement of voters' knowledge about a candidate's possible allegiances and interests; and provision of the data and means necessary to detect violations of any statutory limitations on contributions or expenditures. The precise challenge that the Buckley Court faced, however, was the overbreadth of the Act's requirements "insofar as they apply to contributions to minor parties and independent candidates." Id., at 68-69, 96 S.Ct., at 658-659 (emphasis added).1 Since the appellants in Buckley did not challenge the application of requirements of disclosure of expenditures to minor parties, the Court had no occasion to consider directly the First Amendment interests of a minor political party in preventing disclosure of expenditures, much less to weigh them against the governmental interests in disclosure. The test adopted by Buckley, quoted by the majority, ante, at 93, reflects this limitation, for it contemplates only assessing possible harassment of contributors, without a word about considering the harassment of recipients of expenditures if their names are disclosed or any effects this harassment may have on the party. 38 This is not to say that Buckley provides no guidance for resolving this claim. I agree with the majority that appellants have overstated their argument in declaring that Buckley has no application to the disclosure of recipients of expenditures. Certainly, Buckley enunciates the general governmental interest in regulating minor parties, who, although unlikely to win, can often affect the outcome of an election. Id., at 70, 96 S.Ct., at 659. Buckley also emphasizes the sensitive associational rights of minor parties. 39 Nevertheless, there are important differences between disclosure of contributors and disclosure of recipients of campaign expenditures—differences that the Buckley Court had no occasion to address, but that compel me to conclude that the balance should not necessarily be calibrated identically. First, unlike the government's interest in disclosure of contributions, its interest in disclosure of expenditures does not decrease significantly for small parties. The Court in Buckley recognized that knowing the identity of contributors would not significantly increase the voters' ability to determine the political ideology of the minor-party candidate, for the stance of the minor-party candidate is usually well known. Ibid.2 Nor would identifying a minor party's contributors further the interest in preventing the "buying" of a candidate, because of the improbability of the minor party candidate's winning the election. Ibid. Thus, these two major government interests in disclosure of contributions are significantly reduced for minor parties.3 40 In sharp contrast, however, the governmental interest in disclosure of expenditures remains significant for minor parties. The purpose of requiring parties to disclose expenditures is to deter improper influencing of voters. Corruption of the electoral process can take many forms: the actual buying of votes; the use of "slush funds;" dirty tricks; and bribes of poll watchers and other election officials. Certainly, a "persuasive" campaign worker on election day can corral voters for his minor-party candidate with even a modest "slush fund."4 Even though such improper practices are unlikely to be so successful as to attract enough votes to elect the minor-party candidate, a minor party, whose short-term goal is merely recognition, may be as tempted to resort to impermissible methods as are major parties, and the resulting deflection of votes can determine the outcome of the election of other candidates.5 The requirement of a full and verifiable report of expenditures is important in deterring such practices, for otherwise the party could hide the improper transactions through an accounting sleight-of-hand.6 41 On the other side of the balance, disclosure of recipients of expenditures will have a lesser impact on a minority party's First Amendment interests than will disclosure of contributors. As the majority states, ante, at 91, the First Amendment interest here is "[t]he right to privacy in one's political associations and beliefs." We have never drawn sharp distinctions between members and contributors, Buckley, 424 U.S., at 66, 96 S.Ct., at 657. As we recognized in Buckley, the privacy rights of contributors are especially sensitive, since many seek to express their political views privately through their pocketbook rather than publicly through other means. Disclosure of contributors directly implicates the contributors' associational rights. 42 The impact on privacy interests arising from disclosure of expenditures is of a quite different—and generally lesser dimension. Many expenditures of the minority party will be for quite mundane purposes to persons not intimately connected with the organization. Payments for such things as office supplies, telephone service, bank charges, printing and photography costs would generally fall in this category. The likelihood that such business transactions would dry up if disclosed is remote at best. Unlike silent contributors, whom disclosure would reveal to the public as supporters of the party's ideological positions, persons providing business services to a minor party are not generally perceived by the public as supporting the party's ideology, and thus are unlikely to be harassed if their names are disclosed. Consequently, the party's associational interests are unlikely to be affected by disclosure of recipients of such expenditures. 43 Other recipients of expenditures may have closer ideological ties to the party. The majority suggests that campaign workers receiving per diem, travel, or room expenses may fit in this category. Ante, at 97, n. 12. It is certainly conceivable that such persons may be harassed or threatened for their conduct. Laws requiring disclosure of recipients of expenditures, however, are not likely to contribute to this harassment. Once an individual has openly shown his close ties to the organization by campaigning for it, disclosure of receipt of expenditures is unlikely to increase the degree of harassment so significantly as to deter the individual from campaigning for the party. Further, in striking the balance, the governmental concerns are greatest precisely for the actions of campaign workers that might improperly influence voters. Thus, whatever marginal deterrence that may arise from disclosure of expenditures is outweighed by the heightened governmental interest. 44 In sum, the heightened governmental interest in disclosure of expenditures and the reduced marginal deterrent effect on associational interests demand a separately focused inquiry into whether there exists a reasonable probability that disclosure will subject recipients or the party itself to threats, harassment, or reprisals.7 II 45 Turning to the evidence in this case, it is important to remember that, even though proof requirements must be flexible, Buckley, supra, at 74, 96 S.Ct., at 661, the minor party carries the burden of production and persuasion to show that its First Amendment interests outweigh the governmental interests. Additionally, the application of the Buckley standard to the historical evidence is most properly characterized as a mixed question of law and fact, for which we normally assess the record independently to determine if it supports the conclusion of unconstitutionality as applied.8 46 Here, there is no direct evidence of harassment of either contributors or recipients of expenditures. Rather, as the majority accurately represents it, the evidence concerns harassment and reprisals of visible party members, including violence at party headquarters and loss of jobs. I concur in the majority's conclusion that this evidence, viewed in its entirety, supports the conclusion that there will be a reasonable probability of harassment of contributors if their names are disclosed. This evidence is sufficiently linked to disclosure of contributors in large part because any person publicly known to support the SWP's unpopular ideological position may suffer the reprisals that this record shows active party members suffer, and the disclosure of contributors may lead the public to presume these people support the party's ideology. 47 In contrast, the record, read in its entirety, does not suggest that disclosure of recipients of expenditures would lead to harassment of recipients or reprisals to the party or its members. Appellees gave no breakdown of the types of expenditures they thought would lead to harassment if disclosed. The record does contain the expenditure statements of the SWP, which itemize each expenditure with its purpose while usually omitting the name and address of the recipient. The majority of expenditures, both in number and dollar amount, are for business transactions such as office supplies, food, printing, photographs, telephone service, and books. There is virtually no evidence that disclosure of the recipients of these expenditures will impair the SWP's ability to obtain needed services.9 Even if we assume that a portion of expenditures went to temporary campaign workers or others whom the public might identify as supporting the party's ideology,10 these persons have already publicly demonstrated their support by their campaign work. There is simply no basis for inferring that such persons would thereafter be harassed or threatened or otherwise deterred from working for the party by virtue of inclusion of their names in later expenditure reports, or that if any such remote danger existed, it would outweigh the concededly important governmental interests in disclosure of recipients of expenditures. 48 It is plain that appellees did not carry their burden of production and persuasion insofar as they challenge the expenditure disclosure provisions. I would therefore uphold the constitutionality of those portions of the Ohio statute that require the SWP to disclose the recipients of expenditures.11 1 The plaintiff class as eventually certified included all SWP candidates for political office in Ohio, their campaign committees and treasurers, and people who contribute to or receive disbursements from SWP campaign committees. The defendants are the Ohio Secretary of State and other state and local officials who administer the disclosure law. 2 Section 3517.10 provides in relevant part: (A) Every campaign committee, political committee, and political party which made or received a contribution or made an expenditure in connection with the nomination or election of any candidate at any election held in this state shall file, on a form prescribed under this section, a full, true, and itemized statement, made under penalty of election falsification, setting forth in detail the contributions and expenditures. . . . * * * * * (B) Each statement required by division (A) of this section shall contain the following information: * * * * * (4) A statement of contributions made or received, which shall include: (a) The month, date, and year of the contribution; (b) The full name and address of each person, including any chairman or treasurer thereof if other than an individual, from whom contributions are received. The requirement of filing the full address does not apply to any statement filed by a state or local committee of a political party, to a finance committee of such committee, or to a committee recognized by a state or local committee as its fund-raising auxiliary. (c) A description of the contribution received, if other than money; (d) The value in dollars and cents of the contribution; (e) All contributions and expenditures shall be itemized separately regardless of the amount except a receipt of a contribution from a person in the sum of twenty-five dollars or less at one social or fund-raising activity. An account of the total contributions from each such social or fund-raising activity shall be listed separately, together with the expenses incurred and paid in connection with such activity. No continuing association which makes a contribution from funds which are derived solely from regular dues paid by members of the association shall be required to list the name or address of any members who paid such dues. (5) A statement of expenditures which shall include: (a) The month, day, and year of expenditure; (b) The full name and address of each person to whom the expenditure was made, including any chairman or treasurer thereof if a committee, association, or group of persons; (c) The object or purpose for which the expenditure was made; (d) The amount of each expenditure. * * * * * . . . . All such statements shall be open to public inspection in the office where they are filed, and shall be carefully preserved for a period of at least six years. If the candidate is running for a state-wide office, the statement shall be filed with the Ohio secretary of state; otherwise, the statement shall be filed with the appropriate county board of elections. § 3517.11(A). 3 Section 3517.10(B)(5)(c). 4 The order restrained various state officials from "applying to or enforcing against plaintiffs . . . the disclosure provisions of the Ohio Campaign Expense Reporting Law and the penalty provision of that law, the effect of which will be to postpone the beginning of any possible period of violation of that law by plaintiffs, . . . until such time as the case is decided by the three judge panel, which is hereby convened." (Citations omitted). 5 Apparently none of the parties throughout the six-year period questioned whether the extended duration of the temporary restraining order conformed to the requirements of Rule 65(b) of the Fed.Rules Civ.Proc. 6 Because it invalidated the Ohio statute as applied to the Ohio SWP, the District Court did not decide appellees' claim that the statute was facially invalid. The Ohio statute requires disclosure of contributions and expenditures no matter how small the amount. Ohio Rev.Code Ann. § 3517.10(B)(4)(e). Appellees contended that the absence of a monetary threshold rendered the statute facially invalid since the compelled disclosure of nominal contributions and expenditures lack a substantial nexus with any claimed government interest. See Buckley v. Valeo, 424 U.S. 1, 82-84, 96 S.Ct. 612, 664-665, 46 L.Ed.2d 659 (1976). The District Court's opinion is unreported. 7 2 U.S.C. §§ 432, 434 and 438 require each political committee to keep detailed records of both contributions and expenditures, including the names of campaign contributors and recipients of campaign disbursements, and to file reports with the Federal Election Commission which are made available to the public. 8 The government interest in enforcing limitations is completely inapplicable in this case, since the Ohio law imposes no limitations on the amount of campaign contributions. 9 We believe that the question whether the Buckley test applies to the compelled disclosure of recipients of expenditures is properly before us. Throughout this litigation Ohio has maintained that it can constitutionally require the SWP to disclose the names of both campaign contributors and recipients of campaign expenditures. In invalidating both aspects of the Ohio statute as applied to the SWP, the District Court necessarily held (1) that the Buckley standard, which permits flexible proof of the reasonable probability of threats, harassment, or reprisals, applies to both contributions and expenditures, and (2) that the evidence was sufficient to show a reasonable probability that disclosure would subject both contributors and recipients to public hostility and harassment. In their jurisdictional statement, appellants appealed from the entire judgment entered below and presented the following question for review: "Whether, under the standards set forth by this Court in Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976), the provisions of Sections 3517.10 and 3517.11 of the Ohio Revised Code, which require that the campaign committee of a candidate for public office file a report disclosing the full names and addresses of persons making contributions to or receiving expenditures from such committee, are consistent with the right of privacy of association guaranteed by the First and Fourteenth Amendments of the Constitution of the United States when applied to the committees of candidates of a minority party which can establish only isolated instances of harassment directed toward the organization or its members within Ohio during recent years." Juris. Statement i. We think that the correctness of both holdings of the District Court is "fairly included" in the question presented in the jurisdictional statement. Sup.Ct. Rule 15.1(a). See Procunier v. Navarette, 434 U.S. 555, 559, n. 6, 98 S.Ct. 855, 858, n. 6, 55 L.Ed.2d 24 (1978) ("[O]ur power to decide is not limited by the precise terms of the question presented."). 10 This is one of three government interests identified in Buckley, supra. Appellants do not contend that the other two interests, enhancing voters' ability to evaluate candidates and enforcing contribution limitations, support the disclosure of the names of recipients of campaign disbursements. 11 The dissent suggests that the government interest in the disclosure of recipients of expenditures is not significantly diminished in the case of minor political parties, since parties with little likelihood of electoral success might nevertheless finance improper campaign activities merely to gain recognition. Post, at 109-110. The dissent relies on Justice WHITE's separate opinion in Buckley, in which he pointed out that "unlimited money tempts people to spend it on whatever money can buy to influence an election." 424 U.S., at 265, 96 S.Ct., at 748 (emphasis in original). An examination of the context in which Justice WHITE made this observation indicates precisely why the state interest here is insubstantial. Justice WHITE was addressing the constitutionality of ceilings on campaign expenditures applicable to all candidates. His point was that such ceilings "could play a substantial role in preventing unethical practices." Ibid. In the case of minor parties, however, their limited financial resources serve as a built-in expenditure ceiling which minimizes the likelihood that they will expend substantial amounts of money to finance improper campaign activities. See id., at 71, 96 S.Ct., at 659. For example, far from having "unlimited money," the Ohio SWP has had an average of roughly fifteen thousand dollars available each year to spend on its election efforts. Most of the limited resources of minor parties will typically be needed to pay for the ordinary fixed costs of conducting campaigns, such as filing fees, travel expenses, and the expenses incurred in publishing and distributing campaign literature and maintaining offices. Thus Justice WHITE's observation that "financing illegal activities is low on the campaign organization's priority list," id., at 265, 96 S.Ct., at 748, is particularly apposite in the case of minor parties. We cannot agree, therefore, that minor parties are as likely as major parties to make significant expenditures in funding dirty tricks or other improper campaign activities. See post, at 110. Moreover, the expenditure by minor parties of even a substantial portion of their limited funds on illegal activities would be unlikely to have a substantial impact. Furthermore, the mere possibility that minor parties will resort to corrupt or unfair tactics cannot justify the substantial infringement on First Amendment interests that would result from compelling the disclosure of recipients of expenditures. In Buckley, we acknowledged the possibility that supporters of a major party candidate might channel money into minor parties to divert votes from other major party contenders, 424 U.S., at 70, and that, as noted by the dissent, post, at 110, and n. 5, occasionally minor parties may affect the outcomes of elections. We thus recognized that the distorting influence of large contributors on elections may not be entirely absent in the context of minor parties. Nevertheless, because we concluded that the government interest in disclosing contributors is substantially reduced in the case of minor parties, we held that minor parties are entitled to an exemption from requirements that contributors be disclosed where they can show a reasonable probability of harassment. Ibid. Because we similarly conclude that the government interest in requiring the disclosure of recipients of expenditures is substantially reduced in the case of minor parties, we hold that the minor party exemption recognized in Buckley applies to compelled disclosure of expenditures as well. 12 For example, the expenditure statements filed by the SWP contain a substantial percentage of entries designated as per diem, travel expenses, room rental, and so on. The Ohio statute makes it particularly easy to identify these individuals since it requires disclosure of the purpose of the disbursements as well as the identity of the recipients. Ohio Rev.Code Ann. § 3517.10(B)(5)(c). 13 "[F]inancial transactions can reveal much about a person's activities, associations, and beliefs." Buckley v. Valeo, supra, 424 U.S., at 66, 96 S.Ct., at 657, quoting California Bankers Assn. v. Shultz, 416 U.S. 21, 78-79, 94 S.Ct. 1494, 1525-1526, 39 L.Ed.2d 812 (1974) (POWELL, J., concurring). The District Court found that the FBI at least until 1976 routinely investigated the financial transactions of the SWP and kept track of the payees of SWP checks. 14 The fact that some or even many recipients of campaign expenditures may not be exposed to the risk of public hostility does not detract from the serious threat to the exercise of First Amendment rights of those who are so exposed. We cannot agree with the dissent's assertion that disclosures of disbursements paid to campaign workers and supporters will not increase the probability that they will be subjected to harassment and hostility. Post, at 111-112. Apart from the fact that individuals may work for a candidate in a variety of ways without publicizing their involvement, the application of a disclosure requirement results in a dramatic increase in public exposure. Under Ohio law a person's affiliation with the party will be recorded in a document that must be kept open to inspection by any one who wishes to examine it for a period of at least six years. Ohio Rev.Code Ann. § 3517.10(B). The preservation of unorthodox political affiliations in public records substantially increases the potential for harassment above and beyond the risk that an individual faces simply as a result of having worked for an unpopular party at one time. 15 See, e.g., Socialist Workers Party v. Attorney General, 458 F.Supp. 895, 904 (SDNY 1978) (FBI interference with SWP travel arrangements and speaker hall rental), vacated on other grounds, 596 F.2d 58 (CA2), cert. denied, 444 U.S. 903, 100 S.Ct. 217, 62 L.Ed.2d 141 (1979). 16 Moreover, it would be hard to think of many instances in which the state interest in preventing vote-buying and improper campaign activities would be furthered by the disclosure of payments for routine commercial services. 17 The District Court was quoting from the Part I of the Final Report of Special Master Judge Breitel in Socialist Workers Party, et al v. The Attorney General of the United States, 73 Civ. 3160 (TPG) (SDNY February 4, 1980), detailing the United States Government's admissions concerning the existence and nature of the Government surveillance of the SWP. 18 The District Court also found the following: "The Government possesses about 8,000,000 documents relating to the SWP, YSA . . . and their members. . . . Since 1960 the FBI has had about 300 informants who were members of the SWP and/or YSA and 1,000 non-member informants. Both the Cleveland and Cincinnati FBI field offices had one or more SWP or YSA member informants. Approximately 2 of the SWP member informants held local branch offices. Three informants even ran for elective office as SWP candidates. The 18 informants whose files were disclosed to Judge Breitel received total payments of $358,648.38 for their services and expenses." (Footnotes omitted). 19 After reviewing the evidence and the applicable law, the District Court concluded: "[T]he totality of the circumstances establishes that, in Ohio, public disclosure that a person is a member of or has made a contribution to the SWP would create a reasonable probability that he or she would be subjected to threats, harassment or reprisals." The District Court then enjoined the compelled disclosures of either contributors' or recipients' names. Although the District Court did not expressly refer in the quoted passage to disclosure of the names of recipients of campaign disbursements, it is evident from the opinion that the District Court was addressing both contributors and recipients. 20 Some of the recent episodes of threats, harassment, and reprisals against the SWP and its members occurred outside of Ohio. Anti-SWP occurrences in places such as Chicago (SWP office vandalized) and Pittsburgh (shot fired at SWP building) are certainly relevant to the determination of the public's attitude toward the SWP in Ohio. In Buckley we stated that "[n]ew parties that have no history upon which to draw may . . . offer evidence of reprisals and threats directed against individuals or organizations holding similar views." 424 U.S., at 74, 96 S.Ct., at 661. Surely the Ohio SWP may offer evidence of the experiences of other chapters espousing the same political philosophy. See 1980 Illinois Socialist Workers Campaign v. State of Illinois Board of Elections, 531 F.Supp. 915, 921 (N.D.Ill.1981). Appellants point to the lack of direct evidence linking the Ohio statute's disclosure requirements to the harassment of campaign contributors or recipients of disbursements. In Buckley, however, we rejected such "unduly strict requirements of proof" in favor of "flexibility in the proof of injury." 424 U.S., at 74, 96 S.Ct., at 661. We thus rejected requiring a minor party to "come forward with witnesses who are too fearful to contribute but not too fearful to testify about their fears" or prove that "chill and harassment [are] directly attributable to the specific disclosures from which the exemption is sought." Ibid. We think that these considerations are equally applicable to the proof required to establish a reasonable probability that recipients will be subjected to threats and harassment if their names are disclosed. While the dissent appears to agree, post, at 112-113 n. 7, its "separately focused inquiry," post, at 112, and n. 7, in reality requires evidence of chill and harassment directly attributable to the expenditure-disclosure requirement. 1 Although the partial dissent agrees that this issue is not properly presented and therefore that the question should not be decided, post, at 112, n. 7, its result and reasoning endorse a different standard of proof. See n. 2, infra. 2 The partial dissent says it agrees that "this is not the appropriate case to determine whether a different test or standard of proof should be employed in determining the constitutional validity of required disclosure of expenditures." Post, at 112, n. 7. If that is so, however, appellees' proof, which the partial dissent agrees established a reasonable probability of threats, harassment, or reprisals against contributors, likewise allowed the District Court to find a reasonable probability of threats, harassment, or reprisals against recipients of expenditures. The Buckley standard permits proof that a particular disclosure creates the requisite likelihood of harassment to be based on a showing of harassment directed at members of the party or at the organization itself. 424 U.S., at 74, 96 S.Ct., at 661. Thus, I do not understand how the partial dissent's "separately focused inquiry" can "plainly require a different result," post, at 113, n. 7, or how it possibly can lead to the conclusion that "appellees did not carry their burden of production and persuasion insofar as they challenge the expenditure disclosure provisions," post, at 115 —unless, despite the partial dissent's uncertain disclaimer, post, at 113, n. 7, its "separate focus" alters Buckley's "reasonable probability" and "flexible proof" standards in the context of expenditures. 1 Of course, the plaintiffs in Buckley challenged many aspects of the federal Act, including expenditure limitations and the disclosure requirements for independent contributions and expenditures. The Court upheld all disclosure requirements, including disclosure of independent expenditures "for communications that expressly advocate the election or defeat of a clearly identified candidate." Id., at 80, 96 S.Ct., at 664. The plaintiffs in Buckley did not challenge, however, the federal requirement that all political parties, including minor political parties, disclose the recipients of their expenditures. 2 Certainly, that is true in this instance. The general political stance of the SWP and its candidates is readily discernible from the most cursory glance at its constitution or literature. 3 The majority is obviously correct in noting that the third governmental interest articulated in Buckley —using disclosures to police limitations on contributions and expenditures—has no application to either contributions or expenditures in Ohio, since the Ohio statute sets no limitations on them. 4 As Justice WHITE noted in partial dissent in Buckley, 424 U.S., at 264-265, 96 S.Ct., at 747-748, citing Burroughs v. United States, 290 U.S. 534, 54 S.Ct. 287, 78 L.Ed. 484 (1934): "[T]he corrupt use of money by candidates is as much to be feared as the corrosive influence of large contributions. There are many illegal ways of spending money to influence elections. One would be blind to history to deny that unlimited money tempts people to spend it on whatever money can buy to influence an election." [Emphasis in original.] 5 Certainly the SWP could have this effect. For example, appellants noted at oral argument that the SWP candidate in the 1974 Ohio gubernatorial election received some 95,000 votes. The Republican candidate's margin of victory over the Democratic candidate was only some 13,500 votes. Tr. of Oral Arg., at 18. The impact of minor parties on elections in the United States is well documented. See generally W. Hesseltine, Third-Party Movements in the United States (1962). 6 I therefore disagree with the majority's suggestion, ante, at 98-99, n. 16, that the government interest in deterring corruption is not furthered by disclosure of all expenditures, including those for commercial services. Even if improprieties are unlikely to occur in expenditures for commercial services, full and verifiable disclosure is needed to ensure that other, improper expenditures are not hidden in commercial accounts. 7 According to the majority, "the question whether the Buckley test applies to the compelled disclosure of recipients of expenditures is properly before us." Ante, at 94, n. 9. The majority declares that, in answering this question, "the District Court necessarily held (1) that the Buckley standard, which permits flexible proof of the reasonable probability of threats, harassment, or reprisals, applies to both contributions and expenditures, and (2) that the evidence was sufficient to show a reasonable probability that disclosure would subject both contributors and recipients to public hostility and harassment." Ibid. (emphasis added). Justice BLACKMUN, ante, at 102, however, more accurately characterizes the District Court's action as assuming that the Buckley standard applies to disclosure of expenditures and holding the evidence sufficient to meet this standard. The District Court's assumption is understandable, since appellants did not question it below. Thus, this is not the appropriate case to determine whether a different test or standard of proof should be employed in determining the constitutional validity of required disclosure of expenditures. Even assuming the general applicability of the Buckley standard, though, the question presented here requires us to inquire whether the evidence of harassment establishes a "reasonable probability" that the Ohio law would trigger "threats, harassment, or reprisals" against recipients of expenditures that in turn may harm the party's associational interests. This inquiry is necessarily distinct from the inquiry whether the evidence establishes a reasonable probability that disclosure would trigger threats, harassment, or reprisals against contributors. Although the proof requirements guiding this separate inquiry remain flexible, and direct proof of harm from disclosure is not required, ultimately the party must prove that the harm to it from disclosure of recipients outweighs the governmental interest in disclosure. This separately focused inquiry does not necessarily alter Buckley's "reasonable probability" test or "flexible proof" standard. It does, however, plainly require a different result. 8 See Pullman-Standard v. Swint, 456 U.S. 273, 289, n. 19, 102 S.Ct. 1781, 1790, n. 19, 72 L.Ed.2d 66 (1982). The majority does not clearly articulate the standard of review it is applying. By determining that the District Court "properly concluded" the evidence established a reasonable probability of harassment, ante, at 100, the majority seems to apply an independent-review standard. 9 The District Court admitted Exhibit 129 into the record, which is a certified copy of Findings of Fact made by the Federal Election Commission pursuant to a 1977 court order in Socialist Workers 1974 National Campaign Committee v. Jennings, No. 74-1338 (D.D.C., stipulated judgment entered Jan. 3, 1979). The FEC in that case analyzed affidavits submitted by SWP members and other documentary evidence of public and private harassment of SWP members. In finding No. 126, the FEC accepted the SWP's proposed finding that in 1971 a landlady in San Francisco rejected the application of two SWP members for an apartment, because the FBI had visited the landlady and warned her of the dangers of the SWP. In finding No. 127, the FEC accepted the SWP's proposed finding that in 1974 a landlady in Chicago evicted a SWP member from her apartment. The landlady explained, "they told me all about you," refusing to identify who "they" were. These two incidents are, of course, remote in time and place, and do not suggest that the party itself has had difficulty in finding office space. Nor do they suggest that the general public is likely to engage in similar activity. Moreover, the FBI's actions against the SWP have long been ended, see Final Report of the Select Committee to Study Government Operations with Respect to Intelligence Activities, S.Rep. No. 755, 94th Cong., 2d Sess., Vol. 4-5, at 3-4 (1976), and Congress has since instituted more rigorous oversight of FBI and other intelligence activities, see 50 U.S.C. § 413 (Supp. IV 1980). An inference from these two incidents that disclosure of recipients of expenditures would increase any difficulty the party might have in obtaining office space would be tenuous, and is plainly outweighed by the "substantial public interest in disclosure," Buckley, supra, at 72, 96 S.Ct., at 660. 10 As the majority notes, ante, at 97, n. 12, some entries in the expenditure forms are designated as per diem, travel expenses, and room rental. At least until 1978, the expenditure statements gave the names of persons receiving per diem funds from the SWP. Apparently, party treasurers and party candidates received per diem payments. There is no evidence that filing these statements with the Ohio Secretary of State caused any harassment of the named persons, and indeed it is highly unlikely that this disclosure would increase the exposure of persons already so publicly identified with the party. 11 In holding a state statute unconstitutional as applied, a court must sever and apply constitutional portions unless the Legislature would not have intended to have applied " 'those provisions which are within its power, independently of that which is not . . .,' " Buckley, supra, at 108, 96 S.Ct., at 677 (severing constitutional portions of Federal Election Campaign Act after holding other portions unconstitutional on their face), quoting Champlin Refining Co. v. Corporation Comm., 286 U.S. 210, 234, 52 S.Ct. 559, 564, 76 L.Ed. 1062 (1932). Clearly, the expenditure disclosure requirements of the Ohio statute should be severed and applied even though the contribution disclosure requirements cannot be applied in this instance, for the two requirements are analytically and practically distinct.

23

459 U.S. 131 103 S.Ct. 514 74 L.Ed.2d 311 BURLINGTON NORTHERN INC. et al., Petitioners,v.UNITED STATES et al. No. 81-1008. Argued Nov. 3, 1982. Decided Dec. 13, 1982. Rehearing Denied Feb. 22, 1983. See U.S., 103 S.Ct. 1238. Syllabus In 1974, San Antonio, Tex., negotiated with petitioner railroads to transport to San Antonio coal purchased under long-term contracts in Wyoming for use in the city's coal-fired electricity generating plants. Because it was not satisfied with the railroads' quoted rate for moving the coal, San Antonio filed a complaint with the Interstate Commerce Commission (ICC). In 1976, the ICC issued a temporary order, subject to modification, establishing a rate of $10.93 per ton. In 1978, on petition of the railroads, the ICC ordered the rate raised to $16.12 per ton. But both San Antonio and the railroads were dissatisfied, and in 1979 the ICC issued a third order resulting in a rate of $17.23 per ton. The railroads then filed tariffs at this rate. Petitions for review of the 1978 and 1979 orders were filed by all parties in the Court of Appeals for the District of Columbia Circuit, which in 1980 decided that both orders were arbitrary and capricious, and accordingly vacated them and remanded to the ICC. The parties disagreed about the effect of this decision on the filed tariffs pending the ICC's decision on remand, the railroads continuing to treat the $17.23 rate as the one San Antonio was required to pay, and San Antonio claiming that the $10.93 rate was revived. The railroads then asked the Court of Appeals for clarification of its decision. Ultimately, after the parties, pending review, had carried on their controversy in other forums, including the ICC, which in 1981 vacated the 1976 order, the Court of Appeals later in 1981 held that since it was without authority to determine interim policy pending remand proceedings in the ICC, the effect of the court's 1980 decision was necessarily to reinstate the 1976 order, which was "revived" by the vacation of the 1978 and 1979 orders, and that therefore tariffs set in excess of the 1976 rate were "unlawful" for the period after the court vacated the 1978 and 1979 orders but before the ICC formally vacated the 1976 order. Held: The Court of Appeals should have deferred to the ICC on questions concerning the applicable rates. Pp. 138-144. (a) Under the Interstate Commerce Act, primary jurisdiction to determine the reasonableness of rates lies with the ICC. Federal court authority to reject ICC rate orders extends to the orders alone and not to the rates. Where there is a dispute about the appropriate rate, the equities favor allowing the carriers' rate to control pending a decision by the ICC, since under the Act the shipper may receive reparation for overpayment while the carrier can never be made whole after underpayment. Pp. 138-142. (b) By declaring that the 1976 rate order was "revived" for the period indicated, the Court of Appeals did what a federal court may not do, i.e., freeze the rate the railroads charge shippers prior to a decision by the ICC as to what a reasonable rate should be. This undermines the ICC's ability to exercise its primary jurisdiction to insure equitable and uniform rates. Moreover, the Court of Appeals' determination requires the railroads to accept a return that was considered temporary when it was approved in 1976, and "below a maximum reasonable rate" when it was modified in 1978. If the court was unsure about the continued vitality of the 1976 order, the more appropriate course would have been to remand to the ICC for explanation rather than to undertake itself to construe the order, and in so doing interfere with the ICC's primary jurisdiction. In striking the 1978 and 1979 orders, the Court's action operated to leave in effect the rates filed under the ICC's authority pending the ICC's redetermination of a reasonable rate and subject to reparations to protect the shipper should the ICC find that these rates were too high. Pp. 142-144. 211 U.S.App.D.C. 111, 655 F.2d 1341, reversed. Robert Eden Martin, Washington, D.C., for petitioners. Elliott Schulder, Washington, D.C., for federal respondent in support of the petitioners. William L. Slover, Washington, D.C., for non-federal respondents. Chief Justice BURGER delivered the opinion of the Court. 1 We granted certiorari to clarify the allocation of authority, as between the federal courts and the Interstate Commerce Commission, to set and review rates for movements of coal by rail. 2 * This case arose as a result of a 1972 decision of San Antonio, Texas, acting through its City Public Service Board, to substitute coal-generated electricity for natural gas. Toward that end, in 1974, San Antonio entered into long-term contracts to purchase coal from two suppliers in Campbell County, Wyoming; began to construct two coal-fired generating units; and initiated negotiations with Burlington Northern, Inc. and Southern Pacific Transportation Company for contracts to transport coal from Wyoming to the new plants. Although the Railroads originally quoted San Antonio a rate of $7.90 per ton for moving coal from Campbell County to San Antonio, economic conditions, which were characterized by rapid inflation, required the Railroads to raise the rate to $11.90 per ton. In May 1975, San Antonio filed a complaint with the Interstate Commerce Commission seeking prescription of a just and reasonable tariff. 3 In October 1976, the Commission rendered a decision, San Antonio, Texas v. Burlington Northern, Inc., 355 ICC 405 (1976) (San Antonio I), establishing a rate of $10.93 per ton for the San Antonio movement. The Commission emphasized that the prescription was temporary by noting that "[t]he public interest requires that, in view of the parties' inability to reach an agreement, a rate be prescribed at this time so that the movement may commence. As actual experience is gained, the parties may petition for modification of the prescription if circumstances warrant." Id., at 417-418. The order was to "continue in full force and effect until the further order of the Commission." Ibid. The Railroads sought review in the United States Court of Appeals for the Eighth Circuit, claiming, inter alia, that the Commission had erred in not considering the Railroad Revitalization and Regulatory Reform Act of 1976, Pub.L. 94-210, 90 Stat. 31 (4-R Act),1 which became effective before San Antonio was announced. The Court of Appeals affirmed the Commission, reasoning that since the rate was temporary and expressly subject to modification, the parties could return to the Commission when guidelines for implementing the 4-R Act were promulgated, Burlington Northern, Inc. v. United States, 555 F.2d 637, 648 (CA8 1977). 4 In June 1977, after six months of operation at the San Antonio I rates, the Railroads petitioned the Commission for a modification of the rate. In October 1977, the Commission reopened the San Antonio proceeding, and one year later, issued a new order, San Antonio, Texas v. Burlington Northern Inc., 359 ICC 1 (1978) (San Antonio II ), finding that when compared to other similar movements, the San Antonio I $10.93 rate was "below a maximum reasonable rate and that modification of that rate [was] warranted." Id., at 7. After making extensive new cost findings and applying the ratemaking guidelines of the 4-R Act, the Commission set the maximum rate level at $16.12 per ton. 5 Both San Antonio and the Railroads were dissatisfied with this rate and petitioned for reconsideration. In June 1979, a third order was issued, San Antonio, Texas v. Burlington Northern Inc., 361 ICC 482 (1979) (San Antonio III ), which made certain modifications in the San Antonio II analysis that resulted in a new maximum rate of $17.23 per ton for the San Antonio movement. The Railroads then filed tariffs at the $17.23 rate. 6 Petitions for review of the San Antonio II and San Antonio III prescriptions were filed in the United States Court of Appeals for the District of Columbia Circuit by all the parties. Without expressing an opinion as to whether the rate was too high, as San Antonio claimed, or too low, as the Railroads urged, in June 1980, the Court of Appeals decided that aspects of both the San Antonio II and the San Antonio III rate orders were "arbitrary and capricious" and "without defensible rationale." San Antonio v. United States, 203 U.S.App.Div. 249, 269, 631 F.2d 831, 851 (1980). The Commission's orders were vacated and the case remanded to the Commission. 7 It is at this point that the present controversy arose, for the parties sharply disagreed about the effect of the Court of Appeals decision on the filed tariffs pending the Commission's decision on remand. Construing the decision as vacating only the Commission's orders in San Antonio II and III but not the rates that were filed, the Railroads continued to treat the $17.23 rate as the one which San Antonio was required to pay pursuant to 49 U.S.C. (Supp.III) § 10761. San Antonio, on the other hand, interpreted the Court of Appeals decision as vacating the $17.23 rate and reviving the rate set by San Antonio I. Accordingly, the shipper unilaterally reduced its payments to the $10.93 per ton rate set in 1976.2 8 Although we might have thought otherwise, it was not clear to the Railroads what legal action should be taken to force San Antonio to pay the filed $17.23 tariff. Several maneuvers were attempted: in its first effort to reestablish San Antonio III as the rate applicable to this period, the carriers filed a new tariff in early November 1980. That tariff, which would have required San Antonio to prepay at the $17.23 rate before coal service would be provided, was suspended by a division of the Commission which agreed with San Antonio that the Court of Appeals' decision precluded any rate except $10.93. 9 The Railroads asked the Court of Appeals for clarification of its decision.3 Pending review, however, the parties carried on their controversy in other forums. The Railroads again attempted to file a tariff in conformity with San Antonio III. Although this time the tariff was not suspended or rejected by the Commission, San Antonio continued to pay at the San Antonio I rate even after the new tariff's December 1980 effective date; in addition, it filed a complaint to enforce the San Antonio I rate in the United States District Court for the Western District of Texas. Before the District Court could rule, the Railroads countered by filing a petition asking the Commission to clarify its refusal to suspend or reject the new tariff by declaring that this action amounted to a modification of San Antonio I. In addition, the carriers filed a second prepayment tariff—which was also accepted by the Commission. Before the Commission could react to the Railroad's request for clarification, however, the Texas District Court ruled in San Antonio's favor on an application to preliminarily enjoin the Railroads from conditioning service on prepayment of rates that did not conform with San Antonio I. The Railroads appealed to the Court of Appeals for the Fifth Circuit. 10 In April 1981, while the Railroads' appeal was pending in the Fifth Circuit, the Commission finally took the step necessary to end the controversy over what rate applied from the time of the June 1980 decision of the Court of Appeals for the District of Columbia Circuit. In the context of considering the Railroads' request for clarification, the Commission formally vacated its San Antonio I prescription. The order stated that in a later proceeding, the Commission would determine "what the maximum reasonable rate should have been . . . for the period during which the vacated maximum rate prescriptions in San Antonio II and San Antonio III were in effect." San Antonio, Texas v. Burlington Northern, Inc., 364 ICC 887, 894 (1981) (San Antonio IV ). Pursuant to § 10327(h), this order became effective 30 days later, in May 1981. 11 It was at this point that the Fifth Circuit decided the Railroads' appeal of the Texas District Court decision. In its holding, that court vacated the preliminary injunction on the ground that only the Commission had jurisdiction to enjoin railroads from collecting their filed tariff rate. In addition, that court denied an application by San Antonio for a stay of the Commission's San Antonio IV decision, San Antonio v. Burlington Northern, Inc., 650 F.2d 49 (CA5), clarified, 652 F.2d 422 (5th Cir.1981). Thus, when the Commission's San Antonio IV decision became effective in May 1981, San Antonio finally began to pay for the shipment of its coal at the carriers' tariff rate of $17.23 per ton.4 12 One month later, on June 30, 1981, the Court of Appeals for the District of Columbia issued the clarification of its 1980 holding. It is this clarification that is under review here. Citing Consolidated Rail Corp. v. National Association of Recycling Industries, Inc., 449 U.S. 609, 101 S.Ct. 775, 66 L.Ed.2d 776 (1981) (per curiam ), and Atchison, T. & S.F. R. Co. v. Wichita Board of Trade, 412 U.S. 800, 93 S.Ct. 2367, 37 L.Ed.2d 350 (1973), the Court of Appeals held that since it was without authority to determine interim policy pending remand proceedings in the Commission, the effect of the court's 1980 decision was necessarily to reinstate San Antonio I, which was "revived" by the vacation of San Antonio II and III. 655 F.2d, at 1344. Tariffs set in excess of the San Antonio I rate were therefore declared "unlawful" for the period after the court vacated San Antonio II and III but before the Commission formally vacated San Antonio I. Id., at 1343. We granted certiorari. 455 U.S. 988, 102 S.Ct. 1609, 71 L.Ed.2d 847 (1982).5 13 We agree that Consolidated Rail and Wichita Board of Trade control this case, but these holdings require federal courts to defer to the Commission on questions concerning the applicable rates; accordingly, we reverse. II 14 In recent years, we have had four occasions to consider federal courts' authority to alter rail rates regulated by the Interstate Commerce Act. In the first of these, Arrow Transportation Co. v. Southern R. Co., 372 U.S. 658, 83 S.Ct. 984, 10 L.Ed.2d 52 (1963), a railroad faced with declining revenues had attempted to lower its rates and the issue before us was whether a federal district court had the power to enjoin this reduction at the request of competitors of the railroad and those who shipped by rail. Affirming the District Court's denial of an injunction, we held that Congress, in the Interstate Commerce Act, meant to "vest in the Commission the sole and exclusive power to suspend" the rates. 372 U.S., at 667, 83 S.Ct., at 988. 15 We noted several reasons for this rule. First, a review of the legislative history of the 1910 amendments to the Interstate Commerce Act demonstrated that Congress was dissatisfied with the nonuniformity in rates and inequities that resulted from the 1887 Interstate Commerce Act's failure to give the Commission power to grant injunctive relief. We noted that the authority to suspend rates granted the Commission by the 1910 amendment would not cure the problem unless the suspension power was exclusive. Id., at 664, 83 S.Ct., at 987. 16 Second, we held that court-ordered injunctive relief would interfere with the careful way in which the Commission's suspension power takes into account the need of the carrier to receive a reasonable rate of return, and the desire of the shipper to pay only what is lawful. Unlike an injunction, a suspension order is limited to seven-months' duration. Id., at 665-666, 83 S.Ct., at 987-88. The shippers, on the other hand, are fully protected by the reparation provision which requires carriers to reimburse shippers if the Commission later determines that the filed tariff was unreasonable. Id., at 666, 83 S.Ct., at 988. 17 Finally, we emphasized that court-ordered injunctions were inconsistent with the congressional intent to vest ratemaking decisions in the Commission, stating: 18 "Congress meant to foreclose a judicial power to interfere with the timing of rate changes which would be out of harmony with the uniformity of rate levels fostered by the doctrine of primary jurisdiction." Id., at 668, 83 S.Ct., at 990. (Emphasis in original.) 19 Ten years later, we again considered a federal court's power to enjoin rail rates in United States v. SCRAP, 412 U.S. 669, 93 S.Ct. 2405, 37 L.Ed.2d 254 (1973). There we reversed a three-judge District Court that had enjoined the Commission from permitting surcharges on shipments of recycled goods. We rejected the argument that injunctive relief could be granted under authority conferred by the National Environmental Policy Act, 42 U.S.C. § 4331 et seq., stating that "to grant an injunction in the present context, even though not based upon a substantive consideration of the rates, would directly interfere with the Commission's decision as to when the rates were to go into effect, and would ignore our conclusion in Arrow. . . ." 412 U.S., at 697, 93 S.Ct., at 2421. (First emphasis added; others in original.) 20 A third case, Wichita Board of Trade, supra, stated our position in even stronger terms. There the Commission had approved certain rate increases but failed, in the District Court's view, to explain its reasoning adequately. In addition to vacating the order and remanding the case for reconsideration by the Commission, the District Court enjoined the railroads from charging the rates that had been approved in the order. Although we affirmed the remand to the Commission, we nevertheless reversed as to the injunction, reiterating the views we expressed in Arrow that a federal court has no jurisdiction to enter an order that operates to fix rates. 21 "The only consequence of suspending [an] order is that the railroads may not rely, in some subsequent proceeding, on a Commission finding that the proposed rates were just and reasonable. . . . 22 "Carriers may put into effect any rate that the Commission has not declared unreasonable. . . . Suspension of the Commission's order thus does not in itself preclude the carriers from implementing a new rate." 412 U.S., at 818-819, 93 S.Ct., at 2380-81. (Emphasis added.) 23 Again we noted that Congress channeled all rate decisions to the Commission in the first instance, id., at 820, 93 S.Ct., at 2381; that court-ordered relief interferes with the delicate balance the Act strikes between the competing interests of shipper and carrier, id.; and that the equities favor allowing the railroads to charge more than the Commission may ultimately find reasonable because the Act gives the shippers a right to reparations while no such protection is given to the carriers, id., at 823, 93 S.Ct., at 2383. 24 We now turn to our recent holding in Consolidated Rail, supra, which both parties appear to concede states the controlling law. There the Commission fixed rates for recycled materials. On review, the Court of Appeals revoked the rate increases, remanded to the Commission to determine a rate structure incorporating the standards set forth in the 4-R Act, and enjoined new rates until after the Commission's reconsideration. In reversing this holding summarily, we held: 25 "The authority to determine when any particular rate should be implemented is a matter which Congress has placed squarely in the hands of the Commission. Arrow Transportation Co. v. Southern R. Co., 372 U.S. 658, 662-672 [83 S.Ct. 984, 986-91, 10 L.Ed.2d 52] (1963). . . . [T]here is no basis in our prior decisions for the revocation order or for the injunction against further increases. 'If a reviewing court cannot discern [the Commission's] policies, it may remand the case to the agency for clarification and further justification. . . . When a case is remanded on the ground that the agency's policies are unclear, an injunction ordinarily interferes with the primary jurisdiction of the Commission.' Atchison, T. & S.F. R. Co. v. Wichita Board of Trade, 412 U.S. 800, 822 [93 S.Ct. 2367, 2382, 37 L.Ed.2d 350] (1973) . . . ." 449 U.S., at 612, 101 S.Ct., at 777. (Emphasis added.) 26 To recapitulate, our cases stand for three propositions: (1) under the Interstate Commerce Act, primary jurisdiction to determine the reasonableness of rates lies with the Commission, see also Arizona Grocery Co. v. Atchison, T. & S.F. R. Co., 284 U.S. 370, 384, 52 S.Ct. 183, 184, 76 L.Ed. 348 (1932); (2) federal court authority to reject Commission rate orders for whatever reason extends to the orders alone, and not to the rates themselves, cf. 28 U.S.C. § 2349 ("[t]he Court of Appeals . . . has jurisdiction to . . . make a judgment determining the validity of, and enjoining . . . the order of the agency") (emphasis added); (3) where there is a dispute about the appropriate rate, the equities favor allowing the carrier's rate to control pending decision by the Commission, since under the Act, the shipper may receive reparation for overpayment while the carrier can never be made whole after underpayment. 49 U.S.C. § 11705(b)(3).6 Cf. Atlantic Coast Line R. v. Florida, 295 U.S. 301, 55 S.Ct. 713, 79 L.Ed. 1451 (1935). III 27 We can discern no basis to distinguish this case from Arrow, SCRAP, Wichita Board of Trade, and Consolidated Rail, supra. By entering an order declaring that the San Antonio I rate order was "revived" for the period June 1980-May 1981, the Court of Appeals did that which we have said a federal court may not do: i.e., freeze the rate that railroads charge shippers prior to a decision by the Commission as to what a reasonable rate should be. That approach undermines the Commission's ability to exercise the primary jurisdiction delegated to it by Congress to insure equitable and uniform rates. More important, the determination requires the Railroads to accept a return that was considered temporary when it was approved in 1976, and "below a maximum reasonable rate" when it was modified in 1978. This result would be inequitable in the best of times, but the impact is particularly acute in a period of high inflation and changing regulatory standards.7 28 Because the reparations provisions do not apply to both shippers and carriers, losses suffered by the carriers cannot be recovered. Carriers are not adequately protected by their authority under §§ 10761 and 10762 to file a new rate or their right under § 10327(g) to petition the Commission to modify its "revived" rate order, as San Antonio urges. It is arguable—and in other proceedings, San Antonio has so claimed, see Brief for Petitioner 38-39—that before either action can take effect, the party adversely affected may ask for a hearing pursuant to Arizona Grocery, supra. A plenary hearing necessarily causes delay, and even if it did not, action by the Commission usually will not be effective until 30 days have elapsed after its order is served, § 10327(h). 29 The claim is made that the Court of Appeals was powerless to achieve a different result because, under § 10704(a)(1), the only rate the Railroads could legally charge was the rate prescribed by the Commission. Since the Commission prescribed a rate in San Antonio I, the argument is that this is the rate the Railroads must charge. We disagree. San Antonio I was by its terms limited to "continue in full force and effect until . . . further order of the Commission," 355 ICC 418. Absent a contrary indication from the Commission, San Antonio II terminated the vitality of San Antonio I.8 30 Moreover, if the court was unsure about the continued vitality of San Antonio I, the more appropriate course would have been to remand to the Commission for explanation rather than to undertake itself to construe the order, and in so doing to interfere with the Commission's primary jurisdiction, contrary to important congressional policies.9 31 The existence of a 1976 rate prescription does not require a result different from the result reached in Consolidated Rail. San Antonio II and III each in turn vacated the prescription which preceded it. In striking the orders in San Antonio II and III, the court's action operated to leave in effect the rates filed under the Commission's authority pending the Commission's redetermination of a reasonable rate and subject always to reparations to protect the shipper should the Commission find that these rates were too high.10 32 The June 30, 1981 judgment of the Court of Appeals is 33 Reversed. 1 The 4-R Act changed the regulatory atmosphere in several key respects. Especially relevant here is § 205, which, as codified at 49 U.S.C. (Supp.III) § 10704(a)(2), instructs the Commission to "make an adequate and continuing effort to assist . . . carriers in attaining revenue levels" that are "adequate, under honest, economical and efficient management, to cover total operating expenses . . . plus a reasonable and economic profit or return (or both) on capital employed in the business." 2 For convenience, we continue to refer to the rates as "San Antonio I," "San Antonio II," and "San Antonio III." In actual fact, general rate increases, which are not in issue here, have taken effect significantly raising each of these rates. See Brief for Petitioner 9, n. 3. 3 Initially, the Commission took the position adopted by the panel, namely that the Court of Appeals' decision required the Railroads to charge at San Antonio I rates. While the petition for clarification was pending, however, our decision in Consolidated Rail Corp. v. Nat'l Ass'n of Recycling Industries, Inc., 449 U.S. 609, 101 S.Ct. 775, 66 L.Ed.2d 776 (1981) (per curiam ), was handed down. At about this time, the Commission revised its view to espouse the Railroad's position. The Federal Government has thus joined the Railroads in asking us to overturn the decision of the Court of Appeals. 4 In the period in dispute, from June 1980, when the Court of Appeals vacated the San Antonio II and III orders, to May 1981, when the Commission formally vacated the San Antonio I prescription, San Antonio's failure to pay the tariff rate resulted in a savings to it—and a loss to the Railroads—of over $19 million. See Brief for Federal Respondents 6. 5 San Antonio argues that the Railroads' failure to petition for certiorari within 90 days after rehearing was denied on the June 1980 judgment deprives this Court of jurisdiction. Because the June 1981 decision "resolve[d] a genuine ambiguity in a judgment previously rendered" and dealt with a question which was not "plainly and properly settled with finality," FTC v. Minneapolis-Honeywell Regulator Co., 344 U.S. 206, 211-212, 73 S.Ct. 245, 248-49, 97 L.Ed. 245 (1952), we plainly have jurisdiction. 6 Under § 207(d)(2) of the Staggers Rail Act of 1980, Pub.L. 96-448, 94 Stat.1906, 49 U.S.C. § 10707(d)(2), the carrier can also receive reparations. This right is limited, however, to underpayments resulting from the Commission's suspension of a tariff; it does not apply where, as here, a court has prevented the carrier from collecting a higher tariff. 7 See, e.g., 4-R Act, discussed in note 1, supra; Staggers Rail Act, Pub.L. 96-448, 94 Stat.1895, supra. Both statutes are directly relevant in the determination of a reasonable rate for the San Antonio coal movement; neither was considered in San Antonio I. 8 San Antonio makes much of the dictionary definitions of "modify" and "vacate." While ordinary meanings are not insignificant in statutory construction, San Antonio has not cited a single case under the Interstate Commerce Act making this distinction. 9 Another way in which the Court of Appeals might have minimized interference with congressional objectives, would have been to construe its own opinion as vacating only the Commission's new rate calculations and not the Commission's conclusion that the San Antonio I rate was too low. See 28 U.S.C. § 2349, allowing the court to enjoin or set aside "in whole or part, the order of the agency." Cf. Wichita Board of Trade, supra, 412 U.S., at 822, 93 S.Ct., at 2382. 10 Because we find that Consolidated Rail mandates this result, we need not reach the Railroad's claim that the decision of the Court of Appeals is inconsistent with the filed rate doctrine.

89

459 U.S. 116 103 S.Ct. 505 74 L.Ed.2d 297 John P. LARKIN et al., Appellants,v.GRENDEL'S DEN, INC. No. 81-878. Argued Oct. 4, 1981. Decided Dec. 13, 1982. Syllabus A Massachusetts statute (§ 16C) vests in the governing bodies of schools and churches the power to prevent issuance of liquor licenses for premises within a 500-foot radius of the church or school by objecting to the license applications. Appellee restaurant operator's application for a liquor license was denied when a church located 10 feet from the restaurant objected to the application. Appellee then sued the licensing authorities in Federal District Court, claiming that § 16C, on its face and as applied, violated, inter alia, the Establishment Clause of the First Amendment. The District Court held that § 16C is facially unconstitutional under the Establishment Clause, and the Court of Appeals affirmed. Held: Section 16C violates the Establishment Clause. Pp. 120-127. (a) Section 16C is not simply a legislative exercise of zoning power but delegates to private, nongovernmental entities power to reject certain liquor license applications, a power ordinarily vested in governmental agencies. Under these circumstances, the deference normally due a legislative zoning judgment is not merited. Pp. 120-122. (b) The valid secular objective of § 16C in protecting schools and churches from the commotion associated with liquor outlets may readily be accomplished by other means. Pp. 123-124. (c) The churches' power under § 16C is standardless, calling for no reasons, findings, or reasoned conclusions, and can be seen as having a "primary" and "principal" effect of advancing religion. Pp. 125-126. (d) Section 16C substitutes the unilateral and absolute power of a church for the reasoned decisionmaking of a public legislative body acting on evidence and guided by standards on issues with significant economic and political implications, and thus enmeshes churches in the processes of government and creates the danger of "[p]olitical fragmentation and divisiveness along religious lines," Lemon v. Kurtzman, 403 U.S. 602, 623, 91 S.Ct. 2105, 2116, 29 L.Ed.2d 745. Few entanglements could be more offensive to the spirit of the Constitution. Pp.126-127 662 F.2d 102 (1st Cir., 1981), affirmed. Gerald J. Caruso, Boston, Mass., for appellants, pro hac vice, by special leave of Court. Laurence H. Tribe, Cambridge, Mass., for appellee. Chief Justice BURGER delivered the opinion of the Court. 1 The question presented by this appeal is whether a Massachusetts statute, which vests in the governing bodies of churches and schools the power effectively to veto applications for liquor licenses within a five hundred foot radius of the church or school, violates the Establishment Clause of the First Amendment or the Due Process Clause of the Fourteenth Amendment. 2 * A. 3 Appellee operates a restaurant located in the Harvard Square area of Cambridge, Massachusetts. The Holy Cross Armenian Catholic Parish is located adjacent to the restaurant; the back walls of the two buildings are ten feet apart. In 1977, appellee applied to the Cambridge License Commission for approval of an alcoholic beverages license for the restaurant. 4 Section 16C of Chapter 138 of the Massachusetts General Laws provides: "Premises . . . located within a radius of five hundred feet of a church or school shall not be licensed for the sale of alcoholic beverages if the governing body of such church or school files written objection thereto."1 5 Holy Cross Church objected to appellee's application, expressing concern over "having so many licenses so near" (emphasis in original).2 The License Commission voted to deny the application, citing only the objection of Holy Cross Church and noting that the church "is within 10 feet of the proposed location." 6 On appeal, the Massachusetts Alcoholic Beverages Control Commission upheld the License Commission's action. The Beverages Control Commission found that "the church's objection under Section 16C was the only basis on which the [license] was denied." 7 Appellee then sued the License Commission and the Beverages Control Commission in United States District Court. Relief was sought on the grounds that § 16C, on its face and as applied, violated the Equal Protection and Due Process Clauses of the Fourteenth Amendment, the Establishment Clause of the First Amendment, and the Sherman Act. 8 The suit was voluntarily continued pending the decision of the Massachusetts Supreme Judicial Court in a similar challenge to § 16C, Arno v. Alcoholic Beverages Control Commission, 377 Mass. 83, 384 N.E.2d 1223 (1979). In Arno, the Massachusetts court characterized § 16C as delegating a "veto power" to the specified institutions, 377 Mass., at 89, 384 N.E.2d 1223, but upheld the statute against Due Process and Establishment Clause challenges. Thereafter, the District Court denied appellants' motion to dismiss. 9 On the parties' cross-motions for summary judgment, the District Court declined to follow the Massachusetts Supreme Judicial Court's decision in Arno, supra. The District Court held that § 16C violated the Due Process Clause and the Establishment Clause and held § 16C void on its face, 495 F.Supp. 761 (Mass.1980). The District Court rejected appellee's equal protection arguments, but held that the state's actions were not immune from antitrust review under the doctrine of Parker v. Brown, 317 U.S. 341, 63 S.Ct. 307, 87 L.Ed. 315 (1942). It certified the judgment to the First Circuit Court of Appeals pursuant to 28 U.S.C. § 1292, and the Court of Appeals accepted certification. 10 A panel of the First Circuit, in a divided opinion, reversed the District Court on the Due Process and Establishment Clause arguments, but affirmed its antitrust analysis, 662 F.2d 88 (CA1 1981). 11 Appellee's motion for rehearing en banc was granted and the en banc court, in a divided opinion, affirmed the District Court's judgment on Establishment Clause grounds without reaching the due process or antitrust claims, 662 F.2d 102 (CA1 1981). B 12 The Court of Appeals noted that appellee does not contend that § 16C lacks a secular purpose, and turned to the question of "whether the law 'has the direct and immediate effect of advancing religion' as contrasted with 'only a remote and incidental effect advantageous to religious institutions,' " 662 F.2d, at 104 (emphasis in original), quoting Committee for Public Education v. Nyquist, 413 U.S. 756, 783 n. 39, 93 S.Ct. 2955, 2971 n. 39, 37 L.Ed.2d 948 (1973). The court concluded that § 16C confers a direct and substantial benefit upon religions by "the grant of a veto power over liquor sales in roughly one million square feet . . . of what may be a city's most commercially valuable sites," 662 F.2d, at 105. 13 The court acknowledged that § 16C "extends its benefit beyond churches to schools," but concluded that the inclusion of schools "does not dilute [the statute's] forbidden religious classification," since § 16C does not "encompass all who are otherwise similarly situated to churches in all respects except dedication to 'divine worship.' " Id., at 106-107 (footnote omitted). In the view of the Court of Appeals, this "explicit religious discrimination," id., at 105, provided an additional basis for its holding that § 16C violates the Establishment Clause. 14 The court found nothing in the Twenty-First Amendment to alter its conclusion, and affirmed the District Court's holding that § 16C is facially unconstitutional under the Establishment Clause of the First Amendment. 15 We noted probable jurisdiction, 454 U.S. 1140, 102 S.Ct. 996, 71 L.Ed.2d 291 (1982), and we affirm. II A. 16 Appellants contend that the State may, without impinging on the Establishment Clause of the First Amendment, enforce what it describes as a "zoning" law in order to shield schools and places of divine worship from the presence nearby of liquor dispensing establishments. It is also contended that a zone of protection around churches and schools is essential to protect diverse centers of spiritual, educational and cultural enrichment. It is to that end that the State has vested in the governing bodies of all schools, public or private, and all churches,3 the power to prevent the issuance of liquor licenses for any premises within 500 feet of their institutions. 17 Plainly schools and churches have a valid interest in being insulated from certain kinds of commercial establishments, including those dispensing liquor. Zoning laws have long been employed to this end, and there can be little doubt about the power of a state to regulate the environment in the vicinity of schools, churches, hospitals and the like by exercise of reasonable zoning laws. 18 We have upheld reasonable zoning ordinances regulating the location of so-called "adult" theaters, see Young v. American Mini Theatres, Inc., 427 U.S. 50, 62-63, 96 S.Ct. 2440, 2448, 49 L.Ed.2d 310 (1976); and in Grayned v. City of Rockford, 408 U.S. 104, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972), we recognized the legitimate governmental interest in protecting the environment around certain institutions when we sustained an ordinance prohibiting willfully making, on grounds adjacent to a school, noises which are disturbing to the good order of the school sessions. 19 The zoning function is traditionally a governmental task requiring the "balancing [of] numerous competing considerations," and courts should properly "refrain from reviewing the merits of [such] decisions, absent a showing of arbitrariness or irrationality." Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 265, 97 S.Ct. 555, 563, 50 L.Ed.2d 450 (1977). See also, e.g., Village of Belle Terre v. Boraas, 416 U.S. 1, 7-9, 94 S.Ct. 1536, 1540-1541, 39 L.Ed.2d 797 (1974). Given the broad powers of states under the Twenty-First Amendment, judicial deference to the legislative exercise of zoning powers by a city council or other legislative zoning body is especially appropriate in the area of liquor regulation. See, e.g., California v. Larue, 409 U.S. 109, 93 S.Ct. 390, 34 L.Ed.2d 342 (1972); California Retail Liquor Dealers Association v. Midcal Aluminum, Inc., 445 U.S. 97, 106-110, 100 S.Ct. 937, 943-945, 63 L.Ed.2d 233 (1980). 20 However, § 16C is not simply a legislative exercise of zoning power. As the Massachusetts Supreme Judicial Court concluded, § 16C delegates to private, nongovernmental entities power to veto certain liquor license applications, Arno v. Alcoholic Beverages Control Commission, supra, 377 Mass., at 89, 384 N.E.2d 1223.4 This is a power ordinarily vested in agencies of government. See, e.g., California v. LaRue, supra, 409 U.S., at 116, 93 S.Ct., at 396, commenting that a "state agency . . . is itself the repository of the State's power under the Twenty-First Amendment." We need not decide whether, or upon what conditions, such power may ever be delegated to nongovernmental entities; here, of two classes of institutions to which the legislature has delegated this important decisionmaking power, one is secular, but one is religious. Under these circumstances, the deference normally due a legislative zoning judgment is not merited.5 B 21 The purposes of the First Amendment guarantees relating to religion were twofold: to foreclose state interference with the practice of religious faiths, and to foreclose the establishment of a state religion familiar in other Eighteenth Century systems. Religion and government, each insulated from the other, could then coexist. Jefferson's idea of a "wall," see Reynolds v. United States, 98 U.S. (8 Otto) 145, 164, 25 L.Ed. 244 (1878), quoting Reply from Thomas Jefferson to an address by a committee of the Danbury Baptist Association (January 1, 1802), reprinted in 8 Works of Thomas Jefferson 113 (Washington ed. 1861), was a useful figurative illustration to emphasize the concept of separateness. Some limited and incidental entanglement between church and state authority is inevitable in a complex modern society, see, e.g., Lemon v. Kurtzman, 403 U.S. 602, 614, 91 S.Ct. 2105, 2112, 29 L.Ed.2d 745 (1971); Walz v. Tax Commission, 397 U.S. 664, 670, 90 S.Ct. 1409, 1412, 25 L.Ed.2d 697 (1970), but the concept of a "wall" of separation is a useful signpost. Here that "wall" is substantially breached by vesting discretionary governmental powers in religious bodies. 22 This Court has consistently held that a statute must satisfy three criteria to pass muster under the Establishment Clause: 23 First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion . . .; finally, the statute must not foster "an excessive government entanglement with religion." 24 Lemon v. Kurtzman, supra, 403 U.S., at 612-613, 91 S.Ct., at 2111, quoting Walz v. Tax Commission, supra, 397 U.S., at 674, 90 S.Ct., at 1414. See also Widmar v. Vincent, 454 U.S. 263, 271, 102 S.Ct. 269, 275, 70 L.Ed.2d 440 (1981); Wolman v. Walter, 433 U.S. 229, 236, 97 S.Ct. 2593, 2599, 53 L.Ed.2d 714 (1977). Independent of the first of those criteria, the statute, by delegating a governmental power to religious institutions, inescapably implicates the Establishment Clause. 25 The purpose of § 16C, as described by the District Court, is to "protect[ ] spiritual, cultural, and educational centers from the 'hurly-burly' associated with liquor outlets." 495 F.Supp., at 766. There can be little doubt that this embraces valid secular legislative purposes.6 However, these valid secular objectives can be readily accomplished by other means either through an absolute legislative ban on liquor outlets within reasonable prescribed distances from churches, schools, hospitals and like institutions,7 or by ensuring a hearing for the views of affected institutions at licensing proceedings where, without question, such views would be entitled to substantial weight.8 26 Appellants argue that § 16C has only a remote and incidental effect on the advancement of religion. The highest court in Massachusetts, however, has construed the statute as conferring upon churches a veto power over governmental licensing authority. Section 16C gives churches the right to determine whether a particular applicant will be granted a liquor license, or even which one of several competing applicants will receive a license. 27 The churches' power under the statute is standardless, calling for no reasons, findings, or reasoned conclusions. That power may therefore be used by churches to promote goals beyond insulating the church from undesirable neighbors; it could be employed for explicitly religious goals, for example, favoring liquor licenses for members of that congregation or adherents of that faith. We can assume that churches would act in good faith in their exercise of the statutory power, see Lemon v. Kurtzman, supra, 403 U.S., at 618-619, 91 S.Ct., at 2113-2114, yet § 16C does not by its terms require that churches' power be used in a religiously neutral way. "[T]he potential for conflict inheres in the situation," Levitt v. Committee for Public Education, 413 U.S. 472, 480, 93 S.Ct. 2814, 2819, 37 L.Ed.2d 376 (1973); and appellants have not suggested any "effective means of guaranteeing" that the delegated power "will be used exclusively for secular, neutral, and nonideological purposes." Committee for Public Education v. Nyquist, supra, 413 U.S., at 780, 93 S.Ct., at 2969.9 In addition, the mere appearance of a joint exercise of legislative authority by Church and State provides a significant symbolic benefit to religion in the minds of some by reason of the power conferred. It does not strain our prior holdings to say that the statute can be seen as having a "primary" and "principal" effect of advancing religion. 28 Turning to the third phase of the inquiry called for by Lemon v. Kurtzman, supra, we see that we have not previously had occasion to consider the entanglement implications of a statute vesting significant governmental authority in churches. This statute enmeshes churches in the exercise of substantial governmental powers contrary to our consistent interpretation of the Establishment Clause; "[t]he objective is to prevent, as far as possible, the intrusion of either [Church or State] into the precincts of the other." Lemon v. Kurtzman, supra, 403 U.S., at 614, 91 S.Ct., at 2112. We went on in that case to state: 29 Under our system the choice has been made that government is to be entirely excluded from the area of religious instruction and churches excluded from the affairs of government. The Constitution decrees that religion must be a private matter for the individual, the family, and the institutions of private choice, and that while some involvement and entanglement are inevitable, lines must be drawn. 30 403 U.S., at 625, 91 S.Ct., at 2117 (emphasis added). 31 Our contemporary views do no more than reflect views approved by the Court more than a century ago: 32 "The structure of our government has, for the preservation of civil liberty, rescued the temporal institutions from religious interference. On the other hand, it has secured religious liberty from the invasion of the civil authority." 33 Watson v. Jones, 80 U.S. (13 Wall.) 679, 730, 20 L.Ed. 666 (1871), quoting Harmon v. Dreher, 2 Speer's Equity Reports 87, 120 (S.C.Ct.App.1843). 34 As these and other cases make clear, the core rationale underlying the Establishment Clause is preventing "a fusion of governmental and religious functions," School District of Abington Township v. Schempp, 374 U.S. 203, 222, 83 S.Ct. 1560, 1571, 10 L.Ed.2d 844 (1963). See, e.g., Walz v. Tax Commission, supra, 397 U.S., at 674-75, 90 S.Ct., at 1414; Everson v. Board of Education, 330 U.S. 1, 8-13, 67 S.Ct. 504, 507-510, 91 L.Ed. 711 (1947).10 The Framers did not set up a system of government in which important, discretionary governmental powers would be delegated to or shared with religious institutions. 35 Section 16C substitutes the unilateral and absolute power of a church for the reasoned decisionmaking of a public legislative body acting on evidence and guided by standards, on issues with significant economic and political implications. The challenged statute thus enmeshes churches in the processes of government and creates the danger of "[p]olitical fragmentation and divisiveness along religious lines," Lemon v. Kurtzman, supra, 403 U.S., at 623, 91 S.Ct., at 2116. Ordinary human experience and a long line of cases teach that few entanglements could be more offensive to the spirit of the Constitution.11 36 The judgment of the Court of Appeals is affirmed. 37 So ordered. 38 Justice REHNQUIST, dissenting. 39 Dissenting opinions in previous cases have commented that "great" cases, like "hard" cases, make bad law. Northern Securities Co. v. United States, 193 U.S. 197 at 400-401, 24 S.Ct. 436 at 468, 48 L.Ed. 679 (1904) (Holmes, J., dissenting); Nixon v. General Service Administrator, 433 U.S. 425, 505, 97 S.Ct. 2777, 2821, 53 L.Ed.2d 867 (1977) (BURGER, C.J., dissenting). Today's opinion suggests that a third class of cases—silly cases—also make bad law. The Court wrenches from the decision of the Massachusetts Supreme Judicial Court the word "veto," and rests its conclusion on this single term. The aim of this effort is to prove that a quite sensible Massachusetts liquor zoning law is apparently some sort of sinister religious attack on secular government reminiscent of St. Bartholemew's Night. Being unpersuaded, I dissent. 40 In its original form, § 16C imposed a flat ban on the grant of an alcoholic beverages licenses to any establishment located within 500 feet of a church or a school. 1954 Mass. Acts c. 569, § 1. This statute represented a legislative determination that worship and liquor sales are generally not compatible uses of land. The majority concedes, as I believe it must, that "an absolute legislative ban on liquor outlets within reasonable prescribed distances from churches, schools, hospitals, and like institutions," ante, at 124 (footnote omitted), would be valid. See California v. LaRue, 409 U.S. 109, 120, 93 S.Ct. 390, 398, 34 L.Ed.2d 342 (1972) (Stewart, J., concurring). 41 Over time, the legislature found that it could meet its goal of protecting people engaged in religious activities from liquor-related disruption with a less absolute prohibition. Rather than set out elaborate formulae or require an administrative agency to make findings of fact, the legislature settled on the simple expedient of asking churches to object if a proposed liquor outlet would disturb them. Thus, under the present version of § 16C, a liquor outlet within 500 feet of a church or school can be licensed unless the affected institution objects. The flat ban, which the majority concedes is valid, is more protective of churches and more restrictive of liquor sales than the present § 16C. 42 The evolving treatment of the grant of liquor licenses to outlets located within 500 feet of a church or a school seems to me to be the sort of legislative refinement that we should encourage, not forbid in the name of the First Amendment. If a particular church or a particular school located within the 500 foot radius chooses not to object, the state has quite sensibly concluded that there is no reason to prohibit the issuance of the license. Nothing in the Court's opinion persuades me why the more rigid prohibition would be constitutional, but the more flexible not. 43 The Court rings in the metaphor of the "wall between church and state," and the "three part test" developed in Walz v. Tax Commission, 397 U.S. 664, 90 S.Ct. 1409, 25 L.Ed.2d 697 (1970), to justify its result. However, by its frequent reference to the statutory provision as a "veto," the Court indicates a belief that § 16C effectively constitutes churches as third houses of the Massachusetts legislature. See ante, at 125-126. Surely we do not need a three part test to decide whether the grant of actual legislative power to churches is within the proscription of the Establishment Clause of the First and Fourteenth Amendments. The question in this case is not whether such a statute would be unconstitutional, but whether § 16C is such a statute. The Court in effect answers this question in the first sentence of its opinion without any discussion or statement of reasons. I do not think the question is so trivial that it may be answered by simply affixing a label to the statutory provision. 44 Section 16C does not sponsor or subsidize any religious group or activity. It does not encourage, much less compel, anyone to participate in religious activities or to support religious institutions. To say that it "advances" religion is to strain at the meaning of that word. 45 The Court states that § 16C "advances" religion because there is no guarantee that objections will be made "in a religiously neutral way." Ante, at 125. It is difficult to understand what the Court means by this. The concededly legitimate purpose of the statute is to protect citizens engaging in religious and educational activities from the incompatible activities of liquor outlets and their patrons. The only way to decide whether these activities are incompatible with one another in the case of a church is to ask whether the activities of liquor outlets and their patrons may interfere with religious activity; this question cannot, in any meaningful sense, be "religiously neutral." In this sense, the flat ban of the original § 16C is no different from the present version. Whether the ban is unconditional or may be invoked only at the behest of a particular church, it is not "religiously neutral" so long as it enables a church to defeat the issuance of a liquor license when a similarly situated bank could not do the same. The state does not, in my opinion, "advance" religion by making provision for those who wish to engage in religious activities, as well as those who wish to engage in educational activities, to be unmolested by activities at a neighboring bar or tavern that have historically been thought incompatible. 46 The Court is apparently concerned for fear that churches might object to the issuance of a license for "explicitly religious" reasons, such as "favoring liquor licenses for members of that congregation or adherents of that faith."* Ante, at 125. If a church were to seek to advance the interests of its members in this way, there would be an occasion to determine whether it had violated any right of an unsuccessful applicant for a liquor license. But our ability to discern a risk of such abuse does not render § 16C violative of the Establishment Clause. The state can constitutionally protect churches from liquor for the same reasons it can protect them from fire, see Walz, supra, at 671, 90 S.Ct., at 1412 (1970), noise, see Grayned v. City of Rockford, 408 U.S. 104, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972), and other harm. 47 The heavy First Amendment artillery that the Court fires at this sensible and unobjectionable Massachusetts statute is both unnecessary and unavailing. I would reverse the judgment of the Court of Appeals. 1 Section 16C defines "church" as "a church or synagogue building dedicated to divine worship and in regular use for that purpose, but not a chapel occupying a minor portion of a building primarily devoted to other uses." "School" is defined as "an elementary or secondary school, public or private, giving not less than the minimum instruction and training required by [state law] to children of compulsory school age." Mass.G.L. ch. 138, § 16C. Section 16C originally was enacted in 1954 as an absolute ban on liquor licenses within 500 feet of a church or school, 1954 Mass.Acts, ch. 569, § 1. A 1968 amendment modified the absolute prohibition, permitting licenses within the 500-foot radius "if the governing body of such church assents in writing," 1968 Mass.Acts, ch. 435. In 1970, the statute was amended to its present form, 1970 Mass.Acts, ch. 192. 2 In 1979, there were 26 liquor licensees in Harvard Square and within a 500-foot radius of Holy Cross Church; 25 of these were in existence at the time Holy Cross Church objected to appellee's application. See App. 69-72. 3 Section 16C defines "church" as: "a church or synagogue building dedicated to divine worship" (emphasis added). Appellee argues that the statute unconstitutionally differentiates between theistic and nontheistic religions. We need not reach that issue. For purposes of this appeal, we assume, as did the original panel of the Court of Appeals, that the Massachusetts courts would apply the protections of § 16C to "any building primarily used as a place of assembly by a bona fide religious group," 662 F.2d, at 97, and thereby avoid serious constitutional questions that would arise concerning a statute that distinguishes between religions on the basis of commitment to belief in a divinity. See Torcaso v. Watkins, 367 U.S. 488, 495, 81 S.Ct. 1680, 1683, 6 L.Ed.2d 982 (1961); Everson v. Board of Education, 330 U.S. 1, 15, 67 S.Ct. 504, 511, 91 L.Ed. 711 (1947). 4 This recent construction of the statute by the highest court in Massachusetts is controlling on the meaning of § 16C. See O'Brien v. Skinner, 414 U.S. 524, 531, 94 S.Ct. 740, 743, 38 L.Ed.2d 702 (1974). 5 For similar reasons, the 21st Amendment does not justify § 16C. The 21st Amendment reserves power to states, yet here the State has delegated to churches a power relating to liquor sales. The State may not exercise its power under the 21st Amendment in a way which impinges upon the Establishment Clause of the First Amendment. 6 In this facial attack, the Court assumes that § 16C actually effectuates the secular goal of protecting churches and schools from the disruption associated with liquor serving establishments. The fact that Holy Cross Church is already surrounded by 26 liquor outlets casts some doubt on the effectiveness of the protection granted, however. 7 See California v. Larue, 409 U.S. 109, 120, 93 S.Ct. 390, 398, 34 L.Ed.2d 342 (1972) (Stewart, J., concurring). Section 16C, as originally enacted, consisted of an absolute ban on liquor licenses within 500 feet of a church or school, see note 1 supra; and 27 states continue to prohibit liquor outlets within a prescribed distance of various categories of protected institutions, with certain exceptions and variations: Ala.Code § 28-3-17 (1975); Alaska Stat.Ann. § 04.11.410 (1980); Ark.Stat.Ann. § 48-345 (1977); Colo.Rev.Stat. § 12-47-138 (1978 & 1982 Supp.); Ga.Code Ann. § 3-3-21 (1982); Idaho Code §§ 23-303, 23-913 (1977); Ill.Rev.Stat., ch. 43, § 127 (1982 Supp.); Ind.Code § 7.1-3-21-11 (1982); Kan.Stat.Ann. § 41-710 (1981); La.Rev.Stat.Ann. § 26-280 (West 1975); Md.Ann.Code, Art. 2B, §§ 46B, 47, 52A, 52C (1981 & 1982 Supp.); Mich.Comp.Laws Ann. §§ 436.17a, 436.17c (1978 & 1982 Supp.); Minn.Stat.Ann. § 340.14 (1972 & 1982 Supp.); Miss.Code Ann. § 67-1-51 (1973 & 1982 Supp.); Mont.Code Ann. § 16-3-306 (1981); Neb.Rev.Stat. § 53-177 (1978); N.H.Rev.Stat.Ann. § 177:1 (1978); N.M.Stat.Ann. § 60-6B-10 (1981); N.C.Gen.Stat. § 18A-40 (1978) (schools); Okla.Stat.Ann., Tit. 37, § 534 (West 1981); R.I.Gen.Laws § 3-7-19 (1976 & 1981 Supp.); S.C.Code § 61-3-440 (1977); S.D.Comp.Laws Ann. § 35-2-6.1 (1982 Supp.); Tex.Alco.Bev.Code Ann., Tit. 4, § 109.33 (1978); Utah Code Ann. § 16.6-13.5 (1981 Supp.); W.Va.Code § 11-16-12 (1974); Wis.Stat.Ann. § 125.68 (West 1982 Supp.). The Court does not express an opinion as to the constitutionality of any statute other than that of Massachusetts. 8 Eleven states have statutes directing the licensing authority to consider the proximity of the proposed liquor outlet to schools or other institutions in deciding whether to grant a liquor license: Cal.Bus. & Prof.Code § 23789 (West 1964 & 1982 Supp.); Conn.Gen.Stat. § 30-46 (1982); Del.Code Ann., Tit. 4, § 543 (1975); Haw.Rev.Stat. § 281-56 (1976); Mich.Comp.Laws Ann. §§ 436.17a, 436.17c (1978 & 1982 Supp.) (certain classes of licenses); N.C.Gen.Stat. § 18A-40 (1978) (churches); Ohio Rev.Code Ann. § 4303.26 (1973 & 1981 Supp.); Pa.Stat.Ann., Tit. 47, §§ 4-404, 4-432(d) (Purdon 1969 & 1982 Supp.); Tenn.Code Ann. § 57-5-105 (1982 Supp.); Va.Code § 4-31 (1982 Supp.); Vt. Liquor Control Bd.Regs. ¶ 39 (1976). 9 Appellants argue that the Beverages Control Commission may reject or ignore any objection made for discriminatory or illegal reasons. This contention appears flatly contradicted by the Massachusetts Supreme Judicial Court's own interpretation of the statute, see Arno v. Alcoholic Beverages Control Commission, 377 Mass. 83, 90, 92 and n. 23, 384 N.E.2d 1223 (1979). In any event, an assumption that the Beverages Control Commission might review the decisionmaking of the churches would present serious entanglement problems. See Lemon v. Kurtzman, 403 U.S. 602, 619, 91 S.Ct. 2105, 2114, 29 L.Ed.2d 745 (1971); National Labor Relations Board v. Catholic Bishop of Chicago, 440 U.S. 490, 99 S.Ct. 1313, 59 L.Ed.2d 533 (1979). 10 At the time of the Revolution, Americans feared not only a denial of religious freedom, but the danger of political oppression through a union of civil and ecclesiastical control. B. Bailyn, Ideological Origins of the American Revolution 98-99 n. 3 (1967). See McDaniel v. Paty, 435 U.S. 618, 622-623, 98 S.Ct. 1322, 1325-1326, 55 L.Ed.2d 593 (1978). In eighteenth century England, such a union of civil and ecclesiastical power was reflected in legal arrangements granting church officials substantial control over various occupations, including the liquor trade. See, e.g., 26 George II ch. 31, § 2 (1753) (church officials given authority to grant certificate of character, a prerequisite for an alehouse license); S. & B. Webb, The History of Liquor Licensing in England, Principally from 1700 to 1830, at 8 n. 1, 62-67, 102-103 (1903). 11 Appellee also challenges the statute as a violation of due process. In light of our analysis we need not and do not reach that claim. * I doubt whether there exists a denomination that considers supporting the liquor license applications of its members to be a part of its theology. However else a church's goal in objecting to issuance of a liquor license on such a basis might be characterized, it would certainly be strictly temporal. I note in passing that § 16C does not confer on any church any power to obtain a liquor license for anyone.

23

459 U.S. 197 103 S.Ct. 552 74 L.Ed.2d 364 FEDERAL ELECTION COMMISSION, et al., Petitionersv.NATIONAL RIGHT TO WORK COMMITTEE et al. No. 81-1506. Argued Nov. 1, 1982. Decided Dec. 13, 1982. Syllabus The Federal Election Campaign Act of 1971, 2 U.S.C. § 441b(a) prohibits corporations and labor unions from making contributions or expenditures in connection with federal elections. The section, however, permits some participation by unions and corporations in the federal electoral process by allowing these organizations to establish and pay the expenses of "separate segregated funds" which may be used for political purposes during federal elections. The Act restricts the operations of such segregated funds in several respects. Of most relevance here, 2 U.S.C. §§ 441b(b)(4)(A) and 441b(b)(4)(C) provide that a corporation without capital stock may solicit contributions to a fund it has established only from "members" of the corporation. During 1976 respondent National Right to Work Committee (NRWC), a corporation without capital stock, solicited some 267,000 persons for contributions to a separate segregated fund that it sponsored. Petitioner, the Federal Election Commission, determined that NRWC's solicitation violated § 441b(b)(4)(C) of the Act, because the persons it had solicited were not its members. Among other things, NRWC's solicitation letters did not mention membership, its articles of incorporation disclaim the existence of members, and members play no part in the operation or administration of the corporation. Held: The persons solicited by NRWC were insufficiently attached to the corporation to qualify as members under § 441b(b)(4)(C) of the Act. Pp. 201-207. This interpretation of the Act does not raise constitutional difficulties. The First Amendment associational rights asserted by respondent are overborne by the interests Congress has sought to protect in enacting § 441b. The provision marks the culmination of a careful legislative adjustment of the federal electoral laws to prevent both actual and apparent corruption and reflects a legislative judgment that the special characteristics of corporations require prophylactic measures. Pp. 207-211. 214 U.S.App.D.C. 215, 665 F.2d 371, reversed. Charles N. Steele, Washington, D.C., for petitioners. Richard H. Mansfield, III, Washington, D.C., for respondents. Justice REHNQUIST delivered the opinion of the Court. 1 The question in the case ultimately comes down to whether respondent National Right to Work Committee ("NRWC") limited its solicitation of funds to "members" within the meaning of 2 U.S.C. § 441b(b)(4)(C).1 2 In April of 1977, petitioner, the Federal Election Commission ("the Commission"),2 determined that there was probable cause to believe that NRWC had violated the above-cited provisions of the Act by soliciting contributions from persons who were not its "members." Shortly thereafter, respondent filed a complaint in the United States District Court for the Eastern District of Virginia seeking injunctive and declaratory relief against the Commission. One month later, the Commission filed an enforcement proceeding against respondent in the United States District Court for the District of Columbia, seeking to establish respondent's violation of 2 U.S.C. § 441b. The actions were consolidated in the latter court, which granted summary judgment in favor of the Commission on the basis of stipulated facts. 501 F.Supp. 422.3 The judgment of the District Court was reversed by the Court of Appeals for the District of Columbia Circuit, 665 F.2d 371 (1981), and we granted certiorari. --- U.S. ----, 102 S.Ct. 1766, 72 L.Ed.2d 172 (1982). 3 Respondent NRWC is a nonprofit corporation without capital stock organized under the laws of the Commonwealth of Virginia. Given the central role of the congressional use of the word "member" in this litigation, it is useful to set forth respondent's organizational history in some detail. In 1975, respondent's predecessor and another corporation merged; the articles of merger filed in the District of Columbia by the successor corporation stated that NRWC "shall not have members." A similar statement is contained in the articles of incorporation of NRWC that are presently filed in Virginia. Likewise, respondent's bylaws make no reference to members or to membership in the corporation. The stated purpose of NRWC, according to its Virginia articles of incorporation, is "[t]o help make the public aware of the fact that American citizens are being required, against their will, to join and pay dues to labor organizations in order to earn a living. . . ." Pet.App. at 17a. In pursuance of this objective, NRWC regularly mails messages to millions of individuals and businesses whose names have found their way onto commercially available mailing lists that the organization has purchased or rented. The letters do not mention membership in NRWC, but seek donations to help NRWC publicize its opposition to compulsory unionism and frequently contain a questionnaire that the recipient is requested to answer and return. 4 In late 1975, in order to comply with § 441b of the Act, NRWC established a separate segregated fund, see 2 U.S.C. § 441b(b)(4)(C),4 "to receive and make contributions on behalf of federal candidates." The fund was denominated the "Employees Rights Campaign Committee" ("ERCC"); its operation was completely subsidized from the NRWC treasury, which paid all the expenses of establishing and administering the fund, and of soliciting contributions. During part of 1976, NRWC sent letters to some 267,000 individuals, who had at one time contributed to it, soliciting contributions to ERCC. As a result of these solicitations, the fund received some $77,000 in contributions. 5 In October, 1976, another lobbying group, the Committee for an Effective Congress, filed a complaint against ERCC with the Commission, alleging violation of 2 U.S.C. § 441b(b)(4). The complaint asserted that NRWC had violated this section of the Act by using corporate funds to solicit contributions to ERCC from persons who were not NRWC's stockholders, executive or administrative personnel, or their families. NRWC did not deny these assertions, but took the position that the recipients of its solicitation letters were "members" of NRWC within the proviso set forth in § 441b(b)(4)(C). The Commission found probable cause to believe that a violation had occurred, and after completing the investigative procedures set out in the statute and unsuccessfully attempting to resolve the matter through conciliation, see 2 U.S.C. § 437g, it authorized the filing of a civil enforcement suit. This litigation followed. 6 Essential to the proper resolution of the case is the interpretation of § 441b(b)(4)(C)'s statement that the prohibition against corporate solicitation contained in § 441b(b)(4)(A) shall not prevent "a . . . corporation without capital stock . . . from soliciting contributions to [a separate segregated fund established by a . . . corporation without capital stock] from members of such . . . corporation. . . ." (Emphasis added). The Court of Appeals rejected the Commission's contentions regarding the meaning of "member," and went on to hold that the term "embraces at least those individuals whom NRWC describes as its active and supporting members." Pet. 10a. The opinion of the Court of Appeals indicates that this construction was reached at least in part because of concern for the constitutional implications of any narrower construction. Pet. 6a-10a. As explained below, we reject this construction. 7 The statutory purpose of § 441b, as outlined above, is to prohibit contributions or expenditures by corporations or labor organizations in connection with federal elections. 2 U.S.C. § 441b(a). The section, however, permits some participation of unions and corporations in the federal electoral process by allowing them to establish and pay the administrative expenses of "separate segregated funds," which may be "utilized for political purposes." 2 U.S.C. § 441b(b)(2)(C). The Act restricts the operations of such segregated funds, however, by making it unlawful for a corporation to solicit contributions to a fund established by it from persons other than its "stockholders and their families and its executive or administrative personnel and their families." 2 U.S.C. § 441b(b)(4)(A). Finally, and of most relevance here, the section just quoted has its own proviso, which states in pertinent part that "[t]his paragraph shall not prevent . . . a corporation without capital stock, or a separate segregated fund established by . . . a corporation without capital stock, from soliciting contributions to such a fund from members" of the sponsoring corporation. 2 U.S.C. § 441b(b)(4)(C). The effect of this proviso is to limit solicitation by nonprofit corporations to those persons attached in some way to it by its corporate structure. 2 U.S.C. § 441b(b)(4)(C). 8 The Court of Appeals, as we have noted, construed the term "member" in § 441b to embrace "at least those individuals whom NRWC describes as its active and supporting members." Pet. 10a. The two categories of members recognized by NRWC were described in the following terms by the Court of Appeals: 9 "NRWC attracts members by publicizing its position on issues relating to compulsory unionism through advertisements, personal contacts, and, primarily, letters. These letters describe the purpose of NRWC, urge the recipient to assist NRWC (by, for example, writing to legislators), request financial support, and ask the recipient to respond to a questionnaire that will determine whether that person shares a similar political philosophy. A person who, through his response, evidences an intention to support NRWC in promoting voluntary unionism qualifies as a member. A person who responds without contributing financially is considered a supporting member; a person who responds and also contributes is considered an active member. NRWC sends an acknowledgement and a membership card to both classes. In the regular course of operations, NRWC's members receive newsletters, action alerts, and responses to individual requests for information. They respond to issue surveys and are asked to communicate with their elected representatives when appropriate. See Joint App., vol. II, at 387 et seq." 10 In respondent's view, both categories satisfy the membership requirement of § 441b(b)(4)(C). 11 The Commission, however, insists that these standards of "membership" are too fluid and insubstantial to come within the statutory term "member," and argues further that they do not comply with the Commission's regulation defining the term: 12 "(e) 'Members' means all persons who are currently satisfying the requirements for membership in a membership organization, trade association, cooperative, or corporation without capital stock. . . . A person is not considered a member under this definition if the only requirement for membership is a contribution to a separate segregated fund." Federal Election Commission Regulations, 11 CFR § 114.1(e). 13 The Commission also contends that NRWC's Virginia articles of incorporation filed by respondent, which state that respondent has no members, are dispositive. While we do not feel sufficiently informed at this time to attempt an exegesis of the statutory meaning of the word "members" beyond that necessary to decide this case, we find it relatively easy to dispose of these arguments that respondent's solicitation was limited to its "members," since in our view this would virtually excise from the statute the restriction of solicitation to "members." 14 Section 441b(b)(4)(C) was one of several amendments to the Act enacted in 1976. The entire legislative history of the subsection appears to be the floor statement of Senator Allen who introduced the provision in the Senate and explained the purpose of his amendment in this language: "Mr. President, all this amendment does is to cure 15 an omission in the bill. It would allow corporations that do not have stock but have a membership organization, such as a cooperative or other corporation without capital stock and, hence, without stockholders, to set up separate segregated political funds as to which it can solicit contributions from its membership; since it does not have any stockholders to solicit, it should be allowed to solicit its members. That is all that amendment provides. It does cover an omission in the bill that I believe all agree should be filled." 16 This statement suggests that "members" of nonstock corporations were to be defined, at least in part, by analogy to stockholders of business corporations and members of labor unions. The analogy to stockholders and union members suggests that some relatively enduring and independently significant financial or organizational attachment is required to be a "member" under § 441b(b)(4)(C). The Court of Appeals' determination that NRWC's "members" include anyone who has responded to one of the corporation's essentially random mass mailings would, we think, open the door to all but unlimited corporate solicitation and thereby render meaningless the statutory limitation to "members." 17 We also assume, since there is no body of federal law of corporations, see Burks v. Lasker 441 U.S. 471, 477, 99 S.Ct. 1831, 1836, 60 L.Ed.2d 404 (1979), that Congress intended at least some reference to the laws of the various states dealing with nonprofit corporations. In an analogous situation, where Congress had authorized state taxation of "real property" of subsidiaries of the Reconstruction Finance Corporation, the Court said: 18 "We think the congressional purpose can best be accomplished by application of settled state rules as to what constitutes 'real property,' so long as it is plain, as it is here, that the state rules do not effect a discrimination against the Government, or patently run counter to the terms of the Act." RFC v. Beaver County, 328 U.S. 204, 210, 66 S.Ct. 992, 995, 90 L.Ed. 1172 (1946). 19 Like property, the structure and powers of nonprofit corporations are defined principally by state law; as in the case of property, state law provides some guidance in deciding whether NRWC's solicitation was confined to its "members." 20 Most states apparently permit nonprofit corporations to have "members" similar to shareholders in a business corporation, although state statutes generally do not seem to require this form of organization, see, e.g., ALI-ABA, Model Nonprofit Corporation Act, § 11 (1964); in many states the Board of Directors of a nonprofit corporation may be an autonomous, self-perpetuating body.5 Given the wide variety of treatment of the subject of membership in state incorporation laws, and the focus of the Commission's regulation on the corporation's own standards, we think it was entirely permissible for the Commission in this case to look to NRWC's corporate charter under the laws of Virginia and the bylaws adopted in accordance with that charter. 21 Applying the statutory language as we interpret it to the facts of this case,6 we think Congress did not intend to allow the 267,000 individuals solicited by NRWC during 1976 to come within the exclusion for "members" in 2 U.S.C. § 441b(b)(4)(C). Although membership cards are ultimately sent to those who either contribute or respond in some other way to respondents' mailings, the solicitation letters themselves make no reference to members. Members play no part in the operation or administration of the corporation; they elect no corporate officials, and indeed there are apparently no membership meetings. There is no indication that NRWC's asserted members exercise any control over the expenditure of their contributions. Moreover, as previously noted, NRWC's own articles of incorporation and other publicly filed documents explicitly disclaimed the existence of members. We think that under these circumstances, those solicited were insufficiently attached to the corporate structure of NRWC to qualify as "members" under the statutory proviso. 22 Unlike the Court of Appeals, we do not think this construction of the statute raises any insurmountable constitutional difficulties. The Court of Appeals expressed the view that the sort of solicitations involved here would neither corrupt officials nor coerce members of the corporation holding minority political views, the two goals which it believed Congress had in mind in enacting the statutory provisions at issue. That being so, the Court of Appeals apparently thought, and respondent argues here, that the term "members" must be given an elastic definition in order to prevent impermissible interference with the constitutional rights enunciated in cases such as NAACP v. Button, 371 U.S. 415, 83 S.Ct. 328, 9 L.Ed.2d 405 (1963), and Village of Schaumburg v. Citizens for a Better Environment, 444 U.S. 620, 100 S.Ct. 826, 63 L.Ed.2d 73 (1980). Similarly, respondent places considerable reliance on our statement in Buckley v. Valeo, 424 U.S. 1, 25, 96 S.Ct. 612, 637, 46 L.Ed.2d 659 (1976), that: 23 "The Court's decisions involving associational freedoms establish that the right of association is a 'basic constitutional freedom,' Kusper v. Pontikes, 414 U.S., [51] at 57 [94 S.Ct., 303 at 307, 38 L.Ed.2d 260], that is 'closely allied to freedom of speech and a right which, like free speech, lies at the foundation of a free society.' Shelton v. Tucker, 364 U.S. 479, 486 [, 81 S.Ct. 247, 251, 5 L.Ed.2d 231] (1960). See, e.g., Bates v. Little Rock, 361 U.S. 516, 522-523 [, 80 S.Ct. 412, 416-417, 4 L.Ed.2d 480] (1960); NAACP v. Alabama, supra, [357 U.S. 449] at 460-461 [78 S.Ct. 1163 at 1170-1171, 2 L.Ed.2d 1488]; NAACP v. Button, supra, [371 U.S.] at 452 [83 S.Ct. at 347] (Harlan, J., dissenting). In view of the fundamental nature of the right to associate, governmental 'action which may have the effect of curtailing the freedom to associate is subject to the closest scrutiny.' " NAACP v. Alabama, supra, [357 U.S.] at 460-461, 78 S.Ct. at 1170-1171. 24 Under this standard, respondent asserts, the Act's restriction of its solicitation cannot be upheld. 25 While we fully subscribe to the views stated in Buckley, in the very next sentence to the passage quoted by the respondent, the Court went on to say: 26 "Yet, it is clear that '[n]either the right to associate nor the right to participate in political activities is absolute.' CSC v. Letter Carriers, 413 U.S. 548, 567 [, 93 S.Ct. 2880, 2891, 37 L.Ed.2d 796] (1973)." 27 In this case, we conclude that the associational rights asserted by respondent may be and are overborne by the interests Congress has sought to protect in enacting § 441b. 28 To place respondent's constitutional claims in proper perspective, we repeat language used in Buckley v. Valeo, supra: 29 "The constitutional power of Congress to regulate federal elections is well established and is not questioned by any of the parties in this case." Buckley v. Valeo, 424 U.S. 1, 13, 96 S.Ct. 612, 631, 46 L.Ed.2d 659 (1976). 30 The first purpose of § 441b, the government states, is to ensure that substantial aggregations of wealth amassed by the special advantages which go with the corporate form of organization should not be converted into political "war chests" which could be used to incur political debts from legislators who are aided by the contributions. See United States v. United Automobile Workers, 352 U.S. 567, 579, 77 S.Ct. 529, 535, 1 L.Ed.2d 563 (1957). The second purpose of the provisions, the government argues, is to protect the individuals who have paid money into a corporation or union for purposes other than the support of candidates from having that money used to support political candidates to whom they may be opposed. See United States v. CIO, 335 U.S. 106, 113, 68 S.Ct. 1349, 1353, 92 L.Ed. 1849 (1948). 31 We agree with the government that these purposes are sufficient to justify the regulation at issue. Speaking of corporate involvement in electoral politics, we recently said: 32 "The overriding concern behind the enactment of statutes such as the Federal Corrupt Practices Act was the problem of corruption of elected representatives through the creation of political debts. The importance of the governmental interest in preventing this occurrence has never been doubted." First National Bank of Boston v. Bellotti, 435 U.S. 765, 788, fn. 26, 98 S.Ct. 1407, 1422, fn. 26, 55 L.Ed.2d 707 (1978) (Citations omitted). 33 Likewise, in Buckley v. Valeo, 424 U.S. 1, 26-27, 96 S.Ct. 612, 638-639, 46 L.Ed.2d 659 (1976), we specifically affirmed the importance of preventing both the actual corruption threatened by large financial contributions and the eroding of public confidence in the electoral process through the appearance of corruption. These interests directly implicate "the integrity of our electoral process, and, not less, the responsibility of the individual citizen for the successful functioning of that process." United States v. UAW, supra, 352 U.S., at 570, 77 S.Ct., at 530. 34 We are also convinced that the statutory prohibitions and exceptions we have considered are sufficiently tailored to these purposes to avoid undue restriction on the associational interests asserted by respondent. The history of the movement to regulate the political contributions and expenditures of corporations and labor unions is set forth in great detail in United States v. UAW, supra, 352 U.S., at 570-584, 77 S.Ct., at 530-537, and we need only summarize the development here. Seventy-five years ago Congress first made financial contributions to federal candidates by corporations illegal by enacting the Tillman Act, 34 Stat. 864 (1907). Within the next few years Congress went further and required financial disclosure by federal candidates following election, Act of July 25, 1910, 36 Stat. 822, and the following year required pre-election disclosure as well. Act of August 19, 1911, 37 Stat. 25. The Federal Corrupt Practices Act, passed in 1925, extended the prohibition against corporate contributions to include "anything of value," and made acceptance of a corporate contribution as well as the giving of such a contribution a crime. 43 Stat. 1070. 35 The first restrictions on union contributions were contained in the second Hatch Act, 54 Stat. 767 (1940), and later, in the War Labor Disputes Act of 1943, 57 Stat. 167, union contributions in connection with federal elections were prohibited altogether. These prohibitions on union political activity were extended and strengthened in the Taft-Hartley Act, 61 Stat. 136 (1947), which broadened the earlier prohibition against contributions to "expenditures" as well. Congress codified most of these provisions in the Federal Election Campaign Act of 1971, 86 Stat. 3, and enacted later amendments in 1974, 88 Stat. 1263, and in 1976, 90 Stat. 475. Section 441b(b)(4)(C) is, as its legislative history indicates, merely a refinement of this gradual development of the federal election statute. 36 This careful legislative adjustment of the federal electoral laws, in a "cautious advance, step by step," NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1, 46, 57 S.Ct. 615, 628, 81 L.Ed. 893 (1937), to account for the particular legal and economic attributes of corporations and labor organizations warrants considerable deference, see Rostker v. Goldberg, 453 U.S. 57, 64, 67, 101 S.Ct. 2646, 2651, 2652, 69 L.Ed.2d 478 (1981). As we discuss below, it also reflects a permissible assessment of the dangers posed by those entities to the electoral process. 37 In order to prevent both actual and apparent corruption, Congress aimed a part of its regulatory scheme at corporations. The statute reflects a legislative judgment that the special characteristics of the corporate structure require particularly careful regulation. See United States v. Morton Salt Co., 338 U.S. 632, 652, 70 S.Ct. 357, 368, 94 L.Ed. 401 (1950). While § 441b restricts the solicitation of corporations and labor unions without great financial resources, as well as those more fortunately situated, we accept Congress's judgment that it is the potential for such influence that demands regulation. Nor will we second guess a legislative determination as to the need for prophylactic measures where corruption is the evil feared. As we said in California Medical Association v. FEC, 453 U.S. 182, 201, 101 S.Ct. 2712, 2724, 69 L.Ed.2d 567 (1982), the "differing structures and purposes" of different entities "may require different forms of regulation in order to protect the integrity of the electoral process."7 38 To accept the view that a solicitation limited only to those who have in the past proved "philosophically compatible" to the views of the corporation must be permitted under the statute in order for the prohibition to be constitutional would ignore the teachings of our earlier decisions. The governmental interest in preventing both actual corruption and the appearance of corruption of elected representatives has long been recognized, First National Bank of Boston v. Bellotti, 435 U.S., at 787, n. 26, 98 S.Ct., at 1422, n. 26, and there is no reason why it may not in this case be accomplished by treating unions, corporations, and similar organizations differently from individuals. California Medical Association v. FEC, 453 U.S. 182, 201, 101 S.Ct. 2712, 69 L.Ed.2d 567 (1981). 39 Respondents also assert a claim of unconstitutional vagueness, relying on such additional cases as Connally v. General Construction Co., 269 U.S. 385, 46 S.Ct. 126, 70 L.Ed. 322 (1935); Grayned v. City of Rockford, 408 U.S. 104, 92 S.Ct. 2294, 33 L.Ed.2d 222 (1972); Speiser v. Randall, 357 U.S. 513, 78 S.Ct. 1332, 2 L.Ed.2d 1460 (1958); and Smith v. California, 361 U.S. 147, 80 S.Ct. 215, 4 L.Ed.2d 205 (1959). We think the vagueness claim is adequately answered by the language quoted earlier from CSC v. Letter Carriers, supra. There may be more than one way under the statute to go about determining who are "members" of a nonprofit corporation, and the statute may leave room for uncertainty at the periphery of its exception for solicitation of "members." However, on this record we are satisfied that NRWC's activities extended in large part, if not in toto, to people who would not be members under any reasonable interpretation of the statute. See Broadrick v. Oklahoma, 413 U.S. 601, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973).8 40 The judgment of the Court of Appeals is reversed. 1 As will appear from the following discussion, the phrasing of this question is but the tip of the statutory iceberg. The Federal Election Campaign Act of 1971 ("the Act") makes it "unlawful . . . for any corporation . . . to make a contribution or expenditure in connection with" certain federal elections. 2 U.S.C. § 441b(a). The term "contribution" is defined broadly, 2 U.S.C. § 441b(b)(2)(C), to include any sort of transfer of money or services to various political entities, but excluded from that definition is "the establishment, administration, and solicitation of contributions to a separate segregated fund to be utilized for political purposes by a . . . corporation without capital stock." The Act goes on to make it unlawful, except as thereinafter provided, "for a corporation, or a separate segregated fund established by a corporation, to solicit contributions to such a fund from any person other than its stockholders and its families and its executive or administrative personnel and their families. . . ." 2 U.S.C. § 441b(b)(4)(A). Finally, 2 U.S.C. § 441b(b)(4)(C) states that the prohibition just quoted "shall not prevent a . . . corporation without capital stock, or a separate segregated fund established by a . . . corporation without capital stock, from soliciting contributions to such a fund from members of such . . . corporation without capital stock." 2 The Commission is an independent administrative agency vested with exclusive jurisdiction over civil enforcement of the Act. See 2 U.S.C. §§ 437c(b)(1) and 437d(a) and (e). 3 The relief awarded the Commission by the District Court included declaratory judgment that 2 U.S.C. § 441b(b)(4) is not unconstitutional, an order that NRWC refund to contributors the funds it had obtained from unlawful solicitations, and an order that the corporation pay a $10,000 civil penalty. Pet. at 54a. 4 The separate segregated fund may be completely controlled by the sponsoring corporation or union, whose officers may decide which political candidates' contributions to the fund will be spent to assist. The "fund must be separate from the sponsoring union [or corporation] only in the sense that there must be a strict segregation of its monies" from the corporation's other assets. Pipefitters Local Union No. 562 v. United States, 407 U.S. 385, 414-417, 92 S.Ct. 2247, 2264-2265, 33 L.Ed.2d 11 (1972). See also Buckley v. Valeo, 424 U.S. 1, 28, n. 31, 96 S.Ct. 612, 639, n. 31, 46 L.Ed.2d 659 (1976). 5 One commentator has stated: "The license provided by the statutes in this respect is further enhanced by their loose use of the term 'member.' The New York statute and the Model Act, for example, offer no meaningful definition of 'member' at all, but instead provide that a corporation's articles or bylaws may designate anybody or nobody as members, or may designate different classes of members, and may freely specify the rights, if any, of the corporation's members or classes of members. The California Act is a bit more carefully drawn in this regard, defining a member, essentially, as anyone entitled to vote in elections either for the corporation's Board of Directors or for certain fundamental corporate changes." Hansmann, Reforming Nonprofit Corporation Law, 129 U.Pa.L.Rev. 497, 578 (1981). 6 We assume, as have the parties and courts below, that ERCC satisfies the statutory requirements of a "separate segregated fund" and that NRWC is a corporation covered by § 441b. 7 Our decision in First National Bank of Boston v. Bellotti, 435 U.S. 765, 98 S.Ct. 1407, 55 L.Ed.2d 707 (1978), is entirely consistent with our conclusion here. Bellotti struck down a prohibition against corporate expenditures and contributions in connection with state referenda. Id., at 768, 98 S.Ct., at 1411. The Court explicitly stated that its decision did not involve "the constitutionality of laws prohibiting or limiting corporate contributions to political candidates or committees, or other means of influencing candidate elections. " Id., at 788, n. 26, 98 S.Ct., at 1422, n. 26. In addition, following its citation of Pipefitters v. United States, 407 U.S. 385, 92 S.Ct. 2247, 33 L.Ed.2d 11 (1972); United States v. UAW, 352 U.S. 567, 77 S.Ct. 529, 1 L.Ed.2d 563 (1957); United States v. CIO, 335 U.S. 106, 68 S.Ct. 1349, 92 L.Ed. 1849 (1948), the Court specifically pointed out that in elections of candidates to public office, unlike in referenda on issues of general public interest, there may well be a threat of real or apparent corruption. As discussed in text, the Government has relied on just this threat in enacting § 441b. 8 We also reject as meritless NRWC's claim that the Commission's actions prior to and during conciliation were so misleading and arbitrary as to constitute a deprivation of Due Process. We leave open for consideration upon remand, inter alia, the propriety of the Commission's imposition of a $10,000 civil penalty against respondent.

23

459 U.S. 159 103 S.Ct. 530 74 L.Ed.2d 334 CITY OF PORT ARTHUR, Texas, Appellant,v.UNITED STATES et al. No. 81-708. Argued Oct. 6, 1982. Decided Dec. 13, 1982. Syllabus Under § 5 of the Voting Rights Act of 1965, a covered State or political subdivision, such as appellant city of Port Arthur, must obtain federal preclearance of a change in its voting practices or procedures either from the Attorney General or by obtaining a declaratory judgment from the District Court for the District of Columbia that the proposed change has neither the purpose nor the effect of denying the right to vote on account of race. In 1977 and 1978, Port Arthur was consolidated with two neighboring cities and annexed an incorporated area, with the result that the percentage of the black population within Port Arthur's borders decreased from 45.21% to 40.56%. Appellant ultimately filed a § 5 suit in the District Court, seeking approval of the consolidations and the annexation, and of a proposed expansion of its City Council from seven members (including a mayor), who had been previously elected at large by majority vote, to a nine-member Council. After the rejection of earlier electoral plans, appellant submitted a plan involving election of councilmen from four single-member districts, two of which included black majorities; at-large election of two members from two other districts, each of which consisted of two of the four single-member districts, and one of which had a black majority; and at-large election of the mayor. All council seats would be governed by a majority-vote rule, requiring run-offs if none of the candidates received a majority of the votes cast. Although concluding that the expansion of Port Arthur's borders could not be denied preclearance as being discriminatory in purpose, the District Court held that the electoral plan could not be approved under § 5 because it insufficiently neutralized the adverse impact upon minority voting strength that resulted from the expansion. However, the court stated that if the plan were modified to eliminate the majority-vote requirement with respect to the two non-mayoral, at-large candidates, and to permit election to those two seats to be made by a plurality vote, the court would consider the defect remedied and would offer its approval. Held: The District Court did not exceed its authority in conditioning clearance of the electoral plan on the elimination of the majority-vote requirement. Pp. 165-168. (a) Section 5 does not forbid all expansion of municipal borders that dilute the voting power of particular groups in the community. However, such an expansion can be approved only if modifications in the electoral plan, calculated to neutralize to the extent possible any adverse effect on the political participation of minority groups, are adopted. Pp. 165-166. (b) The District Court did not err in holding that the majority-vote requirement as to the non-mayoral, at-large council seats must be eliminated in order to sufficiently dispel the impact of Port Arthur's expansion on the relative political strength of the black community. Whether the plan adequately reflected black political strength in the enlarged city is not an issue that is determinable with mathematical precision. Since the plan undervalued to some extent the political strength of the black community, eliminating the majority-vote requirement was an understandable adjustment. And, even if the electoral scheme might otherwise be said to reflect the political strength of the minority community, elimination of the majority-vote element was a reasonable hedge against the possibility that the scheme contained a purposefully discriminatory element. Pp. 166-168. 517 F.Supp. 987, affirmed. Robert Q. Keith, Beaumont, Tex., for appellant. Carter G. Phillips, Washington, D.C., for appellees. Justice WHITE delivered the opinion of the Court. 1 Section 5 of the Voting Rights Act of 1965, 42 U.S.C. § 1973c, requires that when a state or political subdivision covered by the Act1 adopts or seeks to administer any change in its standards, practices, or procedures with respect to voting, it must obtain a preclearance either from the Attorney General of the United States or by obtaining a declaratory judgment from the District Court for the District of Columbia that the proposed change has neither the purpose nor the effect of denying or abridging the right to vote on account of race.2 Perkins v. Matthews, 400 U.S. 379, 91 S.Ct. 431, 27 L.Ed.2d 476 (1971), held that changes in the boundary lines of a city by annexations that enlarge the number of eligible voters are events covered by § 5. The question in this case is whether the District Court for the District of Columbia correctly held that the electoral plan for the Port Arthur, Texas, City Council could not be approved under § 5 because it insufficiently neutralized the adverse impact upon minority voting strength that resulted from the expansion of the city's borders by two consolidations and an annexation. 2 * In December 1977, the city of Port Arthur, Texas, consolidated with the neighboring cities of Pear Ridge and Lakeview. Six months later, the city annexed Sabine Pass, an incorporated area. As a result of these expansions of the City's borders, the percentage of the black population in Port Arthur decreased from 45.21% to 40.56%. Blacks of voting age comprised 35% of the population of the enlarged city.3 3 Prior to the expansions, the City was governed by a seven-member council, including a mayor, each member being elected at large by majority vote. Each member except the mayor was required to reside in a specific district of the city. Members were elected for staggered terms. Following the two consolidations, the City Council passed an ordinance adding an eighth member to the council, while retaining the at-large system with residency requirements. After the annexation of Sabine Pass, the City further proposed that the council be expanded to nine members, with at large elections as before. The two consolidations and the annexation, together with the proposed changes in the governing system, were submitted to the Attorney General for preclearance pursuant to § 5 of the Voting Rights Act. The Attorney General refused preclearance, suggesting, however, that he would reconsider if the council members were elected from fairly drawn single-member districts. 4 As § 5 permitted it to do, the City then filed suit in the United States District Court for the District of Columbia seeking a declaratory judgment that the expansions and the nine-member plan did not have the purpose or effect of denying or abridging the right to vote on account of color or race within the meaning of § 5. While that suit was pending, the city approved by referendum the "4-4-1" plan, calling for four members to be elected from single-member districts, four to be elected at large from residency districts identical to the single-member districts, and the ninth member, the mayor, to be elected at large without any residency requirement.4 That plan, like the previous plans, required a majority vote to elect each council member. The city then moved to amend its complaint so as to seek a declaratory judgment as to the legality of the 4-4-1 plan. 5 The District Court concluded that because there were legitimate purposes behind the annexation and the consolidations, those actions, under City of Richmond v. United States, 422 U.S. 358, 95 S.Ct. 2296, 45 L.Ed.2d 245 (1975), could not be denied preclearance as discriminatory in purpose. Because the expansions had substantially reduced the relative political strength of the black population, however, it was necessary for preclearance that the post-expansion electoral system be found to satisfy the requirements of § 5. The District Court held that neither the first nine-member plan nor the 4-4-1 plan measured up, not only because each was adopted with a discriminatory purpose, but also because in the context of the severe racial bloc voting characteristic of the recent past in the City neither plan adequately reflected the minority's potential political strength in the enlarged community as required under City of Rome v. United States, 446 U.S. 156, 100 S.Ct. 1548, 64 L.Ed.2d 119 (1980); City of Richmond v. United States, supra; and City of Petersburg v. United States, 354 F.Supp. 1021 (DDC1972), aff'd, 410 U.S. 962, 93 S.Ct. 1441, 35 L.Ed.2d 698 (1973). 6 Soon after this decision, the city and the United States jointly submitted to the Court for approval the "4-2-3" electoral plan. Under this scheme, the city would be divided into four single-member districts, Districts 1 through 4. District 5, comprising Districts 1 and 4 would elect another member, as would District 6, which combined Districts 2 and 3. Three additional members would be elected at large, one each from Districts 5 and 6, the third at-large seat to be occupied by the mayor and to have no residency requirement. All council seats would be governed by the majority-vote rule, that is, runoffs would be required if none of the candidates voted on received a majority of the votes cast. Blacks constituted a majority in Districts 1 and 4, 79% and 62.78% respectively, as well as a 70.83% majority of the fifth district combining the two majority black districts. The sixth district was 10.98% black. Although the United States expressed reservations about the at-large and majority-vote features, its position was that neither of these aspects of the plan warranted a denial of preclearance. 7 After response to and oral argument upon the submission, the District Court concluded "that the proposed plan insufficiently neutralizes the adverse impact upon minority voting strength which resulted from the expansion of Port Arthur's borders." The court added, however, that if the plan were modified to eliminate the majority-vote requirement with respect to the two non-mayoral, at-large candidates, and to permit election to these two seats to be made by a plurality vote, the court "would consider the defect remedied and offer our approval." This appeal followed, the basic submission being that under § 5 and the controlling cases the District Court exceeded its authority in conditioning clearance of the 4-2-3 plan on the elimination of the majority-vote requirement.5 We noted probable jurisdiction. --- U.S. ----, 102 S.Ct. 1272, 71 L.Ed.2d 457. II 8 Perkins v. Matthews, 400 U.S. 379, 91 S.Ct. 431, 27 L.Ed.2d 476 (1971), held that annexations by a city are subject to § 5 preclearance because increasing the number of eligible voters dilutes the weight of the votes of those to whom the franchise was limited before the annexation and because the right to vote may be denied by dilution or debasement just as effectively as by wholly prohibiting the franchise. It soon became clear, however, that § 5 was not intended to forbid all expansions of municipal borders that could be said to have diluted the voting power of particular groups in the community. In City of Petersburg v. United States, supra, the annexation of an area with a heavy white majority resulted in reducing the black community from majority to minority status. The District Court held that the annexation could nevertheless be approved but "only on the condition that modifications [in the electoral plan] calculated to neutralize to the extent possible any adverse effect upon the political participation of black voters are adopted, i.e., that the [City] shift from an at-large to a ward system of electing its city councilmen." 354 F.Supp., at 1031. We affirmed summarily. 410 U.S. 962, 93 S.Ct. 1441, 35 L.Ed.2d 698 (1973). 9 Later, in City of Richmond v. United States, supra, we expressly reaffirmed Petersburg, recognizing that the Petersburg annexation enhanced the power of the white majority to exclude Negroes from the city council but stating that such a consequence "would be satisfactorily obviated if at-large elections were replaced by a ward system of choosing councilmen." 422 U.S., at 370, 95 S.Ct., at 2303. It was our view that a fairly-designed ward plan "would not only prevent the total exclusion of Negroes from membership on the council but would afford them representation reasonably equivalent to their political strength in the enlarged community." Ibid. We applied these principles in City of Richmond. There, the annexation of a heavily white area reduced the black population of the City from 52% to 42%, and the electoral proposal submitted for preclearance replaced the prior system of at-large elections with a single-member plan under which blacks would be in a substantial majority in four of the nine councilmanic districts. We held that as long as the ward system fairly reflected the strength of the Negro community as it existed after the annexation, preclearance under § 5 should be granted. Under such a plan, "Negro power in the new city [would not be] undervalued, and Negroes [would] not be underrepresented on the council." Id., at 371, 95 S.Ct., at 2304. The annexation could not, therefore, be said to have the effect of denying or abridging the right to vote on account of race within the meaning of § 5. 10 In the case before us, Port Arthur was a party to two consolidations and an annexation. Because the areas taken into the city were predominantly white, the relative percentage of blacks in the enlarged city was substantially less than it was before the expansions. The District Court refused preclearance because in its view the post-expansion electoral system did not sufficiently dispel the adverse impact of the expansions on the relative political strength of the black community in Port Arthur. The City submits that this judgment was in error under Petersburg and Richmond. 11 Richmond, however, involved a fairly drawn, single-member district system that adequately reflected the political strength of the black community in the enlarged city. The plan was consequently an acceptable response to the annexation's adverse impact on minority voting potential. It does not necessarily follow that the mixed single-member and at-large system at issue in this case sufficiently dispelled the impact of Port Arthur's expansions on the relative political strength of the black community. The District Court concluded that although the 4-2-3 system provided a black majority in three councilmanic districts, it was necessary also to eliminate the majority-vote requirement with respect to the two non-mayoral at-large council positions. For several reasons, we cannot say that the District Court erred in this respect. 12 First, whether the 4-2-3 plan adequately reflected the political strength of the black minority in the enlarged city, is not an issue that is determinable with mathematical precision. Because reasonable minds could differ on the question and because the District Court was sitting as a court of equity seeking to devise a remedy for what otherwise might be a statutory violation, we should not rush to overturn its judgment. Cf. Swann v. Board of Education, 402 U.S. 1, 15, 91 S.Ct. 1267, 1275, 28 L.Ed.2d 554 (1971). 13 Second, the 4-2-3 plan undervalued to some extent the political strength of the black community: one-third of the council seats was to be elected from black majority districts, but blacks comprised 40.56% of the population of the enlarged city and 35% of the voting age population. In light of this fact, eliminating the majority-vote requirement was an understandable adjustment. As the District Court well understood, the majority-vote rule, which forbade election by a plurality, would always require the black candidate in an at-large election, if he survived the initial round, to run against one white candidate. In the context of racial bloc voting prevalent in Port Arthur, the rule would permanently foreclose a black candidate from being elected to an at-large seat. Removal of the requirement, on the other hand, might enhance the chances of blacks to be elected to the two at-large seats affected by the District Court's conditional order but surely would not guarantee that result. Only if there were two or more white candidates running in a district would a black have any chance of winning election under a plurality system. We cannot say that insisting on eliminating the majority-vote rule in the two at-large districts would either overvalue black voting strength in Port Arthur or be inconsistent with Richmond. 14 Third, even if the 4-2-3 electoral scheme might otherwise be said to reflect the political strength of the minority community, the plan would nevertheless be invalid if adopted for racially discriminatory purposes, i.e., if the majority-vote requirement in the two at-large districts had been imposed for the purpose of excluding blacks from any realistic opportunity to represent those districts or to exercise any influence on council members elected to those positions. City of Richmond v. United States, 422 U.S. 358, 378-379, 95 S.Ct. 2296, 2307, 45 L.Ed.2d 245. The District Court made no finding that the 4-2-3 plan was tainted by an impermissible purpose; but it had found that the two preceding plans, the first nine-member plan and the 4-4-1 plan, had been adopted for the illicit purpose of preventing black candidates from winning election. The court had also found that the majority-vote requirement was a major means of effectuating this discriminatory end. When it was then presented with the 4-2-3 plan retaining the requirement for the two non-mayoral at-large seats, the Court conditioned approval on eliminating the majority-vote element. It seems to us that in light of the prior findings of discriminatory purpose such action was a reasonable hedge against the possibility that the 4-2-3 scheme contained a purposefully discriminatory element. On balance, we cannot fault the judgment of the District Court. 15 The judgment of the District Court is accordingly 16 Affirmed. Justice POWELL, with whom Justice REHNQUIST and Justice O'CONNOR join, dissenting. 17 The Court affirms the District Court's order, concluding that although the 4-2-3 plan ensures proportional representation for the black voting age population, a District Court nevertheless is free under § 5—in the exercise of a newly perceived equitable jurisdiction—to require a city to "enhance" the chances of increased minority representation on a city's governing body. In this case, the perceived enhancement would be that a plurality, rather than a majority election requirement, would give black citizens a better chance of capturing—in addition to the three district seats assured them—the two at-large seats. Ante, Pp.167-1681 Because the Court's decision is irreconcilable with City of Richmond v. United States, 422 U.S. 358, 95 S.Ct. 2296, 45 L.Ed.2d 245 (1975), and authorizes a standardless equitable jurisdiction in district courts, I dissent. 18 * In City of Richmond, the city annexed territory reducing the percentage of the city's black population from 52% to 42%. After the Attorney General refused to preclear submitted election plans, he and the city came to an agreement and jointly submitted a plan for approval to the District Court for the District of Columbia. The District Court rejected this plan, because the city had failed to "minimiz[e] the dilution of black voting power to the greatest possible extent." Id., at 367, 95 S.Ct., at 2302. This Court, in an opinion by Justice WHITE, vacated the District Court's order, holding that a District Court must accept a new electoral plan for the enlarged municipality as long as it "fairly reflects the strength of the Negro community as it exists after the annexation" and "would afford [it] representation reasonably equivalent to [its] political strength in the enlarged community." Id., at 370-371, 95 S.Ct., at 2303. See City of Rome v. United States, 446 U.S. 156, 187, 100 S.Ct. 1548, 1566, 64 L.Ed.2d 119 (1980), aff'ing, 472 F.Supp. 221, 245 (D.D.C.1979); City of Rome, 446 U.S., at 188, 100 S.Ct., at 1567 (BLACKMUN, J., concurring); United Jewish Organizations v. Carey, 430 U.S. 144, 160, 97 S.Ct. 996, 1007, 51 L.Ed.2d 229 (1977) (opinion of WHITE, J.); Beer v. United States, 425 U.S. 130, 139 n. 11, 96 S.Ct. 1357, 1363 n. 11, 47 L.Ed.2d 629 (1970). In dissent, Justice BRENNAN stated that he would find the dilutive effect of an annexation cured only by an election plan "calculated to neutralize to the extent possible any adverse effect upon the political participation of black voters." Id., at 389, 95 S.Ct., at 2312. 19 In this case, the city expanded its boundaries by annexation and consolidation.2 This resulted in reducing the percentage of its black population from 45% to 40%. The electoral plan for the enlarged city, submitted to the Attorney General under § 5 of the Voting Rights Act of 1965, was disapproved both by the Attorney General and then by a District Court in the District of Columbia. Following negotiations, the Attorney General and the city reached agreement that the 4-2-3 electoral plan—at issue in this case—complied with the requirements of the Voting Rights Act. Accordingly, the plan was jointly submitted by the Attorney General and the city to the District Court for its approval. Under this plan, the city's 35% black voting age population was assured of 33% of the city council positions, i.e., three of nine members. 20 The District Court rejected the agreed upon plan in a brief order because, in words reminiscent of Justice BRENNAN's dissent in City of Richmond, it "insufficiently neutralizes the adverse impact upon minority voting strength." The court added, however, that it would approve the plan were it modified "so as to provide for the election of the two non-mayoral, at-large representatives by plurality vote," a condition to approval that the Attorney General had expressly considered and found not to be required by the Act. 21 I find the Court's decision in City of Richmond and in this case fundamentally inconsistent, because the proportional representation assured by the 4-2-3 plan must, by definition, "afford [blacks] representation reasonably equivalent to their political strength in the enlarged community." Cf. United Jewish Organizations, 430 U.S., at 169, 97 S.Ct., at 1011 (BRENNAN, J., concurring in part) ("[T]he very definition of proportional representation precludes either underrepresentation or overrepresentation. . . ."). Apparently in an effort to justify its decision, the Court states that the agreed 4-2-3 plan "undervalued to some extent the political strength of the black community." Ante, p. 167. No support for this statement is cited, and none is found in the record.3 The District Court made no such finding and the Government, in its submission to the District Court, expressly asserted that the city's plan "would appear to provide the minority community with a fair opportunity to obtain 'representation reasonably equivalent to their political strength in the enlarged community.' City of Richmond v. United States, 422 U.S. 358, 370, 95 S.Ct. 2296, 2303, 45 L.Ed.2d 245 (1975)." App., p. 79a-80a. The black intervenors also agreed at the time of the submission that "the plan does approach affording blacks representation reasonably equivalent to their voting strength in the at-large community. . . ." App., p. 83a. II 22 Furthermore, the Court's decision finds no support in any prior decision of this Court. The theory that political strength should be enhanced, rather than preserved, is new doctrine. It is a view Congress has never embraced, and indeed one that the 1982 extension of the Voting Rights Act fairly can be viewed as rejecting.4 Moreover, although I do not question the power of a District Court to disagree with the Attorney General's construction of the Act, it does not follow that the District Court was "sitting as a court of equity," ante, at 167, and had the power to require political enhancement. We are interpreting and applying a statute that vests no such open-ended jurisdiction in any court. 23 In the first six months of this year, the Department of Justice received approximately 8,709 applications for preclearance of voting changes under § 5, an average of 66 per working day.5 Congress, with the approval of the President, has recently reaffirmed the authority of Department of Justice personnel to exercise this extensive control over state and local political decisions. The sheer volume of applications for preclearance makes imperative the prescribing of predictable standards. Proportional representation, whatever its theoretical and practical limitations may be in a nation with populations as diverse and mobile as that of the United States, is at least an objective standard, and when it is found to exist in a § 5 case—whether deemed necessary under the Act or not—it should be dispositive. The Court today, however, finds for the first time a standardless equitable discretion in District Courts of the District of Columbia to impose requirements in addition to proportional representation. This leaves the responsible authorities in the state and communities under the Act as well as the Attorney General—without guidance as to the requirements of § 5. III 24 The Court's discussion of discriminatory purpose as providing some support for the District Court's "effects" determination is disquieting for a number of reasons. First, as the Court notes, the District Court made no finding that the 4-2-3 plan was tainted by an impermissible purpose. Second, the District Court expressly found that no discriminatory motive prompted the city's annexation of the three jurisdictions involved. 517 F.Supp., at 1019-1021. Third, the factors that led the District Court to conclude that the earlier 8-0-1 and 4-4-1 plans had been adopted for a discriminatory purpose have no bearing on the question whether the city was similarly motivated when it adopted the 4-2-3 plan at a later time and pursuant to good-faith negotiations with the Attorney General. Finally, the Government concedes that purpose is not a factor in this case.6 Indeed, the Court fails to explain nor can it explain satisfactorily—how a plan negotiated with and acceptable to the Attorney General was adopted for a discriminatory purpose. 25 In my opinion, the city has shown that its 4-2-3 plan has satisfied fully § 5's effect-and-purpose test and the standard adopted in City of Richmond. We now should demand no more. I would reverse the District Court's order. 1 It is undisputed that the city of Port Arthur is a political subdivision to which § 5 is applicable. See 46 Fed.Reg. 870, 880. 2 Section 5 in relevant part provides as follows: Whenever a State or political subdivision with respect to which the prohibitions set forth in section 1973b(a) of this title based upon determinations made under the first sentence of section 1973b(b) of this title are in effect shall enact or seek to administer any voting qualification or prerequisite to voting, or standard, practice, or procedure with respect to voting different from that in force or effect . . . such State or subdivision may institute an action in the United States District Court for the District of Columbia for a declaratory judgment that such qualification, prerequisite, standard, practice, or procedure does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, and unless and until the court enters such judgment no person shall be denied the right to vote for failure to comply with such qualification, prerequisite, standard, practice, or procedure: Provided, That such qualification, prerequisite, standard, practice, or procedure may be enforced without such proceeding if the qualification, prerequisite, standard, practice, or procedure has been submitted by the chief legal officer or other appropriate official of such State or subdivision to the Attorney General and the Attorney General has not interposed an objection within sixty days after such submission, or upon good cause shown, to facilitate an expedited approval within sixty days after such submission, the Attorney General has affirmatively indicated that such objection will not be made. 3 The pre-annexation and post-annexation percentages are based on the 1980 census. The figure for the percentage of blacks in the voting age population is an estimate, which the district court derived by extrapolating from the 1970 census data. The 1970 census showed that at that time 34.6% of the voting age population was black while 40.01% of the general population was black. The district court itself noted the dangers of extrapolation, but explained that both parties had suggested the procedure for determining the percentage of the current voting age population that is black. Port Arthur also has a Hispanic community, which comprises 6.30% of the enlarged city's population. 4 The United States unsuccessfully sought to enjoin the referendum election before a three-judge court in the Eastern District of Texas. United States v. City of Port Arthur, No. B-80-216-CA (ED Tex. Sept. 5, 1980). 5 The City argues that the District Court was required to approve a plan jointly submitted by the City and the Attorney General. The Voting Rights Act, however, assigns primary responsibility to the District Court to determine whether a change in voting procedures violates § 5. Preclearance by the Attorney General may obviate a court suit, but here the Attorney General was acting in the capacity of a litigant when it joined the City in submitting a plan for the court's consideration. In that posture, neither the Attorney General, the City, nor both of them together could dictate the court's conclusion as to the acceptability of the plan under § 5. 1 The Court has recognized that a majority vote requirement in at-large elections, unless adopted as a change for discriminatory purposes, is a valid and long-accepted practice "that is followed by literally thousands of municipalities and other local governmental units throughout the Nation." See City of Mobile v. Bolden, 446 U.S. 55, 60, 100 S.Ct. 1490, 1495, 64 L.Ed.2d 47 (1980) (plurality opinion). 2 The District Court acknowledged benefits for the entire population from consolidation: "Port Arthur . . . was extremely interested in maintaining a population in excess of 50,000 so as to remain entitled as a matter of right to funds from federal agencies including the Department of Urban Development ("HUD"). Were the population to decrease below the 50,000 level, HUD would diminish the amount of the direct grant by one-third each year; in the fourth year, the City would have to compete with other applicants for discretionary awards. Since 1975, there was evidence that the municipal population was [declining towards] the 50,000 mark. . . . Having already annexed all of the adjacent black communities, the City turned to Pear Ridge, Lakeview and Griffing Park. * * * * * Although the City would be required to provide services to the new residents, it was anticipated that the additional cost would be minimal and greatly outweighed by the increased tax revenue. . . . Furthermore, Port Arthur hoped that the increased visibility resulting from consolidation would attract new businesses and thereby create new jobs." 517 F.Supp. 987, 999. 3 In interim elections held in 1981, the city's electorate chose three black council members. In fact, the city notes that it is now governed by a council consisting of four blacks and five whites. Reply Brief for Appellant 6. The Court seems to rely on two factors for its conclusion: a slight differential between the percentage of black seats and the percentage of black voting age population; and a larger differential between the percentage of black seats and the percentage of the black population. There is a preference for voting age population statistics, see United Jewish Organizations, 430 U.S., at 164 n. 23, 97 S.Ct., at 1009 n. 23 (opinion of WHITE, J.), because they are more "probative" of the "electoral potential of the minority community," City of Rome, 446 U.S., at 186 n. 22, 100 S.Ct., at 1566 n. 22, than population statistics. Even if the Court were to rely on population statistics here, this Court's formulations reflect the recognition that it would be unreasonable, if not impossible, to require cities to devise voting plans that afford minorities representation precisely proportional to their political strength in the jurisdiction. Indeed, the Court has indicated that proportional representation would be found in circumstances quite similar to those presented here. See Beer, 425 U.S., at 159 n. 19, 96 S.Ct., at 1372 n. 19 (1976) (MARSHALL, J., dissenting) (approving representation/voting age population differential of 6%). Moreover, the Court's conclusion that the 4-2-3 plan will "permanently foreclose" blacks from being elected to either of the at-large seats, ante, at 167, ignores the dynamics of the region, to which the facts of this case attest. With 35% of the voting age population composed of black citizens, it is politically naive to think that these citizens will not have significant—and indeed often decisive—influence in the election of at-large council members. The results in numerous state and local elections demonstrate the political power of such a large and cohesive segment of the electorate. See J. Wilkinson, Harry Byrd 346 (1968) ("By the middle of the 1960's . . . Negroes provided balance-of-power ballots [in Virginia and] elsewhere in the South....."). 4 Section 2(b) of the Voting Rights Act of 1982, Pub.L. 97-205, 96 Stat. 131, states that a violation has been established if it is shown, "based on the totality of circumstances," that the political processes "are not equally open to blacks." The amendment expressly provides that "[t]he extent to which members of a protected class have been elected to office . . . is one circumstance which may be considered. . . ." The Senate committee report stated that: "Electoral devices, including at-large elections, per se would not be subject to attack under Section 2. They would only be vulnerable if, in the totality of circumstances, they resulted in the denial of equal access to the electoral process. [T]he presence of minority elected officials is a recognized indicator of access to the process. . . ." S.Rep. 97-417, 97th Cong., 2d Sess., 16 (1982), U.S.Code Cong. & Admin.News 1982, pp. 177, 193. 5 See Department of Justice, Civil Rights Division, Voting Rights Section, Number of Changes Submitted under Section 5 and Reviewed by the Department of Justice, By State and Year, 1965 June 30, 1982 (unpublished). 6 The following exchange took place at oral argument: The Court: And may I get clear, is purpose still in this case at this level? The Government: Not in terms of the submission to this Court, no, Your Honor. The Court: So we consider only the effect? The Government: Yes, Your Honor. I don't believe that the district court's opinion or order can fairly be read to cast any doubt on the purpose of the plan as adopted. Tr. of Oral Arg. 30. * * *

12

459 U.S. 145 103 S.Ct. 523 74 L.Ed.2d 323 XEROX CORPORATION, Appellantv.COUNTY OF HARRIS, TEXAS and City of Houston, Texas No. 81-1489. Argued Nov. 10, 1982. Decided Dec. 13, 1982. Syllabus Appellant, a New York corporation that manufactures and sells copying machines, shipped machine parts manufactured in this country to Mexico City, Mexico, for assembly by its affiliate there. After assembly, the copiers were transported by a customs bonded trucking company to a customs bonded warehouse in Houston, Tex., where they were segregated from other merchandise and stored while awaiting sale and shipment to appellant's affiliates in Latin America. None of these copiers were ever sold to customers for domestic use, but remained under the continuous control and supervision of the United States Customs Service from the time they entered the warehouse until they cleared United States Customs at their export ports. In 1977, both the city of Houston and Harris County (appellees) assessed ad valorem personal property taxes on the copiers stored in the Houston warehouse. Appellant sought declaratory and injunctive relief in state court, claiming that the taxes were unconstitutional. Appellees counterclaimed for the taxes assessed. The trial court entered judgment for appellant, holding that the taxes violated the Import-Export and Commerce Clauses of the Constitution. The Texas Court of Civil Appeals reversed and granted judgment to appellees on their counterclaims, holding that the taxes violated neither Clause of the Constitution and, alternatively, that the trial court had violated state law in granting injunctive relief. Held: 1. This Court has jurisdiction over the appeal under 28 U.S.C. § 1257(2). Notwithstanding appellees' argument that this Court lacks jurisdiction because the Court of Civil Appeals' decision reversing the grant of an injunction rested on an independent and adequate state ground, an indispensable predicate to the award of judgment to appellees on their counterclaims was a determination that the taxes were permissible under the Federal Constitution. P.149 2. State property taxes, such as those involved here, on goods stored under bond in a customs warehouse are pre-empted by Congress' comprehensive regulation of customs duties. Under the customs system, established pursuant to Congress' powers under the Commerce Clause, imports may be stored duty free in Government-supervised bonded warehouses for prescribed periods, and during such periods, may be withdrawn and reexported without payment of duty. Only if the goods are withdrawn for domestic sale or stored beyond the prescribed period does any duty become due. Congress created such duty-free enclaves under federal control in order to encourage merchants here and abroad to make use of American ports as transshipment centers for goods in foreign trade. It would not be compatible with the comprehensive scheme Congress enacted to effect these goals if the states were free to tax such goods while they were lodged temporarily in Government-regulated bonded storage in this country. Cf. McGoldrick v. Gulf Oil Corp., 309 U.S. 414, 60 S.Ct. 664, 84 L.Ed. 840. Pp. 150-154. 619 S.W.2d 402, reversed and remanded. Alfred H. Hoddinott, Jr., for appellant. Cheryl Helena Chapman, Houston, Tex., for appellees. Chief Justice BURGER delivered the opinion of the Court. 1 We noted probable jurisdiction to decide whether a state may impose nondiscriminatory ad valorem personal property taxes on imported goods stored under bond in a customs warehouse and destined for foreign markets. The Texas Court of Civil Appeals held such taxes constitutional. 2 * Appellant Xerox Corporation is a New York corporation engaged in the business of manufacturing and selling business machines. Its operations span the globe, and it has established affiliates in foreign countries to facilitate foreign sales. It has assembly plants and production facilities in Mexico. 3 Xerox manufactured parts for copying machines in Colorado and New York which were shipped to Mexico City, Mexico for assembly by its affiliate there. The copiers assembled in Mexico were designed for sale in the Latin American market and all printing on the machines and instructions accompanying them were in Spanish or Portuguese. Most of the copiers operated on electric current of 50 cycles per second, rather than the 60 cycles per second that is standard in the United States. Many of the copiers assembled by its affiliate in Mexico City were not approved by either United Laboratories or Canadian Standards Association, as required for sale in the United States. To convert the copiers for domestic sale would have cost approximately $100 per copier. 4 After assembly in Mexico, the copiers were transported by a customs bonded trucking company to the Houston Terminal Warehouse in Houston, Texas,1 a Class 3 customs bonded warehouse. There they were stored for periods ranging from a few days to three years while awaiting sale and shipment to Xerox affiliates in Latin America. The copiers remained in the warehouse, segregated from other merchandise, until a shipment order was received. When Xerox received an order, it would transport the copiers under bond to either the Port of Houston or the Port of Miami, where they were loaded on board vessels for shipment to Latin America. The copiers remained under the continuous control and supervision of the United States Customs Service from the time they entered the bonded warehouse until they cleared United States Customs at the Port of Houston or the Port of Miami for export.2 5 None of the copiers assembled in Mexico and stored in Houston were ever sold to customers for domestic use; all were ultimately sold abroad. Consequently, Xerox paid no import duties on them.3 6 The local authorities did not assess any taxes on the copiers stored under customs bond in 1974 and 1975. In 1977, the City of Houston4 assessed ad valorem personal property taxes of $156,728 on the copiers stored in the Houston warehouse during 1977.5 The County of Harris6 followed suit, assessing $55,969 in taxes for 1977 and also assessing $48,426 in back taxes for 1976, for a total of $104,395. 7 As soon as Xerox learned that the local authorities intended to tax its Mexican-assembled copiers, it shipped all the machines to a foreign trade zone in Buffalo, New York, from which it continued to fill orders for shipment to Latin America. 8 Declaratory and injunctive relief was sought in state court both from the taxes already assessed and such taxes as appellees might impose in the future. Xerox claimed that the taxes in question were unconstitutional because they violated the Import-Export Clause and the Commerce Clause of the Constitution. Art. I., § 10, cl. 2; Art. I, § 8, cl. 3. Appellees counterclaimed for the taxes assessed. 9 The trial court held that the taxes assessed by appellees violated both the Import-Export and Commerce Clauses, and it granted judgment to Xerox. The Texas Court of Civil Appeals, 1st District, reversed, holding that the taxes violated neither the Import-Export Clause nor the Commerce Clause. 619 S.W.2d 402 (1981). Alternatively, it held that the trial court had violated Tex.R.Civ.P. 683 in granting injunctive relief. Finally, it granted judgment to appellees on their counterclaims in the amount of $131,311 plus penalties and interest to Harris County and $156,728 plus penalties and interest to City of Houston. The Texas Supreme Court denied Xerox's application for a writ of error and this appeal followed. We noted probable jurisdiction, --- U.S. ----, 102 S.Ct. 1766, 72 L.Ed.2d 172 (1982), and we reverse. II A. 10 A preliminary question is whether this Court has jurisdiction over the appeal. Appellees argue that this Court lacks jurisdiction since the decision of the Texas court reversing the grant of an injunction rested on an independent and adequate state ground. However, an indispensable predicate to an award of judgment to the appellees on their counterclaims was a determination that the taxes at issue were permissible under the United States Constitution; the Texas Court of Civil Appeals so held. It is not claimed that any independent and adequate state law ground supports that holding. We therefore have jurisdiction to review that judgment. 28 U.S.C. § 1257(2). B 11 Pursuant to its powers under the Commerce Clause, Congress established a comprehensive customs system which includes provisions for government supervised bonded warehouses where imports may be stored duty-free for prescribed periods. At any time during that period the goods may be withdrawn and reexported without payment of duty. Only if the goods are withdrawn for domestic sale or stored beyond the prescribed period does any duty become due. 19 U.S.C. § 1557(a) (1976). While the goods are in bonded warehouses they are in the joint custody of the United States Customs Service and the warehouse proprietor and under the continuous control and supervision of the local customs officers. 19 U.S.C. § 1555 (1976). Detailed regulations control every aspect of the manner in which the warehouses are to be operated. 19 CFR §§ 19.1-19.6 (1981). 12 Government regulated, bonded warehouses have been a link in the chain of foreign commerce since "a very early period in our history." Fabbri v. Murphy, 95 U.S. 191, 197, 94 L.Ed. 468 (1877). A forerunner of the present statute was the Warehousing Act of 1846, ch. 84, 9 Stat. 53 (1846). A major objective of the warehousing system was to allow importers to defer payment of duty until the goods entered the domestic market or were exported. The legislative history explains that Congress sought to reinstate: 13 "the sound though long neglected maxim of Adam Smith, 'that every tax ought to be levied at the time and in the manner most convenient for the contributor to pay it;' [by providing] that the tax shall only be paid when the imports are entered for consumption. . . ." H.R.Rep. No. 411, 26th Cong., 1st Sess. 3 (1846). 14 The Act stimulated foreign commerce by allowing goods in transit in foreign commerce to remain in secure storage, duty free, until they resumed their journey in export. The geographic location of the country made it a convenient place for transshipment of goods within the Western Hemisphere and across both the Atlantic and the Pacific. A consequence of making the United States a center of world commerce was that: 15 ". . . our carrying trade would be vastly increased; that shipbuilding would be stimulated; that many foreign markets would be supplied, wholly or in part, by us with merchandise now furnished from the warehouses of Europe; that the industry of our seaports would be put in greater activity; [and] that the commercial transactions of the country would be facilitated. . . ." App. to Cong.Globe, 29th Cong., 1st Sess. 792 (1846) (remarks of Sen. Dix). 16 To these ends, Congress was willing to waive all duty on goods that were reexported from the warehouse, and to defer, for a prescribed period, the duty on goods destined for American consumption. This was no small sacrifice at a time when customs duties made up the greater part of federal revenues,7 but its objective was to stimulate business for American industry and work for Americans. 17 In short, Congress created secure and duty free enclaves under federal control in order to encourage merchants here and abroad to make use of American ports. The question is whether it would be compatible with the comprehensive scheme Congress enacted to effect these goals if the states were free to tax such goods while they were lodged temporarily in government regulated bonded storage in this country. 18 In McGoldrick v. Gulf Oil Co., 309 U.S. 414, 60 S.Ct. 664, 84 L.Ed. 840 (1939), the City of New York sought to impose a sales tax on imported petroleum that was refined into fuel oil in New York and sold as ships' stores to vessels bound abroad. The crude oil was imported under bond and refined in a customs bonded manufacturing warehouse and was free from all duties. We struck down the state tax, finding it preempted by the Congressional scheme. Id., at 429, 60 S.Ct., at 669. 19 The Court determined that the purpose of the exemption from the tax normally laid upon importation of crude petroleum was "to encourage importation of the crude oil for [refinement into ships' stores] and thus to enable American refiners to meet foreign competition and to recover trade which has been lost by the imposition of the tax." 309 U.S., at 427, 60 S.Ct., at 668; see also id., at 428, 60 S.Ct., at 669. The Court went on to note that, in furtherance of this purpose, 20 "Congress provided for the segregation of the imported merchandise from the mass of goods within the state, prescribed the procedure to insure its use for the intended purpose, and by reference confirmed and adopted customs regulations prescribing that the merchandise, while in customs bonded warehouse, should be free from state taxation." Id., at 428-429, 60 S.Ct., at 669. The Court concluded that 21 "the purpose of the Congressional regulation of commerce would fail if the state were free to impose a tax which would lessen the competitive advantage conferred on the importer by Congress, and which might equal or exceed the remitted import duty." Ibid. 22 In so deciding, the Court expressly declined to rely on the customs regulation "prescribing the exemption from state taxation," holding that the regulation merely stated "what is implicit in the Congressional regulation of commerce presently involved." Ibid.8 23 The analysis in McGoldrick applies with full force here. First, Congress sought, in the statutory scheme reviewed in McGoldrick, to benefit American industry by remitting duties otherwise due. The import tax on crude oil was remitted to benefit oil refiners employing labor at refineries within the United States, whose products would not be sold in domestic commerce. Here, the remission of duties benefited those shippers using American ports as transshipment centers. Second, the system of customs regulation is as pervasive for the stored goods in the present case as it was in McGoldrick for the refined petroleum. In both cases, the imported goods were segregated in warehouses under continual federal custody and supervision. Finally, the state tax was large enough in each case to offset substantially the very benefits Congress intended to confer by remitting the duty.9 In short, freedom from state taxation is as necessary to the Congressional scheme here as it was in McGoldrick. 24 Although there are factual distinctions between this case and McGoldrick, they are distinctions without a legal difference. We can discern no relevance to the issue of Congressional intent in the fact that the fuel oil in McGoldrick could be sold only as ships' stores whereas Xerox had the option to pay the duty and withdraw the copiers for domestic sale, or that in McGoldrick the City sought to impose a sales tax and here appellees assessed a property tax. 25 A similar conclusion was reached in District of Columbia v. International Distributing Corp., 331 F.2d 817 (CADC 1964). There, the Court of Appeals held that a wholesaler of imported alcoholic beverages was not liable for District of Columbia excise taxes on the sale of such beverages to foreign embassies while the beverages were stored in a customs bonded warehouse. The court reasoned that Congress intended to make customs bonded warehouses federal enclaves free of state taxation and that although the imported goods were physically within the District of Columbia, they were not subject to its taxing jurisdiction until they were removed from the warehouse. Since the sales took place prior to removal, the District could not tax those sales. The Court of Appeals quoted with approval the following language of the Tax Court: 26 "The idea of bonded warehouses and their use by the United States customs authorities negatives the proposition that at the time of sale the alcoholic beverages were in the possession of the petitioner [the corporation.] True it is that the private bonded warehouse was physically in the District of Columbia; and the liquors were stored therein; and in that sense they were in the District. In law, however, they were still without that jurisdiction, and did not become subject thereto until they had been withdrawn from the private bonded warehouse and removed from the control of the customs official." 331 F.2d, at 820. 27 International Distributing Corp. merely confirms what this Court said in 1877 in Fabbri v. Murphy, supra, 95 U.S., at 197-198: "Congress did not regard the importation as complete while the goods remained in the custody of the proper officers of customs." 28 Accordingly, we hold that state property taxes on goods stored under bond in a customs warehouse are preempted by Congress's comprehensive regulation of customs duties. III 29 It is unnecessary for us to consider whether, absent Congressional regulation, the taxes here would pass muster under the Import-Export Clause or the Commerce Clause. 30 The judgment of the Texas Court of Civil Appeals is reversed, and the case is remanded for proceedings not inconsistent with this opinion. 31 Reversed and Remanded. 32 Justice POWELL, dissenting. 33 Since 1799 the United States has permitted importers to post a customs bond in lieu of immediate payment of customs duties on imported goods. Today the Court holds that these goods stored in customs-bonded warehouses also are exempt from state property taxation. This holding would be unremarkable were it based on any evidence of congressional intent, but such support is lacking. The Court instead finds that state taxation is incompatible with the purposes of the federal customs-bonded warehousing system. 34 Customs-bonded storage enables importers to defer paying customs duties until the goods are ready for domestic sale or to avoid paying duties altogether if the goods are reexported. The Court correctly observes that Congress's ultimate purpose has been to encourage imports and enhance the position of the United States as a center of international trade. I am not persuaded, however, that nondiscriminatory state taxation of customs-bonded goods is incompatible with this purpose. 35 The Court attributes significance to the "pervasive" system of customs regulation of stored goods, ante, at 153, but fails to explain how this affects a state's power to tax. The purpose of the regulation is to guarantee the security of federal revenues. The owner of customs-bonded goods eventually must pay the customs duties or reexport the goods. The warehousing system enables the Federal Government to monitor the removal of bonded goods for sale or export and ensure that duties are paid when due. A state's imposition of an ad valorem tax does not impair this function. The "pervasive" regulation of the manner in which customs-bonded goods are stored and withdrawn, therefore, is simply immaterial to the validity of state taxation of those goods. 36 The Court also argues that state taxation of customs-bonded goods would frustrate the congressional purpose of encouraging foreign trade with the United States. It asserts that appellees' taxes are large enough "to offset substantially the very benefits Congress intended to confer by remitting the duty." Ibid. . It seems to me that the word "offset" in this context is misused. If a state were to impose a special tax on property stored in customs-bonded warehouses, perhaps such a tax could be viewed as "offsetting" the benefits of storage. An importer deciding whether to use the warehouses would weigh the amount saved through remission of duties against the amount expended to pay the property tax. In this case, however, the County and City, acting pursuant to state law, impose the same ad valorem taxes no matter where the property is stored. An importer deciding whether to bring imported goods into Texas therefore knows that while the goods are in storage he will have to pay the property tax whether or not he uses a customs-bonded warehouse. The value to him of using customs-bonded storage is the full amount of the savings from deferral or avoidance of duties—precisely the benefit Congress expressly has provided in order to encourage merchants to bring business to the United States. 37 The Court accepts appellant's argument that a tax exemption for goods in customs-bonded warehouses reduces importers' costs and thereby furthers the federal interest in encouraging trade. But the Court itself acknowledges that state legislation should be preempted only where "necessary" to achieve a congressional purpose. Ibid. No showing has been made that this standard is met here. Duty-free storage and exemption from state property taxation are independent policies for promoting foreign trade. Congress quite reasonably may choose one policy, as it has done, without choosing the other. 38 The Court relies primarily on McGoldrick v. Gulf Oil Corp., 309 U.S. 414, 60 S.Ct. 664, 84 L.Ed. 840 (1940), in which the Court invalidated a city tax on the sale of fuel oil from a customs-bonded manufacturing warehouse to foreign-bound vessels. McGoldrick does not control this case. As the Court concedes, see ante, at 152, McGoldrick did not hold that the warehousing system and customs regulations alone mandated preemption of state taxation. Rather, a key factor was that Congress expressly had exempted from federal taxation the imported petroleum that was refined in the bonded warehouses for sale to foreign-bound vessels as ships' stores. The explicit congressional purpose "to enable American refiners to meet foreign competition," 309 U.S., at 427, 60 S.Ct., at 668, provided a basis on which to infer congressional intent to preempt state taxation. There is no analogous federal tax exemption here, nor any evidence of congressional intent to encourage meeting of foreign competition. All that exists is the warehousing system itself, and for the reasons stated above I find this insufficient. 39 Nor do I find merit in appellant's constitutional arguments. Appellees' taxes do not violate the Commerce Clause, as they are "applied to an activity with a substantial nexus with the taxing State, [are] fairly apportioned, [do] not discriminate against interstate commerce, and [are] fairly related to the services provided by the State." Complete Auto Transit, Inc. v. Brady, 430 U.S. 274, 279, 97 S.Ct. 1076, 1079, 51 L.Ed.2d 326 (1977). Nor do these nondiscriminatory ad valorem taxes violate the Import-Export Clause, Art. I, § 10, cl. 2. See Michelin Tire Corp. v. Wages, 423 U.S. 276, 96 S.Ct. 535, 46 L.Ed.2d 495 (1976). 40 Appellant notes that Michelin Tire left open the possibility that even nondiscriminatory property taxes may not be imposed on goods that still are in transit. But appellant's copiers were stored for up to three years, and under current law they could have been stored for up to five years. 19 U.S.C. § 1557(a) (Supp. V 1981). The only conceivable basis for the view that these goods remain "in transit" is that Congress has so provided. I cannot agree that Congress has endowed customs-bonded goods with indefinite immunity from nondiscriminatory state-authorized local property taxation. During their prolonged period of storage, appellant's goods benefited from police and fire protection and the various other services provided by the County and City. "[T]he State is simply making the imported goods pay their own way, as opposed to exacting a fee merely for 'the privilege of moving through a State.' " Department of Revenue of Washington v. Association of Washington Stevedoring Cos., 435 U.S. 734, 764, 98 S.Ct. 1388, 1406, 55 L.Ed.2d 682 (1978) (POWELL, J., concurring in part and concurring in the result) (quoting Michelin Tire Corp. v. Wages, 423 U.S. 276, 290, 96 S.Ct. 535, 543, 46 L.Ed.2d 495 (1976)). The Import-Export Clause never was intended to exempt imported goods in these circumstances. 41 I would affirm the judgment of the Texas Court of Civil Appeals. 1 Until 1974, Xerox shipped its Mexican-assembled copiers to the Free Trade Zone of Panama, where they were stored tax free. In 1974, rising anti-American sentiment in Panama led Xerox to seek another storage facility. It settled on the Houston warehouse because of the excellent port facilities in the Port of Houston. 2 Goods stored in customs bonded warehouses are under the "joint custody" of the warehouse proprietor and the U.S. Customs Service. The U.S. Customs Service is "in charge" of the warehouse and all work performed there is under its "supervision." 19 U.S.C. § 1555 (1976). 3 19 U.S.C. § 1557(a) (1976) permits the storage of imported goods for up to three years in a customs bonded warehouse without payment of an import duty. The importer is required to post a bond for the value of the duty. At the end of the three years, the goods may be withdrawn for domestic sale upon payment of the duty owed, or may be withdrawn for reexport without payment of any duty. In 1978, the time limit on the storage period was extended from three years to five years. Customs Procedural Reform and Simplification Act of 1978, Pub.L. No. 95-410, § 108, 92 Stat. 888, 892 (1978). 4 Houston assesses and collects taxes for itself and the Houston Independent School District. 5 In 1976 and 1977, Xerox paid a total of approximately $1,817,000 in ad valorem taxes to appellees for copiers located in Texas for domestic use. 6 Harris County assesses and collects taxes for itself, the State of Texas, and other local taxing authorities. 7 In 1846, approximately 90% of all federal revenues were derived from customs duties. U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970, Bicentennial Edition, Part 2, p. 1106 (1975) (customs accounted for $26,713,000 out of total federal revenues of $29,700,000). 8 Here, a footnote in the regulations governing customs bonded warehouses specifically provided that "Imported goods in bonded warehouses are exempt from taxation or judicial process of any State or subdivision thereof." 19 CFR § 19.6(c) n. 11 (1981). A recent amendment to the regulations deleted this footnote on November 1, 1982, effective December 1, 1982. 47 Fed.Reg. 49370. The Treasury Department offered no explanation for the amendment. The deletion of footnote 11 without explanation does not alter our conclusion that the ad valorem taxes here are preempted by the statutory scheme. 9 The fair market value of the copiers located in the Houston warehouse on January 1, 1977, was $9,015,690. The duty remitted by the United States on these copiers amounted to $540,000. By comparison, the appellees here sought to impose taxes on the copiers for the year 1977 amounting to $211,112. J.A. 12a-15a, 25a.

78

459 U.S. 176 103 S.Ct. 539 74 L.Ed.2d 348 State of COLORADO, Plaintiffv.State of NEW MEXICO and Paul G. Bardacke, Attorney General of New Mexico. No. 80, Orig. Argued Oct. 4, 1982. Decided Dec. 13, 1982. Syllabus The Vermejo River—which originates in southern Colorado but is located primarily in New Mexico—is at present fully appropriated by users in New Mexico. Colorado seeks an equitable apportionment of the river's water in order to divert water for proposed uses. The Special Master, after a trial, recommended in his report that Colorado be permitted a diversion of 4,000 acre-feet per year. The Special Master recognized that strict application of the rule of prior appropriation would not permit any diversion. In applying the principle of equitable apportionment, however, he did not focus exclusively on the rule of priority, but apparently rested his recommendation on the alternative grounds that New Mexico could compensate for some or all of the Colorado diversion through reasonable water conservation measures, and that the injury, if any, to New Mexico would be outweighed by the benefit to Colorado from the diversion. New Mexico filed exceptions to the Special Master's report. Held: 1. The flexible principle of equitable apportionment applies to a State's claim to divert water for future uses, and the criteria relied upon by the Special Master comport with this Court's prior cases. Pp. 183-188. (a) When, as in this case, both States recognize the doctrine of prior appropriation, priority becomes the guiding principle, but not the sole criterion, in determining an equitable apportionment. Pp. 183-184. (b) While the equities supporting the protection of established, senior uses are substantial, it is also appropriate to consider additional factors relevant to a just apportionment, such as the conservation measures available to both States here and the balance of harms and benefits to the States that might result from the diversion sought by Colorado. Pp. 184-187. (c) A State seeking a diversion for future uses must demonstrate by clear and convincing evidence that the benefits of the diversion substantially outweigh the harm that might result. Pp. 187-188. 2. However, the Special Master's report does not contain sufficient factual findings to enable this Court to assess the correctness of his application of the principle of equitable apportionment to the facts of this case. Accordingly, this Court remands for additional findings, including specific findings relating to the Special Master's reliance on the factors of the availability of conservation measures and the weighing of the harms and benefits that would result from the diversion. Pp. 547-549. Remanded for further findings. Richard A. Simms, Santa Fe, N.M., for defendants. Robert F. Welborn, Denver, Colo., for plaintiff. Justice MARSHALL delivered the opinion of the Court. 1 This case concerns the proper apportionment between New Mexico and Colorado of the water of an interstate river. The water of the Vermejo River is at present fully appropriated by users in New Mexico. Colorado seeks to divert water for future uses. Invoking this Court's original jurisdiction under Article 3, § 2 of the Constitution, Colorado brought this action for an equitable apportionment of the water of the Vermejo River. A Special Master appointed by the Court recommended that Colorado be permitted a diversion of 4,000 acre-feet per year. The case is before us on New Mexico's exceptions to the Special Master's report. 2 * The Vermejo River is a small, nonnavigable river that originates in the snow-belt of the Rocky Mountains in southern Colorado and flows southeasterly into New Mexico for a distance of roughly 55 miles before it joins the Canadian River. The major portion of the river is located in New Mexico. The Colorado portion consists of three main tributaries that combine to form the Vermejo River proper approximately one mile below the Colorado-New Mexico border. At present there are no uses of the water of the Vermejo River in Colorado, and no use or diversion has ever been made in Colorado. In New Mexico, by contrast, farmers and industrial users have diverted water from the Vermejo for many years. In 1941 a New Mexico state court issued a decree apportioning the water of the Vermejo River among the various New Mexico users.1 3 In 1975, a Colorado corporation, Colorado Fuel and Iron Steel Corporation ("C.F. & I."), obtained in Colorado state court a conditional right to divert 75 cubic feet per second from the headwaters of the Vermejo River.2 C.F. & I. proposed a transmountain diversion of the water to a tributary of the Purgatoire River in Colorado to be used for industrial development and other purposes. Upon learning of this decree, the four principal New Mexico users—Phelps Dodge Corporation ("Phelps Dodge"), Kaiser Steel Corporation ("Kaiser Steel"), Vermejo Park Corporation ("Vermejo Park"), and the Vermejo Conservancy District ("Conservancy District")—filed suit in United States District Court for the District of New Mexico, seeking to enjoin any diversion by C.F. & I. that would violate their senior rights. On January 16, 1978, the District Court enjoined C.F. & I. from diverting any water from the Vermejo River in derogation of the senior water rights of New Mexico users.3 The court found that under the doctrine of prior appropriation, which both New Mexico and Colorado recognize,4 the New Mexico users were entitled to have their needs fully satisfied because their appropriation was prior in time. C.F. & I. filed a notice of appeal, and the Court of Appeals for the Tenth Circuit has stayed its proceedings during the pendency of this case before us. 4 In June 1978 Colorado moved for leave to file an original complaint in this Court. New Mexico opposed the motion. On April 16, 1979, we granted Colorado's motion and appointed the Honorable Ewing T. Kerr, Senior Judge of the United States District Court for the District of Wyoming, as Special Master in this case. After a lengthy trial involving an extensive presentation of evidence, the Special Master submitted a report to the Court on January 9, 1982. The report was accepted for filing on February 22, 1982. 5 The Special Master found that most of the water of the Vermejo River is consumed by the New Mexico users and that very little, if any, reaches the confluence with the Canadian River. He thus recognized that strict application of the rule of priority would not permit Colorado any diversion since the entire available supply is needed to satisfy the demands of appropriators in New Mexico with senior rights. Nevertheless, applying the principle of equitable apportionment established in our prior cases, he recommended permitting Colorado a transmountain diversion of 4,000 acre-feet5 of water per year from the headwaters of the Vermejo River. He stated: 6 "It is the opinion of the Master that a transmountain diversion would not materially affect the appropriations granted by New Mexico for users downstream. A thorough examination of the existing economies in New Mexico convinces the Master that the injury to New Mexico, if any, will be more than offset by the benefit to Colorado." Report of the Special Master, p. 23. 7 Explaining his conclusion, the Special Master noted that any injury to New Mexico would be restricted to the Conservancy District, the user in New Mexico furthest downstream, since there was sufficient water in the Vermejo River for the three other principal New Mexico water users. Vermejo Park, Kaiser Steel, and Phelps Dodge.6 He further found that the "Vermejo Conservancy District has never been an economically feasible operation." Ibid. 8 The Special Master's recommendation appears to rest on two alternative grounds: first, that New Mexico could compensate for some or all the Colorado diversion through reasonable water conservation measures;7 and second, that the injury, if any, to New Mexico would be outweighed by the benefit to Colorado from the diversion.8 In its various exceptions to his report, New Mexico challenges the Special Master's interpretation of the law of equitable apportionment. New Mexico maintains that the rule of priority should be strictly applied in this case to preclude Colorado from diverting any water from the Vermejo River. New Mexico also challenges the factual bases of the Special Master's conclusions that the recommended diversion would not materially affect New Mexico users and that any harm to New Mexico would be offset by the benefits to Colorado.9 9 We conclude that the criteria relied upon by the Special Master comport with the doctrine of equitable apportionment as it has evolved in our prior cases. We thus reject New Mexico's contention that the Special Master was required to focus exclusively on the rule of priority. However, the Report of the Special Master does not contain sufficient factual findings to enable us to assess the correctness of the Special Master's application of the principle of equitable apportionment to the facts of this case. We therefore remand with instructions to the Special Master to make further findings of fact. II 10 Equitable apportionment is the doctrine of federal common law that governs disputes between states concerning their rights to use the water of an interstate stream. Kansas v. Colorado, 206 U.S. 46, 98, 27 S.Ct. 655, 667-668, 51 L.Ed. 956 (1907); Connecticut v. Massachusetts, 282 U.S. 660, 670-671, 51 S.Ct. 286, 289-290, 75 L.Ed. 602 (1931). It is a flexible doctrine which calls for "the exercise of an informed judgment on a consideration of many factors" to secure a "just and equitable" allocation. Nebraska v. Wyoming, 325 U.S. 589, 618, 65 S.Ct. 1332, 1350-1351, 89 L.Ed. 1815 (1945). We have stressed that in arriving at "the delicate adjustment of interests which must be made," ibid., we must consider all relevant factors, including: 11 "physical and climatic conditions, the consumptive use of water in the several sections of the river, the character and rate of return flows, the extent of established uses, the availability of storage water, the practical effect of wasteful uses on downstream areas, [and] the damage to upstream areas as compared to the benefits to downstream areas if a limitation is imposed on the former." Nebraska v. Wyoming, 325 U.S. 589, 618, 65 S.Ct. 1332, 1351, 89 L.Ed. 1815 (1945). 12 Our aim is always to secure a just and equitable apportionment "without quibbling over formulas." New Jersey v. New York, 283 U.S. 336, 343, 51 S.Ct. 478, 479, 75 L.Ed. 1104 (1931). 13 The laws of the contending states concerning intrastate water disputes are an important consideration governing equitable apportionment. When, as in this case, both states recognize the doctrine of prior appropriation, priority becomes the "guiding principle" in an allocation between competing states. Nebraska v. Wyoming, 325 U.S., at 618, 65 S.Ct., at 1350-1351. But state law is not controlling. Rather, the just apportionment of interstate waters is a question of federal law that depends "upon a consideration of the pertinent laws of the contending States and all other relevant facts." Connecticut v. Massachusetts, 282 U.S., at 670-671, 51 S.Ct., at 289 (emphasis added). 14 In reaching his recommendation the Special Master did not focus exclusively on the rule of priority, but considered other factors such as the efficiency of current uses in New Mexico and the balance of benefits to Colorado and harm to New Mexico. New Mexico contends that it is improper to consider these other factors. It maintains that this Court has strictly applied the rule of priority when apportioning water between states adhering to the prior appropriation doctrine, and has departed from that rule only to protect an existing economy built upon junior appropriations. Since there is no existing economy in Colorado dependent upon the use of water from the Vermejo River, New Mexico contends that the rule of priority is controlling. We disagree with this inflexible interpretation of the doctrine of equitable apportionment. 15 Our prior cases clearly establish that equitable apportionment will protect only those rights to water that are "reasonably acquired and applied." Wyoming v. Colorado, 259 U.S. 419, 484, 42 S.Ct. 552, 564, 66 L.Ed. 999 (1922). Especially in those Western states where water is scarce, "[t]here must be no waste . . . of the 'treasure' of a river. . . . Only diligence and good faith will keep the privilege alive." Washington v. Oregon, 297 U.S. 517, 527, 56 S.Ct. 540, 544-545, 80 L.Ed. 837 (1936). Thus, wasteful or inefficient uses will not be protected. See ibid.; Nebraska v. Wyoming, supra, 325 U.S., at 618, 65 S.Ct., at 1350-1351. Similarly, concededly senior water rights will be deemed forfeited or substantially diminished where the rights have not been exercised or asserted with reasonable diligence. Washington v. Oregon, 297 U.S., at 527-528, 56 S.Ct., at 544-545 (1936); Colorado v. Kansas, 320 U.S. 383, 394, 64 S.Ct. 176, 181, 88 L.Ed. 116 (1943). 16 We have invoked equitable apportionment not only to require the reasonably efficient use of water, but also to impose on states an affirmative duty to take reasonable steps to conserve and augment the water supply of an interstate stream. In Wyoming v. Colorado, Wyoming brought suit to prevent a proposed diversion by Colorado from the Laramie River. This Court calculated the dependable supply available to both states, subtracted the senior Wyoming uses, and permitted Colorado to divert an amount not exceeding the balance.10 In calculating the dependable supply we placed on each state the duty to employ "financially and physically feasible" measures "adapted to conserving and equalizing the natural flow." 259 U.S., at 484, 42 S.Ct., at 564 (emphasis added). Adopting a position similar to New Mexico's in this case, Wyoming objected to a requirement that it employ conservation measures to facilitate Colorado's proposed uses. The answer we gave is especially relevant to this case: 17 "The question here is not what one state should do for the other, but how each should exercise her relative rights in the waters of this interstate stream. . . . Both states recognize that conservation within practicable limits is essential in order that needless waste may be prevented and the largest feasible use may be secured. This comports with the allpervading spirit of the doctrine of appropriation and takes appropriate heed of the natural necessities out of which it arose. We think that doctrine lays on each of these states a duty to exercise her right reasonably and in a manner calculated to conserve the common supply." Ibid.11 18 We conclude that it is entirely appropriate to consider the extent to which reasonable conservation measures by New Mexico might offset the proposed Colorado diversion and thereby minimize any injury to New Mexico users. Similarly, it is appropriate to consider whether Colorado has undertaken reasonable steps to minimize the amount of diversion that will be required. 19 In addition, we have held that in an equitable apportionment of interstate waters it is proper to weigh the harms and benefits to competing states. In Kansas v. Colorado, supra, where we first announced the doctrine of equitable apportionment, we found that users in Kansas were injured by Colorado's upstream diversions from the Arkansas River. 206 U.S., at 113-114, 117, 27 S.Ct., at 674, 675-676. Yet we declined to grant any relief to Kansas on the ground that the great benefit to Colorado outweighed the detriment to Kansas. Id., at 100-101, 113-114, 117, 27 S.Ct., at 668-669, 674, 675-676. Similarly, in Nebraska v. Wyoming, supra, we held that water rights in Wyoming and Nebraska, which under state law were senior, had to yield to the "countervailing equities" of an established economy in Colorado even though it was based on junior appropriations. 325 U.S., at 622, 65 S.Ct., at 1352. We noted that the rule of priority should not be strictly applied where it "would work more hardship" on the junior user "than it would bestow benefits" on the senior user. Id., at 619, 65 S.Ct., at 1351. See also Washington v. Oregon, supra, 297 U.S., at 522, 56 S.Ct., at 542. The same principle is applicable in balancing the benefits of a diversion for proposed uses against the possible harms to existing uses. See, e.g., Wyoming v. Colorado, supra (placing upon Wyoming, the state with senior water rights, a duty to conserve water in order to facilitate a diversion for a proposed use in Colorado); Connecticut v. Massachusetts, supra; New Jersey v. New York, supra.12 20 We recognize that the equities supporting the protection of existing economies will usually be compelling. The harm that may result from disrupting established uses is typically certain and immediate, whereas the potential benefits from a proposed diversion may be speculative and remote. Under some circumstances, however, the countervailing equities supporting a diversion for future use in one state may justify the detriment to existing users in another state. This may be the case, for example, where the state seeking a diversion demonstrates by clear and convincing evidence that the benefits of the diversion substantially outweigh the harm that might result.13 In the determination of whether the state proposing the diversion has carried this burden, an important consideration is whether the existing users could offset the diversion by reasonable conservation measures to prevent waste. This approach comports with our emphasis on flexibility in equitable apportionment and also accords sufficient protection to existing uses. 21 We conclude, therefore, that in the determination of an equitable apportionment of the water of the Vermejo River the rule of priority is not the sole criterion. While the equities supporting the protection of established, senior uses are substantial, it is also appropriate to consider additional factors relevant to a just apportionment, such as the conservation measures available to both states and the balance of harm and benefit that might result from the diversion sought by Colorado. III 22 Applying the doctrine of equitable apportionment, the Special Master recommended that Colorado be permitted to divert 4,000 acre-feet of water per year from the headwaters of the Vermejo River. Because all of the water of the Vermejo River is currently consumed by New Mexico appropriators, the recommended diversion would necessarily reduce the amount of water available to New Mexico. 23 In explaining the basis for his recommendation, the Special Master stated that the diversion would not "materially affect" existing New Mexico appropriations. This conclusion appears to reflect certain assumptions about the ability of New Mexico users to implement water conservation measures. See supra, at 181, and n. 7. The Special Master also concluded that any injury to New Mexico would be "more than offset" by the benefits to Colorado. Report of Special Master, p. 23. Both the availability of conservation measures and a weighing of the harm and benefits that would result from the diversion are factors relevant to the determination of a just and equitable apportionment. However, the Special Master did not clearly state the factual findings supporting his reliance on these factors. Accordingly, we remand for additional factual findings. In particular, we request specific findings concerning the following areas: 24 (1) the existing uses of water from the Vermejo River, and the extent to which present levels of use reflect current or historical water shortages or the failure of existing users to develop their uses diligently; 25 (2) the available supply of water from the Vermejo River, accounting for factors such as variations in streamflow, the needs of current users for a continuous supply, the possibilities of equalizing and enhancing the water supply through water storage and conservation, and the availability of substitute sources of water to relieve the demand for water from the Vermejo River; (3) the extent to which reasonable conservation measures in both states might eliminate waste and inefficiency in the use of water from the Vermejo River; 26 (4) the precise nature of the proposed interim and ultimate use in Colorado of water from the Vermejo River, and the benefits that would result from a diversion to Colorado; 27 (5) the injury, if any, that New Mexico would likely suffer as a result of any such diversion, taking into account the extent to which reasonable conservation measures could offset the diversion.14 IV 28 The flexible doctrine of equitable apportionment clearly extends to a state's claim to divert water for future uses. Whether such a diversion should be permitted will turn on an examination of all factors relevant to a just apportionment. It is proper, therefore, to consider factors such as the extent to which reasonable conservation measures by existing users can offset the reduction in supply due to diversion, and whether the benefits to the state seeking the diversion substantially outweigh the harm to existing uses in another state. We remand for specific factual findings relevant to determining a just and equitable apportionment of the water of the Vermejo River between Colorado and New Mexico. 29 It is so ordered. 30 Chief Justice BURGER, with whom Justice STEVENS joins, concurring in the opinion and the judgment. 31 This case arises from an understandably intense competition between two states over rights to a small, nonnavigable, interstate river. Because on the record before it this Court cannot make an appropriate apportionment of the water, the Court remands the case to the Special Master for further factual findings. 32 I emphasize that under our prior holdings these two states come to the Court on equal footing. See Kansas v. Colorado, 206 U.S. 46, 27 S.Ct. 655, 51 L.Ed. 956 (1907). Neither is entitled to any special priority over the other with respect to use of the water. Colorado cannot divert all of the water it may need or can use simply because the river's headwaters lie within its borders, Wyoming v. Colorado, 259 U.S. 419, 466, 42 S.Ct. 552, 558, 66 L.Ed. 999 (1922). Nor is New Mexico entitled to any particular priority of allocation or undiminished flow simply because of first use. See, e.g. Colorado v. Kansas, 320 U.S. 383, 393, 64 S.Ct. 176, 180, 88 L.Ed. 116 (1943). Each state through which rivers pass has a right to the benefit of the water but it is for the Court, as a matter of discretion, to measure their relative rights and obligations and to apportion the available water equitably. As the Court's opinion states, in the process of apportioning the water, prior dependence and inefficient uses may be considered in balancing the equities. But no state has any priority over any other state. It is on this understanding of the Court's holding that I join the opinion and the judgment. 33 Justice O'CONNOR, with whom Justice POWELL joins, concurring in the judgment. 34 The doctrine of prior appropriation includes the requirement that the appropriator's use of water be beneficial and reasonable. What is reasonable, of course, does not admit of ready definition, being dependent upon the particular facts and circumstances of each case. In this case, the Special Master has cast an accusatory finger at the Vermejo Conservancy District, concluding that "[t]he system of canals used to transport the water to the fields is inefficient." Report of the Special Master, p. 8. 35 Undoubtedly, there is evidence in the record indicating that large losses of water occur through seepage and evaporation in transporting waters of the Vermejo through open ditches for irrigation and stock watering. Tr. 1315. It is a leap, however, from observing that large losses occur to concluding, as Colorado would have the Court do, that the practices of the Conservancy District are wasteful or unreasonable. As the Court observes, ante, at 185, the extent of the duty to conserve that may be placed upon the user is limited to measures that are "financially and physically feasible," Wyoming v. Colorado, 259 U.S. 419, 484, 42 S.Ct. 552, 564, 66 L.Ed. 999 (1922), and "within practicable limits." Ibid.1 Nevertheless, in concluding that the Conservancy District's distribution system is "inefficient," the Special Master made no factual finding that improved economy in that system is within the practicable means available to the District.2 36 Colorado would have the Court assess the Conservancy District's "waste" and "inefficiency" by a new yardstick—i.e., not by comparing the economic gains to the District with the costs of achieving greater efficiency, but by comparing the "inefficiency" of New Mexico's uses with the relative benefits to Colorado of a new use. The Special Master has succumbed to this suggestion. His recommendation that Colorado be permitted a diversion embodies the judgment that, because Colorado can, in some unidentified sense, make "better" use of the waters of the Vermejo, New Mexico may be forced to change her present uses. 37 Today the Court has also gone dangerously far toward accepting that suggestion. The Court holds, ante, at 186, that it is appropriate in equitable apportionment litigation to weigh the harms and benefits to the competing States. It does so notwithstanding its recognition, ante, at 187, that the potential benefits from a proposed diversion are likely to be speculative and remote, and therefore difficult to balance against any threatened harms, and its concession, ibid., that the equities supporting protection of an existing economy will usually be compelling. 38 In equitable apportionment litigation between two prior appropriation States concerning the waters of a fully appropriated river, this Court has never undertaken that balancing task outside the concrete context of either two established economies in the competing States dependent upon the waters to be apportioned3 or of a proposed diversion in one State to satisfy a demonstrable need for a potable supply of drinking water.4 In the former context, the Court may assess the relative benefit and detriment by reference to the actual fruits of use of the waters in the respective States.5 In the latter context, the compelling nature of the proposed use reduces the speculation that might otherwise attend assessment of the benefits of a proposed diversion. Where, as here, however, no existing economy in Colorado depends on the waters of the Vermejo and the actual uses in New Mexico rank in equal importance with the proposed uses in Colorado,6 the difficulty of arriving at the proper balance is especially great. 39 This case therefore highlights the restraint with which the Court should proceed in apportioning interstate waters between a State seeking a future use and a State with an existing economy dependent upon the waters to be apportioned. The Court can only invite litigation within its original jurisdiction if it permits one State to obtain a diversion for a new use upon that State's allegation that the second State is engaging in "wasteful" practices or that she can make "better" use of the waters, even if the second State's uses are entirely reasonable. 40 I do not suggest, of course, that the Court must blind itself to compelling evidence of waste by one State. Protection of existing economies does not require that users be permitted to continue in unreasonably wasteful or inefficient practices. But the Court should be moved to exercise its original jurisdiction to alter the status quo between States only where there is clear and convincing evidence, ante, at 188, n. 13, that one State's use is unreasonably wasteful. To allow Colorado a diversion upon a lesser showing comports neither with the equality of rights of the litigants before us, see Connecticut v. Massachusetts, 282 U.S. 660, 670, 51 S.Ct. 286, 289, 75 L.Ed. 602 (1931), nor with the sparing use that should be made of the Court's equitable powers, see id., at 669, 51 S.Ct., at 289. Further, such action would seriously undermine the Court's affirmation, ante, at 184, that priority of appropriation is the "guiding principle" in allocating waters between two prior appropriation States. 41 The Court's remand reflects its judgment that the paucity of the factual findings before us furnishes an inadequate basis upon which to make "the delicate adjustment of interests" at stake, Nebraska v. Wyoming, 325 U.S. 589, 618, 65 S.Ct. 1332, 1350-1351, 89 L.Ed. 1815 (1945). I concur in that disposition insofar as the Special Master's findings and conclusions do not provide a basis for determining whether Colorado has demonstrated by clear and convincing evidence that the Conservancy District has engaged in unreasonably wasteful practices. 1 Phelps Dodge Corp. v. W.S. Land and Cattle Co., No. 7201 (D.C.Cty.Colfax 1941). 2 In re the Application for Water Rights of C.F. & I. Corp., No. W-3961 (Dist.Ct., W.Div. No. 2, June 20, 1975). 3 Kaiser Steel Corporation et al. v. C.F. & I. Steel Corporation, Civil No. 76-244 (D.N.M.1978). The injunction was not based on a determination of the right of the two states under the law of equitable apportionment, since neither Colorado nor New Mexico was a party to the action. 4 N.M. Const. Art. XVI, § 2; Colo. Const. Art. XVI, § 5, 6. The administration of water rights in each state is governed by statute. Colo.Rev.Stat. § 37-92-101 et seq. (1973); N.M.Stat.Ann. § 72-1-1 et seq. (1978). The prior appropriation doctrine and the riparian doctrine are the two basic doctrines governing the rights to the use of water. Under the prior appropriation doctrine, recognized in most of the western states, water rights are acquired by diverting water and applying it for a beneficial purpose. A distinctive feature of the prior appropriation doctrine is the rule of priority, under which the relative rights of water users are ranked in the order of their seniority. Under the riparian doctrine, recognized primarily in the eastern, midwestern and southern states, the owner of land contiguous to a watercourse is entitled to have the stream flow by or through his land undiminished in quantity and unpolluted in quality, except that any riparian proprietor may make whatever use of the water that is reasonable with respect to the needs of other appropriators. Appropriative rights do not depend on land ownership and are acquired and maintained by actual use. Riparian rights, by contrast, originate from land ownership and remain vested even if unexercised. Appropriative rights are fixed in quantity; riparian rights are variable depending on streamflow and subject to the reasonable uses of others. See generally 1 R. Clark, Waters and Water Rights (1967); W. Hutchins, Selected Problems in the Law of Water Rights in the West (U.S. Dept. of Agriculture, Misc.Pub. No. 418) (1942); 1 W. Hutchins, Water Rights Laws in the Nineteen Western States (U.S. Dept. of Agriculture, Misc.Pub. No. 1206) (1971). 5 An acre-foot is a volumetric measurement which means the amount of water required to cover one acre of ground one foot deep. One acre foot equals 43,560 cubic feet or 325,900 gallons of water. 6 The Conservancy District is the largest user of water from the Vermejo River in New Mexico. It consists of over 60 farms irrigated by an extensive system of canals and reservoirs. The U.S. Maxwell Wildlife Refuge is also located within the District. In the early 1950's the District was part of a large reclamation project funded by the federal Government. Vermejo Park diverts water primarily to irrigate land used to grow hay for its cattle operation. Kaiser Steel uses water primarily for its coal facilities. Phelps Dodge leases its rights to Kaiser Steel and to the C.S. Springer Cattle Company. 7 This is a fair reading of the Special Master's conclusion that New Mexico users would not be "materially affected" by the recommended diversion. While the report does not expressly state that Colorado's diversion might be offset by reasonable conservation efforts, it does refer specifically to the waste and inefficiency of the Conservancy District's system of water canals. Report of the Special Master, pp. 8, 23. In addition, in its second exception to the report New Mexico acknowledges that the Special Master based his conclusion that New Mexico users would not be materially affected on certain findings concerning waste and inefficiency within the Conservancy District. 8 New Mexico contends that the Special Master relied on a third ground, namely, that the mere fact that the Vermejo River originates in Colorado automatically entitles Colorado to a share of the water of the Vermejo River. See Report of the Special Master, p. 8. To the extent that the Special Master applied such a per se rule of apportionment, we reject it as inconsistent with our emphasis on flexibility in equitable apportionment. 9 New Mexico also contends that Colorado is improperly suing directly and solely for the benefit of a private individual—C.F. & I.—in violation of the Eleventh Amendment, and that Colorado's suit is barred by laches. We find no merit to these claims. Because the State of Colorado has a substantial interest in the outcome of this suit, New Mexico may not invoke its Eleventh Amendment immunity from federal actions by citizens of another State. The portion of the Vermejo River in Colorado is owned by the State in trust for its citizens. Colo. Const. Article XVI, Section 5. While C.F. & I. will most likely be the primary user of any water diverted from the Vermejo River, other Colorado citizens may jointly use the water or purchase water rights in the future. In any event, Colorado surely has a sovereign interest in the beneficial effects of a diversion on the general prosperity of the State. Faced with a similar set of circumstances in Kansas v. Colorado, 206 U.S. 46, 99, 27 S.Ct. 655, 668, 51 L.Ed. 956 (1907), we concluded that "[t]he controversy rises . . . above a mere question of local private right and involves the matter of state interest and must be considered from that standpoint." We also conclude that Colorado is not barred by laches from seeking an equitable apportionment. For the reasons that we elaborate below, infra, at 186-188, we hold that under some circumstances the countervailing equities supporting a diversion of water for a future use in one state may justify the detriment suffered by existing users in another state. Therefore the mere fact that Colorado has no existing uses of the waters of the Vermejo River and that current users in New Mexico may suffer some detriment from a diversion does not bar Colorado's suit for an equitable apportionment for future uses. These circumstances, however, do bear on the burden of proof that Colorado must satisfy to justify the possible disruption of existing uses. See infra, at 187-188, and n. 13. A contrary conclusion is not dictated by Washington v. Oregon, 297 U.S. 517, 528, 56 S.Ct. 540, 545, 80 L.Ed. 837 (1936), or Colorado v. Kansas, 320 U.S. 383, 394, 64 S.Ct. 176, 181, 88 L.Ed. 116 (1943) (dictum), which merely require established users or holders of water rights to exercise diligence in protecting their rights and putting them to beneficial uses. See infra, at 184. 10 This description is only roughly accurate, since we did not rigidly follow this procedure in apportioning the Laramie River, but instead departed from a strict application of the rule of priority in numerous respects. For instance, our decree in Wyoming v. Colorado granted Colorado an unqualified right to divert 22,500 acre-feet, even though there were Wyoming appropriations senior to the Colorado appropriations underlying the 22,500 acre-feet grant. 259 U.S., at 489-490, 42 S.Ct., at 565-566. In addition, we granted to Colorado priority to divert a total of 37,750 acre-feet, even though some of the underlying appropriations were junior to a number of Wyoming appropriations. Id., at 495-496, 42 S.Ct., at 567-568. The effect was to guarantee water to junior appropriators in Colorado to the potential detriment of senior appropriators downstream in Wyoming. See 2 R. Clark, Waters and Water Rights § 132.4 (1967). 11 We thus required Wyoming to enhance and equalize the water supply through "practicable storage and conservation" measures, such as the use of storage facilities similar to those already in use in Wyoming. 259 U.S., at 485, 42 S.Ct., at 564. 12 In Connecticut v. Massachusetts we declined to enjoin Massachusetts' proposed diversion for future uses. We took into account the impending "serious water shortage" in the Boston area and the absence of "real or substantial injury or damage" to Connecticut. 282 U.S. 660, 664, 672, 51 S.Ct. 286, 287, 290, 75 L.Ed. 602 (1931). Although Connecticut v. Massachusetts, as well as New Jersey v. New York, 283 U.S. 336, 51 S.Ct. 478, 75 L.Ed. 1104 (1931), involved states that follow the riparian rather than the prior appropriation doctrine, see note 4, supra, our allocation of water for future uses rested on the federal common law of equitable apportionment, which, as we made clear, "is not governed by the same rules of [state] law that are applied for the solution of similar questions of private right." Connecticut v. Massachusetts, 282 U.S., at 670, 51 S.Ct., at 289; see also New Jersey v. New York, supra, 283 U.S., at 342-343, 51 S.Ct., at 479-480. Nothing in those two cases suggested that the apportionment of water for future uses in any way depended on the adherence of both states to the riparian doctrine. 13 Our cases establish that a state seeking to prevent or enjoin a diversion by another state bears the burden of proving that the diversion will cause it "real or substantial injury or damage." Connecticut v. Massachusetts, supra, 282 U.S., at 672, 51 S.Ct., at 290. See also New Jersey v. New York, 283 U.S., at 344-345, 51 S.Ct., at 480; Kansas v. Colorado, 206 U.S., at 117, 27 S.Ct., at 675-676; Colorado v. Kansas, 320 U.S., at 393-394, 64 S.Ct., at 180-181. This rule applies even if the state seeking to prevent or enjoin a diversion is the nominal defendant in a lawsuit. In Colorado v. Kansas, for instance, Colorado sued Kansas seeking to enjoin further lawsuits by Kansas water users against Colorado users. Although Kansas was the defendant, we granted Colorado an injunction based on Kansas' failure to sustain the burden of showing that the Colorado diversions had "worked a serious detriment to the substantial interests of Kansas." 320 U.S., at 400, 64 S.Ct., at 184; see also id., at 389-390, 64 S.Ct., at 178-179. New Mexico must therefore bear the initial burden of showing that a diversion by Colorado will cause substantial injury to the interests of New Mexico. In this case New Mexico has met its burden since any diversion by Colorado, unless offset by New Mexico at its own expense, will necessarily reduce the amount of water available to New Mexico users. The burden has therefore shifted to Colorado to establish that a diversion should nevertheless be permitted under the principle of equitable apportionment. Thus, with respect to whether reasonable conservation measures by New Mexico will offset the loss of water due to Colorado's diversion, or whether the benefit to Colorado from the diversion will substantially outweigh the possible harm to New Mexico, Colorado will bear the burden of proof. It must show, in effect, that without such a diversion New Mexico would be using "more than its equitable share of the benefits of a stream." Id., at 394, 64 S.Ct., at 181. Moreover, Colorado must establish not only that its claim is of a "serious magnitude," but also that its position is supported by "clear and convincing evidence." Connecticut v. Massachusetts, 282 U.S., at 669, 51 S.Ct., at 289. See also Colorado v. Kansas, 320 U.S., at 393, 64 S.Ct., at 180-181; Washington v. Oregon, 297 U.S., at 522, 56 S.Ct., at 542. 14 The Special Master may make any other factual findings that he considers relevant. Additional hearings may be held, although they may be unnecessary in light of the extensive evidence already presented at trial. Upon remand, the Special Master is free to reaffirm his original recommendation or make a different recommendation on the basis of the evidence and applicable principles of equitable apportionment. 1 It is significant to note that in Wyoming v. Colorado, 259 U.S. 419, 42 S.Ct. 552, 66 L.Ed. 999 (1922), upon which the Court relies for the proposition that an affirmative duty to conserve may be imposed on the States, ante, at 185, the Wyoming appropriators already had storage facilities in place for equalizing the river's natural flow. In answering Wyoming's objection that she should not be burdened with conservation measures in order to permit a diversion by Colorado, the Court observed: "We think [the] doctrine [of appropriation] lays on each of these States a duty to exercise her right reasonably and in a manner calculated to conserve the common supply. Notwithstanding her present contention, Wyoming has in fact proceeded on this line, for, as the proof shows, her appropriators, with her sanction, have provided and have in service reservoir facilities which are adapted for the purpose and reasonably sufficient to meet its requirements." Id., at 484-485, 42 S.Ct., at 564 (emphasis added). 2 Evidence in the record indicates that the Conservancy District has employed an engineering firm to investigate the feasibility of constructing an enclosed system to deliver stock water to the District's land owners. Tr. 1318. 3 See Nebraska v. Wyoming, 325 U.S. 589, 65 S.Ct. 1332, 89 L.Ed. 1815 (1945); Washington v. Oregon, 297 U.S. 517, 56 S.Ct. 540, 80 L.Ed. 837 (1936); Kansas v. Colorado, 206 U.S. 46, 27 S.Ct. 655, 51 L.Ed. 956 (1907). 4 See New Jersey v. New York, 283 U.S. 336, 51 S.Ct. 478, 75 L.Ed. 1104 (1931); Connecticut v. Massachusetts, 282 U.S. 660, 51 S.Ct. 286, 75 L.Ed. 602 (1931). It is also significant to note that these disputes occurred between two riparian States. Wyoming v. Colorado, supra, does not represent an exception to the pattern stated in the text. The Court did not engage in any wholesale balancing of the relative harms and benefits to the two States from the proposed diversion. Rather, the Court imposed a very limited duty on Wyoming to make use of the storage facilities her appropriators already had in place, see note 1 supra, for the purpose of calculating the dependable supply of water available to Wyoming. 259 U.S., at 484, 42 S.Ct., at 564. The Court was thereby able to determine that the waters of the Laramie River were not fully appropriated and that a share of the waters was available for Colorado's proposed use. 5 For example, in Kansas v. Colorado, supra, Kansas sought to restrain Colorado from diverting waters of the Arkansas River for the irrigation of lands in Colorado. Colorado had diverted waters from the river since the 1880's. As a result of irrigation, the population of the irrigated areas, the number of acres cultivated, and the value of farm products produced in these areas escalated dramatically. 206 U.S., at 108-109, 27 S.Ct., at 671-672. The Court compared this demonstrated salutary effect of the irrigation on the economy of Colorado with the corresponding population changes and changes in acreage and production of corn and wheat in the affected Kansas counties for the same period. Id., at 110-113, 27 S.Ct., at 672-674. Using these concrete data, the Court was able to discern some minimal injury to Kansas as a result of the diminution of the flow of the Arkansas River. Id., at 113-114, 27 S.Ct., at 674. Viewing the overall impact of the available water on the two economies, however, the Court concluded: "[W]hen we compare the amount of this detriment [to Kansas] with the great benefit which has obviously resulted to the counties in Colorado, it would seem that equality of right and equity between the two States forbids any interference with the present withdrawal of water in Colorado for purposes of irrigation." Ibid. Quite clearly, the Court was not forced to speculate about the benefit and detriment of the diversion to the competing States. Similarly, in Washington v. Oregon, supra, the Court was equipped to assess the balance of harm and benefit to the economies from the diversion at issue. Washington sought an injunction against Oregon's diversion of waters of the Walla Walla River for irrigation in Oregon. On the one hand, Oregon had an existing agricultural economy dependent upon irrigation from the Walla Walla. On the other hand, the evidence revealed that there would be absolutely no benefit to Washington in prohibiting Oregon's diversion during periods of water shortage; the nature of the river channel was such that even if the water was not diverted by Oregon users, it would be absorbed by the gravel beneath the channel and never reach Washington users. 297 U.S., at 522-523, 56 S.Ct., at 542-543. The Court therefore concluded that "[t]o limit the long established use in Oregon would materially injure Oregon users without a compensating benefit to Washington users." Id., at 523, 56 S.Ct., at 543. 6 According to Colorado, the diverted water would be used "in industrial operations at coal mines, agriculture, timbering, power generation, domestic needs and other industrial operations . . . ." Reply Brief for Colorado 8.

1011

459 U.S. 212 103 S.Ct. 588 74 L.Ed.2d 402 Charles V. BOWEN, Petitionerv.UNITED STATES POSTAL SERVICE et al. No. 81-525. Argued Oct. 6, 1982. Decided Jan. 11, 1983. Syllabus After petitioner employee was discharged by respondent United States Postal Service (USPS) as a result of an altercation with another employee, he filed a grievance with respondent Union as provided by the applicable collective-bargaining agreement. When the Union declined to take his grievance to arbitration, petitioner sued respondents in Federal District Court, claiming that he had been wrongfully discharged and seeking damages and injunctive relief. Entering judgment on a jury verdict against both respondents, the District Court held that the USPS had discharged petitioner without just cause and that the Union had handled his grievance in an arbitrary manner. Accordingly, the court upheld the jury's apportionment of damages between the USPS and the Union. The Court of Appeals affirmed except for the award of damages against the Union, holding that because petitioner's compensation was payable only by the USPS, reimbursement for his lost earnings continued to be the USPS's exclusive obligation, and that hence no portion of the deprivations was chargeable to the Union. Held: Where the District Court's findings, accepted by the Court of Appeals, established that petitioner's damages were caused initially by the USPS's unlawful discharge and were increased by the Union's breach of its duty of fair representation, apportionment of the damages was required. Vaca v. Sipes, 386 U.S. 171, 87 S.Ct. 903, 17 L.Ed.2d 842. Pp. 218-230. (a) The governing principle of Vaca is that where an employee proves that his employer violated the collective-bargaining agreement and that his union breached its duty of fair representation, liability is to be apportioned between the employer and the union according to the damages caused by the fault of each. To interpret this principle as requiring that an employer be solely liable for damages resulting from a wrongful discharge treats the relationship between the employer and employee, created by the collective-bargaining agreement, as if it were a simple contract of hire governed by traditional common-law principles. Such a reading fails to recognize that a collective-bargaining agreement is much more than traditional common-law employment terminable at will. Rather, it is an agreement creating relationships and interests under the federal common law of labor policy. Pp. 218-220. (b) Of paramount importance is the right of the employee, who has been injured by both the employer's and the union's breach, to be made whole. Even though both the employer and the union have caused the damage suffered by the employee, the union is responsible for the increase in damages resulting from breach of its duty of fair representation having caused the grievance procedure to malfunction, and, as between the two wrongdoers, the union should bear its portion of the damages. Pp. 220-224. (c) When the union, as the employee's exclusive agent, waives arbitration or fails to seek review of an adverse decision, the employer should be in substantially the same position as if the employee had had the right to act on his own behalf and had done so. In the absence of damages apportionment where the default of both the employer and the union contributes to the employee's injury, incentives to comply with the grievance proceeding would be diminished, and to impose total liability solely on the employer could affect the willingness of employers to agree to arbitration clauses. To require the union to pay damages does not impose a burden on the union inconsistent with national labor policy, but rather provides an additional incentive for the union to process its members' claims where warranted. Pp. 224-228. (d) Czosek v. O'Mara, 397 U.S. 25, 90 S.Ct. 770, 25 L.Ed.2d 21, is not inconsistent with Vaca § recognition that each party should bear the damages attributable to its fault. Pp. 228-230. 642 F.2d 79, reversed and remanded. William B. Poff, Roanoke, Va., for petitioner. Barbara E. Etkind, Philadelphia, Pa., for federal respondent in support of the petitioner. Asher W. Schwartz, New York City, for respondent Union. Justice POWELL delivered the opinion of the Court. 1 The issue is whether a union may be held primarily liable for that part of a wrongfully discharged employee's damages caused by his union's breach of its duty of fair representation. 2 * On February 21, 1976, following an altercation with another employee, petitioner Charles V. Bowen was suspended without pay from his position with the United States Postal Service. Bowen was a member of the American Postal Workers Union, AFL-CIO, the recognized collective bargaining agent for Service employees. After Bowen was formally terminated on March 30, 1976, he filed a grievance with the Union as provided by the collective-bargaining agreement. When the Union declined to take his grievance to arbitration, he sued the Service and the Union in the United States District Court for the Western District of Virginia, seeking damages and injunctive relief. 3 Bowen's complaint charged that the Service had violated the collective bargaining agreement by dismissing him without "just cause" and that the Union had breached its duty of fair representation. His evidence at trial indicated that the responsible Union officer, at each step of the grievance process, had recommended pursuing the grievance but that the national office, for no apparent reason, had refused to take the matter to arbitration. 4 Following the parties' presentation of evidence, the court gave the jury a series of questions to be answered as a special verdict.1 If the jury found that the Service had discharged Bowen wrongfully and that the Union had breached its duty of fair representation, it was instructed to determine the amount of compensatory damages to be awarded and to apportion the liability for the damages between the Service and the Union.2 In explaining how liability might be apportioned, the court instructed the jury that the issue was left primarily to its discretion. The court indicated, however, that the jury equitably could base apportionment on the date of a hypothetical arbitration decision the date at which the Service would have reinstated Bowen if the Union had fulfilled its duty. The court suggested that the Service could be liable for damages before that date and the Union for damages thereafter. Although the Union objected to the instruction allowing the jury to find it liable for any compensatory damages, it did not object to the manner in which the court instructed the jury to apportion the damages in the event apportionment was proper.3 5 Upon return of a special verdict in favor of Bowen and against both defendants, the District Court entered judgment, holding that the Service had discharged Bowen without just cause and that the Union had handled his "apparently meritorious grievance . . . in an arbitrary and perfunctory manner . . . ." 470 F.Supp. 1127, 1129 (WD Va.1979). In so doing, both the Union and the Service acted "in reckless and callous disregard of [Bowen's] rights."4 Ibid. The court found that Bowen could not have proceeded independently of the Union5 and that if the Union had arbitrated Bowen's grievance, he would have been reinstated. Ibid. 6 The court ordered that Bowen be reimbursed $52,954 for lost benefits and wages. Although noting that "there is authority suggesting that only the employer is liable for damages in the form of back pay," it observed that "this is a case in which both defendants, by their illegal acts, are liable to plaintiff. . . . The problem in this case is not one of liability but rather one of apportionment. . . ." Id., at 1130-1131. The jury had found that the Union was responsible for $30,000 of Bowen's damages. The court approved that apportionment, ordering the Service to pay the remaining $22,954.6 7 On appeal by the Service and the Union, the Court of Appeals for the Fourth Circuit overturned the damage award against the Union. 642 F.2d 79 (CA4 1981). It accepted the District Court's findings of fact, but held as a matter of law that, "[a]s Bowen's compensation was at all times payable only by the Service, reimbursement of his lost earnings continued to be the obligation of the Service exclusively. Hence, no portion of the deprivations . . . was chargeable to the Union. Cf. Vaca v. Sipes, 386 U.S. 171, 195 [87 S.Ct. 903, 919, 17 L.Ed.2d 842] . . . (1967)." Id., at 82. The court did not alter the District Court's judgment in any other respect, but "affirmed [it] throughout" except for the award of damages against the Union. Id., at 83. 8 Thus, the Court of Appeals affirmed the District Court's apportionment of fault and its finding that both the Union and the Service had acted "in reckless and callous disregard of [Bowen's] rights."7 Indeed, the court accepted the District Court's apportionment of fault so completely that it refused to increase the $22,954 award against the Service to cover the whole of Bowen's injury. Bowen was left with only a $22,954 award, whereas the jury and the District Court had awarded him lost earnings and benefits of $52,954—the undisputed amount of his damages. II 9 In Vaca v. Sipes, 386 U.S. 171, 87 S.Ct. 903, 17 L.Ed.2d 842 (1967), the Court held that an employee such as Bowen, who proves that his employer violated the labor agreement and his union breached its duty of fair representation, may be entitled to recover damages from both the union and the employer. The Court explained that the award must be apportioned according to fault: 10 "The governing principle, then, is to apportion liability between the employer and the union according to the damage caused by the fault of each. Thus, damages attributable solely to the employer's breach of contract should not be charged to the union, but increases if any in those damages caused by the union's refusal to process the grievance should not be charged to the employer." Id., at 197-198, 87 S.Ct., at 920-921. 11 Although Vaca § governing principle is well established, its application has caused some uncertainty.8 The Union argues that the Court of Appeals correctly determined that it cannot be charged with any damages resulting from a wrongful discharge. Vaca § "governing principle," according to the Union, requires that the employer be solely liable for such damages. The Union views itself as liable only for Bowen's litigation expenses resulting from its breach of duty. It finds support for this view in Vaca § recognition that a union's breach of its duty of fair representation does not absolve an employer of all the consequences of a breach of the collective-bargaining contract. See id., at 196, 87 S.Ct., at 919. The Union contends that its unrelated breach of the duty of fair representation does not make it liable for any part of the discharged employee's damages; its default merely lifts the bar to the employee's suit on the contract against his employer. 12 The difficulty with this argument is that it treats the relationship between the employer and employee, created by the collective-bargaining agreement, as if it were a simple contract of hire governed by traditional common law principles. This reading of Vaca fails to recognize that a collective-bargaining agreement is much more than traditional common law employment terminable at will. Rather, it is an agreement creating relationships and interests under the federal common law of labor policy. A. 13 In Vaca, as here, the employee contended that his employer had discharged him in violation of the collective-bargaining agreement and that the union had breached its duty of fair representation by refusing to take his claim to arbitration. He sued the union in a Missouri state court for breach of its duty. On finding that both the union and the employer were at fault, the jury decided—and the Missouri Supreme Court agreed—that the union was entirely liable for the employee's lost backpay. See id., at 195, 87 S.Ct., at 919. 14 On appeal, this Court was required to resolve a number of issues. One was whether an employee who had failed to exhaust the grievance procedure prescribed in the bargaining agreement could bring suit for a breach of that agreement.9 In Republic Steel Corp. v. Maddox, 379 U.S. 650, 85 S.Ct. 614, 13 L.Ed.2d 580 (1965), the Court had held that "federal labor policy requires that individual employees wishing to assert contract grievances must attempt use of the contract grievance procedure agreed upon by employer and union as the mode of redress."10 Id., at 652, 85 S.Ct., at 616 (emphasis in original). Because the employee in Republic Steel had made no attempt to exhaust the grievance procedure, it was necessary for the Court to consider only the union's interest in participating in the administration of the contract and the employer's interest in limiting administrative remedies. The Court noted, however, that if "the union refuses to press or only perfunctorily presses the individual's claim," federal labor policy might require a different result. Ibid. 15 Vaca presented such a situation. The union, which had the "sole power under the contract to invoke the higher stages of the grievance procedure," had chosen not to take the employee's claim to arbitration. See 386 U.S., at 185, 87 S.Ct., at 914. Thus the Court faced a strong countervailing interest: the employee's right to vindicate his claim. Vaca resolved these conflicting interests by holding that an employee's failure to exhaust the contractual grievance procedures would bar his suit except when he could show that the union's breach of its duty of fair representation had prevented him from exhausting those remedies. See id., at 185, 87 S.Ct., at 914. The Vaca Court then observed: 16 "It is true that the employer in such a situation may have done nothing to prevent exhaustion of the exclusive contractual remedies to which he agreed in the collective bargaining agreement. But the employer has committed a wrongful discharge in breach of that agreement, a breach which could be remedied through the grievance process to the employee-plaintiff's benefit were it not for the union's breach of its statutory duty of fair representation to the employee. To leave the employee remediless in such circumstances would, in our opinion, be a great injustice." Id., at 185-186, 87 S.Ct., at 914-915. 17 The interests thus identified in Vaca provide a measure of its principle for apportioning damages. Of paramount importance is the right of the employee, who has been injured by both the employer's and the union's breach, to be made whole. In determining the degree to which the employer or the union should bear the employee's damages, the Court held that the employer should not be shielded from the "natural consequences" of its breach by wrongful union conduct. Id., at 186, 87 S.Ct., at 914. The Court noted, however, that the employer may have done nothing to prevent exhaustion. Were it not for the union's failure to represent the employee fairly, the employer's breach "could [have been] remedied through the grievance process to the employee-plaintiff's benefit." The fault that justifies dropping the bar to the employee's suit for damages also requires the union to bear some responsibility for increases in the employee's damages resulting from its breach. To hold otherwise would make the employer alone liable for the consequences of the union's breach of duty. 18 Hines v. Anchor Motor Freight, Inc., 424 U.S. 554, 96 S.Ct. 1048, 47 L.Ed.2d 231 (1976), presented an issue analogous to that in Vaca: whether proof of a breach of the duty of fair representation would remove the bar of finality from an arbitral decision. We held that it would, in part because a contrary rule would prevent the employee from recovering 19 "even in circumstances where it is shown that a union has manufactured the evidence and knows from the start that it is false; or even if, unbeknownst to the employer, the union has corrupted the arbitrator to the detriment of disfavored union members." Id., at 570, 96 S.Ct., at 1059. 20 It would indeed be unjust to prevent the employee from recovering in such a situation. It would be equally unjust to require the employer to bear the increase in the damages caused by the union's wrongful conduct.11 It is true that the employer discharged the employee wrongfully and remains liable for the employee's backpay. See Vaca, supra, at 197, 87 S.Ct., at 920. The union's breach of its duty of fair representation, however, caused the grievance procedure to malfunction resulting in an increase in the employee's damages. Even though both the employer and the union have caused the damage suffered by the employee, the union is responsible for the increase in damages and, as between the two wrongdoers, should bear its portion of the damages.12 21 Vaca § governing principle reflects this allocation of responsibility. As the Court stated, "damages attributable solely to the employer's breach of contract should not be charged to the union, but increases if any in those damages caused by the union's refusal to process the grievance should not be charged to the employer." 386 U.S., at 197-198, 87 S.Ct., at 920-921 (emphasis added). The Union's position here would require us to read out of the Vaca articulation of the relevant principle the words emphasized above.13 It would also ignore the interests of all the parties to the collective agreement—interests that Vaca recognized and Hines illustrates. B 22 In approving apportionment of damages caused by the employer's breach of the collective-bargaining agreement and the union's breach of its duty of fair representation, Vaca did not apply principles of ordinary contract law. For, as the Court has noted, a collective-bargaining agreement "is more than a contract; it is a generalized code to govern a myriad of cases which the draftsmen cannot wholly anticipate." United Steelworkers of America v. Warrior & Gulf Navigation Co., 363 U.S. 574, 578, 80 S.Ct. 1347, 1351, 4 L.Ed.2d 1409 (1960). In defining the relationships created by such an agreement, the Court has applied an evolving federal common law grounded in national labor policy. See United Steelworkers of America v. American Manufacturing Co., 363 U.S. 564, 567, 80 S.Ct. 1343, 1346, 4 L.Ed.2d 1403 (1960); Textile Workers v. Lincoln Mills, 353 U.S. 448, 456-457, 77 S.Ct. 912, 917-918, 1 L.Ed.2d 972 (1957). 23 Fundamental to federal labor policy is the grievance procedure. See John Wiley & Sons v. Livingston, 376 U.S. 543, 549, 84 S.Ct. 909, 914, 11 L.Ed.2d 898 (1964); Warrior & Gulf Navigation Co., supra, at 578, 80 S.Ct., at 1350. It promotes the goal of industrial peace by providing a means for labor and management to settle disputes through negotiation rather than industrial strife. See John Wiley & Sons, supra, at 549, 84 S.Ct., at 914. Adoption of a grievance procedure provides the parties with a means of giving content to the collective bargaining agreement and determining their rights and obligations under it. See Warrior & Gulf Navigation, supra, at 581, 80 S.Ct., at 1352. 24 Although each party participates in the grievance procedure, the union plays a pivotal role in the process since it assumes the responsibility of determining whether to press an employee's claims.14 The employer, for its part, must rely on the union's decision not to pursue an employee's grievance. For the union acts as the employee's exclusive representative in the grievance procedure, as it does in virtually all matters involving the terms and conditions of employment. Just as a nonorganized employer may accept an employee's waiver of any challenge to his discharge as a final resolution of the matter, so should an organized employer be able to rely on a comparable waiver by the employee's exclusive representative. 25 There is no unfairness to the union in this approach. By seeking and acquiring the exclusive right and power to speak for a group of employees, the union assumes a corresponding duty to discharge that responsibility faithfully—a duty which it owes to the employees whom it represents and on which the employer with whom it bargains may rely. When the union, as the exclusive agent of the employee, waives arbitration or fails to seek review of an adverse decision, the employer should be in substantially the same position as if the employee had had the right to act on his own behalf and had done so. Indeed, if the employer could not rely on the union's decision, the grievance procedure would not provide the "uniform and exclusive method for [the] orderly settlement of employee grievances," which the Court has recognized is essential to the national labor policy."15 See Clayton v. International Union Automobile, Aerospace & Agricultural Implement Workers, 451 U.S. 679, 686-687, 101 S.Ct. 2088, 2093-2094, 68 L.Ed.2d 538 (1981). 26 The principle announced in Vaca reflects this allocation of responsibilities in the grievance procedure—a procedure that contemplates that both employer and union will perform their respective obligations. In the absence of damages apportionment where the default of both parties contributes to the employee's injury, incentives to comply with the grievance procedure will be diminished. Indeed, imposing total liability solely on the employer could well affect the willingness of employers to agree to arbitration clauses as they are customarily written. 27 Nor will requiring the union to pay damages impose a burden on the union inconsistent with national labor policy.16 It will provide an additional incentive for the union to process its members' claims where warranted. See Vaca, supra, at 187, 87 S.Ct., at 915. This is wholly consistent with a union's interest. It is a duty owed to its members as well as consistent with the union's commitment to the employer under the arbitration clause. See Republic Steel, supra, at 653, 85 S.Ct., at 616. III 28 The Union contends that Czosek v. O'Mara, 397 U.S. 25, 90 S.Ct. 770, 25 L.Ed.2d 21 (1970), requires a different reading of Vaca and a different weighing of the interests our cases have developed. Czosek, however, is consistent with our holding today.17 In Czosek, employees of the Erie Lackawanna Railroad were placed on furlough and not recalled. They brought suit against the railroad for wrongful discharge and against their union for breaching its duty of fair representation. They alleged that the union had arbitrarily and capriciously refused to process their claims against the railroad. See id., at 26, 90 S.Ct., at 771. The District Court dismissed the claim against the railroad because the employees had not pursued the administrative remedies provided by the Railway Labor Act.18 It dismissed the claim against the union because the employees' ability to pursue an administrative remedy on their own absolved the union of any duty. The Court of Appeals for the Second Circuit affirmed the dismissal of the claim against the railroad but found that the employees had stated a claim against the union. Even though the employees had a right to seek full redress from an administrative board, the union still had a duty to represent them fairly. See Conley v. Gibson, 355 U.S. 41, 78 S.Ct. 99, 2 L.Ed.2d 80 (1957). 29 This Court affirmed. In so doing, it addressed the union's concern that if the railroad were not joined as a party, the union might be held responsible for damages for which the railroad was wholly or partly responsible. The Court stated: 30 "[J]udgment against [the union] can in any event be had only for those damages that flowed from [its] own conduct. Assuming a wrongful discharge by the employer independent of any discriminatory conduct by the union and a subsequent discriminatory refusal by the union to process grievances based on the discharge, damages against the union for loss of employment are unrecoverable except to the extent that its refusal to handle the grievances added to the difficulty and expense of collecting from the employer." 397 U.S., at 29, 90 S.Ct., at 773. 31 Although the statement is broadly phrased, it should not be divorced from the context in which it arose. The Railway Labor Act provided the employees in Czosek with an alternate remedy, which they could have pursued when the union refused to process their grievances. Because the union's actions did not deprive the employees of immediate access to a remedy, it did not increase the damages that the employer otherwise would have had to pay. The Court therefore stated that the only damages flowing from the union's conduct in this case were the added expenses the employees incurred. This is consistent with Vaca § recognition that each party should bear the damages attributable to its fault. IV 32 In this case, the findings of the District Court, accepted by the Court of Appeals, establish that the damages sustained by petitioner were caused initially by the Service's unlawful discharge and increased by the Union's breach of its duty of fair representation. Accordingly, apportionment of the damages was required by Vaca.19 We reverse the judgment of the Court of Appeals and remand for entry of judgment allocating damages against both the Service and the Union consistent with this opinion. 33 It is so ordered. 34 Justice WHITE, with whom Justice MARSHALL, Justice BLACKMUN, and Justice REHNQUIST (except as to part IV), join, concurring in part in the judgment and dissenting in part. 35 The Court holds that an employer who wrongfully discharges an employee protected by a collective bargaining agreement with an arbitration clause is only responsible for backpay that accrues prior to the hypothetical date upon which an arbitrator would have issued an award had the employee's union taken the matter to arbitration. All backpay damages that accrue after this time are the sole responsibility of the union, even where, as here, the union is in no way responsible for the employer's decision to terminate the employee. This rationale, which heretofore has been rejected by every Court of Appeals that has squarely considered it,1 does not give due regard to our prior precedents, to equitable principles, or to the national labor policy. I therefore respectfully dissent. For the following reasons, I believe that the employer should be primarily liable for all backpay. 36 * In Smith v. Evening News Ass'n., 371 U.S. 195, 200-201, 83 S.Ct. 267, 270-271, 9 L.Ed.2d 246 (1962), we held for the first time that an individual employee may bring a § 3012 suit against his employer for breach of a collective bargaining agreement. If, as in Smith, the agreement does not contain an arbitration provision, the employee's right to bring suit is unqualified, and, in such a case, the employer unquestionably is liable for any and all backpay that is due. 37 On the other hand, if, as in the present case, the agreement does contain an arbitration provision, it is much more difficult for an employee to maintain a § 301 action against his employer for any backpay whatsoever. This is because Republic Steel Corp. v. Maddox, 379 U.S. 650, 85 S.Ct. 614, 13 L.Ed.2d 580 (1965), established that contractual grievance and arbitration procedures must be exhausted before an employee files a § 301 suit. The Republic Steel rule was adopted to protect the integrity of the collective-bargaining process, and to further the national labor policy of encouraging private resolution of disputes arising over the interpretation and implementation of collective-bargaining agreements. See Clayton v. Automobile Workers, 451 U.S. 679, 686-687, 101 S.Ct. 2088, 2093-2094, 68 L.Ed.2d 538 (1981). 38 Noting that contractual remedies sometimes prove to be "unsatisfactory or unworkable for the individual grievant," we considered in Vaca v. Sipes, 386 U.S. 171, 185, 87 S.Ct. 903, 914, 17 L.Ed.2d 842 (1967), the question "under what circumstances the individual employee may obtain judicial review of his breach-of-contract claim despite his failure to secure relief through the contractual remedial procedures." We found that one situation in which "the employee may seek judicial review of his contractual rights" is where the union has the sole power to invoke the higher stages of the grievance procedure, and "the employee-plaintiff has been prevented from exhausting his contractual remedies by the union's wrongful refusal to process the grievance." Ibid. An employee may maintain a § 301 suit under these circumstances because, in enacting the laws imposing a duty of fair representation on unions, Congress did not intend "to shield employers from the natural consequences of their breaches of bargaining agreements by wrongful union conduct in the enforcement of such agreements." Id., at 186, 87 S.Ct., at 914. 39 Vaca made clear that, with respect to an employer, the only consequence of a union's breach of a fair-representation duty to an employee is that it provides the employee with the means of defeating the employer's "defense based upon the failure to exhaust contractual remedies," ibid., in a § 301 suit. The Court explicitly stated that the union's violation of its statutory duty in no way "exempt[ed] the employer from contractual damages which he would otherwise have had to pay," id., at 196, 87 S.Ct., at 920, and that the employer could not "hide behind the union's wrongful failure to act." Id., at 197, 87 S.Ct., at 920. 40 In Hines v. Anchor Motor Freight, Inc., 424 U.S. 554, 96 S.Ct. 1048, 47 L.Ed.2d 231 (1976), we reiterated that a union's breach of duty to an employee does not shield an employer from damages that it would otherwise owe. Hines involved employees whose grievances had been fully arbitrated. The arbitrator had upheld the discharge as rightful. Nevertheless, the Court held that the employee might still maintain a § 301 action if he could establish that his union had breached its duty of representing him fairly during the arbitral proceedings, even though the employer was in no way responsible for the alleged union malfeasance. Id., at 569, 96 S.Ct., at 1058. The employer protested that, since its conduct during the arbitration was blameless, it should be able to rely on the finality of the arbitral award. We rejected this argument, pointing out that the employer had "surely played its part in precipitating [the] dispute" by discharging the plaintiff-employee in the first place. Ibid. As in Vaca, with respect to the employer, the only consequence of the union's breach was that it "remove[d] the bar" to the employee's right to bring a § 301 action.3 Id., at 567, 96 S.Ct., at 1057. 41 Thus, under our previous holdings, as far as the employer is concerned, a union's breach of a fair-representation duty does no more than remove the procedural exhaustion-of-remedies bar to a § 301 suit by an aggrieved employee. The union's breach does not affect the employer's potential liability, including backpay liability, if the employee prevails in the § 301 judicial proceedings by showing that the employer had breached its contract in discharging him. 42 That the union is not primarily liable for backpay is readily apparent upon close inspection of the facts in Vaca. The employee in that case had been discharged in January 1960. Sometime after February 1961, the union refused to take the matter to arbitration, and, in February 1962, the employee filed suit, claiming that the union's refusal to go to arbitration violated his rights. The trial began in June 1964, and the matter was not finally adjudicated until this Court rendered its decision in February 1967. See Vaca, 386 U.S., at 173-176, 87 S.Ct., at 907-909. Had the union opted in favor of arbitration, an award almost certainly would have been forthcoming long before the judicial suit had even proceeded to trial.4 Nevertheless, the Vaca Court commented that "all or almost all" of the employee's damages would be attributable to the employer, not the union. Id., at 198, 87 S.Ct., at 921. Had the Court intended to hold the union responsible for backpay accruing after the hypothetical arbitration date, presumably well over half of this liability would have been attributed to the union. 43 Of course, this does not mean that the union escapes liability for the "natural consequences," Vaca, 386 U.S., at 186, 87 S.Ct., at 914, of its wrongful conduct. The damages that an employee may recover upon proof that his union has breached its duty to represent him fairly are simply of a different nature than those recoverable from the employer. This is why we found in Vaca that "damages attributable to the employer's breach of contract should not be charged to the union, but increases if any in those damages caused by the union's refusal to process the grievance should not be charged to the employer." 386 U.S., at 197-198, 87 S.Ct., at 920-921. 44 What, then, is the proper measure of the union's damages in a hybrid § 301/breach-of-duty suit? We considered this question in Czosek v. O'Mara, 397 U.S. 25, 29, 90 S.Ct. 770, 773, 25 L.Ed.2d 21 (1970), and concluded that, under the Vaca rule, the union is liable in damages to the extent that its misconduct "add[s] to the difficulty and expense of collecting from the employer."5 Czosek reassured unions that they would not be forced to pay damages "for which the employer is wholly or partly responsible." Id., at 28-29, 90 S.Ct., at 772-773 (emphasis added). 45 It is true that, under the Vaca-Czosek rule, the union may sometimes only have de minimis liability, and we unanimously acknowledged this fact in Electrical Workers v. Foust, 442 U.S. 42, 48, 50, 99 S.Ct. 2121, 2125, 2126, 60 L.Ed.2d 698 (1979). "The damages a union will have to pay in a typical unfair representation suit are minimal; under Vaca § apportionment formula, the bulk of the award will be paid by the employer, the perpetrator of the wrongful discharge, in a parallel § 301 action." Id., at 57, 99 S.Ct., at 2130 (BLACKMUN, J., concurring in result; joined by BURGER, C.J., REHNQUIST, and STEVENS, JJ.). The Foust majority nevertheless reaffirmed Vaca and, moreover, further insulated unions from liability by holding that punitive damages could not be assessed in an action for breach of the duty of fair representation. In reaching these conclusions, the Court relied on the policy of affording individual employees redress for injuries caused by union misconduct without compromising the collective interests of union members in protecting limited union funds. As in Vaca, considerations of deterrence were deemed insufficient to risk endangering union "financial stability." Id., at 50-51, 99 S.Ct., at 2126-2127.6 II 46 Our precedents notwithstanding, the Court today abandons the Vaca rationale and holds that a union's breach of duty does far more than simply remove the exhaustion defense in an employee's § 301 suit against his employer. The union's breach, even if totally unrelated to the employer's decision to terminate the employee, now serves to insulate the employer from further backpay liability, as of the hypothetical arbitration date, even though the employer, unlike the union, can stop backpay accretion at any moment it desires, simply by reinstating the discharged employee. 47 It cannot be denied that, contrary to Vaca and its progeny, under the Court's new rule, the "bulk of the award" for backpay in a hybrid § 301/breach-of-duty suit will have to be borne by the union, not the employer. In the present case, for example, the jury, which was instructed in accordance with the Court's new test, assessed $30,000 in compensatory damages against the union, and only $17,000 against the employer. The union should well consider itself fortunate that this dispute proceeded to trial less than three years after the cessation of petitioner Bowen's employment. Most of the cases of this nature that have been reviewed by this Court have taken the better part of a decade to run their course.7 Because the hypothetical arbitration date will usually be less than one year after the discharge, see note 4 supra, it is readily apparent that, under the Court's rule, in many cases the union will be subject to large liability, far greater than that of the employer, the extent of which will not be in any way related to the union's comparative culpability. Nor will the union have any readily apparent way to limit its constantly increasing liability.8 48 Bowen and the Postal Service argue that the employer is not the "cause" of an employee's lost earnings after the date on which an arbitral decision would have reinstated or otherwise compensated the employee. In the "but for" sense, of course, this is patently false, as the Court concedes. Ante, at 223. But for the employer's breach of contract, there would be no occasion for anyone to reimburse the plaintiff for lost wages accumulated either before or after a hypothetical arbitration. Furthermore, the consequences of the breach—the discharge without cause continue to accumulate as long as the employer refuses to reinstate. The union's failure to arbitrate does not make the discharge and the refusal to reinstate any less wrongful. 49 Thus, there is no reason why the matter should not be governed by the traditional rule of contract law that a breaching defendant must pay damages equivalent to the total harm suffered, "even though there were contributing factors other than his own conduct." 5 A. Corbin, Corbin on Contracts § 999 (1964). The plaintiff need not show the proportionate part played by the defendant's breach of contract among all the contributing factors causing the injury, and his loss need not be "segregated proportionately." Ibid. We followed this rule in Czosek, supra, when we determined that an employer must pay the damages if it is "wholly or partly" responsible for the plaintiff-employee's loss. 397 U.S., at 29, 90 S.Ct., at 773. Even if the union did not stop the employer from persisting in its breach of contract, as it might have done, conduct of this nature is hardly sufficient to exonerate the employer. 50 It bears re-emphasizing that both before and after the hypothetical arbitration date, the union did not in any way prevent the employer from reinstating Bowen, and that the employer could reinstate him. Under these circumstances, it is bizarre to hold, as the Court does, that the relatively impotent union is exclusively liable for the bulk of the backpay. The Court, in effect, sustains the employer's protest to the union that "you should be liable for all damages flowing from my wrong from and after a certain time, because you should have caught and rectified my wrong by that time." Seymour v. Olin Corp., 666 F.2d 202, 215 (CA5 1982). The employer's wrongful conduct clearly was the generating cause of Bowen's loss, and only the employer had the continuing ability to right the wrong and limit liability by reinstating Bowen. The employer has the sole duty to pay wages, and it should be responsible for all back wages to which Bowen is entitled. 51 The Court finds that its apportionment rule "is consistent with the union's commitment to the employer under the arbitration clause" of the collective bargaining agreement. Ante, at 227-228. However, the Court in no way identifies a legitimate source of the union's "commitment under the arbitration clause" that it will bear exclusive liability for post-arbitration-date backpay. The Court's finding is grounded on the assumption that the collective bargaining agreement somehow entitles the employer to rely on the union to bring any wrongful discharge to its attention within the context of the grievance machinery. But the typical collective agreement, including the one here, contains no language entitling the employer to such reliance. The agreement gives the union the right to raise grievances, but it does not obligate it to do so. And, most assuredly, the agreement in no way expressly or impliedly grants the employer any rights against the union if the union fails to bring a meritorious grievance to its attention. 52 Indeed, it is only the union's statutory duty—implied by the judiciary9—to employees to provide them with fair representation that in any way obliges the union to take certain grievances to the employer for consideration. The duty of fair representation obliges a union "to make an honest effort to serve the interests of all [bargaining unit] members" fairly and impartially. Ford Motor Co. v. Huffman, 345 U.S. 330, 337, 73 S.Ct. 681, 685, 97 L.Ed. 1048 (1953); Wallace Corp. v. NLRB, 323 U.S. 248, 255, 65 S.Ct. 238, 241, 89 L.Ed. 216 (1944); Steele v. Louisville & Nashville Ry. Co., 323 U.S. 192, 202-203, 65 S.Ct. 226, 231-232, 89 L.Ed. 173 (1944). It serves as a "bulwark to prevent arbitrary union conduct against individuals stripped of traditional forms of redress by federal labor law." Vaca, 386 U.S., at 182, 87 S.Ct., at 912 (emphasis added). The union owes this duty of fair representation to the employees it represents—the duty does not run to the employer, and the Court does not contend otherwise. 53 Accordingly, neither the collective bargaining agreement nor the union's duty of fair representation provides any support for the Court's conclusion that the union has somehow committed itself to protect the employer, and that the employer has the right to rely on the union to cut off its liability. Contrary to our past cases construing the federal labor law, the Court in effect reads an indemnification provision into the collective bargaining agreement, even though the employer can and more properly should be required to bargain for such a provision, if desired.10 It is a basic tenet of national labor policy that "when neither the collective bargaining process nor its end product violates any command of Congress, a federal court has no authority to modify the substantive terms of a collective bargaining agreement." Mine Workers Health & Retirement Funds v. Robinson, --- U.S. ----, ----, 102 S.Ct. 1226, 1234, 71 L.Ed.2d 419 (1982). See also Carbon Fuel Co. v. Mine Workers, 444 U.S. 212, 218-219, 100 S.Ct. 410, 414-415, 62 L.Ed.2d 394 (1979); Porter Co., Inc. v. NLRB, 397 U.S. 99, 108, 90 S.Ct. 821, 826, 25 L.Ed.2d 146 (1970).11 54 The Court also contends, ante, at 226-227, that its rule will better enable grievance procedures to provide the uniform and exclusive method for the orderly settlement of employee grievances, because a contrary rule "could well affect the willingness of employers to agree to arbitration clauses as they are customarily written." Why the Court's rule will not "affect the willingness" of unions to agree to such clauses is left unexplained. More importantly, since the practical consequence of today's holding is that unions will take many unmeritorious grievances to arbitration simply to avoid exposure to the new breach-of-duty liability, the Court's rule actually impairs the ability of the grievance machinery to provide for orderly dispute resolution. 55 I thus cannot agree with the Court's judgment imposing backpay liability on the union. Lost wages are among the "natural consequences," Vaca, 386 U.S., at 186, 87 S.Ct., at 914, of an employer's wrongful discharge of an employee. Precedent, equity, and national labor policy do not impose on the union primary responsibility for all backpay accruing after its failure to arbitrate.12 III 56 There are at least two situations in which a union should bear some liability for backpay. First, as recognized in Vaca, the union and the employer may be jointly and severally liable where the union has affirmatively induced the employer to commit the alleged breach of contract. 386 U.S., at 197 n. 18, 87 S.Ct., at 920 n. 18. Second, even in a case such as this one, in which the union is not responsible for the discharge, the union should be secondarily liable. That is, if, due to a breach of duty by his union, an employee is unable to collect the backpay to which he is entitled from his employer, the entity primarily liable, he should then be entitled to collect from the union.13 57 This rule of primary and secondary liability prevails in the law of trusts, and should be equally applicable in the present context.14 Just as an individual employee may bring a suit for breach of contract against his employer if, but only if, the union has breached its duty of fair representation in determining not to pursue the grievance on the employee's behalf, a trust beneficiary may sue to enforce a contract entered into on his behalf by the trustee if, but only if, the trustee "improperly refuses or neglects to bring an action against the third person." ALI, Restatement (2d) of Trusts § 282(2) (1959); G. Bogert & G. Bogert, The Law of Trusts and Trustees § 869 (2d ed. 1982); 4 A. Scott, The Law of Trusts § 282.1 (3d ed. 1967). If the beneficiary is able to collect in full from the primary obligor, the trustee should not be monetarily liable. See, e.g., Pollard v. Pollard, 166 Cal.App.2d 698, 333 P.2d 356, 357 (1959). However, the trustee must pay if his wrongful action causes a loss to the beneficiary, such as where the claim was originally enforceable, but the obligor has become insolvent, or where the claim has become barred by the statute of limitations. 2 Scott, supra, § 177. 58 The Court of Appeals for the Fourth Circuit correctly applied a similar rule in the labor context in Harrison v. Transportation Union, 530 F.2d 558 (CA4 1976), cert. denied, 425 U.S. 958, 96 S.Ct. 1739, 48 L.Ed.2d 203 (1976). In that case, the plaintiff-employee's union breached its duty of fair representation by allowing the plaintiff's claim against the employer to become time-barred. The court held that, under these circumstances, the union should be responsible for the lost wages the plaintiff might have recovered from the employer but for the union's misconduct. Id., at 562. See also Nedd v. Mine Workers, 400 F.2d 103, 106-107 (CA3 1968). 59 No such exception to the rule I would apply is applicable in this case. The union did not incite Bowen's discharge, and Bowen is able to recover in full from the Postal Service. Therefore, I would affirm the judgment of the Court of Appeals, to the extent that it holds the union not liable for the backpay to which Bowen is entitled. IV 60 I disagree with the Court of Appeals, however, to the extent that it holds the Service not liable for the $30,000 assessed by the District Court against the union, thus precluding Bowen from recovering this amount from either defendant. The parties stipulated that Bowen lost approximately $47,000 in wages prior to trial.15 The District Court, based upon the jury's special verdict,16 found the employer liable for $17,000 of these damages, and the union liable for the remainder. Although the Court of Appeals held that Bowen's lost earnings were an exclusive obligation of the Service, the court, in a footnote belatedly added to its opinion, refused to amend the judgment and assess the Service for the $30,000 that the District Court erroneously charged against the union. This was done on the erroneous ground that Bowen did not file a cross-appeal against the Postal Service for the $30,000. 642 F.2d 79, 82 n. 6 (CA4 1981). 61 The purport of Vaca v. Sipes is to provide employees with effective remedies to make them whole. 386 U.S., at 185-186, 87 S.Ct., at 914-915. See Electrical Workers v. Foust, 442 U.S., at 48-49, 99 S.Ct., at 2125-2126 (1979); id., at 54, 99 S.Ct., at 2129 (BLACKMUN, J., concurring in result). The footnote added by the Court of Appeals had exactly the opposite effect: it deprived Bowen of his full recovery and did so in a procedurally questionable manner. The Court of Appeals found no infirmity in the total quantum of the District Court's judgment in Bowen's favor. It reversed only that aspect of the judgment that was of no real concern to Bowen, the apportionment of the burden of the award between the union and the Postal Service. Yet the Court of Appeals frustrated Bowen's entitlement to complete recovery by holding that Bowen's failure to appeal prevented the reopening of the award against the Postal Service. 62 Bowen had no cause to challenge this judgment. Under the law, he had no right to a joint and several liability award against the defendants. Because the "Union played no part in [the Postal Service's] alleged breach of contract and since [the Postal Service] took no part in the Union's alleged breach of duty, joint liability for either wrong would be unwarranted." Vaca v. Sipes, 386 U.S., at 197, n. 18, 87 S.Ct., at 920, n. 18. Thus, from Bowen's standpoint, apart from collectibility, the legal effect of the judgment could not have been improved. Whether or not he had some technical basis for appealing a judgment in his favor, neither the facts of this case nor the concerns of national labor policy required him to appeal to protect his judgment. To rule otherwise would impose upon appellate courts the burden of additional appeals from favorable decisions prosecuted by litigants attempting to insulate their judgments from actions like that taken here by the Court of Appeals. In this respect, the Court and I are in agreement. See ante, at 217-218, n. 7. 63 Accordingly, I would affirm the Court of Appeals' judgment that the union was not liable for backpay damages, but I would reverse the remainder of the judgment and remand the case with instructions that the District Court be directed to enter judgment against the Postal Service for the entire amount of Bowen's backpay loss. 64 Justice REHNQUIST, dissenting. 65 I have joined Parts I, II, and III of Justice WHITE's dissenting opinion. However I have some doubt about the proposition advanced by Part IV of the dissent and by the Court, ante, at 217-218, n. 7. 66 The District Court entered judgment for Bowen in the amount of $52,954. It apportioned $30,000 of this amount against the Union, and $22,954 against the Postal Service. When it reversed the judgment against the Union, the Court of Appeals declined to increase the award against the Postal Service or to remand for a new trial. Because this Court has reversed the judgment of the Court of Appeals, its assertion that Bowen "should not have been deprived of the full amount of his compensatory damages because of his failure to cross appeal," id., is dictum. Although the issue is not before us, I am writing separately to express my doubts about the soundness of this proposition. 67 The District Court observed that "there is authority suggesting that only the employer is liable for damages in the form of back pay", ante, at 218, and the decisions of the Courts of Appeals discussed both by Justice White's opinion and the Court's opinion show at the very least that there was substantial doubt that a union could be held liable for damages such as those awarded by the District Court. Under these circumstances, Bowen could not reasonably think that he was in the sort of "safe harbor" which the Court's opinion and Justice White's opinion suggest. Appellate courts review judgments, and Bowen's judgment against the Postal Service was for $22,954. 68 Prudent plaintiff's counsel would have filed a conditional cross-appeal, seeking to increase the amount of that judgment if the Union were held not liable. This is because an "appellee may not attack the decree with a view either to enlarging his own rights thereunder or of lessening the rights of his adversary, whether what he seeks is to correct an error or to supplement the decree with respect to a matter not dealt with below." United States v. American Railway Express Co., 265 U.S. 425, 435, 44 S.Ct. 560, 563, 68 L.Ed. 1087 (1924). 69 It is not clear to me, and in light of the Court's disposition of the case I need not decide, whether the Court of Appeals acted properly in refusing to alter the judgment against the Postal Service, or whether it should have remanded to the District Court for further proceedings on the damage issue. It seems to me, however, that the disposition suggested by the Court and by Part IV of Justice WHITE's dissent would permit plaintiffs to "attack" the judgment of the Court of Appeals in a way prohibited by authorities such as American Railway Express, supra. 1 The jury sat only as an advisory panel on Bowen's claims against the Service. See 28 U.S.C. § 2402 ("Any action against the United States under section 1346 shall be tried by the court without a jury"). 2 Question 3 of the Special Verdict stated, "If [you find that the Union breached its duty of fair representation and/or the Service discharged Bowen without just cause], state from a preponderance of the evidence or with reasonable certainty the amount of compensatory damages to which [Bowen] is entitled." Question 8 stated, "If compensatory damages are awarded by your answer to Question 3, state the amount, if any, that should be attributable to the defendant Union and the amount, if any, that should be attributable to the defendant Postal Service." App. to Pet. for Cert. A21-A22. 3 Counsel for the Union stated, "Your Honor, in respect to this special verdict form, the [Union] would object to any verdict or any question here which would allow the jury to return a judg[ ]ment against the [Union] for any [ ] form of wages. Traditionally, the Union does not pay wages. And these damages are wholly assessable to the [Service], if at all." Record 611-612. In a Motion for Judgment Notwithstanding the Verdict, counsel for the Union reasserted that the "amount of back wages awarded [Bowen] by the jury against the [Union] is as a matter of law wholly assess[a]ble against the employer." Record, Vol. I, Item 37, ¶ 2. 4 The District Court had instructed the jury that both the Union and the Service could be liable for punitive damages if either had acted "maliciously or recklessly or in callous disregard of the rights of the Plaintiff [Bowen]." Record 597. The jury found that the Service and the Union were liable for punitive damages of $30,000 and $10,000, respectively. App. to Pet. for Cert. A22. The District Court determined, however, that punitive damages could not be assessed against the Service because of sovereign immunity. 470 F.Supp., at 1131. Although the court found that the Union's actions supported the jury's award of punitive damages, it set the award aside. It concluded that it would be unfair to hold the Union liable when the Service was immune. Ibid. Bowen did not appeal the District Court's decision on this point. 5 The grievance-arbitration clause contained in the contract between the Service and the Union provides for a four step grievance procedure. The employee may initiate the grievance by discussing it with his supervisor. The Union has discretion to appeal on the employee's behalf and can elect to pursue the grievance through the next three steps. If the grievance is not settled, the Union may refer the grievance to arbitration. See Record, Vol. I., Item 25, Exhibit 1. Although Bowen could have appealed his discharge to the Civil Service Commission, his right to do so expired 15 days after notice of the Service's action. Moreover, by choosing to pursue his administrative remedies, Bowen would have "waive[d] access to any procedures under the National Agreement beyond Step 2B of the Grievance-Arbitration Procedures." App. 90-91. By choosing the remedy provided by the grievance procedure, he was prevented from presenting his claim to the Civil Service Commission. 6 The District Court found as a fact that if Bowen's grievance had been arbitrated he would have been reinstated by August, 1977. Lost wages after that date were deemed the fault of the Union: "While the [Service] set this case in motion with its discharge, the [Union's] acts, upon which [Bowen] reasonably relied, delayed the reinstatement of [Bowen] and it is a proper apportionment to assign fault to the [Union] for approximately two-thirds of the period [Bowen] was unemployed up to the time of trial." 470 F.Supp., at 1131. 7 In a footnote added after the opinion was first filed, the court noted that it made "no revision in the judgment of $22,954.12 against the Postal Service. In this connection we note that no appeal was entered by [Bowen] from the judgment against the Service in the amount of $22,954.12." 642 F.2d, at 82 n. 6. The court's view that the judgment against the Service could not be increased because of Bowen's failure to appeal is erroneous. Bowen won an unambiguous victory in the District Court. He established that he had been discharged by the employer without just cause and that the Union had breached its duty of fair representation. The amount of lost wages and benefits was not in dispute, and the jury and the District Court awarded him all of his damages, apportioning them between the Union and the Service. Bowen had no reason to be unhappy with the award and should not have been deprived of the full amount of his compensatory damages because of his failure to cross appeal. 8 The dissenting opinion asserts that the "rationale" of apportioning damages, applied by the Court today, "has been rejected by every Court of Appeals that has squarely considered it." See post, at 231 and n. 1. Apart from the fact that we apply the rationale—the "governing principle"—articulated in Vaca, few Courts of Appeals have stated a rationale nor has there been the consistency in result perceived by the dissent. Only one case cited by the dissent has declined to apportion damages after considering the issue fully. See Seymour v. Olin Corp., 666 F.2d 202 (CA5 1982). Others, such as the opinion below, have rejected apportionment after giving the issue only minimal consideration. See Bowen v. United States Postal Service, 642 F.2d 79, 82 (CA4 1981) (simply citing Vaca, but not Vaca § governing principle); Milstead v. International Brotherhood of Teamsters, 649 F.2d 395, 396 (CA6 1981) (finding that damages may not be apportioned on the basis of St. Clair v. Local 515, 422 F.2d 128 (CA6 1969), which found that damages may be apportioned). Some courts have not apportioned damages, but have articulated apparently conflicting rationales. See Wyatt v. Interstate & Ocean Transport Co., 623 F.2d 888, 892-893 (CA4 1980) (refusing to hold union liable for portion of damages caused by its breach but stating that damages can be apportioned when the union "exacerbate[s the employee's] loss or diminution of wages, beyond that for which the employer could be charged"); De Arroyo v. Sindicato de Trabajadores Packinghouse, 425 F.2d 281, 289-290 (CA1), cert. denied, 400 U.S. 877, 91 S.Ct. 121, 27 L.Ed.2d 115 (1970) (refusing to hold union liable for portion of damages caused by its default but stating that apportionment would be proper where there was evidence "that but for the Union's conduct the plaintiffs would have been reinstated or reimbursed at an earlier date"). While it is true these cases reach the same result as the dissent, they do not represent an affirmation of its reasoning. Other Courts of Appeals have recognized that damages should be apportioned between the union and the employer. See Smart v. Ellis Trucking Co., 580 F.2d 215, 219 n. 6 (CA6 1978) (on remand, trial court to determine "the extent to which employer's liability for any backpay may be limited" because of its reliance on arbitration proceeding); Harrison v. Chrysler Corp., 558 F.2d 1273, 1279 (CA7 1977) ("union which breaches its duty of fair representation may be sued by an employee for lost pay attributable to the breach"); Ruzicka v. General Motors Corp., 523 F.2d 306, 312 (CA6 1975) ("Union liable for that portion of Appellant's injury representing 'increases if any in those damages [chargeable to the employer] caused by the union's refusal to process the grievance' ") (brackets in circuit court opinion); St. Clair v. Local 515, supra, at 132 (holding that union "liable for nothing more [than damages measured by backpay] and perhaps for less" because Vaca requires those damages to be apportioned between the employer and union according to each party's fault). See also Feller, A General Theory of the Collective Bargaining Agreement, 61 Cal.L.Rev. 663, 817-824 (1973) (employer's liability should not be increased by union's default); Comment, Apportionment of Damages in DFR/Contract Suits: Who Pays for the Union's Breach, 1981 Wis.L.Rev. 155 (same). In sum, a fair reading of these cases reveals that, contrary to the dissent's assertion, the Courts of Appeals have been far from unanimous in either their results or their rationales. 9 The Court had previously held in Smith v. Evening News Association, 371 U.S. 195, 83 S.Ct. 267, 9 L.Ed.2d 246 (1962), that an employee may sue his employer for a breach of the collective-bargaining agreement under § 301 of the Labor Management Relations Act. See 29 U.S.C. § 185. Because the contract in Smith did not contain a grievance-arbitration procedure that required exhaustion, Smith did not reach the issue presented in Vaca. See id., at 196 n. 17, 87 S.Ct., at 920 n. 17. 10 The Court based its decision on Congress's express approval of contract grievance procedures as a preferred method of settling disputes, the union's interest in actively participating in the continuing administration of the contract, and the employer's interest in limiting the choice of remedies available to aggrieved employees. See 379 U.S., at 653, 85 S.Ct., at 616. 11 We note that this is not a situation in which either the union or the employer has participated in the other's breach. See Vaca, supra, at 197 n. 18, 87 S.Ct., at 920 n. 18. 12 Although the union remains primarily responsible for the portion of the damages resulting from its default, Vaca made clear that the union's breach does not absolve the employer of liability. Thus if the petitioner in this case does not collect the damages apportioned against the Union, the Service remains secondarily liable for the full loss of backpay. 13 In Vaca, the jury had found the union responsible for the entire amount of damages suffered by the employee. The judgment upholding the verdict therefore was reversed. Justice White's dissent reasons that because Vaca found that the employer is not absolved from liability by the union's breach, the employer must be solely responsible. The first proposition, however, does not require the second. Thus, Vaca § recognition that the employer "may not hide behind the union's wrongful act" does not answer the question posed by this case, how damages should be apportioned as between the two wrongdoers, the union and the employer. On this point, the explicit language of Vaca § governing principle makes clear that the union is responsible for increases in the employee's damages flowing from the wrongful discharge, a point which the dissent glosses over. Although the Court in Vaca concluded that the union had not breached its duty, it observed that "[i]n this case, even if the Union had breached its duty, all or almost all of [the employee's] damages would still be attributable to his allegedly wrongful discharge." 386 U.S., at 198, 87 S.Ct., at 921. Assuming that such a breach did occur, the facts are not sufficiently clear to determine when the breach would have occurred or the portion of damages attributable to each party's fault. Thus this speculative observation is not inconsistent with the Court's precisely worded statement of the governing principle. 14 The parties to the collective-bargaining agreement, of course, may choose not to include a grievance procedure supervised by the union or, if they do, may choose not to make the procedure exclusive. See Vaca, supra, at 184 n. 9, 87 S.Ct., at 913 n. 9; Republic Steel, supra, at 657-658, 85 S.Ct., at 618-619, cf. 29 U.S.C. § 159(a) (employee may present grievances to his employer "without the intervention of the bargaining representative, as long as the adjustment is not inconsistent with the terms of a collective-bargaining contract or agreement then in effect. . . ."). Most collective-bargaining agreements, however, contain exclusive grievance-arbitration procedures and give the union power to supervise the procedure. See Feller, A General Theory of the Collective Bargaining Agreement, 61 Cal.L.Rev. 663, 742, 752-753 (1973). When the collective bargaining agreement provides the union with sole authority to press an employee's grievance, the union acts as the employee's exclusive representative in the grievance-arbitration procedure. See Vaca, supra, at 191-192, 87 S.Ct., at 917-918. 15 Under the dissent's analysis, the employer may not rely on the union's decision not to pursue a grievance. Rather it can prevent continued liability only by reinstating the discharged employee. See post, at 238-239. This leaves the employer with a dubious option: it must either reinstate the employee promptly or leave itself exposed to open-ended liability. If this were the rule, the very purpose of the grievance procedure would be defeated. It is precisely to provide the exclusive means of resolving this kind of dispute that the parties agree to such a procedure and national labor policy strongly encourages its use. See Republic Steel, supra, at 653, 85 S.Ct., at 616. When the union has breached its duty of fair representation, the dissent justifies its rule by arguing that "only the employer ha[s] the continuing ability to right the wrong by reinstating" the employee, an ability that the union lacks. See post, at 239. But an employer has no way of knowing that a failure to carry a grievance to arbitration constitutes a breach of duty. Rather than rehiring, as the dissent suggests, the employer reasonably could assume that the union had concluded the discharge was justified. The union would have the option, if it realized it had committed an arguable breach of duty, to bring its default to the employer's attention. Our holding today would not prevent a jury from taking such action into account. See infra, n. 19. Moreover, the rule urged by the dissenting opinion would allow the union and the employee, once the case goes to trial, to agree to a settlement pursuant to which the union would acknowledge a breach of its duty of fair representation in exchange for the employee's undertaking to look to his employer for his entire recovery. Although we may assume that this would not occur frequently, the incentive the dissent's rule would provide to agree to such a settlement demonstrates its unsoundness. 16 Requiring the union to pay its share of the damages is consistent with the interests recognized in International Brotherhood of Electrical Workers v. Foust, 442 U.S. 42, 99 S.Ct. 2121, 60 L.Ed.2d 698 (1979). In Foust, we found that a union was not liable for punitive damages. The interest in deterring future breaches by the union was outweighed by the debilitating impact that "unpredictable and potentially substantial" awards of punitive damages would have on the union treasury and the union's exercise of discretion in deciding what claims to pursue. Id., at 50-52, 99 S.Ct., at 2126-2127. An award of compensatory damages, however, normally will be limited and finite. Moreover, the union's exercise of discretion is shielded by the standard necessary to prove a breach of the duty of fair representation. Thus, the threat that was present in Foust is absent here. 17 In cases following Vaca and Czosek, the Court has not had occasion to address the question presented here. In Hines v. Anchor Motor Freight, Inc., 424 U.S. 554, 96 S.Ct. 1048, 47 L.Ed.2d 231 (1976), we held that proof of a breach of the duty of fair representation will remove the bar of finality from arbitral decisions. We did not consider the scope of the remedy available, but stated that if the employer had wrongfully discharged the employee and the union had breached its duty, the employee was "entitled to an appropriate remedy against the employer as well as the Union." Id., at 572, 96 S.Ct., at 1060. In Foust, we reviewed the principles announced in Vaca as bearing on the question of whether a union can be held liable for punitive damages. Although Foust § discussion of Vaca could be read as suggesting a contrary principle to that stated in Vaca, the holding in Foust —that a union may not be held liable for punitive damages—is consistent with our holding here. Finally, in Clayton v. International Union, United Automobile, Aerospace & Agricultural Implement Workers of America, 451 U.S. 679, 690, n. 15, 101 S.Ct. 2088, 2096 n. 15, 68 L.Ed.2d 538 (1981), the Court reiterated Vaca § governing principle. Although the Court did not apply that principle, the context in which it was discussed bears on the question presented here. In considering whether the remedies available to the employee in an internal union procedure were equivalent to the remedies available to an employee in a § 301 suit, the Court found it significant that the union procedures allowed an employee to recover backpay against the union. This was a recognition of Vaca § explicitly announced governing principle. 18 See 45 U.S.C. § 153 First (i), (j). These sections provide that an employee who is unsuccessful at the grievance level can seek relief on his own from the National Railroad Adjustment Board. The Board is authorized to provide remedies similar to those available in a court suit. See Republic Steel, supra, at 657, n. 14, 85 S.Ct., at 619, n. 14. 19 We need not decide whether the District Court's instructions on apportionment of damages were proper. The Union objected to the instructions only on the ground that no back wages at all could be assessed against it. It did not object to the manner of apportionment if such damages were to be assessed. Nor is it necessary in this case to consider whether there were degrees of fault, as both the Service and the Union were found to have acted in "reckless and callous disregard of [Bowen's] rights." 1 In addition to the opinion below in the present case, Bowen v. Postal Service, 642 F.2d 79 (CA4 1981), see Seymour v. Olin Corp., 666 F.2d 202 (CA5 1982); and Milstead v. Teamsters, 649 F.2d 395 (CA6 1981), cert. denied, 454 U.S. 896, 102 S.Ct. 394, 70 L.Ed.2d 211 (1982). These three are the only circuit court decisions rendering square holdings on the issue. However, also consistent with the view advanced in this dissent are Wyatt v. Interstate & Ocean Transport Co., 623 F.2d 888 (CA4 1980) (assessing the employer for all backpay); and Soto Segura v. Sea-Land Service, Inc., 581 F.2d 291, 298 (CA1 1978), where the First Circuit specifically noted that, in accordance with Vaca v. Sipes, 386 U.S. 171, 87 S.Ct. 903, 17 L.Ed.2d 842 (1967), and Czosek v. O'Mara, 397 U.S. 25, 90 S.Ct. 770, 25 L.Ed.2d 21 (1970), the District Court "did not charge the union for any of the back pay due appellee but instead awarded $5,750 in attorney's fees proximately caused by the Union's failure to process his grievance." See also DeArroyo v. Sindicata de Trabajadores Packinghouse, 425 F.2d 281, 289-290 (CA1), cert. denied, 400 U.S. 877, 91 S.Ct. 121, 27 L.Ed.2d 115 (1970), where the employer was held liable for all backpay, even though a jury had found that 40% of this amount accrued because of the union's wrongful conduct. See Feller, A General Theory of the Collective Bargaining Agreement, 61 Cal.L.Rev. 663, 671-672 (1973). The Court incorrectly states, ante, at 218-220 n. 8, that the Courts of Appeals have not been consistent on this issue. No circuit court has ever required a union to pay backpay in a case such as this. In fact, the only two cases the Court cites that even suggest the possibility of union liability for backpay are St. Clair v. Local No. 515, 422 F.2d 128 (CA6 1969), and Harrison v. Chrysler Corp., 558 F.2d 1273 (CA7 1977). In St. Clair, the court did not purport to decide the issue; it stated only that the union certainly would not be liable for anything more than backpay less interim earnings, "and perhaps for less," because, in light of Vaca, "the Supreme Court has strongly implied that . . . the increment of damages caused by the union's breach of duty is virtually de minimis." 422 F.2d, at 132. In Harrison, the court did make the comment that "a union which breaches its duty of fair representation may be sued by the employee for lost pay attributable to its breach," 558 F.2d, at 1279, but no union was even a party to the litigation, so this dicta can hardly be regarded as authoritative. 2 Section 301 of the Labor Management Relations Act, 29 U.S.C. § 185. Because the employer in the present case is the U.S. Postal Service, petitioner Bowen's action technically arises under § 1208(b) of the Postal Reorganization Act, 39 U.S.C. § 1208(b), which is identical to § 301 in all relevant respects. 3 Justice Stewart filed a two-paragraph concurring opinion in Hines, in which he stated that the employer should not be liable for backpay accruing between the time of the "tainted" arbitral decision and a subsequent "untainted" determination that the discharges were, after all, wrongful. 424 U.S. at 572-573, 96 S.Ct., at 1060-1061. No other member of the Court joined Justice Stewart's observations, and his opinion was founded on the employer's good-faith reliance on a favorable arbitral decision. Here, the Court goes far beyond Justice Stewart and grants the employer the right to rely on a nonexistent arbitration, even though the union is by no means under a duty to the employer to take any grievances to arbitration. See infra, at 239-241. 4 Statistics developed by the Federal Mediation and Conciliation Service ("FMCS") show that, in 1981, the average time between the filing of a grievance and the rendering of an arbitral award was 230.26 days. For the years between 1972 and 1980, the average varied from a high of 268.3 in 1977 to a low of 223.5 in 1975 and 1978. FMCS, Thirty-Fourth Annual Report 39 (1981). See generally Ross, The Well-Aged Arbitration Case, 11 Indus. & Lab.Rel.L.Rev. 262 (1958); Seitz, Delay: The Asp in the Bosom of Arbitration, 36 Arb.J. 29 (Sept. 1981). Also, in some industries, labor and management have agreed to expedited arbitral proceedings that can further reduce the average time. See Sandver, Blaine & Woyar, Time and Cost Savings through Expedited Arbitration Procedures, 36 Arb.J. 11 (Dec. 1981). 5 Czosek arose under the Railway Labor Act ("RLA"), 45 U.S.C. (and Supp. III) §§ 151 et seq., which permits an employee whose union fails to process his grievance to press it himself. Id. §§ 153 First (i), (j). The Court seeks to limit Czosek to the RLA context, on the theory that, because the employee in Czosek could have filed a grievance without union assistance, the union's default in that case did not "increase the damages that the employer otherwise would have had to pay." Ante, at 228-230. However, the Czosek opinion nowhere suggests that this distinction is relevant, and it cites only Vaca in support of its finding on this point. We reaffirmed in Electrical Workers v. Foust, 442 U.S. 42, 50 n. 13, 99 S.Ct. 2121, 2127 n. 13, 60 L.Ed.2d 698 (1979), that the Czosek rule was an application of "Vaca 's apportionment principle." 6 Even though Foust requires that punitive damages not be assessed against a union, the Vaca rule nevertheless provides for a credible deterrent against wrongful union conduct. Attorney's fees and other litigation expenses have been assessed as damages against unions, because such damages measure the extent by which the union's breach of duty adds to the difficulty and expense of collecting from the employer. See, e.g., Seymour v. Olin Corp., 666 F.2d, at 215; Scott v. Teamsters Local 377, 548 F.2d 1244 (CA6), cert. denied, 431 U.S. 968, 97 S.Ct. 2927, 53 L.Ed.2d 1064 (1977). 7 See, e.g., Clayton v. ITT Gilfillan, 623 F.2d 563, 565 (CA9 1980), rev'd in part sub nom., Clayton v. Automobile Workers, 451 U.S. 679, 101 S.Ct. 2088, 68 L.Ed.2d 538 (1981) (discharge in February 1975; we remand for trial in May 1981); Electrical Workers v. Foust, 442 U.S. at 43-45, 99 S.Ct. at 2123-2124 (discharge in February 1971; trial in May 1976; Court of Appeals' judgment in 1978; this Court rules in 1979); Hines v. Anchor Motor Freight, 424 U.S. at 556-559, 96 S.Ct. at 1052-1053 (discharges in 1967; district court grants summary judgment in 1973; we remand for trial in March 1976); Czosek v. O'Mara, 397 U.S. at 26, 90 S.Ct. at 772 (discharge in 1962; we remand for trial in February 1970); Vaca v. Sipes, 386 U.S., at 175-176, 87 S.Ct., at 908-909 (discharge in January 1960; trial begins in June 1964). 8 While remaining disturbingly vague about the point, the Court at least concedes that a union may shift some or all backpay responsibility back to the employer by "bring[ing] its default to the employer's attention." Ante, at 227 n. 15. 9 Although no statute expressly imposes a duty of fair representation upon unions, we have held, beginning with Steele v. Louisville & Nashville Ry. Co., 323 U.S. 192, 65 S.Ct. 226, 89 L.Ed. 173 (1944), that "the exclusive agent's statutory authority to represent all members of a designated unit includes a statutory obligation to serve the interests of all members without hostility or discrimination toward any, to exercise its discretion with complete good faith and honesty, and to avoid arbitrary conduct." Vaca, 386 U.S., at 177, 87 S.Ct., at 909. See generally Aaron, The Duty of Fair Representation: An Overview, in The Duty of Fair Representation 8 (J. McKelvey ed. 1977). 10 See Edwards, Employers' Liability for Union Unfair Representation: Fiduciary Duty or Bargaining Reality?, 27 Lab.L.J. 686, 691-692 (1976). 11 The Court correctly reaffirms, ante, at 224, that a collective bargaining agreement "is more than a contract; it is a generalized code to govern a myriad of cases which the draftsmen cannot wholly anticipate." Steelworkers v. Warrior & Gulf Navigation Co., 363 U.S. 574, 578, 80 S.Ct. 1347, 1350, 4 L.Ed.2d 1409 (1960). This means that "[g]aps may be left to be filled in by reference to the practices of the particular industry and of the various shops covered by the agreement," because "[m]any of the specific practices which underlie the agreement may be unknown, except in hazy form, even to the negotiators." Id., at 580-581, 80 S.Ct., at 1351-52. The Court does not suggest that the union is obliged by any "industry practice" to protect the employer from back pay liability, or that such an obligation can be inferred in any other way from the "gaps" in the agreement. The Court's holding therefore inserts a new substantive term into the agreement, which is precisely what we have forbidden the lower courts from doing in our previous holdings. The Court's mere belief that an employer "should" be able to rely on the union because there is "no unfairness to the union in this approach," ante, at 226, is not a valid justification for the holding. 12 The Court asserts, ante, at 227, n. 15, that the view advanced in this dissent would allow the union and the employee "to agree to a settlement pursuant to which the union would acknowledge a breach of its duty of fair representation in exchange for the employee's undertaking to look to his employer for his entire recovery." I seriously doubt, however, that a union will lightly "concede" a breach of its fair-representation duty to bargaining unit employees, particularly since it may be liable for the plaintiff's costs of collection, including attorney's fees in not inconsiderable amounts. See n. 6, supra. Furthermore, the Court's position, by exposing the union to even greater liability, may well exert correspondingly greater pressure on the union to settle, with or without an acknowledgement of breach of duty, leaving the employer to defend the action alone. 13 The Court takes the exact opposite tack. It holds that the union is primarily responsible for post-hypothetical-arbitration-date back pay, and that the employer is secondarily liable for this amount. Ante, at 223 n. 12. 14 It is not always proper to import common law principles into federal labor law. See NLRB v. Hearst Publications, Inc., 322 U.S. 111, 120-129, 64 S.Ct. 851, 855-859, 88 L.Ed. 1170 (1944). In the present context, however, the trust analogy appears to be appropriate. See Cox, Rights under a Labor Agreement, 69 Harv.L.Rev. 601, 652 (1956); Jenkins v. Wm. Schluderberg-T.J. Kurdle Co., 217 Md. 556, 144 A.2d 88, 90 (1958). 15 The parties stipulated that Bowen lost $45,389.87 in back wages and fringe benefits from the time of discharge until trial. Record 315. Bowen's counsel misstated this figure as $47,000 in his closing argument, and the jury apparently acted on this basis. Record 550. No party has complained of the $1,610.13 discrepancy. 16 The union timely objected to the District Court's instructions to the extent they allowed the jury to apportion any compensatory damages to the union. Record 611-612. The union renewed its objection in its motion for judgment notwithstanding the verdict, or in the alternative to alter or amend the judgment. Record, Vol. I, Item 37, ¶ 2. The Court quotes the relevant material. See ante, at 215 n. 3.

67

459 U.S. 297 103 S.Ct. 634 74 L.Ed.2d 465 DIRECTOR, OFFICE OF WORKERS' COMPENSATION PROGRAMS, UNITED STATES DEPARTMENT OF LABOR, Petitionerv.PERINI NORTH RIVER ASSOCIATES et al. No. 81-897. Argued Oct. 4, 1982. Decided Jan. 11, 1983. Syllabus Before 1972, coverage under the Longshoremen's and Harbor Workers' Compensation Act (LHWCA or Act) extended only to injuries sustained by workers on the actual "navigable waters of the United States (including any dry dock)." In 1972, the Act was amended by expanding the "navigable waters" situs to include certain adjoining land and by adding a status requirement that employees covered by the Act be "engaged in maritime employment" within the meaning of § 2(3) of the Act. In this case, an employee (Churchill) of respondent construction firm was injured while performing his job on the deck of a cargo barge being used in the construction of a sewage treatment plant extending over the Hudson River in New York. Churchill's claim for compensation under the LHWCA was administratively denied on the ground that he was not "engaged in maritime employment" under § 2(3). On Churchill's petition for review, in which the Director of the Office of Workers' Compensation Programs (Director) (petitioner here) participated as respondent in support of Churchill, the Court of Appeals held that Churchill was not in "maritime employment" because his employment lacked a "significant relationship to navigation or to commerce on navigable waters." Held: 1. Where Churchill is a party respondent under this Court's Rule 19.6 and has filed a brief arguing for his coverage under the Act, there is a justiciable controversy before the Court. Accordingly, it is unnecessary to consider whether the Director, as the official responsible for administration and enforcement of the Act, has Art. III standing as an aggrieved party to seek review of the decision below. The Director's petition under 28 U.S.C. § 1254(1) brings Churchill before the Court, and he, as the injured employee, has a sufficient interest in the question at issue to give him standing to urge consideration of the merits of the Court of Appeals' decision. Pp. 302-305. 2. Churchill, as a marine construction worker injured while performing his job upon actual navigable waters, was "engaged in maritime employment" within the meaning of § 2(3), and thus was covered by the amended Act. Pp. 305-325. (a) There is no doubt that Churchill would have been covered by the Act before it was amended in 1972. Pp. 305-312. (b) There is nothing in the legislative history or in the 1972 amendments themselves to indicate that Congress intended to withdraw coverage from employees injured on navigable waters in the course of their employment as that coverage existed before the 1972 amendments, or that it intended the status language of § 2(3) to require that such an employee show that his employment possessed a direct or substantial relation to navigation or commerce in order to be covered. On the contrary, the legislative history indicates that Congress did not intend to "exclude employees traditionally covered." Moreover, Congress explicitly deleted language from the Act that was found in Calbeck v. Travelers Insurance Co., 370 U.S. 114, 82 S.Ct. 1196, 8 L.Ed.2d 368, to be responsible for the "jurisdictional dilemma" created by the "maritime but local" doctrine whereby a maritime worker was often required to make a perilous jurisdictional "guess" as to which of the two mutually exclusive compensation schemes, i.e., the federal or the state scheme, was applicable to cover his injury. Pp. 313-325. 652 F.2d 255, reversed and remanded. Richard G. Wilkins, Dept. of Justice, Washington, D.C., for petitioner, pro hac vice, by special leave of Court. Martin Krutzel, New York City, for respondents. Justice O'CONNOR delivered the opinion of the Court. 1 In 1972, Congress amended the Longshoremen's and Harbor Workers' Compensation Act, 44 Stat. 1424, as amended, 86 Stat. 1251, 33 U.S.C. § 901, et seq., (hereinafter LHWCA or Act). Before 1972, LHWCA coverage extended only to injuries sustained on the actual "navigable waters of the United States (including any dry dock)." 44 Stat. 1426. As part of its 1972 amendment of the Act, Congress expanded the "navigable waters" situs to include certain adjoining land areas, § 3(a), 33 U.S.C. § 903(a) (1978). At the same time, Congress added a status requirement that employees covered by the Act must be "engaged in maritime employment" within the meaning of § 2(3) of the Act.1 We granted certiorari in this case to consider whether a marine construction worker, who was injured while performing his job upon actual navigable waters,2 and who would have been covered by the Act before 1972, is "engaged in maritime employment" and thus covered by the amended Act.3 We hold that the worker is "engaged in maritime employment" for purposes of coverage under the amended LHWCA. Accordingly, we reverse the decision below. 2 * The facts are not in dispute. Respondent Perini North River Associates ("Perini") contracted to build the foundation of a sewage treatment plant that extends approximately 700 feet over the Hudson River between 135th and 145th Streets in Manhattan. The project required that Perini place large, hollow circular pipes called caissons in the river, down to embedded rock, fill the caissons with concrete, connect the caissons together above the water with concrete beams, and place precast concrete slabs on the beams. The caissons were delivered by rail to the shore, where they were loaded onto supply barges and towed across the river to await unloading and installation. 3 The injured worker, Raymond Churchill, was an employee of Perini in charge of all work performed on a cargo barge used to unload caissons and other materials from the supply barges and to set caissons in position for insertion into the embedded rock. Churchill was on the deck of the cargo barge giving directions to a crane operator engaged in unloading a caisson from a supply barge when a line used to keep the caissons in position snapped and struck Churchill. He sustained injuries to his head, leg, and thumb.4 4 Churchill filed a claim for compensation under the LHWCA. Perini denied that Churchill was covered by the Act, and after a formal hearing pursuant to § 19 of the Act, 33 U.S.C. § 919, an Administrative Law Judge determined that Churchill was not "engaged in maritime employment" under § 2(3) of the Act because his job lacked "some relationship to navigation and commerce on the navigable waters." Petn., at 31a. Churchill and the Director, Office of Workers' Compensation Programs ("Director") appealed to the Benefits Review Board, pursuant to § 21(b)(3) of the Act, 33 U.S.C. § 921(b)(3). The Board affirmed the Administrative Law Judge's denial of coverage, on the theory that marine construction workers involved in building facilities not ultimately used in navigation or commerce upon navigable waters are not engaged in "maritime employment." Petn., at 13a.5 One Board Member dissented, arguing that "all injuries sustained in the course of employment by employees over 'navigable waters' as that term was defined prior to the 1972 Amendments, are covered under the [amended] Act." Petn., at 17a.6 5 Churchill then sought review of the Board's decision in the Second Circuit Court of Appeals, under § 21(c) of the Act, 33 U.S.C. § 921(c).7 The Director participated as respondent, and filed a brief in support of Churchill's position. The Second Circuit denied Churchill's petition, relying on its decision in Fusco v. Perini North River Associates, 622 F.2d 1111 (1980), cert. denied, 449 U.S. 1131, 101 S.Ct. 953, 67 L.Ed.2d 119 (1981). According to the Second Circuit, Churchill was not in "maritime employment" because his employment lacked a " 'significant relationship to navigation or to commerce on navigable waters.' " Churchill v. Perini North River Associates, 652 F.2d 255, 256 n. 1 (CA2 1981). The Director now seeks review of the Second Circuit denial of Churchill's petition. The Director agrees with the position taken by the dissenting member of the Benefits Review Board: the LHWCA does not require that an employee show that his employment possesses a "significant relationship to navigation or commerce," where, as here, the employee is injured while working upon the actual navigable waters in the course of his employment, and would have been covered under the pre-1972 LHWCA.8 II 6 Before we consider whether Churchill is covered by the Act, we must address Perini's threshold contention that the Director does not have standing to seek review of the decision below. According to Perini, the Director's only interest in this case is in furthering a different interpretation of the Act than the one rendered by the Administrative Law Judge, the Benefits Review Board, and the Court of Appeals.9 7 Perini's claim ignores the procedural posture in which this case comes before the Court. That posture makes it unnecessary for us to consider whether the Director, as the agency official "responsible for the administration and enforcement" of the Act,10 has standing as an aggrieved party to seek review of the decision below.11 The Director is not alone in arguing that Churchill is covered under the LHWCA. Churchill, the injured employee, is before the Court as well. He has filed a brief in support of the Director's request for a writ of certiorari, and a brief addressing the merits of his claim, in which he presents the same arguments presented by the Director. But, for some reason that is not entirely clear, Churchill has not elected to seek review as a petitioner, and by virtue of the rules of this Court, he is considered a party respondent.12 It is in this procedural context that Perini's challenge to Article III standing must be considered. Perini concedes that the Director was a proper party respondent before the Court of Appeals in this litigation.13 As party respondent below, the Director is entitled under 28 U.S.C. § 1254(1) to petition for a writ of certiorari. Although the Director has statutory authority to seek review in this Court, he may not have Article III standing to argue the merits of Churchill's claim because the Director's presence does not guarantee the existence of a justiciable controversy with respect to the merits of Churchill's coverage under LHWCA. However, the Director's petition makes Churchill an automatic respondent under our Rule 19.6, and in that capacity, Churchill "may seek reversal of the judgment of the Court of Appeals on any ground urged in that court." O'Bannon v. Town Court Nursing Center, 447 U.S. 773, 783-84 n. 14, 100 S.Ct. 2467, 2474-2475 n. 14, 65 L.Ed.2d 506 (1980). The director's petition, filed under 28 U.S.C. § 1254(1), brings Churchill before this Court, and there is no doubt that Churchill, as the injured employee, has a sufficient interest in this question to give him standing to urge our consideration of the merits of the Second Circuit decision. 8 The constitutional dimension of standing theory requires, at the very least, that there be an "actual injury redressable by the court." Simon v. Eastern Kentucky Welfare Rights Org., 426 U.S. 26, 39, 96 S.Ct. 1917, 1924, 48 L.Ed.2d 450 (1976). This requirement is meant "to assure that the legal questions presented to the court will be resolved, not in the rarified atmosphere of a debating society, but in a concrete factual context conducive to a realistic appreciation of the consequences of judicial action," as well as to assure "an actual factual setting in which the litigant asserts a claim of injury in fact." Valley Forge Christian College v. Americans United for Separation of Church and State, 454 U.S. 464, 472, 102 S.Ct. 752, 758, 70 L.Ed.2d 700 (1982). The presence of Churchill as a party respondent arguing for his coverage under the Act assures that an admittedly justiciable controversy is now before the Court. III 9 The question of Churchill's coverage is an issue of statutory construction and legislative intent. For reasons that we explain below, there is no doubt that Churchill, as a marine construction worker injured upon actual navigable waters in the course of his employment upon those waters, would have been covered by LHWCA before Congress amended it in 1972. In deciding whether Congress intended to restrict the scope of coverage by adding the § 2(3) status requirement, we must consider the scope of coverage under the pre-1972 Act and our cases construing the relevant portions of that Act. We must then focus on the legislative history and purposes of the 1972 amendments to the LHWCA to determine their effect on pre-existing coverage. 10 * Beginning with our decision in Southern Pacific Co. v. Jensen, 244 U.S. 205, 37 S.Ct. 524, 61 L.Ed. 1086 (1917), we held that there were certain circumstances in which states could not, consistently with Article III, Section 2 of the Constitution, provide compensation to injured maritime workers.14 If the employment of an injured worker was determined to have no "direct relation" to navigation or commerce, and "the operation of local law [would not] materially affect" the uniformity of maritime law, then the employment would be characterized as "maritime but local," and the state could provide a compensation remedy. Grant Smith-Porter v. Rohde, 257 U.S. 469, 477, 42 S.Ct. 157, 158, 66 L.Ed. 321 (1922). See also Western Fuel Co. v. Garcia, 257 U.S. 233, 242, 42 S.Ct. 89, 90, 66 L.Ed. 210 (1921). If the employment could not be characterized as "maritime but local," then the injured employee would be left without a compensation remedy. 11 After several unsuccessful attempts to permit state compensation remedies to apply to injured maritime workers whose employment was not local,15 Congress passed the LHWCA in 1927, 44 Stat. 1424. Under the original statutory scheme, a worker had to satisfy five primary conditions in order to be covered under the Act. First, the worker had to satisfy the "negative" definition of "employee" contained in § 2(3) of the 1927 Act in that he could not be a "master or member of a crew of any vessel, nor any person engaged by the master to load or unload or repair any small vessel under eighteen tons net." 44 Stat. 1424.16 Second, the worker had to suffer an "injury" defined by § 2(2) as "accidental injury or death arising out of and in the course of employment . . . ." 44 Stat. 1424. Third, the worker had to be employed by a statutory "employer," defined by § 2(4) as "an employer any of whose employees are employed in maritime employment, in whole or in part, upon the navigable waters of the United States (including any dry dock)." 44 Stat. 1424.17 Fourth, the worker had to meet a "situs" requirement contained in § 3(a) of the Act that limited coverage to workers whose "disability or death results from an injury occurring upon the navigable waters of the United States (including any dry dock)." 44 Stat. 1426. Fifth, § 3(a) precluded federal compensation unless "recovery for the disability or death through workmen's compensation proceedings may not validly be provided by State law." 44 Stat. 1426. 12 Federal compensation under the LHWCA did not initially extend to all maritime employees injured on the navigable waters in the course of their employment. As mentioned, § 3(a) of the 1927 Act permitted federal compensation only if compensation "may not validly be provided by State law." 44 Stat. 1426. This language was interpreted to exclude from LHWCA coverage those employees whose employment was "maritime but local." See, e.g., Crowell v. Benson, 285 U.S. 22, 52 S.Ct. 285, 76 L.Ed. 598 (1932). Application of the "maritime but local" doctrine required case-by-case determinations, and a worker was often required to make a perilous jurisdictional "guess" as to which of two mutually exclusive compensation schemes was applicable to cover his injury. Employers faced uncertainty as to whether their contributions to a state insurance fund would be sufficient to protect them from liability. 13 In Davis v. Department of Labor, 317 U.S. 249, 63 S.Ct. 225, 87 L.Ed. 246 (1942), this Court recognized that despite its many cases involving the "maritime but local" doctrine, it had "been unable to give any guiding, definite rule to determine the extent of state power in advance of litigation . . . ." Id., at 253, 63 S.Ct., at 227. Employees and employers alike were thrust on "[t]he horns of a jurisdictional dilemma." Id., at 255, 63 S.Ct., at 228.18 Davis involved an employee who was injured while dismantling a bridge from a standing position on a barge. We upheld the application of the state compensation law in Davis not because the employee was engaged in "maritime but local" employment, but because we viewed the case as in a "twilight zone" of concurrent jurisdiction where LHWCA coverage was available and where the applicability of state law was difficult to determine. We held that doubt concerning the applicability of state compensation acts was to be resolved in favor of the constitutionality of the state remedy. Relying in part on Davis, the Court in Calbeck v. Travelers Insurance Co., 370 U.S. 114, 82 S.Ct. 1196, 8 L.Ed.2d 368 (1962), created further overlap between federal and state coverage for injured maritime workers. In Calbeck, we held that the LHWCA was "designed to ensure that a compensation remedy existed for all injuries sustained by employees [of statutory employers] on navigable waters, and to avoid uncertainty as to the source, state or federal, of that remedy." Id., at 124, 82 S.Ct., at 1202. Our examination in Calbeck of the "complete legislative history" of the 1927 LHWCA revealed that Congress did not intend to incorporate the "maritime but local" doctrine in the Act. Id., at 120, 82 S.Ct., at 1200. "Congress used the phrase 'if recovery . . . may not validly be provided by State law' in a sense consistent with the delineation of coverage as reaching injuries occurring on navigable waters." Id., at 126, 82 S.Ct., at 1203.19 14 Before 1972, there was little litigation concerning whether an employee was "in maritime employment" for purposes of being the employee of a statutory employer: "Workers who are not seamen but who nevertheless suffer injury on navigable waters are no doubt (or so the courts have been willing to assume) engaged in 'maritime employment.' " G. Gilmore & C. Black, The Law of Admiralty 428 (2d ed. 1975). One case in which we did discuss the maritime employment requirement was Parker v. Motor Boat Sales, Inc., 314 U.S. 244, 62 S.Ct. 221, 86 L.Ed. 184 (1941). In Parker, the injured worker, hired as a janitor, was drowned while riding in one of his employer's motor boats keeping lookout for hidden objects under the water. When the employee's beneficiary sought LHWCA compensation, the employer argued that the employment was " 'so local in character' " that the State could validly have provided a remedy, and the § 3(a) language ("if recovery . . . may not validly be provided by State law") precluded federal relief. Id., at 246, 62 S.Ct., at 223. A unanimous Court rejected the employer's argument, and held that the employee was engaged in maritime employment and that LHWCA coverage extended to an employee injured on the navigable waters in the course of his employment, without any further inquiry whether the injured worker's employment had a direct relation to navigation or commerce.20 In abolishing the "jurisdictional dilemma" created by the "maritime but local" doctrine, Calbeck relied heavily on Parker, see 370 U.S., at 127-128, 82 S.Ct., at 1203-1204. 15 It becomes clear from this discussion that the 1927 Act, as interpreted by Parker, Davis, and Calbeck, provided coverage to those employees of statutory "employers," injured while working upon navigable waters in the course of their employment. Indeed, the consistent interpretation given to LHWCA before 1972 by the Director, the deputy commissioners, the courts, and the commentators was that (except for those workers specifically excepted in the statute), any worker injured upon navigable waters in the course of employment was "covered . . . without any inquiry into what he was doing (or supposed to be doing) at the time of his injury." G. Gilmore & C. Black, supra, at 429-430.21 As a marine construction worker required to work upon navigable waters, and injured while performing his duties on navigable waters, there can be no doubt that Churchill would have been covered under the 1927 LHWCA. B 16 In its "first significant effort to reform the 1927 Act and the judicial gloss that had been attached to it," Congress amended the LHWCA in 1972. 44 Stat. 1424, as amended, 86 Stat. 1251, 33 U.S.C. § 901 et seq. Northeast Marine Terminal Co., supra, 432 U.S. at 261, 97 S.Ct., at 2356. The purposes of the 1972 amendments were to raise the amount of compensation available under the LHWCA, to extend coverage of the Act to include certain contiguous land areas, to eliminate the longshoremen's strict-liability seaworthiness remedy against shipowners, to eliminate shipowner's claims for indemnification from stevedores, and to promulgate certain administrative reforms. See S.Rep. No. 92-1125, at 1 (1972) (hereinafter S.Rep.); H.Rep. No. 92-1441 (1972) (hereinafter H.Rep.), U.S.Code Cong. & Admin.News 1972, p. 4698. 17 For purposes of the present inquiry, the important changes effected by the 1972 amendments concerned the definition of "employee" in § 2(3), 33 U.S.C. § 902(3), and the description of coverage in § 3(a), 33 U.S.C. § 903(a). These amended sections provide: 18 "The term 'employee' means any person engaged in maritime employment, including any longshoreman or other person engaged in longshoring operations, and any harborworker including a ship repairman, shipbuilder, and shipbreaker, but such term does not include a master or member of a crew of any vessel, or any person engaged by the master to load or unload or repair any small vessel under eighteen tons net." Section 2(3), 33 U.S.C. § 902(3). 19 "Compensation shall be payable under this chapter in respect of disability or death of an employee, but only if the disability or death results from an injury occurring upon the navigable waters of the United States (including any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other adjoining area customarily used by an employer in loading, unloading, repairing, or building a vessel). . . ." Section 3(a), 33 U.S.C. § 903(a).22 20 "The 1972 Amendments thus changed what had been essentially only a 'situs' test of eligibility for compensation to one looking to both the 'situs' of the injury and the 'status' of the injured." Northeast Marine Terminal Co., supra, 432 U.S. at 264-265, 97 S.Ct., at 2357-2358. In expanding the covered situs in § 3(a), Congress also removed the requirement, present in § 3(a) of the 1927 Act, that federal compensation would be available only if recovery "may not validly be provided by State law." The definition of "injury" remained the same,23 and the definition of "employer" was changed to reflect the new definition of "employee" in § 2(3).24 21 The Director and Churchill claim that when Congress added the status requirement in § 3(a), providing that a covered employee must be "engaged in maritime employment," it intended to restrict or define the scope of the increased coverage provided by the expanded situs provision in § 3(a), but that Congress had no intention to exclude from coverage workers, like Churchill, who were injured upon actual navigable waters, i.e., navigable waters as previously defined, in the course of their employment upon those waters. 22 According to Perini, Congress intended to overrule legislatively this Court's decision in Calbeck, and the status requirement was added to ensure that both the landward coverage and seaward coverage would depend on the nature of the employee's duties at the time he was injured. Perini's theory, adopted by the court below, is that all coverage under the amended LHWCA requires employment having a "significant relationship to navigation or commerce on navigable waters."25 Perini argues further that Churchill cannot meet the status test because he was injured while working on the construction of a foundation for a sewage treatment plant—an activity not typically associated with navigation or commerce on navigable waters. 23 We agree with the Director and Churchill. We are unable to find any congressional intent to withdraw coverage of the LHWCA from those workers injured on navigable waters in the course of their employment, and who would have been covered by the Act before 1972. As we have long held, "This Act must be liberally construed in conformance with its purpose, and in a way which avoids harsh and incongruous results." Voris v. Eikel, 346 U.S. 328, 333, 74 S.Ct. 88, 91, 98 L.Ed. 5 (1953). See also Baltimore & Philadelphia Steamboat Co. v. Norton, 284 U.S. 408, 414, 52 S.Ct. 187, 189, 76 L.Ed.2d 366 (1932); Northeast Marine Terminal Co., supra, 432 U.S. at 268, 97 S.Ct., at 2359. 24 It is necessary to consider the context in which the 1972 amendments were passed, especially as that context relates directly to the coverage changes that were effected. Despite the fact that Calbeck extended protection of the LHWCA to all employees injured upon navigable waters in the course of their employment, LHWCA coverage still stopped at the water's edge—a line of demarcation established by Jensen. In Nacirema Operating Co. v. Johnson, 396 U.S. 212, 90 S.Ct. 347, 24 L.Ed.2d 371 (1969), we held that the LHWCA did not extend to longshoremen whose injuries occurred on the pier attached to the land. We recognized that there was much to be said for the uniform treatment of longshoremen irrespective of whether they were performing their duties upon the navigable waters (in which case they would be covered under Calbeck ), or whether they were performing those same duties on a pier. We concluded, however, that although Congress could exercise its authority to cover land-based maritime activity, "[t]he invitation to move that [Jensen ] line landward must be addressed to Congress, not to this Court." Id., at 224, 90 S.Ct., at 354. See Victory Carriers, Inc. v. Law, 404 U.S. 202, 216, 92 S.Ct. 418, 427, 30 L.Ed.2d 383 (1971). 25 "Congress responded with the Longshoremen's and Harbor Workers' Compensation Act Amendments of 1972 (1972 Act)." P.C. Pfeiffer Co. v. Ford, 444 U.S. 69, 73, 100 S.Ct. 328, 332, 62 L.Ed.2d 225 (1979). The 1972 amendments were enacted after committees in both the House and Senate prepared full reports that summarized the general purposes of the legislation and contained an analysis of the changes proposed for each section. See S.Rep., supra; H.Rep., supra. These legislative reports indicate clearly that Congress intended to "extend coverage to protect additional workers." S.Rep., at 1 (emphasis added).26 Although the legislative history surrounding the addition of the status requirement is not as clear as that concerning the reasons for the extended situs, it is clear that "with the definition of 'navigable waters' expanded by the 1972 Amendments to include such a large geographical area, it became necessary to describe affirmatively the class of workers Congress desired to compensate." Northeast Marine Terminal Co., supra, 432 U.S., at 264, 97 S.Ct., at 2357. This necessity gave rise to the status requirement: "The Committee does not intend to cover employees who are not engaged in loading, unloading, repairing, or building a vessel, just because they are injured in an area adjoining navigable waters used for such activity." S.Rep., at 13; H.Rep., at 11, U.S.Code Cong. & Admin.News 1972, p. 4708. This comment indicates that Congress intended the status requirement to define the scope of the extended landward coverage.27 26 There is nothing in these comments, or anywhere else in the legislative reports, to suggest, as Perini claims, that Congress intended the status language to require that an employee injured upon the navigable waters in the course of his employment had to show that his employment possessed a direct (or substantial) relation to navigation or commerce in order to be covered. Congress was concerned with injuries on land, and assumed that injuries occurring on the actual navigable waters were covered, and would remain covered.28 In discussing the added status requirement, the Senate report states explicitly that the "maritime employment" requirement in section 3(a) was not meant "to exclude other employees traditionally covered." S.Rep., at 16. We may presume "that our elected representatives, like other citizens, know the law," Cannon v. University of Chicago, 441 U.S. 677, 696-697, 99 S.Ct. 1946, 1957-1958, 60 L.Ed.2d 560 (1979), and that their use of "employees traditionally covered" was intended to refer to those employees included in the scope of coverage under Parker, Davis, and Calbeck.29 27 Other aspects of the statutory scheme support our understanding of the "maritime employment" status requirement. Congress removed from § 3(a) the requirement that, as a prerequisite to federal coverage, there can be no valid recovery under state law.30 As we noted in our discussion in section A, supra, the continued use of the "maritime but local" doctrine occurred after passage of the 1927 Act because the original coverage section contained this requirement that Congress explicitly deleted in 1972. Surely, if Congress wished to repeal Calbeck and other cases legislatively, it would do so by clear language and not by removing from the statute the exact phrase that Calbeck found was responsible for continued emphasis on the "maritime but local" doctrine.31 28 Congressional intent to adhere to Calbeck is also indicated by the fact that the legislative reports clearly identified those decisions that Congress wished to overrule by the 1972 amendments. As mentioned above, the 1972 amendments had other purposes apart from an expansion of coverage to shoreside areas. Two other purposes involved the elimination of a strict liability unseaworthiness remedy against a vessel owner afforded to longshoremen by Seas Shipping Co. v. Sieracki, 328 U.S. 85, 66 S.Ct. 872, 90 L.Ed. 1099 (1946), and an indemnity claim against the stevedore by the vessel owner afforded by Ryan Stevedoring Co. v. Pan-Atlantic Steamship Corp., 350 U.S. 124, 76 S.Ct. 232, 100 L.Ed. 133 (1956). The legislative reports explicitly identified these decisions as intended to be overruled legislatively by the 1972 amendments. See S.Rep., at 8-12; H.Rep., at 4-8. It is, therefore, highly unlikely that Congress would have intended to return to the "jurisdictional monstrosity" that Calbeck sought to lay to rest without at least some indication of its intent to do so. 29 In considering the scope of the status test as applied to land-based employees in Northeast Marine Terminal Co., we rejected the "point of rest" theory proposed by the employer, under which landward coverage under the 1972 amendments would include only the portion of the unloading process that takes place before longshoremen place the cargo onto the dock. We reasoned that the "point of rest" concept is "[a] theory that nowhere appears in the Act, that was never mentioned by Congress during the legislative process, that does not comport with Congress' intent, and that restricts coverage of a remedial Act designed to extend coverage . . . ." The absence of the concept, "claimed to be so well known in the industry is both conspicuous and telling." 432 U.S., at 278-279, 275, 97 S.Ct., at 2365, 2363. In the same sense, the absence of even the slightest congressional allusion to the "maritime but local" doctrine, a concept that plagued maritime compensation law for over forty years and that would have the effect of restricting coverage in the face of congressional intent not to "exclude other employees traditionally covered," is equally conspicuous and telling. 30 Finally, we note that our conclusion concerning the continued coverage of employees injured on actual navigable waters in the course of their employment is consistent with, and supported by, our recent decision in Sun Ship, Inc. v. Pennsylvania, 447 U.S. 715, 100 S.Ct. 2432, 65 L.Ed.2d 458 (1980). In Sun Ship, the issue before the Court was whether extended shoreside coverage under the 1972 amendments had the effect of displacing concurrent state remedies for landward injuries. After a review of the development of the "maritime but local" doctrine, and review of certain portions of the legislative history of the 1972 amendments, we conclude that those amendments were not intended to resurrect the dilemma, created by mutually exclusive spheres of jurisdiction, that Calbeck and Davis eliminated. Our reasoning was based, in part, on the removal by Congress of the language in the 1927 Act that made federal compensation available if recovery could not validly be provided by state law: "[T]he deletion of that language in 1972—if it indicates anything—may logically only imply acquiescence in Calbeck [ ]. . . ." Id., at 721, 100 S.Ct., at 2436. 31 Sun Ship held that with respect to land-based injuries, "the extension of federal jurisdiction supplements, rather than supplants, state compensation law." Id., at 720, 100 S.Ct., at 2436. If we were to hold that the addition of the status requirement was meant to exclude from coverage some employees injured on the actual navigable waters in the course of their employment, a most peculiar result would follow. Concurrent jurisdiction will exist with respect to the class of employees to whom Congress extended protection in 1972, while employees "traditionally covered" before 1972 would be faced with a hazardous pre-Davis choice of two exclusive jurisdictions from which to seek compensation. Such an anomalous result could not have been intended by Congress. We also note that a return to exclusive spheres of jurisdiction for workers injured upon the actual navigable waters would be inconsistent with express congressional desire to extend LHWCA jurisdiction landward in light of the inadequacy of most state compensation systems. See S.Rep., at 12; H.Rep., at 10. 32 In holding that we can find no congressional intent to affect adversely the pre-1972 coverage afforded to workers injured upon the actual navigable waters in the course of their employment, we emphasize that we in no way hold that Congress meant for such employees to receive LHWCA coverage merely by meeting the situs test, and without any regard to the "maritime employment" language.32 We hold only that when a worker is injured on the actual navigable waters in the course of his employment on those waters, he satisfies the status requirement in § 2(3), and is covered under the LHWCA, providing, of course, that he is the employee of a statutory "employer," and is not excluded by any other provision of the Act.33 We consider these employees to be "engaged in maritime employment" not simply because they are injured in a historically maritime locale, but because they are required to perform their employment duties upon navigable waters.34 IV 33 In conclusion, we are unable to find anything in the legislative history or in the 1972 amendments themselves that indicate that Congress intended to withdraw coverage from employees injured on the navigable waters in the course of their employment as that coverage existed before the 1972 amendments. On the contrary, the legislative history indicates that Congress did not intend to "exclude other employees traditionally covered." Moreover, Congress explicitly deleted the language from § 3(a) that we found in Calbeck to be responsible for the "jurisdictional dilemma" caused by two mutually exclusive spheres of jurisdiction over maritime injuries. Accordingly, the decision of the Court of Appeals is hereby reversed, and the case is remanded to the Court of Appeals for further proceedings consistent with this opinion. 34 It is so ordered. 35 Justice REHNQUIST, concurring in the judgment. 36 At the time of his injury, Churchill was engaged in unloading materials from a supply barge to a cargo barge. This work is very much like the work of longshoremen, who typically load and unload vessels. Therefore Churchill was "engaged in maritime employment" within the meaning of § 2(3) the Act, and was within its coverage. Accordingly, I concur in the judgment of the Court. 37 Justice STEVENS, dissenting. 38 Neither the legislative history nor the judicial history on which the Court relies today justifies a departure from the language of the statute defining the post-1972 coverage of the Longshoremen and Harbor Workers Compensation Act (LHWCA). Indeed, when the issue is viewed in its proper historical perspective, it becomes even more clear that a literal reading of the Act will avoid anomalies that troubled Congress in 1972 as well as unnecessary litigation and duplicate insurance coverage in the post-1972 period. I shall first comment on the statutory language and then discuss its history. 39 * The principal focus of the statute is identified by its title as well as its text. It provides workers' compensation benefits for injuries to longshoremen and harbor workers.1 The coverage of the statute is defined by two basic tests—a situs test focusing on the place where the injury occurred, and a status test focusing on the character of the injured employee's occupation. An injured person is entitled to compensation under the Act only if he satisfied both tests at the time of the injury. The two tests work together to provide comprehensive coverage for a large class of workers who perform hazardous longshore and ship repair work. 40 The requisite occupational status is defined in § 2(3) of the Act. It provides: "The term 'employee' means any person engaged in 41 maritime employment, including any longshoreman or other person engaged in longshoring operations, and any harborworker including a ship repairman, shipbuilder, and shipbreaker, but such term does not include a master or member of a crew of any vessel, or any person engaged by the master to load or unload or repair any small vessel under eighteen tons net." 33 U.S.C. § 902(3). 42 The term "maritime employment" expressly includes two important subcategories, both of which are defined with reasonable clarity. The question of construction that is presented is what, if any, additional categories of employment are included within the term "maritime employment." There are several independent reasons for not giving the term an expansive, essentially open-ended reading. 43 First, one of the oldest and most respected rules of statutory construction teaches us that general terms should be construed in the light of the specific examples that are expressly identified as included therein. In this statute, the subcategories—longshoremen and harbor workers—are both described in detail, and no other subcategory is even mentioned, giving rise to an especially strong inference that Congress intended a snug fit between "maritime employment" and the two subcategories.2 44 This inference is corroborated by the fact that Congress took the trouble to add language making it clear that the statutory concept of "maritime employment" was not intended to describe either the master or a member of the crew of any vessel.3 In short, the ordinary meaning of the words "maritime employment" is actually excluded from the description of the occupational categories that Congress intended the LHWCA to cover. 45 It is also clear that the definition of "employee" is entirely unaffected by where he may be injured; if a worker is not an "employee" when ashore, he is not an "employee" when afloat. Therefore, it is critically significant that the definition of where "employees" are covered—the situs provision reveals the same limited concern for the same key occupations as the status provision. An "employee" is covered only while on navigable waters and on "any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other adjoining area customarily used by an employer in loading, unloading, repairing, or building a vessel." 33 U.S.C. § 903(a) (emphasis added). 46 If we ignore history, and merely concentrate on the text of this statute, the conclusion is inescapable that it merely provides coverage for people who do the work of longshoremen and harbor workers—amphibious persons who are directly involved in moving freight onto and off of ships, or in building, repairing, or destroying ships. A "checker" is such a worker.4 So are "terminal laborers," Northeast Marine Terminal Co. v. Caputo, 432 U.S. 249, 97 S.Ct. 2348, 53 L.Ed.2d 320 (1977), "cotton headers," P.C. Pfeiffer Co. v. Ford, 444 U.S. 69, 100 S.Ct. 328, 62 L.Ed.2d 225 (1979), and "warehousemen," ibid. A construction worker on a sewage treatment plant plainly lacks this direct link to maritime commerce, regardless of where he may have been working at the time of his injury. II 47 If we examine the legislative history of the 1972 Amendments5—without regard to the text of the statute or judicial decisions that are unmentioned in that history—we must reach the same conclusion. I cannot find a single word6 in the Committee Hearings, the Committee Reports, or the Legislative Debates that even suggests that any Congressman or Senator believed that the statute provided coverage for anyone other than longshoremen, harbor workers, and persons in the entirely separate categories that had been included by special statutory enactment.7 48 At the opening of the House Subcommittee Hearings, Congressman Daniels explained his understanding of the existing scope of the LHWCA8 and the need for amendments: 49 "This Act provides workmen's compensation protection to longshoremen, ship repairmen, workers at U.S. defense bases outside the United States and workers employed in private industry in the District of Columbia. 50 "Amendments to the Longshoremen's and Harbor Workers' Compensation Act are long overdue. Benefits under this act have not been increased for 12 years, and the cost to the injured workers of inadequate benefits has become a serious matter. 51 "For example, the law now allows a totally disabled worker to receive two-thirds of his average weekly wages at the time of his injury. However, since 1961 there has been a limitation of $70 per week as the maximum payment for a permanent disability. This statutory maximum results in a substantially lower payment than two-thirds of the weekly wage for most longshoremen and District of Columbia workers covered by this statute. 52 "More than 270,000 longshoremen and ship repairmen are covered by this statute. In addition, another 300,000 employees of private employers within the District of Columbia are protected by this law as well as an additional 200,000 workers in defense bases and work on Outer Continental Shelf projects. "Last year there were more than 109,000 injuries 53 under this statute; 240 of them fatal. 68,000 of them related to longshore work, and another 27,000 involved District of Columbia workers. 54 "The benefits under this act have not been increased in 12 years. The cost to the injured workers of inadequate benefits has become a serious matter. . . ." Hearings on H.R. 247 et al. before the Select Subcommittee on Labor of the House Committee on Education and Labor, 92nd Cong., 2d Sess., 46 (1972) (hereinafter "House Hearings"). 55 Throughout the hearings, the legislators were told over and over again how important it was to increase the Act's benefits for workers in the categories identified by Congressman Daniels.9 It seems plain that these were the categories of employment that were understood by Congress to define the traditional coverage of the Act. 56 When the House and Senate Committees reported out their respective bills, they had granted the sought-after increase in benefits. They had also amended the provisions defining the scope of coverage, including the language of "status" and "situs" discussed in the previous section. They had done so in response to a problem in the scope of prior coverage. Before 1972, longshoremen's and harbor workers' federal coverage had stopped at the water's edge. Because their duties regularly took them off the vessel and onto the pier, they were constantly "walking in and out of coverage." On the House side, Joseph Leonard, the international safety director of the International Longshoremen's Association, spoke about the hardship this system imposed: 57 "Federal compensation law stops at the gangplank to the pier. When you come off of the gangplank you come under a different law; you come under the State. Thirty-six States cover these docks and maybe more now with the inland waterways. 58 "The longshoremen are the only workers in the United States who must worry about their injury to determine the compensation. . . . It is time for a Federal law for compensation for all longshoremen." House Hearings, at 297. 59 And on the Senate side, the Minority Counsel brought this problem to the Senators' attention.10 60 The Committee amendments responded to this problem by defining the protected situs to encompass the entire area in which members of the protected class customarily perform their regular duties. This definition of situs clearly precludes coverage for a construction worker standing on a sewage treatment plant or a bridge. Yet if one accepts the view of the claimant in this case, the statute grants him coverage while aboard a floating vessel and therefore expects him to walk in and out of coverage during a typical workday. Such a view is flatly inconsistent with the explicit intent of Congress to "permit a uniform compensation system to apply to employees who would otherwise be covered by this Act for [only] part of their activity." H.R.Rep. No. 92-1441, pp. 10-11, U.S.Code Cong. & Admin.News 1972, p. 4708. S.Rep. No. 92-1125, p. 13.11 Only if we adhere to the language used by Congress to define the relevant status harmoniously with the relevant situs can the congressional purpose be achieved. III 61 The pre-1972 judicial history of the LHWCA confirms my construction of the 1972 Amendment and also explains why the work of longshoremen and harbor workers is described as "maritime employment" in the statute. Only once during the 45-year interval between the enactment of the LHWCA in 1927 and its amendment in 1972, in Parker v. Motor Boat Sales, Inc., 314 U.S. 244, 62 S.Ct. 221, 86 L.Ed. 184 (1941), did this Court uphold an award of benefits under the LHWCA for a worker who was neither a longshoreman nor a harbor worker.12 That lonely decision rested on a concern that is no longer significant, and surely provides an insufficient predicate for the Court's all-inclusive interpretation of "maritime employment." Before commenting specifically on the Parker case, however, I shall briefly identify the two principal chapters in the pre-1972 history of the LHWCA. 62 The first chapter (which covers the period from 1917 to 1927) explains why there was a need for federal legislation to provide compensation for injured longshoremen and harbor workers. Prior to 1917, it was assumed that these workers were adequately protected by whatever state legislation existed. In that year, however, this Court held that the national interest in the uniform regulation of maritime commerce precluded state jurisdiction over injuries occurring on navigable waters. Southern Pacific Rail Co. v. Jensen, 244 U.S. 205, 37 S.Ct. 524, 61 L.Ed. 1086 (1917).13 63 Over the classic dissents of two of our greatest justices, the Court adhered to that view even though Congress twice attempted to authorize the exercise of state jurisdiction over these "maritime" injuries.14 The so-called "Jensen line" thus developed as a constitutional limit on the exercise of state power over maritime employment. 64 The reasoning of the Jensen case originally appeared to foreclose the application of state workmen's compensation schemes to any injury occurring on navigable waters. The Court soon made it clear, however, that there was a somewhat vaguely defined area an area that became known as the "maritime but local" area—in which state jurisdiction survived. Thus, in 1922, five years before the enactment of the LHWCA, the Court held that a carpenter injured at work aboard an uncompleted ship that had been launched in the Willamette River could recover under the Oregon Workmen's Compensation law. Grant Smith-Porter Co. v. Rohde, 257 U.S. 469, 42 S.Ct. 157, 66 L.Ed. 321 (1922). The national interest in uniformity that had been considered paramount in Jensen was not thought to be materially prejudiced by Oregon's regulation of "certain local matters."15 65 Unlike the work of the carpenter in Rohde, the work of the longshoreman was considered by the Court to have a character that required regulation by a uniform federal scheme. That much was made clear by the Court's opinion in Northern Coal and Dock Co. v. Strand, 278 U.S. 142, 49 S.Ct. 88, 73 L.Ed. 232 (1928),16 a case involving a fatal shipboard injury to a longshoreman. 66 "The unloading of a ship is not a matter of purely local concern. It has direct relation to commerce and navigation, and uniform rules in respect thereto are essential. The fact that Strand worked for the major portion of the time upon land is unimportant. He was upon the water in pursuit of his maritime duties when the accident occurred." Id., at 144, 49 S.Ct., at 89. 67 The LHWCA was enacted in 1927 to remedy this inability of the States to provide adequate protection for longshoremen injured on navigable waters. The fact that these workers had been characterized as "maritime" in the cases that had denied them adequate state protection explains why Congress later used the same term in the LHWCA. 68 The second chapter (which covers the period from 1927 to 1972) explains why it was necessary for Congress to limit the coverage of the LHWCA to a defined category of employees. As originally enacted in 1927, the LHWCA was merely intended to fill the gap in state coverage that had been created by Jensen and its progeny.17 The provision that defined the scope of coverage, § 902(3) (which remained unchanged until 1972), purported to exclude federal coverage if recovery may "validly be provided by state law." At first, the statutory language was taken literally, and state and federal coverage were thought to be purely complementary and mutually exclusive. See Crowell v. Benson, 285 U.S. 22, 41 and, n. 3, 52 S.Ct. 285, 288 and, n. 3, 76 L.Ed. 598 (1932). But given the imprecision of the Jensen-Rohde line, that system risked serious unfairness: If an injured employee asked for state benefits and was seaward of the line, a literalist interpretation of the LHWCA would bar recovery. An employee close to the line might easily misguess, miss the statute of limitations, and end up with no benefits at all. 69 This Court responded to this potential for injustice in two ways. Notwithstanding the plain language of the statute which purported to describe mutually exclusive spheres of state and federal jurisdiction,18 the Court first upheld a state award in a case in which it was assumed that the federal statute would also apply, Davis v. Department of Labor, 317 U.S. 249, 63 S.Ct. 225, 87 L.Ed. 246 (1942), and then upheld a federal award in a case in which the Court assumed that recovery could "validly be provided by state law." Calbeck v. Travelers Insurance Co., 370 U.S. 114, 82 S.Ct. 1196, 8 L.Ed.2d 368 (1962).19 But the Court's mechanism for ensuring that no employee would go entirely unprotected created a twilight zone of overlapping jurisdiction in which many employers were required to obtain duplicate insurance coverage.20 Moreover, the practice of defining coverage entirely by reference to the place where an accident occurred gave rise to the anomalous circumstance that longshoremen regularly walked in and out of coverage during the performance of their routine duties. 70 Whatever force the Jensen rule may once have had, it is now perfectly clear that a shore-based worker who is normally covered by a state compensation program may still recover state benefits even though he is injured over navigable waters. Surely no member of this Court would question the fact that the construction worker injured in this case could have received a state award even though he was on a barge in the Hudson River when he was injured. The concern about the inability of the States to protect land-based workers who may temporarily cross the Jensen line is no longer significant—surely that concern provided no motivation whatsoever for the action Congress took in 1972 when it amended the LHWCA. 71 On the other hand, the 1972 Congress clearly did have reason to be concerned about the cost of duplicate insurance coverage and the unpredictability of coverage that depends entirely on the happenstance of where an accident occurs. As I have mentioned above,21 the unpredictability of coverage was mentioned explicitly in the legislative history. And the burdens of duplicate insurance for employees who might occasionally walk into federal coverage became substantially more onerous as a result of the 1972 changes that made federal LHWCA benefits significantly higher than state workers' compensation benefits.22 Both of these concerns are alleviated by defining the scope of the statutory coverage in terms of the status of the covered employee. And both of these concerns can only be aggravated by indiscriminately extending coverage to an undefined group of workers who plainly do not "load, unload, build, or repair ships." 72 All that remains to support the Court's rewriting of the statute is the absence of an expressed intent to withdraw pre-1972 coverage. As I have already noted, that intent is adequately demonstrated by the changes in the text of the statute itself.23 Even if that were not sufficient, however, the Court is really objecting to nothing more than a failure to mention a single case decided in 1942—Parker v. Motor Boat Sales, Inc.—during the hearings or the debates. But when one considers the highly unusual facts of that case, it is unlikely that any member of Congress had it in mind and virtually inconceivable that Congress would have wanted to provide federal coverage for similar future cases. The employee in the Parker case—a janitor for a small boat concern located on the James River—was not protected by a state workmen's compensation program for a reason that had nothing to do with the character of his employment or the place of his injury. The employer did not have the minimum number of employees to bring it under the Virginia statute. See 116 F.2d 789, 793. The happenstance that the janitor was riding in a motorboat at the time of his injury enabled the Court to find a basis for sustaining an award under the LHWCA as it was then written.24 Even if the presumption that Congress understands the legal context in which it legislates justifies the inference that it remembered this isolated case decided three decades earlier, it by no means follows that Congress had a duty to disavow the case explicitly in order to give effect to its otherwise plainly expressed purpose.25 73 This case presents us with a straightforward problem of statutory construction. The Court should begin its analysis with the language of the statute itself. "Absent a clearly expressed legislative intention to the contrary, that language must ordinarily be regarded as conclusive." Consumer Product Safety Commission v. GTE Sylvania, 447 U.S. 102, 108, 100 S.Ct. 2051, 2056, 64 L.Ed.2d 766. In this case the statutory language plainly encompasses longshoremen and harbor workers; there is no affirmative evidence of a legislative intent to provide coverage for any other type of occupation. Surely there is no evidence of an intent to classify the work of a janitor or a builder of sewage treatment plants as "maritime employment."26 Because the claimant in this case was neither a longshoreman nor a harbor worker, I would affirm the judgment of the United States Court of Appeals for the Second Circuit 1 Section 2(3) of the Act, 33 U.S.C. § 902(3), provides: "(3) The term 'employee' means any person engaged in maritime employment, including any longshoreman or other person engaged in longshoring operations, and any harborworker including a ship repairman, shipbuilder, and shipbreaker, but such term does not include a master or member of a crew of any vessel, or any person engaged by the master to load or unload or repair any small vessel under eighteen tons net." 2 We use the expression "actual navigable waters" to describe the covered situs as it existed in the 1927 LHWCA, 44 Stat. 1424: "navigable waters of the United States (including any dry dock)." 44 Stat. 1426. As explained below, the 1972 amendments to the LHWCA expanded the concept of "navigable waters" to include certain adjoining shoreside areas. Section 3(a), 33 U.S.C. § 903(a). 3 In Northeast Marine Terminal Co. v. Caputo, 432 U.S. 249, 97 S.Ct. 2348, 53 L.Ed.2d 320 (1977), we examined the scope of the section 2(3) status requirement as it applied to injuries that occurred on the newly-covered landward situs. In that case, we expressly declined to speculate whether Congressional addition of the status requirement meant that "Congress excluded people who would have been covered before the 1972 Amendments; that is, workers who are injured on the navigable waters as previously defined." Id., at 265 n. 25, 97 S.Ct., at 2358 n. 25. 4 At the time Churchill was injured, he was working on a barge in actual navigable waters. There is no claim that he was standing on the foundation of the sewage treatment plant. 5 The Board also determined that Churchill's duties did not make him a "person engaged in longshoring operations" under § 2(3) of the LHWCA. 6 The dissenting Board member also relied on this Court's decision in Sun Ship, Inc. v. Pennsylvania, 447 U.S. 715, 100 S.Ct. 2432, 65 L.Ed.2d 458 (1980), to support his position. 7 33 U.S.C. 921(c) provides in pertinent part: "(c) Any person adversely affected or aggrieved by a final order of the Board may obtain a review of that order in the United States court of appeals for the circuit in which the injury occurred, by filing in such court within sixty days following the issuance of such Board order a written petition praying that the order be modified or set aside. . . ." 8 The Ninth Circuit is in agreement with the Second Circuit position. See Weyerhaeuser Co. v. Gilmore, 528 F.2d 957 (CA9 1975), cert. denied, 429 U.S. 868, 97 S.Ct. 179, 50 L.Ed.2d 148 (1976). The Fifth Circuit takes a position contrary to that of the Second Circuit and Ninth Circuit. See Boudreaux v. American Workover, Inc., 680 F.2d 1034 (CA5 1982) (en banc) (Tate, J.). 9 Perini bases its standing argument on § 21(c) of the Act, 33 U.S.C. § 921(c). See note 7, supra. According to Perini, the Director is not "adversely affected or aggrieved" by the decision below, and does not have standing before this Court. Perini relies on several circuit court decisions which, in construing § 21(c), have held the Director to be without statutory standing in cases before the Courts of Appeals. See Fusco v. Perini North River Associates, 601 F.2d 659 (CA2 1979), vacated and remanded, 444 U.S. 1028, 100 S.Ct. 697, 62 L.Ed.2d 664 (1980), rev'd on remand, 622 F.2d 1111 (CA2 1980), cert. denied, 449 U.S. 1131, 101 S.Ct. 953, 67 L.Ed.2d 119 (1981); Director, OWCP v. Donzi Marine, Inc., 586 F.2d 377 (CA5 1978); and I.T.O. Corp. of Baltimore v. Benefits Review Board, 542 F.2d 903 (CA4 1976), vacated sub nom., Adkins v. I.T.O. Corp. of Baltimore, 433 U.S. 904, 97 S.Ct. 2967, 53 L.Ed.2d 1088 (1977), rev'd on remand on other grounds, 563 F.2d 646 (1977). Section 21(c) is not relevant to our present inquiry. Perini concedes that § 21(c) applies on its face to statutory review before the Courts of Appeals. Moreover, the cases on which Perini relies do not purport to address the Article III standing issue. 10 20 CFR § 802.410(b) (1981). Section 39 of the Act provides that "the Secretary [of Labor] shall administer the provisions of this chapter, and for such purpose the Secretary is authorized (1) to make such rules and regulations . . . as may be necessary in the administration of this chapter." 33 U.S.C. § 939. The Secretary has assigned enforcement and administration responsibilities to the Director. 11 We acknowledge that on three occasions, this Court has granted petitions for certiorari to review cases brought by the Director. See Director, OWCP v. Walter Tantzen, Inc., 446 U.S. 905, 100 S.Ct. 1829, 64 L.Ed.2d 257 (1980), vacating and remanding, Walter Tantzen, Inc. v. Shaughnessy, 601 F.2d 670 (CA2 1979), rev'd on remand, 624 F.2d 5 (1980); Director, OWCP v. Rasmussen, 436 U.S. 955, 98 S.Ct. 3068, 57 L.Ed.2d 1120 (1978); and Director, OWCP v. Jacksonville Shipyards, Inc., 433 U.S. 904, 97 S.Ct. 2967, 53 L.Ed.2d 1088 (1977), vacating and remanding, Jacksonville Shipyards, Inc. v. Perdue, 539 F.2d 533 (CA5 1976), aff'd on remand, 575 F.2d 79 (1978), cert. denied, 440 U.S. 967, 99 S.Ct. 1520, 59 L.Ed.2d 784 (1979). Tantzen and Jacksonville Shipyards were both summary dispositions, and Rasmussen was decided on the merits, see Director, OWCP v. Rasmussen, 440 U.S. 29, 99 S.Ct. 903, 59 L.Ed.2d 122 (1979). In none of these cases did we have occasion to consider whether the Director had standing in his own right to seek review of a decision of the Benefits Review Board with which the Director disagreed. In Rasmussen, the employer and the insurer also petitioned for certiorari, and the cases were consolidated. It was not necessary to consider the issue of the Director's standing in that case because a justiciable controversy was before the Court by virtue of the petition of the employer and insurer. In both Tantzen and Jacksonville, the Director had defended a Board decision in the Courts of Appeals as the federal respondent, and continued to defend the Board decision before this Court. See note 13, infra. 12 Supreme Court Rule 19.6 provides in part: "All parties other than petitioners shall be respondents, but any respondent who supports the position of a petitioner shall meet the time schedule for filing papers which is provided for that petitioner . . . ." Under Rule 19.6, Churchill is a party in this Court by virtue of his being a party in the proceedings below. Moreover, he has demonstrated his continued stake in the outcome of this case by filing in support of the Director at both the certiorari and merits stages of the proceedings. 13 The fact that Perini concedes that the Director was a proper party respondent before the Court of Appeals in this case means that no question is thereby presented concerning whether the Director, as party respondent below, is a "party" for purposes of 28 U.S.C. § 1254(1), which states that a writ of certiorari may be "granted upon the petition of any party" below. Given that the parties do not question the identity of the federal respondent, it is not necessary for us to decide the issue explicitly left open by the Court in Northeast Marine Terminal Co. v. Caputo, 432 U.S. 249, 256 n. 11, 97 S.Ct. 2348, 2353, n. 11, 53 L.Ed.2d 320 (1977), as to whether the Director is a proper party respondent in the Courts of Appeals. Although we declined to address this issue because the parties did not raise it in Northeast Marine Terminal, we noted that "[t]he Department of Labor has recently promulgated a regulation making it clear that the Director of OWCP is the proper federal party in a case of this nature. 42 Fed.Reg. 16133 (Mar. 1977)." Ibid. 14 Article III, Section 2 extends the federal power "to all Cases of admiralty and maritime Jurisdiction." In Jensen, we held that state compensation acts could not cover longshoremen injured seaward of the water's edge. The line of demarcation between land and water became known as the "Jensen line." 15 See Knickerbocker Ice Co. v. Stewart, 253 U.S. 149, 40 S.Ct. 438, 64 L.Ed. 834 (1920); Washington v. W.C. Dawson & Co., 264 U.S. 219, 44 S.Ct. 302, 68 L.Ed. 646 (1924). 16 Section 3(a), 44 Stat. 1426, also excluded from coverage "[a]n officer or employee of the United States or any agency thereof or of any State or foreign government, or of any political subdivision thereof." 17 The 1927 Act did not contain any provision that an injured employee must be "engaged in maritime employment" at the time of injury in order to be covered. Rather, the Act employed the expression "maritime employment" only as part of the definition of a statutory "employer." 18 In Davis, our concern for the employer's dilemma was related to the fact that because the employer did not know with any certainty whether his employee would be covered under the LHWCA, "[t]he employer's contribution to a state insurance fund may wholly fail to protect him against the liabilities for which it was specifically planned." 317 U.S., at 255, 63 S.Ct., at 228. We resolved that dilemma in Calbeck by making it clear to employers that if they required their employees to work upon actual navigable waters, those employees would be covered by the LHWCA. The dissent takes this certainty in favor of LHWCA coverage to mean that in 1972, Congress wanted to ensure that employers like Perini would have only to pay for state compensation benefits, and would not have to obtain more costly LHWCA protection. The dissent's concern about duplicative insurance seems exaggerated for two reasons. First, even under the dissent's view of coverage, both state and federal remedies are available to injured workers, and employers with employees working on the shore would have to contribute to state compensation funds in the event that an employee covered by LHWCA's shoreside extension sought state compensation, or an employee was deemed for whatever reason not to be eligible for LHWCA relief. "[T]he 1972 extension of federal jurisdiction supplements, rather than supplants, state compensation law." Sun Ship, Inc. v. Pennsylvania, 447 U.S. 715, 720, 100 S.Ct. 2432, 2436, 65 L.Ed.2d 458 (1980). We also note that the dissent argues that before 1972, the financial burden of duplicative coverage was not heavy because LHWCA benefits were lower than they now are, and insurance carriers would cover LHWCA operations for a nominal addition to state compensation program premiums. There is nothing in the record in this case, in the legislative history, or, for that matter, in the dissent, concerning whether the relative spread between state and federal insurance premiums is higher now than before 1972. Second, the dissent's view clearly does not result in any certainty whatsoever for employers like Perini with respect to whether those employers have to pay for LHWCA coverage. If any Perini employee (including Churchill) were to engage in loading, unloading, or repairing of the barge on which Churchill was working, the employee would be covered. Indeed, if Churchill himself had to make some minor mechanical adjustment on the barge and was injured while doing so, he would be covered under the dissent's view. 19 We noted in Sun Ship, Inc. v. Pennsylvania, 447 U.S. 715, 100 S.Ct. 2432, 65 L.Ed.2d 458 (1980), that in extending LHWCA coverage into the "maritime but local" zone, Calbeck did not overturn Davis "by treating the federal statute as exclusive." Id., at 718-719, 100 S.Ct., at 2435-2436. Rather, Calbeck eliminated the "jurisdictional dilemma" that resulted from the existence of two spheres of exclusive jurisdiction, by making injuries within the "maritime but local" sphere compensable under either state or federal law. 20 The majority opinion in Davis assumed that if the claimant in that case sought federal relief, and such relief was awarded at the administrative stage of the proceedings, the Court would have sustained the award under Parker. In his dissent in Davis, Chief Justice Stone argued that the federal act applied to give exclusive relief in that case: "after our decision in Parker v. Motor Boat Sales, . . . [the Davis claimant's] right of recovery under the federal act can hardly be doubted." 317 U.S., at 260, 63 S.Ct., at 231. Professor Robertson has noted that "Parker should have meant the abolition of the 'maritime but local' exception," but that Davis indicated that the doctrine had continued vitality. D. Robertson, Admiralty and Federalism 210 (1970). Professor Robertson also states that if the claimant in Davis had sought federal, rather than state, compensation, "the Parker case would certainly have said that [the claimant] could get it." Id., at 211. 21 The dissent attempts to carve a new "maritime but local" area in which the exclusive remedy is state compensation. The dissent argues that Congress meant to exclude from LHWCA coverage all employees who are not longshoremen or harbor workers, and that only longshoremen and harbor workers possess the "direct link to maritime commerce" necessary for LHWCA coverage. According to the dissent, the pre-1972 case-law, with the exception of Parker, supports its position. The dissent's view rests on a misreading of our decisions in Davis and Calbeck, and a failure to consider the impact of Parker, Davis, and Calbeck on the scope of pre-1972 coverage. The dissent points out that Davis involved an employee who sought state compensation, and it concludes that Davis says nothing about LHWCA coverage. The employee in Davis was standing on a barge and assisting in the dismantling of a bridge, an activity that would clearly not have the "direct link to maritime commerce" that the dissent suggests is required. Although the Davis employee sought state compensation, both the Davis majority and the Davis dissent assumed that if the Davis employee sought LHWCA coverage, Parker would require that he get it. In Calbeck, the claimants were welders performing work on vessels, but our holding in Calbeck was clearly predicated on Parker and Davis, and cannot properly be characterized as a case where LHWCA coverage was predicated on the existence of some "direct link to maritime commerce" or "traditional" LHWCA employment. The dissent claims that since Churchill could be covered by a state compensation remedy, it is consistent with Calbeck to deny LHWCA coverage. This, of course, neglects the fact that Calbeck made clear that "Congress brought under the coverage of the Act all such injuries [suffered by employees working on the navigable waters] whether or not a particular one was also within the constitutional reach of a state workmen's compensation law." 370 U.S., at 127, 82 S.Ct., at 1203. Parker, Davis, and Calbeck were read by the lower federal and state courts not to limit LHWCA coverage only to "traditional" maritime activities, but to cover injuries that occurred on the navigable waters in the course of employment. See, e.g., Nalco Chemical Corporation v. Shea, 419 F.2d 572 (CA5 1969) (a pilot salesman travelling to offshore platform); Interlake Steamship Co. v. Nielsen, 338 F.2d 879 (CA6 1964), cert. denied, 381 U.S. 934, 85 S.Ct. 1765, 14 L.Ed.2d 699 (1965) (watchman); Radcliff Gravel Co. v. Henderson, 138 F.2d 549 (CA5 1943), cert. denied, 321 U.S. 782, 64 S.Ct. 638, 88 L.Ed. 1074 (1944) (workers who trimmed sand and gravel loaded on barges after being dredged from water bed); Rex Investigative and Patrol Agency, Inc. v. Collura, 329 F.Supp. 696 (EDNY 1971) (land-based employee sent temporarily onto vessel to act as watchman); Standard Dredging Corp. v. Henderson, 57 F.Supp. 770 (Ala.1944) (employee engaged in dredging bed of intracoastal canal); Ford v. Parker, 52 F.Supp. 98 (Md.1943) (night watchman); Perry v. Baltimore Contractors, Inc., 202 So.2d 694 (La.Ct.App.1967), cert. denied, 390 U.S. 1028, 88 S.Ct. 1419, 20 L.Ed. 280 (1968) (worker injured while diving in order to assist in construction of a tunnel under intracoastal canal). This list is by no means exhaustive, and does not include various administrative decisions. In another case, Pennsylvania Railroad Co. v. O'Rourke, 344 U.S. 334, 73 S.Ct. 302, 97 L.Ed. 367 (1953), we held that a statutory "employer" existed as long as the employer had any employee engaged in "maritime employment," and that it was not necessary that the injured employee be the one employee that made his employer a statutory "employer." However, we also held in that case that the injured employee was, in fact, engaged in maritime employment when he was working as a railway brakeman, removing railroad cars from a car float by the use of an ordinary switch engine. Id., at 340, 73 S.Ct., at 305. Although Pennsylvania Railroad Co. involved a question as to which of two federal statutes applied to cover the employee's injury (the LHWCA or the Federal Employers' Liability Act), and did not involve an application of the "maritime but local" doctrine, the Deputy Commissioners had interpreted Pennsylvania Railroad Co. to mean "that injury over the water means, without much more inquiry, that they ought to grant [LHWCA] awards." Robertson, n. 19, supra, at 220. In the two cases that came to us in Calbeck, the Deputy Commissioners had granted LHWCA awards on the basis of Pennsylvania Railroad Co. See Robertson, n. 20, supra, at 219-220. 22 We note that the new coverage section still provides that no compensation shall be paid to "[a]n officer or employee of the United States or any agency thereof or of any State or foreign government, or of any political subdivision thereof." Section 3(a), 33 U.S.C. § 903(a). 23 "Injury" is defined in § 2(2), 33 U.S.C. § 902(2), as "accidental injury or death arising out of and in the course of employment, and such occupational disease or infection as arises naturally out of such employment or as naturally or unavoidably results from such accidental injury, and includes an injury caused by the willful act of a third person directed against an employee because of his employment." 24 "Employer" is defined in § 2(4), 33 U.S.C. § 902(4) as "an employer any of whose employees are employed in maritime employment, in whole or in part, upon the navigable waters of the United States (including any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other adjoining area customarily used by an employer in loading, unloading, repairing, or building a vessel)." The reports also add: "[T]he Committee has no intention of extending coverage under the Act to individuals who are not employed by a person who is an employer, i.e., a person at least some of whose employees are engaged, in whole or in part in some form of maritime employment. Thus, an individual employed by a person none of whose employees work, in whole or in part, on the navigable waters, is not covered even if injured on a pier adjoining navigable waters." S.Rep., at 13; H.Rep., at 11, U.S.Code Cong. & Admin.News 1972, p. 4708. We note that there is an apparent inconsistency between the actual wording of section 2(4) and the expression in the legislative history. Section 2(4) defines an "employer" to be the employer of any employee engaged in maritime employment on the "navigable waters" as defined by the 1972 Amendments to include the expanded landward situs. The legislative history, however, appears to contemplate that a statutory employer must have at least one employee working over the actual navigable waters before any employee injured on the new land situs can be covered. 25 We see no real distinction between the "direct relationship" test used to articulate the "maritime but local" doctrine, and the "significant relationship" test urged by Perini. In support of the use of this test, Perini relies on the "maritime but local" cases. 26 The reasons for the extended landward coverage in report sections labeled "Extension of Coverage to Shoreside Areas": "The present [1927] Act, insofar as longshoremen and ship builders and repairmen are concerned, covers only injuries which occur 'upon the navigable waters of the United States.' Thus coverage of the present Act stops at the water's edge; injuries occurring on land are covered by State Workmen's Compensation laws. The result is a disparity in benefits payable for death or disability for the same type of injury depending on which side of the water's edge and in which State the accident occurs." "To make matters worse, most State Workmen's Compensation laws provide benefits which are inadequate. . . ." * * * * * "The Committee believes that the compensation payable to a longshoreman or a ship repairman or builder should not depend on the fortuitous circumstance of whether the injury occurred on land or over water. Accordingly, the bill would amend the Act to provide coverage of longshoremen, harbor workers, ship repairmen, ship builders, shipbreakers, and other employees engaged in maritime employment (excluding masters and members of the crew of a vessel) if the injury occurred either upon the navigable waters of the United States or any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other area adjoining such navigable waters customarily used by an employer in loading, unloading, repairing, or building a vessel." "The intent of the Committee is to permit a uniform compensation system to apply to employees who would otherwise be covered by this Act for part of their activity. . . ." S.Rep., at 12-13; H.Rep., at 10-11, U.S.Code Cong. & Admin.News 1972, p. 4708. 27 Perini argues that Congress' intent to eliminate the problem associated with movement from covered to noncovered areas will be frustrated by our holding because some employees may be deemed to satisfy the status test while working upon the navigable waters, but be deemed not to satisfy the status test when performing the same activity on land. We have had two opportunities to examine the scope of landward coverage under the 1972 amendments. See Northeast Marine Terminal Co., supra, and P.C. Pfeiffer, Co., supra. In neither case did we interpret the "maritime employment" status provision to require an examination into whether the employment had a "direct" or "significant relationship to navigation or commerce." Rather, in both cases, we decided that the employees were covered because they were "engaged in longshoring operations," and thus fit one of the categories explicitly enumerated by Congress as part of "maritime employment." See 432 U.S., at 271, 273, 97 S.Ct., at 2360, 2361; 444 U.S., at 82, 100 S.Ct., at 337. We have had no occasion as yet to determine other possible applications of the status test to activities performed on the expanded landward situs. Although we do not maintain that landward coverage could never be determined by reference to anything but the explicitly enumerated categories of activities in the section 2(3) definition of "employee," we note that our cases to date have focused on these explicit categories because the legislative history indicates that Congress intended to extend landward coverage to those specifically included occupations. See S.Rep., at 13; H.Rep., at 10-11. See also Northeast Marine Terminal Co., supra, 432 U.S., at 273, 97 S.Ct., at 2361. Regardless of the potential difficulties that may arise in the future in applying the status test to land-based injuries, it is clear that in extending coverage landward, Congress sought to make available LHWCA compensation to those who, before the 1972 amendments, regularly did move from covered to noncovered areas, but did not intend to withdraw coverage from those employees, traditionally covered by the Act, who were injured in the course of their employment on navigable waters as previously defined. 28 Ignoring the references in the Committee reports to the fact that in 1972 Congress merely sought to extend benefits landward, the dissent focuses instead on passages in the legislative history which indicate that Congress wanted to extend benefits to certain employees who regularly did (in Congress' view) walk in and out of coverage, and who performed the same tasks on land as they performed over the actual navigable waters. The dissent concludes from this that Congress sought to withdraw coverage from those employees injured over the actual navigable waters in the course of employment who would have been covered before 1972 and who we now hold are "engaged in maritime employment" for purposes of the amended LHWCA. The fact that Congress desired to extend coverage landward for a certain group of employees does not tend to prove that Congress sought to withdraw coverage from another group of employees who were customarily covered before the 1972 amendments. The dissent's view would relegate a number of employees to state compensation remedies in the face of express and extensive congressional findings that "most State Workmen's Compensation laws provide benefits which are inadequate." S.Rep., at 12-13; H.Rep., at 10, U.S.Code Cong. & Admin.News 1972, p. 4707. The dissent claims that it "cannot find a single word" in the legislative history to support LHWCA coverage of any employee who is not a longshoreman or harbor worker. Post, at 330. The word that the dissent overlooks is "maritime" in section 2(3) of the Act. Before 1972, employees such as Churchill were considered to be engaged in "maritime" employment. In order to withdraw coverage from employees, such as Churchill, who are maritime employees injured in the course of their employment upon the actual navigable waters, Congress would have had to ignore the consistent interpretation given the Act before 1972, by the Director, the Deputy Commissioners, the courts, and the commentators. See n. 21, supra. 29 Perini cites our decision in Executive Jet Aviation, Inc. v. City of Cleveland, 409 U.S. 249, 93 S.Ct. 493, 34 L.Ed.2d 454 (1972), and argues that the LHWCA is premised upon admiralty jurisdiction, which requires a connection between an employee and traditional maritime activity. Perini's reliance on Executive Jet is misplaced. In that case, the only issue before the Court was whether federal admiralty jurisdiction extended to tort claims arising out of the crash of an airplane into navigable waters on a flight "within the continental United States, which [is] principally over land." Id., at 266, 93 S.Ct., at 503. Jurisdiction in Executive Jet was predicated on 28 U.S.C. § 1333(1), which provides that the federal district courts have original and exclusive jurisdiction of "[a]ny civil case of admiralty or maritime jurisdiction." The explicit language of Executive Jet makes it clear that our discussion was occasioned by "the problems inherent in applying a locality-alone test of admiralty tort jurisdiction to crashes of aircraft . . . ." in a situation where "the fact that an aircraft happens to fall in navigable waters, rather than on land, is wholly fortuitous." Id., at 265, 266, 93 S.Ct., at 503, 504. Although the term "maritime" occurs both in 28 U.S.C. § 1333(1) and in § 2(3) of the Act, these are two different statutes "each with different legislative histories and jurisprudential interpretations over the course of decades." Boudreaux v. American Workover, Inc., 680 F.2d 1034, 1035, 1050 (CA5 1982) (footnote omitted). In addition, Churchill, as a marine construction worker, was by no means "fortuitously" on the water when he was injured. 30 The dissent argues that it is "now perfectly clear" that Churchill (or any other "shore-based worker" injured upon actual navigable waters) could have received a state compensation award, and there should be no concern about such an employee being left without a remedy. This position is by no means "perfectly clear." See, e.g., Holcomb v. Robert W. Kirk and Associates, Inc., 655 F.2d 589 (CA5 1981) (watchman injured on while working on vessel sought compensation under state scheme, and was denied because injury was covered under LHWCA—Court of Appeals granted LHWCA compensation, holding when Congress passed the 1972 amendments, it took for granted that injuries occurring on the actual navigable waters were covered under Parker, Davis, Pennsylvania Railroad Co., Calbeck, and the myriad lower court cases applying our decisions); Rex Investigative and Patrol Agency, supra, at 698 (the court found that the injured watchman's state compensation claim had been dismissed because the "claim properly belonged before a federal, rather than a [New York] state, agency"). 31 Certain comments made in the debates preceding passage of the 1972 amendments in the House indicate support for our view that Congress intended to extend protection in 1972, and not to withdraw protection. For example, Representative Steiger posed the following question and answer to explain the coverage provision: "Q. The present law covers employees working on navigable waters. Do the amendments change the scope of coverage? " "A. Yes. The present law's coverage is limited to employees working on navigable waters, including those working on dry docks. The amendments will extend coverage to wharfs, terminals, marine railways and other adjoining areas . . . ." 118 Cong.Rec. 36385 (1972) (emphasis in original). See also 118 Cong.Rec. 36270-36271 (1972) (remarks of Sen. Williams); id., at 36381-36382 (remarks of Rep. Daniels). 32 In both Northeast Marine Terminal Co., supra, 432 U.S., at 263-264, 97 S.Ct., at 2356-2357, and in P.C. Pfeiffer, supra, 444 U.S., at 78-79, 100 S.Ct., at 334-335, we recognized that the status requirement is occupational and the situs test is geographic. 33 See also, e.g., 1A Benedict on Admiralty §§ 17, 19 (7th ed. 1981); Gilmore & Black, supra, at 428-430; Robertson, Injuries to Maritime Petroleum Workers: A Plea for Radical Simplification, 55 Texas L.Rev. 973, 986-987 (1977); Comment, Broadened Coverage under the LHWCA, 33 La.L.Rev. 683, 694 (1973); Note, 54 N.C.L.Rev. 925, 940 (1976). But see 4 Larson, The Law of Workmen's Compensation §§ 89.27, 89.41 (1982); Tucker, Coverage and Procedure under the Longshoremen's and Harbor Workers' Compensation Act Subsequent to the 1972 Amendments, 55 Tulane L.Rev. 1056, 1062 (1981). 34 Our holding, of course, extends only to those persons "traditionally covered" before the 1972 amendments. We express no opinion whether such coverage extends to a worker injured while transiently or fortuitously upon actual navigable waters, or to a land-based worker injured on land who then falls into actual navigable waters. Our decision today should not be read as exempting water-based workers from the new status test. Rather, our holding is simply a recognition that a worker's performance of his duties upon actual navigable waters is necessarily a very important factor in determining whether he is engaged in "maritime employment." Contrary to the suggestion by the dissent, post, at n. 26, there is no inconsistency in our failing to decide the question of coverage as to these employees, and our reliance in Parker. In Parker, we held that the injured employee was engaged in "maritime employment" in a situation where we did not discuss whether the employer was a statutory "employer." 1 By reason of several specific statutory enactments, the LHWCA's compensation scheme is, or has been, also applied to: (a) employees on defense bases, Act of Aug. 16, 1941, ch. 357, § 1, 55 Stat. 622 (codified as amended at 42 U.S.C. §§ 1651-1654), (b) employees of nonappropriated fund instrumentalities such as post exchanges, Act of June 19, 1952, ch. 444, § 2, 66 Stat. 139 (codified as amended at 5 U.S.C. §§ 8171-8173), (c) employees of government contractors injured overseas by war-risk hazards, Act of Dec. 2, 1942, ch. 668, Title I, § 102, 56 Stat. 1031 (codified as amended at 42 U.S.C. § 1702), (d) workers in the District of Columbia, Act of May 17, 1928, 45 Stat. 600, ch. 612, repealed by Act of July 1, 1980, D.C. Law 3-77, § 3, 27 D.C.R. 2503, and (e) workers on oil drilling rigs on the outer continental shelf, Act of Aug. 7, 1953, ch. 345, § 4(c), 67 Stat. 462 (codified as amended at 43 U.S.C. § 1333(b)). In this case, however, we are concerned with the coverage provided by the LHWCA itself. 2 Coincidentally, two authors named Sutherland have made this point in language that is strikingly suitable to this case. See W. Sutherland, The Shipbuilders' Assistant 77 (1755 ed.) ("The straiter and snuger the Sheer lies, the less Wind is held to hinder the Motion of the Ship") (emphasis added); J. Sutherland, Sutherland on Statutory Construction § 273 (1891) ("The words 'other persons,' following in a statute the words 'warehouseman' and 'wharfinger,' must be understood to refer to other persons ejusdem generis, viz., those who are engaged in a like business, or who conduct the business of warehousemen or wharfingers with some other pursuit, such as shipping, grinding, or manufacturing"). 3 Seamen are protected under the Jones Act. See 46 U.S.C. § 688. 4 See H.R.Rep. No. 92-1441, p. 11; S.Rep. No. 92-1125, p. 13; Northeast Marine Terminal Co. v. Caputo, 432 U.S. 249, 97 S.Ct. 2348, 53 L.Ed.2d 320 (1977). 5 The 1972 Amendments made two changes that are relevant here. First, they modified the definitions in 33 U.S.C. § 902(3) and 33 U.S.C. § 902(4). Before the Amendments, the definitions read: (3) The term "employee" does not include a master or member of a crew of any vessel, nor any person engaged by the master to load or unload or repair any small vessel under eighteen tons net. (4) The term "employer" means an employer any of whose employees are employed in maritime employment, in whole or in part, upon the navigable waters of the United States (including any dry dock). As amended in 1972, the definitions read: "(3) The term 'employee' means any person engaged in maritime employment, including any longshoreman or other person engaged in longshoring operations, and any harborworker including a ship repairman, shipbuilder, and shipbreaker, but such term does not include a master or member of a crew of any vessel, or any person engaged by the master to load or unload or repair any small vessel under eighteen tons net. (4) The term 'employer' means an employer any of whose employees are employed in maritime employment, in whole or in part, upon the navigable waters of the United States (including any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other adjoining area customarily used by an employer in loading, repairing, or building a vessel)." Second, the Amendments modified the section defining covered injuries, 33 U.S.C. § 903(a). Before the Amendments, it read: "Compensation shall be payable under this Act in respect of disability or death of an employee, but only if the disability or death results from an injury occurring upon the navigable waters of the United States (including any dry dock) and if recovery for the disability or death through workmen's compensation proceedings may not validly be provided by State law. No compensation shall be payable in respect of the disability or death of— (1) A master or member of a crew of any vessel, nor any person engaged by the master to load or unload or repair any small vessel under eighteen tons net; or (2) An officer or employee of the United States or any agency thereof or of any State or foreign government, or of any political subdivision thereof." As amended in 1972, the section reads: "Compensation shall be payable under this Act in respect of disability or death of an employee, but only if the disability or death results from an injury occurring upon the navigable waters of the United States (including any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other adjoining area customarily used by an employer in loading, repairing, or building a vessel). No compensation shall be payable in respect of the disability or death of— (1) A master or member of a crew of any vessel, nor any person engaged by the master to load or unload or repair any small vessel under eighteen tons net; or (2) An officer or employee of the United States or any agency thereof or of any State or foreign government, or of any political subdivision thereof." 86 Stat. 1251, Longshoremen's and Harbor Workers' Compensation Act Amendments of 1972. 6 The Court assumes that the words "traditionally covered" in the Committee reports are intended to refer to employees who are not longshoremen or harbor workers. Ante, at 319, quoting S.Rep. No. 92-1125, p. 16. See n. 1, supra. In particular, the Court assumes that the Committee was referring to the claimants in Parker v. Motor Boat Sales, Inc., 314 U.S. 244, 62 S.Ct. 221, 86 L.Ed. 184 (1941), Davis v. Department of Labor, 317 U.S. 249, 63 S.Ct. 225, 87 L.Ed. 246 (1942), and Calbeck v. Travelers Insurance Co., 370 U.S. 114, 82 S.Ct. 1196, 8 L.Ed.2d 368 (1962). As I point out in Part III, infra, the Calbeck claimants were shipbuilders, a subcategory of the statutorily defined class of harbor workers, who are of course still covered under the 1972 Act; Davis held only that the claimant was entitled to state benefits; and Parker was plainly not a "traditional" LHWCA case. None of these cases was cited at any time in the hearings or the reports. In my opinion the reference to the "traditional" coverage of the Act was intended to identify the coverage of longshoremen and harbor workers as opposed to the special categories of coverage defined by specific statutory enactment. 7 See n. 1, supra. 8 None of the original bills proposing Amendments to the LHWCA in 1972 embodied any change in the scope of coverage. See H.R. 247; H.R. 3505; H.R. 12006; H.R. 15023; S. 2318; S. 525; S. 1547 (all in 92nd Cong., 2d Sess.). The changes were incorporated between the hearings and the final committee action. See H.R. 12006, 92nd Cong., 2d Sess.; S. 2318, 92nd Cong., 2d Sess. (as reported). The hearings are nonetheless relevant because they give more direct evidence of what groups the legislators intended to protect than does the history of pre-1972 Supreme Court decisions. 9 E.g., Statement of James Hodgson, Secretary of Labor, House Hearings at 47-64 (referring throughout to "longshoremen" and the "longshore industry"); Statement of Ralph Hartman, Bethlehem Steel Corp., id., at 47 ("reference to the Longshoremen's and Harbor Workers' Compensation Act seems to suggest that the only industry involved is 'longshoring,' which fails to recognize that the act is also applicable to shipbuilding and ship repair yards—and to the District of Columbia"); Exhibits D1, D2, E, and F to Statement of James Flynn, New York Shipping Association, id., at 100 (pointing out how hazardous longshoring is); Statement of Howard McGuigan, AFL-CIO, id., at 255-258 (pointing out how LHWCA benefits were far below 662/3% of current wage levels in the longshore industry, in the shipbuilding and ship repair industry, and in the District of Columbia). Cf. Statement of John J. O'Donnell, Air Line Pilots Association, id., at 327-329 (suggesting that coverage be extended to flight crews). 10 The Minority Counsel, Eugene Mittelman, had the following exchange with a representative of the AFL-CIO: "MR. MITTELMAN. My last question concerns the fact that the longshoreman [sic] applies only when the man is over the navigable waters of the United States, and under whole series of court decisions there has been established a line where the provisions of the Longshore Act apply when the man is over the water, and yet the provisions of the State workmen's compensation law applies if the man is injured on land. * * * * * "Do you have any position on this, concerning whether the Federal law should be extended, really, so that a uniform system of benefits is applicable to longshoremen, regardless of which side of the water-line the injury occurred on? "MR. McGUIGAN. The first position we would have is that obviously there would be no incentive to cover him under the act until we know the act gives him benefits superior to the State workmen's compensation laws. "MR. MITTELMAN. I appreciate that. But assuming we would amend the act to provide a reasonable schedule of benefits as proposed in this bill, would you favor the principle of extending of the Longshore Act to cover all longshore workers whether performed on land or over water? * * * * * "MR. O'BRIEN. . . . [I]f the act were amended to take up its former place of prominence in the field of workmen's compensation, we would certainly like to see the coverage of the act extended." Hearings on S. 2318 et al. before the Subcommittee on Labor of the Senate Committee on Labor and Public Finance, 92nd Cong., 2d Sess., pp. 73-74. 11 The language of the Committee Reports shows how clearly Congress understood who was to be covered: "The present Act, insofar as longshoremen and ship builders and ship repairmen are concerned, covers only injuries which occur 'upon the navigable waters of the United States.' Thus, coverage of the present Act stops at the water's edge; injuries occurring on land are covered by State Workmen's Compensation laws. The result is a disparity in benefits payable for death or disability for the same type of injury depending on which side of the water's edge and in which State the accident occurs." * * * * * "The Committee believes that the compensation payable to a longshoreman or a ship repairman or a builder should not depend on the fortuitous circumstance of whether the injury occurred on land or over water. Accordingly, the bill would amend the Act to provide coverage of longshoremen, harbor workers, ship repairmen, ship builders, shipbreakers, and other employees engaged in maritime employment (excluding masters and members of the crew of a vessel) if the injury occurred either upon the navigable waters of the United States or any adjoining pier, wharf, dry dock, terminal, building way, marine railway, or other area adjoining such navigable waters customarily used by an employer in loading, unloading, repairing, or building a vessel." H.R.Rep. No. 92-1441, p. 10, U.S.Code Cong. & Admin.News 1972, p. 4708; S.Rep. No. 92-1125, pp. 12-13. 12 Arguably one other case, mentioned in a footnote of the Court's opinion, ante, at n. 644, echoed Parker's broad construction of the scope of LHWCA coverage. Pennsylvania R.R. v. O'Rourke, 344 U.S. 334, 73 S.Ct. 302, 97 L.Ed. 367 (1953). There, this Court struck down an award of benefits under the Federal Employers' Liability Act, reasoning that the employee in that case a brakeman who worked moving freight cars onto "car floats"—could have recovered under the LHWCA. The opinion in O'Rourke is somewhat cloudy, however, since it does not explicitly state that the particular employee was engaged in maritime employment, but only that his employer had such employees. Id., at 339-340, 73 S.Ct., at 305-306. Like the cases on which the Court relies, O'Rourke was not mentioned in the 1972 legislative history. 13 The Court reasoned: "The work of a stevedore in which the deceased was engaging is maritime in its nature; his employment was a maritime contract; the injuries which he received were likewise maritime; and the rights and liabilities of the parties in connection therewith were matters clearly within the admiralty jurisdiction. "If New York can subject foreign ships coming into her ports to such obligations as those imposed by her Compensation Statute, other States may do likewise. The necessary consequence would be destruction of the very uniformity in respect to maritime matters which the Constitution was designed to establish; and freedom of navigation between the States and with foreign countries would be seriously hampered and impeded." 244 U.S., at 217, 37 S.Ct., at 529 (citation omitted). 14 See id., at 218-223, 37 S.Ct., at 529-532 (Holmes, J., dissenting); Knickerbocker Ice Co. v. Stewart, 253 U.S. 149, 166-170, 40 S.Ct. 438, 442-443, 64 L.Ed. 834 (1920) (Holmes, J., dissenting); Washington v. W.C. Dawson & Co., 264 U.S. 219, 228-239, 44 S.Ct. 302, 305-309, 68 L.Ed. 646 (1924) (Brandeis, J., dissenting). 15 In explaining why the holding in Rohde was consistent with Jensen and subsequent cases, the Court stated: "In each of them the employment or contract was maritime in nature and the rights and liabilities of the parties were prescribed by general rules of maritime law essential to its proper harmony and uniformity. Here the parties contracted with reference to the state statute; their rights and liabilities had no direct relation to navigation, and the application of the local law cannot materially affect any rules of the sea whose uniformity is essential." 257 U.S. 469, 477, 42 S.Ct. 157, 158, 66 L.Ed. 321. 16 The Strand case was decided in 1928 but arose out of an injury that had occurred in 1924, prior to the enactment of the LHWCA. 17 "The main impetus for the Longshoremen's and Harbor Workers' Compensation Act was the need to correct a gap made plain by decisions of this Court. We believe that there is only one interpretation of the proviso in § 3(a) which would accord with the aim of Congress; the field in which a state may not validly provide for compensation must be taken, for the purposes of the Act, as the same field which the Jensen line of decision excluded from state compensation laws. Without affirming or rejecting the constitutional implications of those cases, we accept them as the measure by which Congress intended to mark the scope of the Act they brought into existence." Parker v. Motor Boat Sales, Inc., 314 U.S. 244, 250, 62 S.Ct. 221, 225, 86 L.Ed. 184. 18 See Davis v. Department of Labor, 317 U.S. 249, 261, 63 S.Ct. 225, 231, 87 L.Ed. 246 (1942) (Stone, C.J., dissenting); Calbeck v. Travelers Insurance Co., 370 U.S. 114, 132, 82 S.Ct. 1196, 1206, 8 L.Ed.2d 368 (1962) (Stewart, J., dissenting). 19 The Court relies heavily on the proposition that Congress did not wish "to repeal Calbeck " (ante, at 321). It is, of course, true that the claimants in that case are still covered by the Act. What Congress repealed was the statutory language that appeared to preclude coverage for harbor workers like the Calbeck claimants who were injured in the maritime but local area. The problem confronted by the Court in Calbeck simply no longer exists. 20 In 1942, this Court observed: "The horns of the jurisdictional dilemma press as sharply on employers as on employees. In the face of the cases referred to above, the most competent counsel may be unable to predict on which side of the line particular employment will fall. The employer's contribution to a state insurance fund may therefore wholly fail to protect him against the liabilities for which it was specifically planned. If this very case is affirmed, for example, the employer will not only lose the benefit of the state insurance to which he has been compelled to contribute and by which he has thought himself secured against loss for accidents to his employees; he must also, by virtue of the conclusion that the employee was subject to the federal act at the time of the accident, become liable for substantial additional payments. He will also be subject to fine and imprisonment for the misdemeanor of having failed, as is apparently the case, to secure payment for the employee under the federal act. 33 U.S.C. §§ 938, 932." Davis, supra, 317 U.S., at 255, 63 S.Ct., at 228. On that point, the dissenter was in complete agreement. See id., at 262, 63 S.Ct., at 232 (Stone, C.J., dissenting). 21 See supra, at 332-334. 22 Before 1972, the financial burden of duplicate coverage had not been particularly heavy. LHWCA benefits were low, and insurance carriers offered to cover operations subject to the LHWCA for only a nominal addition to the state workers' compensation premiums. See Note, 50 Calif.L.Rev. 342, 347 (1962); Comment, 30 NACCA L.J. 200, 203, 206 (1964); Gardner, Remedies for Personal Injuries to Seamen, Railroadmen, and Longshoremen, 71 Harv.L.Rev. 438, 449-50, and n. 34 (1958). Today, of course, things are quite different. In 1981, Longshoremen's Act premiums averaged 252 percent higher than California construction worker premiums, and 160 percent higher than Florida premiums. See Testimony of the Associated General Contractors of America, Hearings on S. 1182 before the Subcommittee on Labor of the Senate Committee on Labor and Public Finance, 97th Cong., 1st Sess., pp. 924-936. 23 The "status" provision replaced the "unless recovery may validly be provided by state law" language that was being construed in Parker and Calbeck. 24 In 1942, as it does today, the LHWCA expressly excluded coverage of injuries to members of the crew of any vessel and to persons who load or unload small boats. See n. 5, supra. Thus, a janitor could not recover on the theory that he was a member of the crew of the motor boat, or that he helped to load or unload the motor boat. It is difficult to explain the narrow category of workmen associated with motorboat operations for whom Parker expressed concern or for whom the Court preserves coverage today. 25 The Court cites four cases from the federal district courts, three from the courts of appeals, and one from a state Supreme Court in which workers who were not longshoremen or harbor workers were stated to have been covered by the LHWCA before 1972. Ante, at 312, n. 21. It uses these cases to support its argument that it would have been a radical and unsettling change for the 1972 Congress to limit post-1972 coverage to people who perform the work of longshoremen and harbor workers. I would draw a somewhat different inference. It is hard to believe that Congress had in mind such a light sprinkling of cases during the 45-year interval between 1927 and 1972 when it spoke of the traditional coverage of the Act, especially given Congressman Daniels' reminder that in 1970 there were 68,000 injuries to longshoremen. See supra, p. 332. 26 I note some tension among different components of the Court's opinion with regard to whether the janitor in Parker would be covered after 1972. On the one hand, the Court states: "[B]efore 1972 . . . any worker injured upon navigable waters in the course of employment was 'covered . . . without any inquiry into what he was doing (or supposed to be doing) at the time of his injury' . . . We are unable to find any congressional intent to withdraw coverage of the LHWCA from those workers injured on navigable waters in the course of their employment, and who would have been covered by the Act before 1972. . . . Congress . . . assumed that injuries occurring on the actual navigable waters were covered, and would remain covered." Ante, at 319. On the other hand, it concludes: "Our holding, of course, extends only to those persons 'traditionally covered' before the 1972 amendments. We express no opinion whether such coverage extends to a worker injured while transiently or fortuitously upon navigable waters . . . . Our decision today should not be read as exempting water-based workers from the new status test. Rather, our holding is simply a recognition that a worker's performance of his duties upon actual navigable waters is necessarily a very important factor. . . ." Ante, at 324 n. 33. Similarly, at one point the Court says "[Congress's] use of 'employees traditionally covered' was intended to refer to those employees included in the scope of coverage under Parker, Davis, and Calbeck," ante at 319-320, but at another point it concedes that those very cases were read "not to limit LHWCA coverage only to 'traditional' maritime activities," ante, at 312, n. 20. I agree with the Court that the post-1972 Act provides coverage for "traditional" maritime activities. However, as I have indicated above, pp. 644-649 supra, Congress understood such activities to be those of longshoremen and harbor workers, not janitors and construction workers..

78

459 U.S. 248 103 S.Ct. 608 74 L.Ed.2d 430 The PILLSBURY COMPANY, et al., Petitionersv.John CONBOY. No. 81-825. Argued Oct. 6, 1982. Decided Jan. 11, 1983. Syllabus Title 18 U.S.C. § 6002 provides that "no testimony or other information compelled under the order [of a federal court] (or any information directly or indirectly derived from such testimony or other information) may be used against the witness in any criminal case." When respondent appeared before a grand jury investigating price-fixing activities in the corrugated container industry, he was granted use immunity pursuant to § 6002 for his testimony. Subsequently, in civil antitrust actions brought in Federal District Court by petitioner purchasers of corrugated containers, respondent appeared, pursuant to a subpoena, for a deposition. At the deposition, questions were read from the transcript of his immunized grand jury testimony and rephrased to include the transcript answer, and then respondent was asked if he had "so testified" before the grand jury. He refused to answer each question, asserting his Fifth Amendment privilege against self-incrimination. Granting petitioners' motion to compel respondent to answer, the District Court held him in contempt when he continued to claim his privilege. The Court of Appeals reversed, holding that respondent was entitled to assert his Fifth Amendment privilege, since his deposition testimony was not protected under § 6002 but could be used against him in a subsequent criminal action. Held: A deponent's civil deposition testimony, such as that in question in this case, repeating verbatim or closely tracking his prior immunized testimony, is not, without duly authorized assurance of immunity at the time, immunized testimony within the meaning of § 6002, and therefore may not be compelled over a valid assertion of his Fifth Amendment privilege. Pp. 252-264. (a) To construe § 6002, as petitioners urge, so as to hold that the grant of immunity compelled respondent to give testimony at the civil deposition that repeats verbatim or closely tracks his prior testimony sweeps further than Congress intended and could hinder the Government's enforcement of its criminal laws by turning use immunity into a form of transactional immunity for subjects examined in the immunized proceeding. Use immunity is intended to immunize and exclude from a subsequent criminal trial only that information to which the Government expressly has surrendered future use. The purpose of § 6002 is to limit the scope of immunity to a constitutionally required level, as well as to limit the use of immunity to those cases in which the Government determines that gaining the witness' testimony outweighs the loss of the opportunity for criminal prosecution of that witness. Pp. 255-261. (b) Petitioners' proposed construction of § 6002 also could put the deponent to some risk unless he receives an assurance of immunity or exclusion that the courts cannot properly give. Silence, on the other hand, preserves the deponent's rights and the Government's interests, as well as the judicial resources that otherwise would be required to make the many difficult judgments that petitioners' interpretation of § 6002 would require. Pp. 261-263. 661 F.2d 1145 (7th Cir.1981), affirmed. Francis J. McConnell, Chicago, Ill., for petitioners. Michael W. Coffield, Chicago, Ill., for respondent. Justice POWELL delivered the opinion of the Court. 1 Pursuant to the federal use immunity provisions, 18 U.S.C. §§ 6001-6005 (1976), a United States Attorney may request an order from a federal court compelling a witness to testify even though he has asserted his privilege against self-incrimination. Section 6002 provides, however, that "no testimony or other information compelled under the order (or any information directly or indirectly derived from such testimony or other information) may be used against the witness in any criminal case. . . ." The issue presented in this case is whether a deponent's civil deposition testimony, repeating verbatim or closely tracking his prior immunized testimony, is immunized "testimony" that can be compelled over the valid assertion of his Fifth Amendment privilege. 2 * Respondent John Conboy is a former executive of a defendant in the In re Corrugated Container Antitrust Litigation, M.D.L. 310 (S.D.Tex.). In January 1978, United States Department of Justice attorneys interviewed Conboy following a promise of use immunity. Conboy subsequently appeared before a grand jury investigating price-fixing activities and, pursuant to 18 U.S.C. § 6002, was granted formal use immunity for his testimony. 3 Following the criminal indictment of several companies, numerous civil antitrust actions were filed in various United States district courts. Those actions were consolidated for discovery in the District Court for the Southern District of Texas. Petitioners here are purchasers of corrugated containers who elected to opt out of the class-action proceedings and pursue their own causes of action against manufacturers. The District Court ordered that portions of the immunized government interview and grand-jury testimony of certain witnesses, including that of Conboy, be made available to lawyers for the class and opt-outs.1 4 Pursuant to a subpoena issued by the District Court for the Northern District of Illinois, Conboy appeared in Chicago for a deposition at which he, his counsel, and petitioners' counsel had copies of his immunized testimony. The transcripts were marked as deposition exhibits so that all could follow the intended examination. The questioning fell into the following pattern: a question was read from the transcript; it then was rephrased to include the transcript answer (i.e., "Is it not the fact that. . . ."); finally, Conboy was asked if he had "so testified" in his immunized interview and grand-jury examination.2 Conboy refused to answer each question, asserting his Fifth Amendment privilege against self-incrimination. 5 The District Court granted petitioners' motion to compel Conboy to answer the questions.3 When Conboy continued to claim his privilege, the District Court held him in contempt, but stayed its order pending appeal. A panel of the Court of Appeals for the Seventh Circuit affirmed the contempt order, holding that, "[b]ecause the questions asked in this deposition were taken verbatim from or closely tracked the transcript of Conboy's grand jury testimony, we believe that his answers at the deposition would be 'derived from' the prior immunized [testimony] and therefore unavailable for use in any subsequent criminal prosecution." In re Corrugated Container Antitrust Litigation, Appeal of John Conboy, 655 F.2d 748, 751 (CA7), rev'd en banc, 661 F.2d 1145 (1981). 6 On rehearing en banc, the Court of Appeals reversed the District Court. It first determined that Conboy's alleged fear of prosecution was more than "fanciful," 661 F.2d, at 1152, and that Conboy therefore was entitled to assert his Fifth Amendment privilege unless his deposition testimony could not be used against him in a subsequent criminal action, see id., at 1153.4 The court then held that under § 6002, absent a separate and independent grant of immunity,5 a deponent's civil deposition testimony that repeats verbatim or closely tracks his prior immunized testimony is not protected. While acknowledging that verbatim questions "of course [would be] derived" from the immunized testimony, the court reasoned that the answers to such questions "are derived from the deponent's current, independent memory of events" and thus "necessarily create a new source of evidence" that could be used in a subsequent criminal prosecution against Conboy. Id., at 1155 (emphasis in original). 7 We granted certiorari to resolve the conflict in the Courts of Appeals,6 and now affirm. II 8 It is settled that government must have the power to compel testimony "to secure information necessary for effective law enforcement." Murphy v. Waterfront Commission, 378 U.S. 52, 79, 84 S.Ct. 1594, 1609, 12 L.Ed.2d 678 (1964).7 For many years, however, a person who was compelled to testify under a grant of governmental immunity could not be prosecuted for any conduct about which he had testified. See New Jersey v. Portash, 440 U.S. 450, 457, 99 S.Ct. 1292, 1296, 59 L.Ed.2d 501 (1979). Prosecutors therefore were reluctant to grant such "transactional" immunity to potential targets of criminal investigations. See S.Rep. No. 91-617, p. 53 (1969). 9 The "major purpose" of the Organized Crime Control Act of 1970, Pub.L. No. 91-452, 84 Stat. 927, of which § 6002 was a key provision, was "to provide the criminal justice system with the necessary legal tools to . . . strengthe[n] the evidence gathering process and insur[e] that the evidence will then be available and admissible at trial." 116 Cong.Rec. 35,200 (1970) (statement of Rep. St. Germain). Congress sought to make the grant of immunity more useful for law enforcement officers through two specific changes. First, Congress made the grant of immunity less expansive8 by repealing the authority for transactional immunity and providing for the less comprehensive use immunity authorized in § 6002.9 Second, Congress gave certain officials in the Department of Justice10 exclusive authority to grant immunities.11 10 The Court upheld the constitutionality of the use immunity statute in Kastigar v. United States, 406 U.S. 441, 92 S.Ct. 1653, 32 L.Ed.2d 212 (1972). The power to compel testimony is limited by the Fifth Amendment, and we held that any grant of immunity must be co-extensive with the privilege. We were satisfied, however, that § 6002 provided this measure of protection and thus "removed the dangers against which the privilege protects." Id., at 449, 92 S.Ct., at 1659. In rejecting the argument that use and derivative-use immunity would not adequately protect a witness from various incriminating uses of the compelled testimony, we emphasized that "[t]he statute provides a sweeping proscription of any use, direct or indirect, of the compelled testimony and any information derived therefrom. . . ." Id., at 460, 92 S.Ct., at 1664. We added that once a defendant establishes that he has testified under a grant of immunity, "the prosecution [has] the affirmative duty to prove that the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony." Ibid. Thus, "immunity from use and derivative use 'leaves the witness and the Federal Government in substantially the same position as if the witness had claimed his privilege' in the absence of a grant of immunity." Kastigar, 406 U.S., at 458-459, 92 S.Ct., at 1663-1664 (quoting Murphy, 378 U.S., at 79, 84 S.Ct., at 1610). III 11 With the foregoing statutory history and relevant principles in mind, we turn now to this case. It is not disputed that the questions asked of Conboy were directly or indirectly derived from his immunized testimony. The issue as presented to us is whether the causal connection between the questions and the answers is so direct that the answers also are derived from that testimony and therefore should be excluded under the grant of immunity. 12 Petitioners' argument is based on the language of § 6002 and on a common understanding of the words "derived from." The questions formulated on the basis of immunized testimony are clearly "derived from" the prior testimony. Thus, the answers that repeat verbatim or closely track a deponent's testimony are necessarily also "derived from" and "tainted by" such testimony. Petitioners therefore find no basis for the distinction made by the Court of Appeals between questions and answers responsive to those same questions. An answer by its very nature is evoked by and responds to information contained in a question. 13 Conboy's position is also straightforward: Questions do not incriminate; answers do. Unlike the questions, answers are not directly or indirectly derived from the immunized grand jury or interview transcripts, but from the deponent's current, independent memory of events. Even when a deponent's deposition answers are identical to those he gave to the grand jury, he is under oath to tell the truth, not necessarily as he told it before the grand jury, but as he knows it now. Each new statement of the deponent creates a new "source." In sum, the initial grant of immunity does not prevent the prosecutor from prosecuting; it merely limits his sources of evidence. 14 Although the parties make their arguments in terms tracking those of the statute—whether the deposition testimony is "derived from" the prior testimony—it is clear that the crux of their dispute is whether the earlier grant of immunity itself compelled Conboy to talk.12 Petitioners contend that the prior grant of immunity already had supplanted Conboy's Fifth Amendment privilege at the time of the civil deposition. Petitioners would limit this immunity, of course, to testimony that "closely tracks" his prior immunized testimony. It is argued that this would not threaten the Government's need for admissible evidence or the individual's interest in avoiding self-incrimination. In the absence of such a threat, admissible evidence should be available to civil antitrust plaintiffs. But we cannot accept the assumptions upon which petitioners' conclusion rests. In our view, a District Court cannot compel Conboy to answer deposition questions, over a valid assertion of his Fifth Amendment right, absent a duly authorized assurance of immunity at the time.13 15 We note at the outset that although there may be practical reasons for not testifying,14 as far as the deponent's Fifth Amendment right is concerned he should be indifferent between the protection afforded by silence and that afforded by immunity. A deponent's primary interest is that the protection be certain. The Government's interest, however, may be affected seriously by whether the deponent relies at the civil deposition on his Fifth Amendment privilege or on his prior grant of immunity. With due recognition of petitioners' need for admissible evidence, our inquiry then is whether this need can be met without jeopardizing the Government's interest in limiting the scope of an immunity grant or encroaching upon the deponent's certainty of protection. 16 Questions taken verbatim from a transcript of immunized testimony could evoke one of several responses from a deponent: (i) he could repeat or adopt his immunized answer;15 (ii) he could affirm that the transcript of his immunized answers accurately reflects his prior testimony; (iii) he could recall additional information responsive to the question but not disclosed in his immune testimony; or (iv) he could disclose information that is not responsive to the question. Petitioners do not contend, nor could they, that the prior grant of use immunity affords protection for all self-incriminating information disclosed by the immunized witness on any occasion after the giving of the immunized testimony. Rather, petitioners argue that only the first three responses would be "derived from" his immune testimony and therefore would be unavailable for use against the deponent in any subsequent criminal prosecution. 17 Petitioners' premise is that the deposition of Conboy is designed not to discover new information,16 but to obtain evidence that simply repeats the statements in the immunized transcript.17 Because there will be little opportunity for the grant of immunity to sweep in statements on direct examination that the Government did not intend to immunize, or for the deponent to give responses that may fall outside of the grant of immunity and later be used against him in a subsequent criminal prosecution, petitioners argue that Conboy's deposition will yield only a carbon copy of the grand-jury transcript. In such a situation, it would be desirable for civil plaintiffs, particularly those bringing private suits that supplement the criminal enforcement of the federal antitrust laws, to have access to the available, probative information. 18 But even if the direct examination is limited to the questions and answers in the immunized transcript, there remains the right of cross examination,18 a right traditionally relied upon expansively to test credibility as well as to seek the truth. Petitioners recognize this problem, but maintain that the antitrust defendants "would be entitled to test the accuracy and truthfulness of Conboy's repeated immunized testimony without going beyond the confines of that testimony." Regardless of any limitations that may be imposed on its scope,19 however, cross examination is intended to and often will produce information not elicited on direct. We must assume that, to produce admissible evidence, the scope of cross examination at the deposition cannot easily be limited to the immunized testimony. This assumption implicates both the Government's and the individual's interests embodied in § 6002. B 19 Use immunity was intended to immunize and exclude from a subsequent criminal trial only that information to which the Government expressly has surrendered future use. If the Government is engaged in an ongoing investigation of the particular activity at issue, immunizing new information (e.g., the answers to questions in a case like this one) may make it more difficult to show in a subsequent prosecution that similar information was obtained from wholly independent sources. If a District Court were to conclude in a subsequent civil proceeding that the prior immunity order extended to civil deposition testimony closely tracking the immunized testimony, it in effect could invest the deponent with transactional immunity on matters about which he testified at the immunized proceedings. This is precisely the kind of immunity Congress intended to prohibit. The purpose of § 6002 was to limit the scope of immunity to the level that is constitutionally required, as well as to limit the use of immunity to those cases in which the Attorney General, or officials designated by him, determine that gaining the witness's testimony outweighs the loss of the opportunity for criminal prosecution of that witness.20 C 20 Petitioners' interpretation of § 6002 also places substantial risks on the deponent.21 Unless the grant of immunity assures a witness that his incriminating testimony will not be used against him in a subsequent criminal prosecution, the witness has not received the certain protection of his Fifth Amendment privilege that he has been forced to exchange. No court has authority to immunize a witness. That responsibility, as we have noted, is peculiarly an executive one, and only the Attorney General or a designated officer of the Department of Justice has authority to grant use immunity. See 18 U.S.C. §§ 6002, 6003. Nor should a court, at the time of the civil testimony, pre-determine the decision of the court in a subsequent criminal prosecution on the question whether the Government has met its burden of proving "that the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony." Kastigar, 406 U.S., at 460, 92 S.Ct., at 1664. Yet in holding Conboy in contempt for his Fifth Amendment silence, the District Court below essentially predicted that a court in any future criminal prosecution of Conboy will be obligated to protect against evidentiary use of the deposition testimony petitioners seek. We do not think such a predictive judgment is enough. 21 Petitioners' interpretation of § 6002 imposes risks on the deponent whether or not the deposition testimony properly can be used against him in a subsequent criminal prosecution.22 Accordingly, the District Court's compulsion order in this case, in the absence of statutory authority or a new grant of immunity by the United States Attorney, cannot be justified by the subsequent exclusion of the compelled testimony. As Justice MARSHALL notes in his concurring opinion: "Whatever justification there may be for requiring a witness to give incriminating testimony in aid of a criminal investigation after the government has granted use immunity, there is no similar justification for compelling a witness to give incriminating testimony for the benefit of a private litigant when the government has not chosen to grant immunity." Post, at 267. 22 The result of compelling testimony—whether it is immunized or excluded—is that the Government's interests, as well as the witness's, suffer. Reliance on judicial exclusion of nonimmunized testimony would be inconsistent with the congressional policy of leaving the granting of immunity to the Executive Branch. 23 As the Court stated in Maness v. Meyers, 419 U.S. 449, 95 S.Ct. 584, 42 L.Ed.2d 574 (1975), compelling a witness to testify in "reliance upon a later objection or motion to suppress would 'let the cat out' with no assurance whatever of putting it back." Id., at 463, 95 S.Ct., at 593. We believe Conboy acted properly in maintaining his silence in the face of the District Court's compulsion order and by testing the validity of his privilege on appeal. IV 24 This Court has emphasized the importance of the private action as a means of furthering the policy goals of certain federal regulatory statutes, including the federal antitrust laws. See, e.g., Perma Life Mufflers, Inc. v. International Parts Corp., 392 U.S. 134, 139, 88 S.Ct. 1981, 1984, 20 L.Ed.2d 982 (1968); United States v. Borden Co., 347 U.S. 514, 518-519, 74 S.Ct. 703, 706, 98 L.Ed. 903 (1954). But private civil actions can only supplement, not supplant, the primary responsibility of Government. Petitioners' proposed construction of § 6002 sweeps further than Congress intended and could hinder governmental enforcement of its criminal laws by turning use immunity into a form of transactional immunity for subjects examined in the immunized proceeding. It also puts the deponent in some danger of criminal prosecution unless he receives an assurance of immunity or exclusion that the courts cannot properly give. Silence, on the other hand, preserves the deponent's rights and the Government's interests, as well as the judicial resources that otherwise would be required to make the many difficult judgments that petitioners' interpretation of § 6002 would require.23 V 25 We hold that a deponent's civil deposition testimony, closely tracking his prior immunized testimony, is not, without duly authorized assurance of immunity at the time, immunized testimony within the meaning of § 6002, and therefore may not be compelled over a valid assertion of his Fifth Amendment privilege.24 The judgment of the Court of Appeals accordingly is 26 Affirmed. 27 Justice MARSHALL, concurring. 28 I join the Court's decision that a witness who has given immunized testimony may invoke the Fifth Amendment privilege at a later proceeding in response to questions based on his immunized testimony. Permitting a civil litigant to rely on prior immunized testimony to defeat an otherwise valid claim of privilege would be inconsistent with the purposes of the use-immunity statute, regardless of whether, had the witness answered voluntarily, his answers could have been used against him in a later criminal trial. The Court's decision today does not reach the question whether such answers could later be admitted against the witness. In his dissenting opinion, Justice STEVENS argues that Conboy may not assert the Fifth Amendment privilege precisely because his answers could not properly be used against him in a later criminal trial. Because I agree with Justice STEVENS that such answers could not be properly used in a subsequent criminal trial, I write separately to explain why I believe respondent nevertheless retained his Fifth Amendment privilege. 29 If Conboy had voluntarily answered petitioners' deposition questions, his answers would have been "directly or indirectly derived from" his prior testimony before the grand jury. The questions were based solely on the transcript of respondent's grand jury testimony. There is no suggestion that the same or similar questions would have been asked had petitioners' attorneys not obtained a transcript of the grand jury testimony. Thus, if respondent had answered the questions, his answers would not have been "derived from a legitimate source wholly independent of the compelled testimony." Kastigar v. United States, 406 U.S. 441, 460, 92 S.Ct. 1653, 1664, 32 L.Ed.2d 212 (1972). 30 The admission of such answers at a subsequent criminal prosecution would represent a substantial departure from the fundamental premise of this Court's decision in Kastigar. In upholding the use-immunity statute against an attack based upon the Fifth Amendment privilege against self-incrimination, the Court concluded that use immunity affords a witness protection "as comprehensive as the protection afforded by the privilege." Id., at 449, 92 S.Ct., at 1659. The Court stated that the statute "prohibits the prosecutorial authorities from using the compelled testimony in any respect," id., at 453, 92 S.Ct., at 1661 (emphasis in original), and that it "provides a sweeping proscription of any use, direct or indirect, of the compelled information and any information derived therefrom," id., at 460, 92 S.Ct., at 1664. If the prosecution could introduce answers elicited from a witness by questions that would not have been asked but for the witness's immunized testimony, the protection afforded by use immunity would not be "as comprehensive as the protection afforded by the privilege." Id., at 449, 92 S.Ct., at 1659. 31 I therefore agree with my Brother STEVENS that answers to the questions posed by petitioners' attorneys could not properly have been used at a subsequent criminal trial. It does not follow, however, that respondent can be compelled to answer. In this case it is conceded that, had respondent never given the immunized testimony before the grand jury, he would have been entitled to invoke the Fifth Amendment privilege in response to questions concerning the same subject matter as the questions asked at the deposition. The only question is whether respondent is barred from asserting the Fifth Amendment privilege because he previously testified under a statutory grant of immunity and because his answers to the deposition questions would be "directly or indirectly derived" from his prior immunized testimony. 32 In my view, a trial judge may not constitutionally compel a witness to give incriminating testimony solely upon a finding that the witness's answers could not properly be used against him in a later criminal proceeding.1 This Court's decision in Kastigar v. United States, 406 U.S. 441, 92 S.Ct. 1653, 32 L.Ed.2d 212 (1972), does not support such compulsion. In Kastigar the Court was concerned with a federal statute that permits a United States Attorney, a federal agency, or a duly authorized representative of Congress to grant use immunity and thereby compel a witness to give incriminating testimony. See 18 U.S.C. §§ 6002-6005. Kastigar itself involved a grant of use immunity conferred upon a witness called to testify before a grand jury. In upholding the use-immunity statute against constitutional attack, the Court held only that, pursuant to statutory authority to confer such immunity, the government may constitutionally compel incriminating testimony in exchange for immunity from use or derivative use of that testimony. 406 U.S., at 462, 92 S.Ct., at 1665. Kastigar does not hold that a trial judge, acting without statutory authority to grant immunity, may rely on prior immunized testimony to overrule an otherwise valid assertion of the Fifth Amendment privilege by a deponent in a civil case. 33 Whatever justification there may be for requiring a witness to give incriminating testimony in aid of a criminal investigation after the government has granted use immunity, there is no similar justification for compelling a witness to give incriminating testimony for the benefit of a private litigant when the government has not chosen to grant immunity. Any interest served by compelling the testimony is insufficient to justify subjecting the witness to the risks that attend the compulsion of incriminating testimony. 34 Whenever a witness is forced to give incriminating testimony, there is a significant risk that fruits of that testimony will later be used against him. Further incriminating evidence that is derived from compelled testimony cannot always be traced back to its source: 35 "A witness who suspects that his compelled testimony was used to develop a lead will be hard pressed indeed to ferret out the evidence necessary to prove it, for though the Court puts the burden of proof on the government, the government will have no difficulty in meeting its burden by mere assertion if the witness produces no contrary evidence. The good faith of the prosecuting authorities is thus the sole safeguard of the witness' rights. . . . [E]ven their good faith is not a sufficient safeguard. For the paths of information through the investigative bureaucracy may well be long and winding, and even a prosecutor acting in the best faith cannot be certain that somewhere in the depths of his investigative apparatus, often including hundreds of employees, there was not some prohibited use of the compelled testimony." Kastigar v. United States, 406 U.S., at 469, 92 S.Ct., at 1669 (MARSHALL, J., dissenting). 36 See also Piccirillo v. New York, 400 U.S., at 567-568, 91 S.Ct., at 530 (BRENNAN, J., dissenting from dismissal of certiorari); Speiser v. Randall, 357 U.S. 513, 525, 78 S.Ct. 1332, 1341, 2 L.Ed.2d 1460 (1958). If respondent is not allowed to assert the Fifth Amendment privilege, he may undergo numerous civil depositions, he may be forced to elaborate upon his original testimony,2 and his testimony may be broadly disseminated. As a result, he may face a much greater risk that tainted evidence will be used against him than he initially faced following the compulsion of the grand jury testimony. The opportunity to seek exclusion of tainted evidence is an incomplete protection, for "a court, at the time of the civil testimony, [cannot] pre-determine the decision of the court in a subsequent criminal prosecution on the question whether the Government has met its burden of proving 'that the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony.' " Ante, at 261 (quoting Kastigar v. United States, 406 U.S., at 460, 92 S.Ct., at 1665). Cf. Maness v. Meyers, 419 U.S. 449, 461-463, 95 S.Ct. 584, 592-593, 42 L.Ed.2d 574 (1975). 37 It may be appropriate to subject a witness to these risks when the government has conferred use immunity pursuant to statutory authorization, but the interests supporting compulsion of the testimony are far weaker here. In Kastigar the Court noted that the use immunity statute advanced the government interests in compelling incriminating testimony, id., at 443-444, 446-447, 92 S.Ct., at 1655-1656, 1657, and in leaving open the possibility of prosecuting the witness on the basis of "evidence from legitimate independent sources," id., at 461, 92 S.Ct., at 1665. In this case, however, neither Congress nor the United States Attorney has made a similar expression of government interest.3 The only public interest that would be served by forcing respondent to testify would be that of obtaining testimony relevant to a private antitrust suit.4 Even that interest would not be substantially served.5 38 If he were compelled to answer petitioners' deposition questions, Conboy would face a realistic risk that his testimony would lead to further incriminating evidence that he would be unable to exclude at a subsequent criminal prosecution. The interests underlying the use-immunity statute have no application here, and in my view the general interest in obtaining testimony cannot be considered an adequate substitute for those interests. I therefore join the Court in concluding that the Fifth Amendment does not permit a trial judge in a civil case to compel incriminating testimony solely upon a finding that the testimony would be "directly or indirectly derived from" the witness's previously immunized testimony. 39 Justice BRENNAN, concurring in the judgment. The Court today holds: 40 "We hold that a deponent's civil deposition testimony, closely tracking his prior immunized testimony, is not, without duly authorized assurance of immunity at the time, immunized testimony within the meaning of § 6002, and therefore may not be compelled over a valid assertion of his Fifth Amendment privilege." Ante, at 263-264 (footnote omitted). 41 Justice BLACKMUN's opinion concurring in the judgment likewise states: 42 "In this case, we are asked to decide whether a witness who has testified before a federal grand jury pursuant to a grant of use immunity, 18 U.S.C. §§ 6001-6005, may be forced to testify about the same events in a subsequent civil deposition, despite his assertion of his Fifth Amendment privilege against self-incrimination. I agree with the Court's conclusion that he may not be forced so to testify." Post, at 272. 43 I understand these to be two statements of the same rule,* and I completely agree with both of them. For this reason, I concur in the judgment of the Court. 44 I am not in entire agreement with everything in the majority opinion or in Justice BLACKMUN's opinion. My differences with them, however, are over small matters of approach, and do not go to the substance of their conclusions. Moreover, this case arises in the rather specialized legal setting of use immunity statutes and does not require any broad-ranging analysis beyond the scope of the problem here presented. With these considerations in mind, I do not think it worthwhile to file a lengthy separate opinion setting forth these differences in detail. 45 Justice BLACKMUN, concurring in the judgment. 46 In this case, we are asked to decide whether a witness who has testified before a federal grand jury pursuant to a grant of use immunity, 18 U.S.C. §§ 6001-6005, may be forced to testify about the same events in a subsequent civil deposition, despite his assertion of his Fifth Amendment privilege against self-incrimination. I agree with the Court's conclusion that he may not be forced so to testify. Because I reach this conclusion only by a different route, I write separately to explain my views. 47 * The statute authorizing grants of use immunity, 18 U.S.C. § 6002, provides that a witness may be ordered to testify despite his claim of a Fifth Amendment privilege, but "no testimony or other information compelled under the order (or any information directly or indirectly derived from such testimony or other information) may be used against the witness in any criminal case" (with stated limited exceptions). The Court notes that the parties in this case "make their arguments in terms tracking those of the statute—whether the deposition testimony is 'derived from' the prior testimony." Ante, at 256. In the Court's view, however, "the crux of their dispute is whether the earlier grant of immunity itself compelled Conboy to talk." Ibid. It seems to me that by characterizing the issue in this way, the Court begs the question now before us. The earlier grant of immunity, by itself, obviously does not compel Conboy to testify at a later deposition. It is the District Court that has sought to compel Conboy's testimony. Whether that court may do so is certainly the ultimate issue the Court must decide. But the Court's rephrasing does not bring us closer to the answer. 48 It is, of course, black-letter law that a witness cannot assert a Fifth Amendment privilege not to testify "if the testimony sought cannot possibly be used as a basis for, or in aid of, a criminal prosecution against the witness." Brown v. Walker, 161 U.S. 591, 597, 16 S.Ct. 644, 647, 40 L.Ed. 819 (1896); see Mason v. United States, 244 U.S. 362, 365-366, 37 S.Ct. 621, 622, 61 L.Ed. 1198 (1917). In this case, however, the Court concludes that Conboy has a valid Fifth Amendment privilege "irrespective of whether . . . his [deposition] answers could have been admitted against him at a criminal trial." Ante, at 257, n. 13. The Court never explains the basis for this conclusion, and it seems to me that it is plainly wrong. If Conboy's deposition testimony cannot be used against him in a subsequent criminal prosecution, he cannot assert a Fifth Amendment privilege at his deposition and the District Court may compel him to testify. We must turn to § 6002 to determine whether the testimony can be so used. Section 6002 informs us that when immunity has been granted, the witness is protected against use of "information directly or indirectly derived from [the immunized] testimony." Whether Conboy's deposition testimony is so derived is the real issue before the Court. 49 The Court finds this statutory language irrelevant to its analysis. The Court asserts that petitioners have a "need for admissible evidence," the Government has an "interest in limiting the scope of an immunity grant," and respondent Conboy has an "interest . . . that [his Fifth Amendment] protection be certain." Ante, at 258, 257. The Court then seeks to adjust these interests and arrive at a solution satisfactory to all. While this may be appropriate as a means of setting public policy,1 I cannot agree that it is an appropriate method of statutory interpretation. 50 As with every case involving the construction of a statute, "our starting point must be the language employed by Congress." Reiter v. Sonotone Corp., 442 U.S. 330, 337, 99 S.Ct. 2326, 2330, 60 L.Ed.2d 931 (1979). If we were forced to examine the language of § 6002 without reference to its background and legislative history, the words of the statute might be sufficiently ambiguous so as to require resort to the policy concerns addressed by the Court. In this case, however, "regard for the specific history of the legislative process that culminated in the Act now before us affords more solid ground for giving it appropriate meaning." United States v. Universal C.I.T. Credit Corp., 344 U.S. 218, 222, 73 S.Ct. 227, 230, 97 L.Ed. 260 (1952). II A. 51 This Nation's first use immunity statute was passed by Congress in 1868. It provided that "no answer or other pleading of any party, and no discovery, or evidence obtained by means of any judicial proceeding from any party or witness . . ., shall be given in evidence, or in any manner used against such party or witness . . ., in any court of the United States . . ., in respect to any crime." Act of Feb. 25, 1868, ch. 13, § 1, 15 Stat. 37. In Counselman v. Hitchcock, 142 U.S. 547, 12 S.Ct. 195, 35 L.Ed. 1110 (1892), this Court held that immunity of this type could not be used to compel a witness to testify against himself, because it did not provide protection coextensive with the Fifth Amendment. The Counselman Court reasoned that the statute "protected [the witness] against the use of his 52 testimony against him . . . in any criminal proceeding, in a court of the United States. But it had only that effect. It could not, and would not, prevent the use of his testimony to search out other testimony to be used in evidence against him . . . . It could not prevent the obtaining and the use of witnesses and evidence which should be attributable directly to the testimony he might give under compulsion, and on which he might be convicted, when otherwise, and if he had refused to answer, he could not possibly have been convicted." Id., at 564, 12 S.Ct., at 198. 53 In concluding, the Court stated that "no statute which leaves the party or witness subject to prosecution after he answers the criminating question put to him, can have the effect of supplanting the privilege conferred by the Constitution of the United States." Id., at 585, 12 S.Ct., at 206. 54 Due to this latter statement in the Counselman opinion, Congress and the lower courts assumed that only a broad "transaction" immunity would satisfy the requirements of the Fifth Amendment. Thus, beginning in 1893, Congress enacted a series of statutes giving a witness complete immunity from prosecution for any crime divulged in compelled testimony. This reliance on transaction immunity continued until 1970, when Congress enacted § 6002 as part of the Organized Crime Control Act of 1970, Pub.L. 91-452, 84 Stat. 927. 55 In the meantime, however, the Court decided several cases suggesting that some forms of use immunity would be constitutionally permissible. In Murphy v. Waterfront Comm'n, 378 U.S. 52, 84 S.Ct. 1594, 12 L.Ed.2d 678 (1964), the Court held that a state witness could not be compelled to give testimony that could be incriminating under federal law "unless the compelled testimony and its fruits cannot be used in any manner by federal officials in connection with a criminal prosecution against him." Id., at 79, 84 S.Ct., at 1609. In a footnote, the Court added that once a defendant had been immunized in a state proceeding, "the federal authorities have the burden of showing that their evidence is not tainted by establishing that they had an independent, legitimate source for the disputed evidence." Ibid., n. 18. Several years later, in Gardner v. Broderick, 392 U.S. 273, 276, 88 S.Ct. 1913, 1915, 20 L.Ed.2d 1082 (1968), the Court stated that "[a]nswers may be compelled regardless of the [Fifth Amendment] privilege if there is immunity from federal and state use of the compelled testimony or its fruits in connection with a criminal prosecution against the person testifying." And shortly thereafter, in People v. La Bello, 24 N.Y.2d 598, 602, 301 N.Y.S.2d 544, 547, 249 N.E.2d 412, 414 (1969), the New York Court of Appeals interpreted Murphy and Gardner to hold that Counselman did not bar use immunity statutes, so long as they protected the immunized witness "from the use of his testimony or the fruits thereof." B 56 It was in this context that Congress in 1969 began considering a new type of immunity statute. The House and Senate reports accompanying the Organized Crime Control Act of 1970 make clear that Congress was persuaded by the reasoning of these cases. After quoting from La Bello and discussing Counselman and Murphy at length, see S.Rep. No. 91-617, pp. 52-55 (1969); H.R.Rep. No. 91-1188, pp. 8-11 (1970), the reports state that the statutory immunity provided by § 6002 "is intended to be as broad as, but no broader than, the privilege against self-incrimination. . . . It is designed to reflect the use-restriction immunity concept of Murphy . . . rather [than] the transaction immunity concept of Counselman." S.Rep. No. 91-617, p. 145; H.R.Rep. No. 91-1188, p. 12; see H.R.Rep. No. 91-1549, p. 42 (1970), U.S.Code Cong. & Admin.News 1970, p. 4017. 57 Section 6002's prohibition against the use of compelled testimony or "any information directly or indirectly derived from such testimony" reflected Congress' view of the extent of the Fifth Amendment privilege. According to the House and Senate reports, the phrase was chosen to conform to "present law" on "the use of evidence derivatively obtained." The reports then cite Wong Sun v. United States, 371 U.S. 471, 83 S.Ct. 407, 9 L.Ed.2d 441 (1963), the seminal case on what is commonly known as the "fruits" doctrine, as representing "present law." See S.Rep. No. 91-617, p. 145; H.R.Rep. No. 91-1188, p. 12; H.R.Rep. No. 91-1549, p. 42. In Murphy and Gardner, upon which Congress relied, the Court had used the term "fruits" to describe the constitutional limits on use immunity. References to the "fruits" doctrine are scattered throughout the legislative history, whenever the boundaries of the use immunity statute are discussed.2 In Kastigar v. United States, 406 U.S. 441, 461, 92 S.Ct. 1653, 1665, 32 L.Ed.2d 212 (1972), we recognized that the immunity § 6002 provides is "analogous to the Fifth Amendment requirement in cases of coerced confessions." We noted that § 6002 was modeled on a recommendation from the National Commission on Reform of Federal Criminal Laws, and we quoted with approval a Commission report stating: " 'The proposed immunity is . . . of the same scope as that frequently, even though unintentionally, conferred as the result of constitutional violations by law enforcement officers.' " Id., at 452, n. 36, 92 S.Ct., at 1660, n. 36 (quoting Second Interim Report of the National Commission on Reform of Federal Criminal Laws, Mar. 17, 1969, Working Papers of the Commission 1446 (1970)). 58 In light of this evidence of legislative intent, the phrase "directly or indirectly derived from" in § 6002 cannot be regarded as ambiguous or lacking in meaning. It seems to me that Congress made its intent clear. First, it intended to grant only the minimum protection required by the Constitution. Second, it believed that the protection constitutionally required in cases of compelled testimony was identical to the protection required in cases of coerced statements or evidence otherwise illegally obtained. 59 Respondent Conboy's interpretation of § 6002 is obviously narrower than that offered by petitioners; deposition testimony involving the same subject matter as prior immunized testimony would be protected by the prior grant of use immunity under petitioners' interpretation, but not under Conboy's. Because Congress intended grants of use immunity to be as narrow as possible, we must accept Conboy's interpretation if it is consistent with the Constitution. The question before us, then, is whether a witness' Fifth Amendment rights would be violated if testimony given at a subsequent deposition were not covered by his grant of use immunity. 60 When an incriminating statement has been obtained through coercion, the Fifth Amendment prohibits use of the statement or its "fruits." Congress understood this when it enacted § 6002, and, as the legislative history demonstrates, Congress intended to incorporate the "fruits" doctrine into the statute by use of the phrase "directly or indirectly derived." In order to ascertain whether respondent Conboy's deposition testimony would be "directly or indirectly derived" from his immunized grand jury testimony, and consequently whether Conboy's interpretation of the statute is constitutional, we must determine whether the deposition testimony would be "fruits" as that concept is understood in the context of the Fourth and Fifth Amendments.3 III 61 In Wong Sun v. United States, supra, the Court held that a statement following an illegal arrest must be suppressed as "fruits" of the arrest unless it results from "an intervening independent act of a free will," and is "sufficiently an act of free will to purge the primary taint of the unlawful invasion." 371 U.S., at 486, 83 S.Ct., at 416. In Harrison v. United States, 392 U.S. 219, 222-224, 88 S.Ct. 2008, 2010-2011, 20 L.Ed.2d 1047 (1968), the Court applied a similar standard to statements following an illegally obtained confession. Our more recent cases have adhered to this test. See, e.g., Rawlings v. Kentucky, 448 U.S. 98, 107-110, 100 S.Ct. 2556, 2562-2564, 65 L.Ed.2d 633 (1980); Brown v. Illinois, 422 U.S. 590, 600-604, 95 S.Ct. 2254, 2260-2262, 45 L.Ed.2d 416 (1975). In determining whether this standard is met, we examine a range of factors including the speaker's knowledge of his Fifth Amendment rights; the temporal proximity of the constitutional violation and the subsequent statement; the nature of the violation and of the Government's involvement; and, of course, the voluntariness of the statement. See id., at 603-604, 95 S.Ct., at 2261-2262. In brief, the issue is whether the speaker has voluntarily chosen to make the later statement, uninfluenced by the fact that prior statements have been compelled.4 62 I find little difficulty in concluding that if a witness in Conboy's position were to testify during his civil deposition, his statements would not be "fruits" of his previous immunized testimony.5 In this case, Conboy attended his deposition accompanied by a lawyer. He was obviously aware of his Fifth Amendment rights, and he asserted them with vigor. There is no suggestion that Conboy was under a misapprehension about the relationship between his immunized testimony and his civil deposition. The deposition took place long after the conclusion of the immunized testimony, and Conboy did not remain under the impression that his testimony was being compelled by the Justice Department. From his past experience before the grand jury, Conboy knew that each time the Justice Department required his testimony, it provided a fresh grant of use immunity. Government attorneys were not involved in this civil case, and no fresh grant of immunity had been obtained. Under the circumstances, there was no danger that Conboy would inadvertently incriminate himself under some lingering compulsion of the prosecuting authorities. Any statement he made would have been an independent act of free will. Consequently, had Conboy answered the deposition questions, his testimony would not have been protected by the original immunity grant because it would not have been directly or indirectly derived from his immunized testimony. 63 In my view, a prior grant of use immunity could never justify compelling a witness' testimony over a claim of Fifth Amendment privilege at a subsequent civil deposition. Although not every witness will be as well-informed as Conboy, any witness who asserts the privilege necessarily engages in an independent act of free will. The assertion of the privilege should signal the judge supervising the civil proceedings that the testimony may well not be "derived from" the immunity grant.6 Although the compelled testimony would be inadmissible at a subsequent criminal trial,7 I agree with the Court that a witness should not be forced to rely upon the uncertainties of a later motion to suppress. This would indeed " ' "let the cat out" with no assurance whatever of putting it back.' " Ante, at 262 (quoting Maness v. Meyers, 419 U.S. 449, 463, 95 S.Ct. 584, 593, 42 L.Ed.2d 574 (1975)). 64 I do not mean to suggest, however, that whenever a witness immunized in prior proceedings testifies at a civil deposition without asserting a Fifth Amendment privilege, his testimony automatically should be admissible against him in a subsequent criminal prosecution. If there is a subsequent prosecution and the Government seeks to introduce deposition testimony of this sort, the judge in the criminal case should determine whether, under the circumstances, the deposition testimony is inadmissible as "derived from" the prior immunized statements. If the witness reasonably believed that his prior grant of immunity protected his testimony, the testimony might well be derived from the immunity grant under the standards I have set forth above. If, on the other hand, the deposition testimony was a truly independent act of free will, it would be admissible in any later prosecution. 65 Justice STEVENS, with whom Justice O'CONNOR joins, dissenting. 66 A witness in a judicial proceeding has a duty to answer proper questions. The witness cannot, however, be compelled to incriminate himself. If a witness believes a truthful response to a question could be used against him in a subsequent criminal proceeding, or might lead to the discovery of incriminating evidence, he may assert his constitutional right to remain silent. When such an assertion is made, a judge must evaluate the asserted risk. If it clearly appears that the answer could not be used against the witness in a subsequent criminal proceeding and could not provide a prosecutor with any information that he does not already have, the witness must speak. This case concerns a witness's refusal to give answers that could not incriminate him. 67 The Court today holds that the existence of a valid Fifth Amendment privilege does not depend on whether a truthful answer would be incriminating. The Court does not dispute the fact that neither the respondent's answers during the deposition in this case, nor any information discovered on the basis of those answers, could be used against him in a subsequent criminal proceeding. Ante, at 257, n. 13. Nevertheless, the Court holds that the Fifth Amendment empowers the respondent to refuse to testify. The opinion of the Court stresses two interests: "the Government's need for admissible evidence" in a future effort to prosecute the respondent, and "the individual's interest in avoiding self-incrimination." Ante, at 256. It holds that potential threats to those interests create a Fifth Amendment privilege in this case. 68 I am frankly puzzled by this analysis. The Government's supposed desire to introduce evidence in a future proceeding should be irrelevant if the Government has already forsworn the right to introduce that evidence by a prior grant of immunity. And, as far as the deponent's interest in avoiding self-incrimination is concerned, "he should be indifferent between the protection afforded by silence and that afforded by immunity," ante, at 257. Thus, whether analyzed from the point of view of the prosecutor or the witness, the same question must be answered: whether the statutory immunity that has already attached to respondent's grand jury testimony precludes the Government, or any other prosecutor, from using the respondent's deposition answers against him in any criminal case. That question requires an analysis not of whether the deposition answers are "immunized 'testimony,' " ante, at 250, but rather of whether the answers would be "directly or indirectly derived from [his grand jury] testimony" within the meaning of the use immunity statute. Because I think it clear that they would be so derived, I respectfully dissent. 69 * Respondent has been a witness in two separate proceedings. In January 1978, he was subpoenaed to testify before a federal grand jury investigating a violation of the Sherman Act. Because he was a participant in the price-fixing arrangements under review, he asserted his constitutional privilege against being compelled to be a witness against himself.1 The prosecutor then invoked his authority under the Organized Crime Control Act of 1970,2 and a federal judge ordered the respondent to testify in exchange for a grant of immunity. 70 In May 1981, respondent was subpoenaed to appear in a second proceeding.3 At that deposition proceeding, respondent was asked the same questions that he had been asked before the grand jury. Everyone agrees that the questions were derived from the transcript of his grand jury testimony, and no one disputes the fact that truthful answers to those questions would merely have confirmed information that was already recorded in the grand jury transcript.4 It is therefore logical to inquire, as the Court below did, whether ratification of the prior immunized testimony would subject the respondent to a new risk of prosecution. 71 The plain language of the Organized Crime Control Act protects the witness from that risk. The law provides: 72 "[N]o testimony or other information compelled under the order (or any information directly or indirectly derived from such testimony or other information ) may be used against the witness in any criminal case, . . . ." 18 U.S.C. § 6002 (emphasis added). 73 When a witness appears at a second proceeding and is asked whether the information that he was previously compelled to disclose to the grand jury was true, his responses are quite plainly "information directly or indirectly derived from such testimony." This seems particularly obvious when the interrogator's only basis for his questions is the transcript of the grand jury proceeding. 74 This natural construction of the statute was endorsed by the Government immediately after the Organized Crime Control Act took effect. In a memorandum explaining the statute to United States Attorneys, the Assistant Attorney General in charge of the Criminal Division explained that it allowed an immunized witness to be prosecuted "if it can be clearly established that independent evidence standing alone is in fact the sole basis of the contemplated prosecution." Dept. of Justice Memo No. 595, Supp. 1, Sept. 2, 1971, p. 5. He emphasized that, "Although the Government may prosecute the witness on the basis of similar evidence obtained independently of the witness's testimony in a rare case where such an independent source develops, as a practical matter it will be difficult for the Government to prove an independent derivation, especially if the information first was divulged in the witness's testimony." Id., at 5, n. 4 (emphasis in original). And when the Solicitor General of the United States later appeared before this Court to defend the Act's constitutionality, he based his argument in part on the proposition that the words "directly or indirectly derived" were intended to create an "extended use immunity" and should be construed broadly.5 75 This Court accepted the Solicitor General's argument. It upheld the use immunity statute after construing it to provide protection commensurate with the protection resulting from the invocation of the privilege itself:6 76 "The statute provides a sweeping prohibition of any use, direct or indirect, of the compelled testimony and any information derived therefrom . . . . [The] burden of proof, which we reaffirm as appropriate, is not limited to a negation of taint; rather, it imposes on the prosecution the affirmative duty to prove that the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony." Kastigar v. United States, 406 U.S. 441, 460, 92 S.Ct. 1653, 1664, 32 L.Ed.2d 212 (emphasis added). 77 We held that evidence may be used in a subsequent prosecution only if the Government successfully demonstrates that it would have obtained that evidence even if the witness had never testified before the grand jury. See id., at 458-459, 92 S.Ct., at 1663-1664; Murphy v. Waterfront Commission, 378 U.S. 52, 79, 84 S.Ct. 1594, 1609, 12 L.Ed.2d 678. 78 The questions that were propounded to the respondent at his deposition in this case called for answers that were presumptively within the scope of the statutory immunity. That presumption would protect him from the use in a subsequent criminal prosecution of any of the information contained in his answers unless it could be shown that the information would have been obtained even if the witness had never testified before the grand jury. Nothing in this record suggests that answers to questions based entirely on the grand jury transcript were not "fruits" of the prior testimony.7 79 The district judge properly ruled that the respondent's answers could not have been introduced against him at a subsequent criminal prosecution any more than the original testimony could have been. Moreover, if the respondent's answers would be a necessary link in a chain that led to other information, then that information would also be "derived" from the prior testimony and likewise could not be used at a subsequent criminal prosecution. The witness therefore had no greater right to assert a constitutional privilege against self-incrimination in the second proceeding than he had in the grand jury proceeding itself. II 80 Although the Court does not dispute the fact that respondent's answers were within the scope of the immunity grant, ante, at 257, n. 13, it nevertheless places a great deal of reliance on "the Government's interest in limiting the scope of an immunity grant," ante, at 258. In my judgment the Court commits a triple error in this analysis. First, it uses policy judgments that could at most affect an interpretation of the use immunity statute in other cases to justify its erroneous interpretation of the Fifth Amendment in this case. Second, it misunderstands the prosecutorial interest in how the statute should be interpreted in those other cases. And third, it overlooks the obvious enforcement costs of its holding in this case. The first error does not need elaboration; the second two do. 81 A federal prosecutor does not offer immunity to a suspected criminal unless he expects to obtain important testimony that would not otherwise be available. The prosecutor realizes that, in almost all cases, an offer of immunity—even of use immunity means sacrificing the chance to prosecute the witness for his own role in the criminal enterprise.8 The question is what kind of return society will get on the prosecutor's investment in immunity. Once the prosecutor pays the immunity price, he will normally wish to probe deeply for evidence that will implicate the witness's criminal associates as thoroughly as possible. The primary law enforcement interest is to maximize the amount of information that the witness provides. A broad construction of the immunity grant serves that purpose; a narrow construction can only motivate witnesses to be as unresponsive as possible.9 82 Yet the Court suggests that the Government prosecutors take a different attitude towards immunized witnesses. Even though the Government itself has not promoted such a view in the deposition proceedings in this case or by argument in this Court,10 the opinion of the Court suggests that when a prosecutor immunizes a witness in order to obtain particular information, he harbors an intent to indict the witness afterwards and would therefore prefer that the witness remain in the same peril of prosecution as before being immunized.11 Yet it defies human nature to presume that the witness would be just as cooperative during a 24-hour truce, knowing that hostilities will resume immediately thereafter, as he would be after signing a peace treaty. 83 Nor does the Court explain its assertion that applying the statute as it is written and as it was construed in Kastigar "in effect could invest the deponent with transactional immunity." Ante, at 260. Transactional immunity is not at all the issue here. Transactional immunity would require the prosecutor to forfeit an open-and-shut case that he had already built independently. Use immunity, as explained in Kastigar and as granted to the respondent, allows the prosecutor to retain that case.12 I have found absolutely no evidence, and the Court cites none today, to support the implicit suggestion that Congress substituted "use immunity" for "transactional immunity" in order to allow prosecutors to take advantage of subsequent repetitions of immunized testimony.13 84 The Court's reference to "transactional immunity" suggests a fear that ordering the respondent to answer a deposition question may somehow jeopardize legitimate efforts to prosecute him. Consideration of the facts of this particular case demonstrates that the Court's apparent fear is baseless. Unless some prosecutor already has an independent basis for prosecuting the respondent—and nothing in the record suggests that any such independent basis exists—the Government has already agreed that he will not be prosecuted for engaging in illegal price discussions with Fred Renshaw and Dick Herman of the Alton Box Board. If, at the deposition, he is required to confirm that such discussions took place, how can that confirmation affect his criminal liability? If some prosecutor has a demonstrably independent basis for proving the respondent's participation in the discussions, his confirmation will not make that basis any less demonstrably independent.14 And if that prosecutor has an independent basis for showing that the respondent participated in the discussions, that basis will be no less demonstrably independent if the respondent is required to identify the time, place, and other persons who participated in the discussions. 85 Furthermore, one should not overlook the societal costs—law enforcement costs—of the Court's expansion of the Fifth Amendment. The public interest in obtaining the full and candid testimony of a witness with knowledge of the inner workings of a price-fixing conspiracy is both real and significant.15 Conceivably, a relatively brief account of the basic structure of the conspiracy might have been sufficient to persuade the grand jury to indict other parties and also to persuade those defendants to plead guilty or to enter into some other settlement with the Government.16 Even if a grand jury transcript is confined to a brief description of a pricefixing arrangement, for example, the public interest may well be served by allowing private parties who have been injured thereby to inquire into the details of the arrangement.17 86 The Court assumes that the scope of the Fifth Amendment privilege in this case should be expanded in order to serve society's law enforcement interests. I do not accept this mode of Fifth Amendment interpretation. But even if I did, I would find the theoretical risk that compelled testimony could hamper a potential prosecution to be plainly outweighed by the enforcement interest in allowing the deposition to go forward. And, significantly, even the slight theoretical risk that concerns the Court is not presented by this case, in which no new incriminating information is called for by the deposition questions. III 87 The Court makes the curious argument that the Fifth Amendment privilege must extend to testimony that could not incriminate a witness because otherwise the witness will be put to the risk of "predicting" whether a court in a later criminal proceeding would agree that the testimony was within the scope of the immunity. Ante, at 261-262. I do not agree that the "risk" that troubles the Court is entitled to protection under the Fifth Amendment. 88 A witness in the respondent's chair at a deposition can do one of two things: he can answer or he can assert a Fifth Amendment privilege. If he answers, he is obviously more "at risk" under Justice BLACKMUN's narrow view of the use immunity statute than under the broad one adopted in Kastigar. For that reason, the Court does not dispute the fact that if the respondent had answered the deposition questions in this case, his answers could not be used against him.18 The Court and I part company, however, in reacting to the risks that the witness faces if he asserts a Fifth Amendment privilege. 89 If the court supervising the deposition concludes that an answer is not "directly or indirectly derived" from prior immunized testimony, it must uphold the assertion of the Fifth Amendment privilege under both my analysis and the Court's. If, on the other hand, the supervising court concludes that the answer is "directly or indirectly derived" from immunized testimony, I believe it must reject the asserted privilege. The Court disagrees, for two analytically distinct reasons. 90 First, the Court suggests that the supervising court might make a mistake in deciding whether the testimony is directly or indirectly derived. It suggests that in this case Judge Singleton might not have been able to "pre-determine the decision of the court in a subsequent criminal prosecution on the question whether the Government has met its burden of proving 'that the evidence it proposes to use is derived from a legitimate source wholly independent of the compelled testimony.' " Ante, at 261. The Court does not explain what sort of evidence the prosecutor might subsequently be able to produce that would show the answers to be "wholly independent;" indeed, it is difficult to conceive how such evidence could possibly exist in this case. More importantly, the Court does not explain why the risk of error in this situation is different from the identical risk that exists whenever a privilege is asserted. The Court's argument would require every trial judge always to honor a claim of privilege, no matter how obvious it may be that the claim lacks merit, to guard against being found wrong later.19 91 Second, the Court suggests, with Justice MARSHALL, that it would be unfair to require the witness to answer because " '[f]urther incriminating evidence that is derived from compelled testimony cannot always be traced back to its source.' " Ante, at 262, n. 22, quoting ante, at 268 . Yet such an argument applies with equal force to the entire concept of use immunity. Our holding in Kastigar rests squarely on the proposition that one may not assert a Fifth Amendment privilege on the basis of the risk that evidence might not be traced back to its source. Cf. Kastigar, supra, 406 U.S. at 468-471, 92 S.Ct., at 1668-1670 (MARSHALL, J., dissenting). Even if the Court were now prepared to retreat from that proposition, this case is surely not the proper vehicle. The respondent here was asked only to ratify or confirm facts that were already known. On this record, it clearly appears that the answers to the specific questions asked could not possibly provide any basis for prosecution, or even for investigation, beyond what was already provided by the grand jury testimony.20 92 In summary, it is perfectly clear on this record that the respondent's deposition testimony (a) would be protected by the statutory immunity; (b) could not be used against respondent in a subsequent criminal proceeding; and (c) could not provide a prosecutor with any information he does not already have. A concern that a court might not decide some other case correctly cannot justify an incorrect disposition of the case before us. 93 I respectfully dissent. 1 The propriety of the District Court's release of grand jury materials to the civil parties is not before the Court. 2 An example of this three-question pattern is as follows: Q. Who did you have price communications with at Alton Box Board? * * * * * Q. Is it not the fact that you had price communications with Fred Renshaw and Dick Herman . . .? * * * * * Q. Did you not so testify in your government interview of January 10, 1978? Joint Appendix 29-31. 3 Chief Judge John V. Singleton, Jr. of the District Court for the Southern District of Texas expressly exercised the powers of the District Court for the Northern District of Illinois pursuant to 28 U.S.C. § 1407(b). The contempt hearing was conducted by telephone with his chambers in Houston. 4 The correctness of the Court of Appeals' conclusion that Conboy could assert a Fifth Amendment privilege, absent some immunity, is not before us. 5 A United States Attorney declined to authorize immunity grants in connection with the civil depositions here. 6 Compare In re Corrugated Container Antitrust Litigation, Appeal of Fleischacker, 644 F.2d 70, 75 (CA2 1981) (deposition answers immunized), and Little Rock School District v. Borden, Inc., 632 F.2d 700, 705 (CA8 1980) (same), with In re Corrugated Container Anti-Trust Litigation, Appeal of Franey, 620 F.2d 1086, 1095 (CA5 1980) (answers not immunized), cert. denied, 449 U.S. 1102, 101 S.Ct. 897, 66 L.Ed.2d 827 (1981). 7 See United States v. Calandra, 414 U.S. 338, 345, 94 S.Ct. 613, 618, 38 L.Ed.2d 561 (1974); United States v. Mara, 410 U.S. 19, 41, 93 S.Ct. 774, 785, 35 L.Ed.2d 99 (1973) (MARSHALL, J., dissenting); Kastigar v. United States, 406 U.S. 441, 443-444, 92 S.Ct. 1653, 1655-1656, 32 L.Ed.2d 212 (1973); Murphy, 378 U.S., at 93-94, 84 S.Ct., at 1610-1611 (WHITE, J., concurring); Blackmer v. United States, 284 U.S. 421, 438, 52 S.Ct. 252, 255, 76 L.Ed. 375 (1932); Blair v. United States, 250 U.S. 273, 281, 39 S.Ct. 468, 471, 63 L.Ed. 979 (1919); Brown v. Walker, 161 U.S. 591, 600, 16 S.Ct. 644, 648, 40 L.Ed. 819 (1896). 8 In Murphy, Justice WHITE stated that "[i]mmunity must be as broad as, but not harmfully and wastefully broader than, the privilege against self-incrimination." 378 U.S., at 107, 84 S.Ct., at 1618 (concurring opinion) (quoted with approval in 116 Cong.Rec. 35,291 (1970) (statement of Rep. Poff)). In its committee report, the House explained that § 6002 was not to provide an "immunity bath," but was to be "no broader than" the Fifth Amendment privilege. H.R. No. 91-1549, p. 5 (1970), U.S.Code Cong. & Admin.News 1970, p. 4007. 9 Section 6002 provides: Whenever a witness refuses, on the basis of his privilege against self-incrimination, to testify or provide other information in a proceeding before or ancillary to— (1) a court or grand jury of the United States, (2) an agency of the United States, or (3) either House of Congress, a joint committee of the two Houses, or a committee or a subcommittee of either House and the person presiding over the proceeding communicates to the witness an order issued under this part, the witness may not refuse to comply with the order on the basis of his privilege against self-incrimination; but no testimony or other information compelled under the order (or any information directly or indirectly derived from such testimony or other information) may be used against the witness in any criminal case, except a prosecution for perjury, giving a false statement, or otherwise failing to comply with the order. Section 6001(2) defines "other information" to include "any book, paper, document, record, recording, or other material." 10 Section 6003 states: (a) In the case of any individual who has been or may be called to testify or provide other information at any proceeding before or ancillary to a court of the United States or a grand jury of the United States, the United States district court for the judicial district in which the proceeding is or may be held shall issue, in accordance with subsection (b) of this section, upon the request of the United States attorney for such district, an order requiring such individual to give testimony or provide other information which he refuses to give or provide on the basis of his privilege against self-incrimination, such order to become effective as provided in section 6002 of this part. (b) A United States attorney may, with the approval of the Attorney General, the Deputy Attorney General, or any designated Assistant Attorney General, request an order under subsection (a) of this section when in his judgment— (1) the testimony or other information from such individual may be necessary to the public interest; and (2) such individual has refused or is likely to refuse to testify or provide other information on the basis of his privilege against self-incrimination. 11 Congress foresaw the courts as playing only a minor role in the immunizing process: "The court's role in granting the order is merely to find the facts on which the order is predicated." H.R.Rep. No. 91-1549, supra note 8, at 43, U.S.Code Cong. & Admin.News 1970, p. 4018; H.R.Rep. No. 91-1188, p. 13 (1970). See 116 Cong.Rec. 35,291 (1970) (statement of Rep.Poff). Cf. President's Commission on Law Enforcement and Administration of Justice, The Challenge of Crime in a Free Society 141 (1967) (recommending that "[i]mmunity should be granted only with the prior approval of the jurisdiction's chief prosecuting officer."). 12 See Brief for Petitioners 19 ("Conboy had no Fifth Amendment privilege to assert because of the coextensive protection provided by the immunity statute."); Reply Brief for Petitioners 12 ("[R]equiring a witness to answer questions a second time that were previously answered under a grant of immunity does not result in an expansion of the original immunity grant."). 13 Justice BLACKMUN, concurring in the Court's judgment, assumes that Conboy had a right to remain silent at the deposition, which by definition assumes the immunity order itself does not compel a witness to testify at a civil deposition. He discusses the "fruits" doctrine where a witness's testimony at a deposition is "an independent act of free will" and concludes that "had Conboy answered the deposition questions, his testimony would not have been protected by the original immunity grant. . . ." Post, at 280. We have no occasion to address this hypothetical. The issue is whether Conboy can be compelled to testify—i.e., whether the immunity order compels him to track his prior testimony at the civil deposition—over the assertion of his Fifth Amendment rights. If, as we conclude, the original grant of immunity does not extend to the subsequent civil proceeding, then the trial judge lacks authority to compel Conboy to testify over the assertion of his Fifth Amendment privilege. This is so irrespective of whether, had he testified at the deposition rather than asserting the privilege, his answers could have been admitted against him at a criminal trial. We therefore need not now decide the extent to which civil deposition testimony, freely given by a witness in Conboy's position, is "directly or indirectly derived" from prior grand jury testimony. As Justice BLACKMUN'S opinion makes a factual analysis under the fruits doctrine, it appears to leave open the possibility that the outcome in a subsequent criminal prosecution of the deponent may be different in a future case because of differences in the factual record. He nevertheless concludes, as do we, that District Courts are without power to compel a civil deponent to testify over a valid assertion of his Fifth Amendment right, absent a separate grant of immunity pursuant to § 6002. 14 Besides the costs of testifying against close associates, any witness increases the risk of committing perjury the more he talks. Cf. 18 U.S.C. § 6002 (perjured testimony not immunized). 15 The extreme case would be where petitioners read the entire immunized grand-jury transcript; then ask the witness if that is his testimony; and he answers simply "Yes." 16 Direct examination may not be as limited as petitioners assume. The District Court's civil contempt order stated that the questions asked in the deposition "were taken directly" from the immunized transcripts, but did not define exactly what deposition questions petitioners could ask. Other Courts of Appeals have permitted direct questioning to go beyond mere restatements of the prior testimony. See In re Corrugated Container Antitrust Litigation, Appeal of Fleischacker, 644 F.2d 70, 79 (CA2 1981) (compelling answers to questions "concerning specific subjects that actually were touched upon by questions appearing in the transcript of the immunized testimony"); Little Rock School District v. Borden, Inc., 632 F.2d 700, 705 (CA8 1980) (compelling answers as long as deposition questions confined to " 'the same time, geographical and substantive frame work as the [witness' immunized] grand jury testimony' ") (quoting Appeal of Starkey, 600 F.2d 1043, 1048 (CA8 1979)). The dissenting opinion of Justice STEVENS apparently does not attempt to indicate when questioning will exceed proper limits. 17 For purposes of this case, we assume that the grand jury transcripts are inadmissible as evidence in a civil trial because the testimony is not subject to cross examination. Cf. Fed.R.Evid. 803(8) (hearsay exception for certain public records); Fed.R.Evid. 804(a)(1) (witness unavailable when exempted from testifying on ground of privilege); Fed.R.Evid. 804(b)(1) (former testimony admissible when witness unavailable and the party against whom the testimony is now offered had an opportunity for cross examination). 18 Cf. Fed.R.Civ.P. 26(b)(1) (stating that depositions may be taken "if the information sought appears reasonably calculated to lead to the discovery of admissible evidence"); Fed.R.Civ.P. 30(c) (allowing cross examination at depositions); Fed.R.Civ.P. 32(a) (deposition "admissible under the rules of evidence applied as though the witness were then present and testifying"); Fed.R.Evid. 804(b)(1) (deposition admissible if the party against whom the testimony is now offered in a civil action had an opportunity to develop testimony by cross examination). 19 See United States v. Cardillo, 316 F.2d 606, 611 (CA2 1963) (in determining whether testimony of a witness who invokes the privilege during cross examination may be used against defendant, court draws a distinction between cases in which the assertion of the privilege merely precludes inquiry into collateral matters that bear on credibility of witnesses and those in which assertion prevents inquiry into matters about which witness testified on direct). 20 We need not decide whether United States Attorneys, when designated by the Attorney General, presently have authority to immunize the testimony of a witness in a civil proceeding when the Government determines that the public interest would be served. 21 None of the tests set forth by Courts of Appeals that have adopted petitioners' interpretation of § 6002, provides deponents with certain guidance as to when they must talk and when they must not. See note 16, supra. 22 Cf. post, at 268 (MARSHALL, J., concurring) ("Further incriminating evidence that is derived from compelled testimony cannot always be traced back to its source[.]"); note 14, supra (increasing risk of harm and perjury); note 23, infra (increasing exposure to civil liability). 23 The dissent minimizes the enforcement interest that our construction of § 6002 protects, post, 288-290, contending that we "misunderstood the prosecutorial interest," id., at 288. We note, however, that by conceding that there is some "risk" that the deponent's testimony may hamper a prosecution, id., at 293, the dissent concedes that its interpretation of § 6002 provides at least somewhat broader immunity than Congress intended. Moreover, the dissent overlooks the possible difficulty of securing the cooperation of individuals such as Conboy who may be more reluctant to testify in the immunized proceedings if they know that later deposition testimony may increase their exposure to civil liability. Finally, in the dissent's judgment, "the theoretical risk that compelled testimony could hamper a potential prosecution [is] plainly outweighed by the enforcement interest in allowing the deposition to go forward." Id., . See also id., at 289. This, however, is a judgment reserved for officials of the Department of Justice, not the federal courts, to make on a case-by-case basis. 24 Our holding is limited to precluding District Courts from compelling testimony in a civil deposition over a valid assertion of the Fifth Amendment privilege, absent a specific assurance of immunity for such testimony omission. 1 A witness is generally entitled to invoke the Fifth Amendment privilege against self-incrimination whenever there is a realistic possibility that his answer to a question can be used in any way to convict him of a crime. It need not be probable that a criminal prosecution will be brought or that the witness's answer will be introduced in a later prosecution; the witness need only show a realistic possibility that his answer will be used against him. Moreover, the Fifth Amendment forbids not only the compulsion of testimony that would itself be admissible in a criminal prosecution, but also the compulsion of testimony, whether or not itself admissible, that may aid in the development of other incriminating evidence that can be used at trial. See Hoffman v. United States, 341 U.S. 479, 486, 71 S.Ct. 814, 818, 95 L.Ed. 1118 (1951). The privilege is inapplicable only "if the testimony sought cannot possibly be used as a basis for, or in aid of, a criminal prosecution against the witness." Brown v. Walker, 161 U.S. 591, 597, 16 S.Ct. 644, 647, 40 L.Ed. 819 (1896). It has long been recognized that the court may require a witness to give testimony, including testimony that admits to involvement in a criminal act, when there is no possibility of future criminal charges being brought against the witness. For example, a witness may be compelled to testify concerning his involvement in a crime when he is protected from later prosecution by the Double Jeopardy Clause, see, e.g., Reina v. United States, 364 U.S. 507, 513, 81 S.Ct. 260, 264, 5 L.Ed.2d 249 (1960) (dictum), by the applicable statute of limitations, see, e.g., United States v. Goodman, 289 F.2d 256, 259 (CA4 1961), or by a pardon, see Brown v. Walker, 161 U.S. 591, 599-600, 16 S.Ct. 644, 647-648, 40 L.Ed. 819 (1896). As Justice BRENNAN indicated in his dissenting opinion in Piccirillo v. New York, 400 U.S. 548, 564-565, 91 S.Ct. 520, 528-529, 27 L.Ed.2d 596 (1971) (dissenting from dismissal of certiorari), this limitation upon the privilege against self-incrimination is derived from the language of the Fifth Amendment: "Implicitly, of course, 'in any criminal case' suggests a limitation upon the reach of the privilege . . . . [I]f there is no possibility of a criminal case, then the privilege would not apply. And that is precisely the basis on which this Court has consistently upheld grants of immunity from Brown v. Walker, 161 U.S. 591, 16 S.Ct. 644, 40 L.Ed. 819 (1896), to Ullman v. United States, 350 U.S. 422, 76 S.Ct. 497, 100 L.Ed. 511 (1956)." It has also been recognized that a court may compel a witness to testify when his answers could neither implicate him in any criminal conduct nor possibly lead to the discovery of past criminal conduct. See Hoffman v. United States, 341 U.S. 479, 71 S.Ct. 814, 95 L.Ed. 1118 (1951); Heicke v. United States, 227 U.S. 131, 142-145, 33 S.Ct. 226, 227-229, 57 L.Ed. 450 (1913). This limitation, too, is implicit in the language of the constitutional guarantee, since a witness who has been forced to provide testimony that cannot incriminate him has not in any meaningful sense been "compelled in any criminal case to be a witness against himself." In this case, the Fifth Amendment privilege is fully applicable. Respondent remains subject to criminal prosecution, and his answers to the deposition questions asked by petitioners' attorneys would both implicate him in criminal conduct and tend to lead to further incriminating information. 2 The questioning at the deposition went well beyond mere ratification of the accuracy of the grand jury transcript. Conboy was also called upon to answer again the identical questions asked before the grand jury. While it may be true that petitioners expected Conboy "only to ratify or confirm facts that were already known," post, at 295 (STEVENS, J., dissenting), there was no assurance in this case that Conboy's answers, based upon his current recollection of events, would not provide details that were absent from his prior grand jury testimony. 3 As the Court observes, the government interests that supported the compulsion of incriminating testimony in Kastigar would be undermined by the compulsion of respondent's testimony in this case. The government interest in preserving the chance to prosecute respondent in the future based on "legitimate independent" evidence would be compromised by the creation of additional immunized testimony. Ante, at 260-261. 4 Even if the U.S. Attorney consented to the trial judge's compulsion of respondent's answers, the judge's action might be improper. As the Court notes, it is an open question whether the Government has statutory authority "to immunize the testimony of a witness in a civil proceeding when the Government determines that the public interest would be served." Ante, at n. 20 (emphasis added). Moreover, the constitutionality of such a statutory authorization remains open to doubt. Cf. Garrity v. New Jersey, 385 U.S. 493, 496, 87 S.Ct. 616, 618, 17 L.Ed.2d 562 (1967) (declining to consider constitutionality of forfeiture-of-office statute which, in effect, allowed the authorities to compel a public officer, under threat of removal from office, to provide incriminating testimony in exchange for immunity from use or derivative use of that testimony at a criminal proceeding). Indeed, this Court has not yet spoken as to the circumstances under which a trial court in a criminal case may compel a defense witness to testify concerning questions as to which he had previously testified before the grand jury or may compel the Government to secure such a witness's testimony by granting him immunity. Cf. United States v. Praetorius, 622 F.2d 1054, 1064 (CA2 1980); United States v. Morrison, 535 F.2d 223, 229 (CA3 1976); United States v. Alessio, 528 F.2d 1079 (CA9 1976); Earl v. United States, 361 F.2d 531, 534, n. 1 (CADC 1966) (Burger, J.). 5 It is questionable whether the deposition testimony would be admissible at trial, in light of the limits that might have to be placed on cross-examination by the other civil litigants. Ante, at 259-260. Nor would respondent's answers help petitioners obtain further relevant information, since petitioners already have access to respondent's grand jury testimony. * While the majority's statement of the holding is formally limited to the situation where a deponent's deposition testimony "closely track[s] his prior immunized testimony," ante, at 263, I do not take that to be a substantive difference between its formulation and that of Justice BLACKMUN. As both the majority's opinion and Justice STEVENS' dissenting opinion, post, p. 282 make clear, the "closely tracking" situation is the strongest possible case for finding that the deposition testimony is derived from the prior immunized testimony. Hence, to hold that a deponent may assert his privilege in this case is necessarily to hold that he may do so in all cases, as Justice BLACKMUN states explicitly. 1 As Justice STEVENS' dissent demonstrates, the interests of the Government and the parties are not at all as clear as the Court asserts. Reliance on these interests is particularly inappropriate in a case such as this one, where the Government is not a party and we can only speculate about which interpretation of the statute would best serve the Government's interest in law enforcement. 2 See, e.g., S.Rep. No. 91-617, p. 108 (§ 6002 "is a restriction against use of incriminating disclosures or their fruits"); Hearings on S. 30, et al. before the Subcommittee on Criminal Laws and Procedures of the Senate Committee on the Judiciary, 91st Cong., 1st Sess., 216 (1969) (report of New York County Lawyers' Association) (under § 6002, the "compelled testimony or its fruits may not be used against the witness"); id., at 281 (statement of Rep. Poff) (rule of § 6002 is "similar to the exclusionary rule which is now applied to evidence assembled in violation of various constitutional rights"); id., at 506 (statement of Sen. McClellan) (use immunity statutes can be made constitutional "through the use of the fruit of the poisonous tree process of derivative suppression, an analogy borrowed from fourth amendment illegally obtained evidence cases"). 3 The considerations underlying the Fifth Amendment "fruits" doctrine are not necessarily the same as those relevant in the Fourth Amendment context. With respect to the issue before us, however, Fourth Amendment "fruits" cases provide us with guidance in determining whether a witness' deposition testimony is "derived from" prior immunized testimony within the meaning of § 6002. 4 In Kastigar v. United States, 406 U.S. 441, 459, 92 S.Ct. 1653, 1664, 32 L.Ed.2d 212 (1972), we recognized that Congress intended § 6002 to provide the minimum protection required by the Constitution. Wong Sun and its progeny establish that the "fruits" doctrine provides all the protection the Constitution requires. Thus, although my analysis is framed in terms of constitutional standards, the issue here of what the Constitution requires is not different from the issue of what Congress intended. 5 My analysis is necessarily limited to the choices facing a witness prior to the threat of contempt by the District Court. The witness cannot be held in contempt unless the testimony sought is protected by the grant of use immunity or, in other words, unless it would be "fruits." The question whether the testimony would be "fruits" thus cannot turn on whether the District Court has issued a contempt order. 6 I agree with Justice STEVENS that the existence of a witness' Fifth Amendment privilege does not depend on his decision to assert the privilege. See post, at 287, n. 7. Nevertheless, the state of mind of the witness is relevant to a "fruits" inquiry, because a witness' statements are "fruits" only if they do not result from an independent act of free will. Cf. Harrison v. United States, 392 U.S. 219, 222-224, 88 S.Ct. 2008, 2010-2011, 20 L.Ed.2d 1047 (1968). A witness' assertion of the privilege is strong evidence of that state of mind; the witness has demonstrated that he feels free to decide whether or not to speak. 7 It seems to me beyond question that deposition testimony compelled by means of a contempt order, over the assertion of a Fifth Amendment privilege, would be inadmissible at a subsequent criminal trial whether or not it was later held to be within the scope of the original grant of immunity. If the testimony was within the grant of immunity (i.e., if it was "fruits"), it would be inadmissible under § 6002. If the testimony was not within the grant of immunity, the witness should have been permitted to assert his privilege and the testimony wrongfully compelled should be excluded. See Maness v. Meyers, 419 U.S. 449, 474, 95 S.Ct. 584, 599, 42 L.Ed.2d 574 (1975) (WHITE, J., concurring). 1 The Fifth Amendment provides: "No person . . . shall be compelled in any criminal case to be a witness against himself. . . ." U.S. Const.Amdt. V. 2 See 18 U.S.C. §§ 6002-6003, quoted by the Court, ante, at 253-254, nn. 9 and 10. 3 This second proceeding happens to have been a pretrial deposition in a civil case, but the issue before us would be no different if the second proceeding had been a criminal trial of respondent's co-conspirators, or a coroner's inquest. Respondent happens to have been represented by able counsel at the second proceeding, but again the scope of his immunity would be no different if he had not had a lawyer and had simply answered the questions that were propounded. Moreover, the fact that respondent asserted his privilege against self-incrimination has nothing to do with the availability of the privilege—a matter which is dependent entirely on whether the content of a truthful answer to the questions that were propounded could be used against him in a later criminal trial. His reluctance or willingness to testify would determine whether he elected to assert his privilege or to waive it, but has nothing to do with the existence or nonexistence of the privilege itself. 4 One insignificant nonincriminating fact would be added. The grand jury transcript establishes (1) that respondent had price communications with Fred Renshaw and Dick Herman and (2) that he remembered those communications at the time of his grand jury testimony; an answer to the deposition question would establish the additional fact that respondent still remembers those communications. That additional fact is not itself incriminating and certainly is information indirectly derived from the grand jury transcript within the meaning of the statute. 5 In relevant part, the argument reads: "MR. GRISWOLD: . . . As to evidence first discovered after immunity has been granted, there should be a heavy burden on the government to show that any such evidence is not the fruit of a lead or clue resulting from or uncovered by the compelled testimony. This should not be a conclusive presumption because there can be cases where the government can demonstrate that such evidence was independently derived. It comes in the mail, for example, the day after the testimony was given and it had been postmarked in France a week before. * * * * * "Q. Well, Mr. Solicitor General, what about the situation . . . where the government does compel a testimony and the testimony is given and this induces the prosecutor not to use the testimony except to launch an investigation and by independent means, wholly unrelated to the testimony except by the fact that it was given, search out, independently— "MR. GRISWOLD: That is a hard question, but I think if it does appear that the investigation was the consequence of the evidence being given, that then the evidence is something which was indirectly derived as a result of the testimony given. "Q. Would you— "MR. GRISWOLD: I would construe directly and indirectly quite broadly and I would put the burden on the government with respect to evidence derived after the testimony is given. "Q. So 'but for,' you put on a 'but for' test in the sense that except for the testimony the government would never have had it? "MR. GRISWOLD: Almost, Mr. Justice. On the other hand, I hate very much to give conclusions about purely hypothetical cases, knowing full well the practical situations that can arise which will make it look differently, but I'm perfectly free to say that I think there should be a heavy burden on the government to show that the evidence it wants to use was not directly or indirectly derived from the testimony." Tr. of Oral Arg. in Kastigar v. United States, O.T. 1971, No. 70-117, pp. 30-32. See also the Solicitor General's brief in Kastigar, quoted in nn. 11, 12 infra. 6 If the grant were not at least that broad, a witness obviously could not be compelled to testify before a grand jury. See Counselman v. Hitchcock, 142 U.S. 547, 12 S.Ct. 195, 35 L.Ed. 1110; Ullmann v. United States, 350 U.S. 422, 436-438, 76 S.Ct. 497, 505-506, 100 L.Ed. 511. 7 Cf. Harrison v. United States, 392 U.S. 219, 88 S.Ct. 2008, 20 L.Ed.2d 1047 (1968). In his concurring opinion Justice BLACKMUN seems to assume that the "fruits" inquiry focuses on the state of mind of the deposition witness rather than on the historical derivation of the evidence. He suggests that if the witness "elects" to answer a question, his response is not a fruit and therefore is not directly or indirectly derived from the prior testimony. Even under that approach, however, I would think the question is whether the witness has any choice in the matter. He is being asked about incriminating testimony that, by hypothesis, he would prefer not to repeat. Nevertheless, since he is under subpoena, he must speak unless he has a valid Fifth Amendment privilege, and neither the Constitution nor the statute vests him with any power to decide whether he does. 8 As the Solicitor General assured us in Kastigar, "The immunity provision involved in this case was not passed for the purpose of enabling law enforcement officials to compel self-incriminating information from witnesses and then prosecute them for routine matters." Brief for United States in Kastigar v. United States, O.T. 1971, No. 70-117, pp. 32-33. This fact was emphasized to the Congress that passed the use immunity provision. The Assistant Attorney General in charge of the Criminal Division of the Department of Justice testified that: "As a practical matter, where the witness has elected to testify under this statute, it would be a most unusual circumstance for the Government that used him to turn around and prosecute him." Hearings on H.R. 11157 and H.R. 12041 before Subcommittee No. 3 of the Committee on the Judiciary of the House of Representatives, 91st Cong., 1st Sess., 47 (1969) (statement of Will Wilson). And a member of the Commission on Reform of the Federal Criminal Law, testifying in support of the statute, stated: "I think there is one other thing about this that probably ought to be pointed out and that is that in most instances a grant of immunity is going to be made to a willing witness who isn't going to be prosecuted at all. That is probably the most important aspect of the whole matter. The prosecution will have just as much of an interest in protecting the interests of the person who has served the purposes of law enforcement in that regard as can be. As a consequence fears for the person who has willingly cooperated under the grant of immunity are, I think, probably more fanciful than real." Hearings on H.R. 11157 and H.R. 12041 before Subcommittee No. 3 of the Committee on the Judiciary of the House of Representatives, 91st Cong., 1st Sess., 53-54 (1969) (statement of Judge George Edwards). As of October 1, 1976, these predictions had proven true. On that date, the Attorney-in-Charge at the Freedom of Information Privacy Unit of the Criminal Division of the Department of Justice wrote a letter to a research scholar. The letter reported that, while the Immunity Unit did not maintain statistics on the number of times witnesses had been subsequently prosecuted for matters disclosed in their immunized testimony, "if any such instances exist, they are rare." Note, 14 Am.Crim.L.Rev. 275, 282, n. 46 (1976). 9 The Solicitor General made this point in a slightly different manner in his Kastigar brief: "A practical reason for refraining from subsequent prosecution of a person who provides information is that the government has a vital interest in assuring the continued and unimpeded flow of information concerning criminal activities, and this interest may be furthered if a witness believes he will not be prosecuted." Brief for United States in Kastigar v. United States, O.T. 1971, No. 70-117, p. 34. 10 The Solicitor General regularly provides us with briefs amicus curiae in cases in which its enforcement interests are implicated. He filed no such brief in this case and apparently asserted no objection to petitioner's use of the grand jury transcript as a basis for questioning of deposition witnesses, including respondent. 11 See ante, at 260. The testimony quoted in n. 8, supra, describes this suggestion as "more fanciful than real." For a view that is more real than fanciful, see the testimony of the Assistant Attorney General for the Criminal Division of the Department of Justice in Hearings on H.R. 11157 and H.R. 12041 before Subcommittee No. 3 of the Committee on the Judiciary of the House of Representatives, 91st Cong., 1st Sess., 41-42 (1969) (statement of Will Wilson). That testimony identified the prototypical situations where use immunity would be valuable: where the prosecutor wants to induce someone who is already in prison to testify about a different conspiracy in exchange for a reduction in the existing sentence; where a suspect's attorney offers his client's assistance "in exchange for some type of immunity from that crime which we are investigating;" where the prosecutor's investigation has focused on an agent of a principal "and we decide as a matter of policy that it is more important to prosecute the principal than the agent;" and where a minor actor refuses to testify out of loyalty to a major actor, as in the case of a bookie's customers—"[o]bviously the Government isn't interested in extensive prosecution of 200 or 300 people who simply placed bets, so you use the immunity grant there to make the case against the central person." 12 As the Solicitor General explained in Kastigar, there may be occasions in which an immunized witness is led unexpectedly (by cross-examination at trial, or by grand juror questions) to testify about a new crime, "with respect to which the prosecution may possess overwhelming evidence." Brief for United States in Kastigar v. United States, O.T. 1971, No. 70-117, p. 36. Although the government was willing to give " 'absolute immunity' as to any matter to which the witness testifie[d]" in "a limited area," the government should not be made to abandon an independent case. Ibid. 13 The United States Attorneys' Manual, Title I, Ch. 11, p. 2 (revised Dec. 15, 1981) explains the real reasons why the government prefers use immunity to transactional immunity: "[T]hey have, under appropriate circumstances, significant advantages over former 'transactional immunity' statutes in that they provide no gratuity to a testifying witness, they encourage the giving of more complete testimony by proscribing the use of everything the witness relates, and they still permit a prosecution of the witness in the rare case where it can be shown that the supporting evidence clearly was obtained only from independent sources." 14 In his argument in Kastigar, the Solicitor General seemed to assume that an adequate demonstration that evidence had an independent source would normally involve proof that the source ante-dated the grand jury testimony. See n. 5, supra. In this case respondent's grand jury testimony was given in 1978 and the deposition was taken in 1981. It would be much easier to prove that the basis for a possible future prosecution had a pre-1981 source than a pre-1978 source. 15 The enforcement interest described in the text supplements the general public interest in accurate fact-finding, an interest that is also hindered by the Court's holding. Cf. Lord Chancellor Hardwicke's oft-quoted phrase, "[T]he public has a right to every man's evidence," 12 T. Hansard, Parliamentary History of England 675, 693 (1812), quoted in Kastigar v. United States, 406 U.S., at 443, 92 S.Ct., at 1655. 16 It is not unusual to accept a civil consent decree or a modest penalty in exchange for the dismissal of criminal charges under the antitrust laws. 17 The Court suggests, ante, at 295-260, that cross-examination somehow poses unique problems in this case. Yet it concedes that it is not unusual for a valid assertion of a Fifth Amendment privilege to inhibit cross-examination as to collateral matters such as credibility. Ante, at 260, n. 19. It is thus concerned only that cross-examination might not be allowed on matters about which the witness testified on direct examination because such cross-examination will produce information not elicited on direct. I do not understand why such cross-examination would not be allowed; even if the information were not itself elicited on direct, it would concern a matter about which the witness was required to testify on direct and would thus be derived from the prior immunized testimony in the same way as the direct examination. But even if it were possible that a valid assertion of the Fifth Amendment privilege might so restrict cross-examination that a deposition answer would be inadmissible at trial, that is surely not a sufficient reason to establish a constitutional privilege against giving the direct testimony. 18 It is true, of course, that a witness will risk having his extended testimony used against him later if he makes statements that are not derived from his grand jury testimony. But the assumption that counsel would not be able to identify those "danger areas" demeans the competence of our trial bar. The problem raised by such testimony is essentially the same as the problem presented when any witness testifies in a manner that might be exploited to uncover evidence against him. When in doubt, prudent counsel can always obtain an authoritative court ruling by having the witness assert the Fifth Amendment privilege. 19 The Court is somewhat misleading when it discusses the risk that a trial judge may erroneously reject an assertion of a Fifth Amendment privilege in a paragraph that discusses risks borne by the witness. Such a risk is obviously borne by the government, which may not make use of testimony that is "wrongfully compelled" by a judge. Maness v. Meyers, 419 U.S. 449, 474, 95 S.Ct. 584, 599, 42 L.Ed.2d 574 (1975) (WHITE, J., concurring). Cf. Garrity v. New Jersey, 385 U.S. 493, 500, 87 S.Ct. 616, 620, 17 L.Ed.2d 562 (1967) (government may not use statements obtained under threat of removal from public office). 20 The Court also notes that requiring the respondent to speak increases the risk that he may reveal that he perjured himself before the grand jury, as well as the risk that he may be exposed to civil liability for his misdeeds. Ante, at 262, n. 22. But potential civil liability has never been held to establish a Fifth Amendment privilege. Cf. Ullmann v. United States, 350 U.S. 422, 430-431, 76 S.Ct. 497, 502, 100 L.Ed. 511; Brown v. Walker, 161 U.S. 591, 605-606, 16 S.Ct. 644, 650, 40 L.Ed. 819 (1896). And respondent has never suggested that he asserted the privilege to avoid the risk of prosecution for perjury; the Court does not explain why that risk could not be evaluated case by case when and if it is asserted.

01

459 U.S. 344 103 S.Ct. 665 74 L.Ed.2d 523 Larry SHEPARD, Petitionerv.NATIONAL LABOR RELATIONS BOARD et al. No. 81-1627. Argued Dec. 6, 1982. Decided Jan. 18, 1983. Syllabus Respondent union entered into a collective-bargaining agreement with respondent contractors associations and their members prohibiting dealings by the contractors with nonunion dump truck operators. Petitioner, the owner and operator of a dump truck, who previously had not been a member of a union, joined the union under protest and paid an initiation fee, dues, and a contribution to a fringe benefit plan. Petitioner and one of respondent contractors associations then filed charges with the National Labor Relations Board, claiming that the agreement violated, inter alia, § 8(e) of the National Labor Relations Act (Act), which prohibits so-called "hot cargo" contracts. An Administrative Law Judge held that the union and the contractors had violated § 8(e) by agreeing not to do business with nonunion owner-operators of dump trucks, and recommended that the Board issue a cease-and-desist order and order the union and the contractors to reimburse the owner-operators who were compelled to join the union for amounts paid as dues, initiation fees, and fringe benefit contributions. The Board affirmed and adopted the recommended order except for the reimbursement provision, holding that a reimbursement order would not effectuate the remedial policies of the Act. The Court of Appeals enforced the Board's order in all respects. Held: The Board acted within its authority in deciding that a reimbursement order would not effectuate the policies of the Act. Congress has delegated to the Board the power to determine when those policies would be effectuated by a particular remedy, and the Board could properly conclude that a remedy such as reimbursement should be reserved for especially egregious situations. There is nothing in the language or structure of the Act that requires the Board to reflexively order "complete relief" for every unfair labor practice. Pp. 349-352. 215 U.S.App.D.C. 373, 669 F.2d 759, affirmed. Robert F. Gore, Springfield, Va., for petitioner. Edwin S. Kneedler, Washington, D.C., for respondents. Justice REHNQUIST delivered the opinion of the Court. 1 This case grows out of a labor dispute in the construction industry in San Diego County, California. The issue is whether the National Labor Relations Board was required to provide a make-whole remedy for a violation of § 8(e) of the National Labor Relations Act (the Act), 29 U.S.C. § 158(e) (1976), which prohibits so-called "hot cargo" contracts.1 2 Petitioner Larry Shepard owns a dump truck, and operates it in the San Diego area to haul materials to and from construction sites. Contractors in this area generally hire dump truck operators through so called "brokers" on a day-to-day basis. Brokers agree with contractors to supply trucks and operators, then refer hauling jobs to individual owner-operators such as Shepard. Brokers handle the owner-operators' billing and perform other coordinating services. They receive commissions based on the amount billed. 3 Before August, 1978, Shepard was not a member of any union. In 1977 respondent Building Material and Dump Truck Drivers, Teamsters Local 36 (the Union) entered into a new master collective bargaining agreement (the Agreement) with respondent contractors' associations and their member contractors (the Contractors). This Agreement accomplished a long sought objective of the Union by prohibiting dealings on the part of contractors with non-union operators. The effect of the Agreement was described by the Court of Appeals in this language: 4 "[T]he Union enlisted the aid of the Contractors to insure that only signatory brokers received subcontracts and only union truck operators performed hauling services for building contractors in the San Diego area." 669 F.2d 759, 762 (CADC 1981). 5 In February 1978 Shepard contracted with Terra Trucking Co., a broker that had subscribed to the Agreement, for brokerage services. Although Shepard was not a member of the Union, he authorized Terra to make deductions from his earnings for several purposes, including the fringe benefit plan created by the Agreement. Terra deducted the appropriate sums when Shepard worked on union jobs and paid them to the Union's fringe benefit funds. 6 In August, 1978, the Union wrote to Terra stating that under the Agreement Terra must not deal with seven non-union owner-operators, including Shepard. Terra informed these owner-operators that they would have to join the Union or find a new broker. Shepard joined under protest and paid an initiation fee and dues. 7 Shepard and Respondent California Dump Truck Owners Association (the Association) filed charges with the National Labor Relations Board, claiming that the Agreement violated both § 8(e) and § 8(b)(4) of the Act,2 29 U.S.C. § 158(b)(4) (1976), the latter of which prohibits secondary boycotts. At the request of the Regional Director of the Board, Shepard filed a new charge alleging only a violation of § 8(e). In 1979 the Regional Director consolidated the two charges and issued a complaint against the Union and the Contractors alleging only a violation of § 8(e). After a hearing, an Administrative Law Judge found that these owner-operators are independent contractors rather than employees, and that the Union and the Contractors had therefore violated § 8(e) by agreeing not to do business with non-union owner-operators. The ALJ recommended that the Board issue a cease and desist order and order the Union and the Contractors to reimburse owner-operators who were compelled to join the Union for amounts paid as dues, initiation fees, and fringe benefit contributions. 8 The Board affirmed the ALJ's findings and adopted his recommended order except for the reimbursement provision. The Board stated: "The Board has on one occasion adopted without comment an [ALJ's] recommended order containing such a remedy. Local 814, International Brotherhood of Teamsters, Chauffeurs, Warehousemen and Helpers of America (Santini Brothers, Inc.), 208 NLRB 184, 201 (1974). In the present case, however, there is insufficient evidence in the record with respect to alleged losses directly attributable to actual coercion by Respondents. Furthermore, we find a reimbursement order, typically used to "make whole" employees for violations of the Act, to be generally overbroad and inappropriate in the context of 8(e) violations. We note that aggrieved owner-operators engaged in business as independent contractors may pursue a damage claim under Sec. 303 of the Act. For the foregoing reasons, we find that the reimbursement of owner-operators ordered by the [ALJ] would not effectuate the remedial policies of the Act. See [Carpenters Local 60] v. N.L.R.B., 365 U.S. 651 [81 S.Ct. 875, 6 L.Ed.2d 1] (1961)." 249 N.L.R.B. 386, n. 2 (1980) (emphasis in original). 9 On petitions for review, the Court of Appeals enforced the Board's order in all respects. It held that "the Board's explanation of its decision is adequate and that, given our limited authority to disturb the Board's exercise of discretion in such matters we may not interfere." Id., at 766. In a similar case involving dump truck owner-operators and a similar collective bargaining agreement, the Court of Appeals for the Ninth Circuit remanded the case to the Board to order reimbursement, or to explain why reimbursement would not effectuate the purposes of the Act. Joint Council of Teamsters No. 42 v. N.L.R.B., 671 F.2d 305, 310-313 (CA9 1981). We granted certiorari in this case, --- U.S. ----, 102 S.Ct. 2232, 72 L.Ed.2d 844, and now affirm the judgment of the Court of Appeals for the District of Columbia Circuit. 10 The Board's authority to issue an order in this case is granted by § 10(c) of the Act, 29 U.S.C. § 160(c) (1976): 11 If . . . the Board shall be of the opinion that any person named in the complaint has engaged in or is engaging in any such unfair labor practice, then the Board . . . shall issue . . . an order requiring such person to cease and desist from such unfair labor practice, and to take such affirmative action including reinstatement of employees with or without back pay, as will effectuate the policies of this Act. 12 Shepard and the Association argue that the Board is required to order a make-whole remedy in this case. They rely on the reasoning of the Ninth Circuit in Joint Council of Teamsters No. 42, supra, that "where money has been collected illegally, the Board should order a refund, absent some rational ground for not doing so." 671 F.2d, at 310. We think the Court of Appeals for the Ninth Circuit took too restricted a view of the Board's discretion in designing a remedy. We conclude that the Board need not order reimbursement because its conclusion that the policies of the Act would not be effectuated by such an order is reasonable. 13 Congress has delegated to the Board the power to determine when the policies of the Act would be effectuated by a particular remedy. "In fashioning its remedies . . . the Board draws on a fund of knowledge and expertise all its own, and its choice of remedy must therefore be given special respect by reviewing courts." N.L.R.B. v. Gissel Packing Co., 395 U.S. 575, 612, n. 32, 89 S.Ct. 1918, 1939, n. 32, 23 L.Ed.2d 547 (1969). See Fibreboard Paper Products Corp. v. N.L.R.B., 379 U.S. 203, 216, 85 S.Ct. 398, 405, 13 L.Ed.2d 233 (1964). In this case, the Board issued a cease and desist order and an order requiring the Union and the contractors to post notices stating that the illegal portions of the Agreement will not be enforced. Shepard insists that the Board should have gone the last mile and ordered reimbursement as well. 14 The Board justified its action in declining to grant this additional remedy by the portion of its order quoted above. This explanatory paragraph strikes us as something less than a model of precise expository prose. Shortly after the enactment of the National Labor Relations Act, this Court had occasion to remind the Board: 15 "The administrative process will best be vindicated by clarity in its exercise. Since Congress has defined the authority of the Board and the procedure by which it must be asserted and has charged the federal courts with the duty of reviewing the Board's orders (§ 10(e) and (f)), it will avoid needless litigation and make for effective and expeditious enforcement of the Board's order to require the Board to disclose the basis of its order. We do not intend to enter the province that belongs to the Board, nor do we do so. All we ask of the Board is to give clear indication that it has exercised the discretion with which Congress has empowered it." Phelps Dodge Corporation v. Labor Board, 313 U.S. 177, 197, 61 S.Ct. 845, 853, 85 L.Ed. 1271 (1941). 16 In this case, we think that the sense of the Board's explanation is that it has decided to treat cases in which there is no finding of "actual" coercion differently from cases in which there is such a finding. By actual coercion, the Board apparently means threats, picketing, a strike, or some other form of coercion that would amount to a violation of § 8(b)(4). See Teamsters Local 20 v. Morton, 377 U.S. 252, 259, 84 S.Ct. 1253, 1258, 12 L.Ed.2d 280 (1964). It can scarcely be doubted that the Board could properly conclude that a remedy such as reimbursement should be reserved for especially egregious situations. 17 In choosing to accord the limited relief that it did, the Board relied on Carpenters Local 60 v. NLRB, 365 U.S. 651, 81 S.Ct. 875, 6 L.Ed.2d 1 (1961), in which this Court held that a showing of coercion was required before the Board could order a union to reimburse dues paid to it by workers who were required by an unlawful "closed shop" contract to join the union. The Board presumably concluded that the reasoning of this case supported, at least by analogy, its decision not to award reimbursement. We think this conclusion was justifiable. 18 Congress has provided a judicial damage remedy for illegal secondary activity in Section 303 of the Labor Management Relations Act of 1947,3 29 U.S.C. § 187 (1976). Shepard and the Board agree that § 303 provides a remedy only for violations of § 8(b)(4) of the Act, which, in turn, requires proof of coercion. Brief for Petitioner 28-37; Brief for National Labor Relations Board 32, n. 19, 38-42. Of course, Congress is free to provide a damage remedy for some violations of federal law, and not for others. It is reasonable for the Board to follow the pattern of the Act and order reimbursement only where Congress chose to permit damages. 19 The crux of the argument against the Board's position made by Shepard and the Association is that actual coercion is not an element of a § 8(e) violation and therefore should not be required as a prerequisite to what they call "complete relief." Brief for Petitioner, at 17. But the very way in which this argument is framed suggests that its proponents misconceive the role of the Board. The Board is not a court; it is not even a labor court; it is an administrative agency charged by Congress with the enforcement and administration of the federal labor laws. While a prayer for "complete relief" might find a receptive ear in a court of general jurisdiction, it is well settled that there are wide differences between administrative agencies and courts. See, e.g., Federal Communications Commission v. Pottsville Broadcasting Co., 309 U.S. 134, 141-144, 60 S.Ct. 437, 440-442, 84 L.Ed. 656. This Court has said that the Board's "power to order affirmative relief under § 10(c) is merely incidental to the primary purpose of Congress to stop and to prevent unfair labor practices. Congress did not establish a remedial scheme authorizing the Board to award full compensatory damages for injuries caused by wrongful conduct." Automobile Workers v. Russell, 356 U.S. 634, 642-643, 78 S.Ct. 932, 937, 2 L.Ed.2d 1030 (1958). 20 We find nothing in the language or structure of the Act that requires the Board to reflexively order that which a complaining party may regard as "complete relief" for every unfair labor practice. We are satisfied for the reasons heretofore stated that the Board acted within its authority in deciding that a reimbursement order in this case would not effectuate the policies of the Act. The judgment of the Court of Appeals is therefore 21 Affirmed. 22 Justice O'CONNOR, dissenting. 23 I agree with the Court that the National Labor Relations Board (NLRB) could reasonably determine in this case that reimbursing the petitioner is not necessary to effectuate the objectives of the National Labor Relations Act (the Act). My disagreement is with the Court's conclusion that the Board provided an adequate explanation for its decision. The Board offered three reasons for its conclusion that reimbursing the petitioner would not effectuate the purposes of the Act. Each of its stated reasons was in error or inadequate to justify its conclusion. I would therefore remand the case to the Board in order to give it an opportunity to determine the appropriateness of reimbursement in light of the Court's opinion. 24 * A brief review of the facts is useful in understanding the inadequacy of the Board's explanation for its decision. 25 For over a decade, there has been a dispute between respondent Building Material and Dump Truck Drivers, Teamsters Local 36 (the Union) and respondent California Dump Truck Owners Association (the Association) over the availability of hauling jobs for non-union truck operators. In June 1977, three contractors' associations, which are respondents in this case (the Contractors), entered into a new master labor agreement (the Agreement) with the Union which required signatory contractors to transport "all materials . . . to or from or on the site of the work by workmen furnished by the appropriate craft [union]. . . ." App. 10. The Agreement also required contractors to obtain the services of dump truck operators only through brokers who had signed an agreement with the Union and provided for penalties for contractors who failed to comply. Thus through the Agreement the Union required the Contractors to ensure that only signatory brokers received subcontracts for hauling and that only union operators performed hauling services. 26 Petitioner Larry Shepard is a self-employed dump truck operator. He accepted referrals from the Terra Trucking Co., a broker. In February 1978, Shepard entered into a subhaul agreement with Terra, under which the broker was authorized to make deductions from his earnings for a number of purposes, including "payroll benefits as required by the Union Agreement." App. 22. When Shepard worked on union jobs, Terra deducted the appropriate amounts for payment to the Union's benefit funds. 27 Terra signed the Agreement and was therefore required to refer only union operators to contractors. In August 1978 Terra's president, Fred ReCupido, received a letter from the Union stating that seven of Terra's "employees," including Shepard, were not members in good standing of the Union. The letter requested that the seven be "removed from [Terra's] employ and not be rehired until properly cleared by [the Union]." App. 27. ReCupido told the seven they would have to join the Union by September 5, 1978, if they wished to work through Terra. Shepard joined the Union in September 1978, and paid initiation fees and union dues "under protest," on advice of counsel. Some of the other operators named in the Union's letter also joined at that time. 28 On August 25, 1978, Shepard's counsel filed unfair labor practice charges on behalf of Terra's non-union operators alleging violations of both § 8(b)(4) of the Act, 29 U.S.C. § 158(b)(4) (1976)—the prohibition against secondary boycotts—and § 8(e), 29 U.S.C. § 158(e) (1976), the hot cargo provision. At the request of the Regional Director of the Board, those charges were withdrawn and replaced in October 1978 by charges alleging only a § 8(e) violation. The Regional Director joined Shepard's unfair labor practice charge with charges previously filed by the Association and issued a consolidated complaint against the Union and the Contractors alleging that the Agreement violated § 8(e). 29 After trial, an Administrative Law Judge found that the Union and the Contractors violated § 8(e) by agreeing not to do business with non-union operators and their brokers. He found that since 1965, the Union had brought economic pressure against the Contractors in order "to achieve its goal of unionization of owner-operators," and that the Agreement was "part of the Union's continuing efforts to achieve its goal. . . ." 249 N.L.R.B. 386, 393. 30 The ALJ found that "Shepard joined the Union because of the letter Local 36 sent ReCupido." Id., at 391. In addition to this specific finding, the ALJ made findings concerning another incident1 and stated that "union membership of owner-operators resulted from illegal provisions of the [Agreement]." Id., at 394. 31 The ALJ recommended that the Board issue a cease and desist order and require that notices of its ruling be posted conspicuously. In addition, the ALJ recommended that the Union and the Contractors be required to reimburse operators for payments to the Union. Id., at 395.2 32 The Board upheld the ALJ's findings and conclusions but deleted his "make-whole" reimbursement order. The Board stated its reasons for doing so in a footnote which reads in relevant part: 33 [T]here is insufficient evidence in the record with respect to alleged losses directly attributable to actual coercion by [the Union and the Contractors]. Furthermore, we find a reimbursement order, typically used to "make whole" employees for violations of the Act, to be generally overbroad and inappropriate in the context of 8(e) violations. We note that aggrieved owner-operators engaged in business as independent contractors may pursue a damage claim under Sec. 303 of the Act. For the foregoing reasons, we find that the reimbursement of owner-operators ordered by the Administrative Law Judge would not effectuate the remedial policies of the Act. See [Carpenters Local 60 v. NLRB], 365 U.S. 651 [81 S.Ct. 875, 6 L.Ed.2d 1] (1961). Id., at 386, n. 2 (emphasis in original). 34 The United States Court of Appeals for the District of Columbia Circuit upheld the Board's refusal to order reimbursement, rejecting the contentions that the Board had failed to explain its decision adequately and that the relief ordered was insufficient as a matter of law. 669 F.2d 759, 766. II 35 The broad language of § 10(c) of the Act, 29 U.S.C. § 160(c) (1976), compels the conclusion that the Board has the authority to order restitution of money unlawfully collected by a union, regardless of whether the money was collected from employees or other persons.3 See Virginia Electric & Power Co. v. NLRB, 319 U.S. 533, 63 S.Ct. 1214, 87 L.Ed. 1568 (1943). Indeed, in a proceeding very similar to the instant case, the Board ordered reimbursement of independent owner-operators who joined a union as a result of the union's successful efforts to coerce an employer to enter into and enforce a hot cargo agreement. Local 814, International Brotherhood of Teamsters, Chauffeurs, Warehousemen and Helpers of America (Santini Brothers, Inc.), 208 N.L.R.B. 184 (1974), enf'd, 178 U.S.App.D.C. 223, 546 F.2d 989 (1976), cert. denied, 434 U.S. 818, 98 S.Ct. 56, 54 L.Ed.2d 73 (1977). 36 The Board's first reason for denying reimbursement was that it found that there was "insufficient evidence in the record with respect to alleged losses directly attributable to coercion by [the Union and the Contractors]." There is however, ample evidence, as found by the Administrative Law Judge, that Shepard and other Terra owner-operators joined the Union and paid initiation fees and dues,4 against their will, as a result of the Union's effort to enforce an agreement which violated § 8(e). 37 The Board's second reason, that a reimbursement order is "generally overbroad and inappropriate in the context of 8(e) violations," cannot withstand scrutiny. Although it would be inappropriate to order reimbursement of persons who would have made payments to a union regardless of whether it had attempted to enforce an illegal provision, an order requiring that Shepard be reimbursed for the initiation fees and dues he paid to the Union would not be "overbroad and inappropriate" in light of the ALJ's finding that Shepard joined the Union as a result of the Union's effort to enforce the hot cargo provision. Cf. Carpenters Local 60 v. NLRB, 365 U.S. 651, 81 S.Ct. 875, 6 L.Ed.2d 1 (1961). 38 As its third reason for refusing to order reimbursement, the Board stated that the owner-operators "may pursue a damage claim under Sec. 303 of the Act." But as the Board conceded, § 303 by its terms only creates a damages remedy for persons harmed by a § 8(b)(4) violation, not a § 8(e) violation. Ante, at 351. See Connell Construction Co. v. Plumbers & Steamfitters, 421 U.S. 616, 649 n. 9, 95 S.Ct. 1830, 1848 n. 9, 44 L.Ed.2d 418 (1975) (Stewart, J., dissenting). Thus in the absence of a finding of a § 8(b)(4) violation, petitioner could not successfully pursue a § 303 action. 39 It is true that the Court "will uphold a decision of less than ideal clarity if the agency's [reasons] may be reasonably discerned." Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 442, 42 L.Ed.2d 447 (1974) (citation omitted). But as the Court ruled in SEC v. Chenery Corp., 318 U.S. 80, 95, 63 S.Ct. 454, 462, 87 L.Ed. 626 (1943), "[a]n administrative order cannot be upheld unless the grounds upon which the agency acted in exercising its powers were those upon which its action can be sustained." See FPC v. Texaco, Inc., 417 U.S. 380, 397, 94 S.Ct. 2315, 2326, 41 L.Ed.2d 141 (1974). The Board's order in this case simply does not support its denial of reimbursement.5 I would therefore reverse the judgment of the Court of Appeals and remand this case to allow the Board to consider whether reimbursement of any or all of the funds paid to the Union by the petitioner is necessary to effectuate the Act's prohibition against hot cargo agreements. See NLRB v. Food Store Employees Union, 417 U.S. 1, 94 S.Ct. 2074, 40 L.Ed.2d 612 (1974). 1 Section 8(e) provides in pertinent part: "It shall be an unfair labor practice for any labor organization and any employer to enter into any contract or agreement . . . whereby such employer ceases or refrains or agrees to cease or refrain from handling, using, selling, transporting or otherwise dealing in any of the products of any employer, or to cease doing business with any other person, and any contract or agreement entered into . . . containing such an agreement shall be to such extent unenforcible [sic] and void. . . ." 2 Section 8(b)(4) provides in pertinent part: "It shall be an unfair labor practice for a labor organization or its agents—(4)(i) to engage in, or to induce or encourage any individual employed by any person engaged in commerce or in an industry affecting commerce to engage in, a strike or a refusal in the course of his employment to use, manufacture, process, transport, or otherwise handle or work on any goods, articles, materials, or commodities or to perform any services; or (ii) to threaten, coerce, or restrain any person engaged in commerce or in an industry affecting commerce, where in either case an object thereof is— (A) forcing or requiring any employer or self-employed person to join any labor or employer organization or to enter into any agreement which is prohibited by subsection (e) of this section; (B) forcing or requiring any person to cease using, selling, handling, transporting, or otherwise dealing in the products of any other producer, processor, or manufacturer, or to cease doing business with any other person. . .;" 3 Section 303 provides in pertinent part: "(a) It shall be unlawful, for the purpose of this section only, in an industry or activity affecting commerce, for any labor organization to engage in any activity or conduct defined as an unfair labor practice in section [8(b)(4) ]. (b) Whoever shall be injured in his business or property by reason [of] any violation of subsection (a) of this section may sue therefor in any district court of the United States . . . without respect to the amount in controversy, or in any other court, and shall recover the damages by him sustained and the cost of the suit." 1 The ALJ found that in November 1977, the Kissinger Trucking Co., a broker, entered into an agreement with a contractor, the Penhaul Co., to supply hauling services. Shortly thereafter, Kissinger's manager was informed by Penhaul's superintendent that the Union had said that Kissinger should be replaced because it was referring non-union operators. Kissinger lost the contract with Penhaul and subsequently signed the 1977 Agreement. 249 N.L.R.B. 386, 390. 2 The recommended order, which was omitted from publication, would have required the Union and the Contractors "[j]ointly and severally [to] make whole all owner-operators . . . for all dues, initiation fees, assessments, and contributions to trust funds which . . . said owner-operators paid to [the] Union or its trust fund as a result of enforcement of the [illegal] provisions of the . . . Agreement." Building Material and Dump Truck Drivers, Teamsters Local Union No. 36, International Brotherhood of Teamsters, Chauffeurs, Warehousemen and Helpers of America, No. 21-CE-197, slip op., at 18 (NLRB, Oct. 30, 1979). 3 I express no opinion as to whether the Board could, as the ALJ recommended here, require an employer to reimburse employees or independent contractors for funds unlawfully collected by a union with the acquiescence of the employer. 4 Because Shepard signed the subhaul agreement, which authorized Terra to make deductions from his earnings for payments to the Union's benefits funds, prior to the Union's efforts to enforce the illegal provisions of the Agreement, it is not clear whether his payments to the benefit funds should be attributed to the Union's attempts to enforce the hot cargo provision. 5 Unlike the majority, I do not believe that it can reasonably be discerned from the terse footnote quoted above that the Board has barred reimbursement in cases in which there is no finding of a § 8(b)(4) violation because "reimbursement should be reserved for especially egregious situations." Ante, at 350.

67