World Politics
http://journals.cambridge.org/WPO
Additional services for World Politics:
Email alerts: Click here
Subscriptions: Click here
Commercial reprints: Click here
Terms of use : Click here
International System And Foreign Policy
Approaches: Implications for Conict Modelling
and Management
Raymond Tanter
World Politics / Volume 24 / Supplement S1 / March 1972, pp 7 - 39
DOI: 10.2307/2010558, Published online: 18 July 2011
Link to this article: http://journals.cambridge.org/abstract_S0043887100002860
How to cite this article:
Raymond Tanter (1972). International System And Foreign Policy Approaches:
Implications for Conict Modelling and Management. World Politics, 24, pp 7-39
doi:10.2307/2010558
Request Permissions : Click here
Downloaded from http://journals.cambridge.org/WPO, IP address: 130.60.206.75 on 06 May 2015
INTERNATIONAL SYSTEM AND
FOREIGN POLICY APPROACHES:
Implications for Conflict Modelling
and Management
By RAYMOND TANTER*
. . . The international system is an expanding version of the notion of
two-actors-in-interaction. . . . Interaction analysis focuses on the outputs
of national systems. The national systems, themselves, are black-boxed.
—Charles A. McClelland1
If a nation performs an action of a certain type today, its organizational
components must yesterday have been performing (or have had established
routines for performing) an action only marginally different from
that action.
—Graham T. Allison2
INTRODUCTION
THE quotations from Charles A. McClelland and Graham T. Allison
represent two distinct approaches to the study of international
relations: (i) international system analysis; and (2) foreign policy
analysis. Essentially, international system analysts seek to explain interactions
between nations by phenomena such as their prior interactions
and the structure of the system. Foreign policy analysts, on the other
hand, seek to explain foreign policy behavior as the output of subnational
organizations following standard operating procedures or engaging
in a problem-solving search. Given the international system and
foreign policy approaches as contrasting points of departure, the goals
of the present study are:
•Acknowledgments to ONR Contract Number Noooi4-67-A-oi8i-oo26, ARPA #1411
for support; to Cheryl Kugler, Hazel Markus, Michael Mihalka, Stephen Shaffer, and
Lewis Snider for research assistance; to Patricia Armstrong for typing; to Graham T.
Allison, Robert R. Beattie, Morton H. Halperin, Nazli Choucri, Robert C. North and
Robert A. Young, whose ideas helped guide this inquiry; to Lutz Erbring, Edward L.
Morse, Richard H. Ullman and Oran R. Young for helpful critique; to Charles A.
McClelland, whose ideas and World Event/Interaction Survey provided a basis for the
modelling and coding procedures used in the study; and to Walter Corson for providing
his data, scaling system, and helpful interpretations.
1 Charles A. McClelland, Theory and the International System (New York 1966),
20, 104.
2 Graham T. Allison, Essence of Decision: Explaining the Cuban Missile Crisis (Boston
1971), 87.
8 RAYMOND TANTER
1. to evaluate models based on an international system approach, a
foreign policy approach, and a combination of both approaches as they
are used to study alliance behavior in conflict situations; and
2. to infer from the evaluation of these models some implications for
conflict modelling and management.
International system approaches may imply interaction models,
whereas foreign policy approaches may suggest decision-making models.
For example, J. D. Singer posits that by focusing on the international
system, we can study the patterns of interaction which the system
reveals,3 while game theoretic approaches to the study of conflicts of
interest blend both interaction and decision-making concepts through
their emphasis on strategic interaction and rational choice behavior.4
Game theory deals with strategic situations in which the consequences
of action are uncertain; several different outcomes may result from a
given action.0 Players in a game confront others who are assumed to be
rational and whose choices also affect the outcome of the game. A game
theoretic approach to conflict thus emphasizes strategic interaction and
bargaining under conditions of risk.0
An alternative set of conflict models widely employed in world politics
concerns arms race processes. The most familiar is the Richardson
process model, named after Lewis Richardson.7 Richardson's model
stresses interaction processes between nations but ignores rational choice
behavior. The outcome of Richardson's model " . . . is what would occur
if instinct and tradition were allowed to act uncontrolled."8 The model
ignores choice processes internal to a state and stresses the automatic
response of one nation to the arms expenditures of another. The model
3 J. David Singer, "The Level-of-Analysis Problem in International Relations," in
Klaus Knorr and Sidney Verba, eds., The International System (Princeton 1961), 80.
It should be noted that the interaction approach is often distinguished from the international
system approach. The latter orientation is based on the assumption that international
politics is more than the sum of converging interactions and transactions;
properties of the system as a whole are assumed to influence the behavior of individual
nations.
4 The term strategic interaction in game theory often refers to the outcome of
competing strategies. Here, interaction means the process where each actor pays attention
to and responds to the prior patterns of his opponent.
5 See Herbert Simon, "Some Strategic Considerations in the Construction of Social
Science Models," in Paul Lazarsfeld, ed., Mathematical Thinking in the Social Sciences
(Glencoe 1954), 388-415. Also see Herbert Simon, Models of Man: Social and Rational;
Mathematical Essays on Rational Human Behavior in a Social Setting (New York
1957), 241-60.
6 See Anatol Rapoport, Two-Person Game Theory (Ann Arbor 1966).
7 Lewis F. Richardson, Arms and Insecurity: A Mathematical Study of the Causes
and Origins of War (Pittsburgh i960).
*lbid., 12.
INTERNATIONAL SYSTEM 9
is deterministic and described in terms of "social physics."9 There are
a variety of arms race models which have attempted to improve on
Richardson's formulation. Martin McGuire's model, for example, incorporates
rational choice behavior.10
Less formal than the game theoretic and Richardson process models
are the mediated stimulus response (S-R) and event/interaction models
of Robert North and Charles McClelland respectively.11 North's model
focuses on perception as an explanatory concept intervening between a
stimulus and a response. McClelland, on the other hand, emphasizes
prior international event/interaction sequences and systemic configurations
as explanations for present international interactions.12
The game theory model assumes rational choice behavior; the mediated
stimulus response, the event/interaction, and Richardson process
models allow for irrational (misperception) or non-rational (recurring
event sequence) behavior.13 Nevertheless, all four classes of models have
in common the interaction theme. That is, each model explains present
interaction on the basis of prior interaction with a minimum of focus
on the internal attributes of the actor.14 Of these four interaction models,
9 Anatol Rapoport, Fights, Games and Debates (Ann Arbor i960), 15-107; and
"Lewis F. Richardson's Mathematical Theory of War," Journal of Conflict Resolution,
1 (September 1957), 249-99. See also Kenneth E. Boulding, Conflict and Defense: A
General Theory (New York 1962); Paul Smoker, "Fear in the Arms Race: A Mathematical
Study," in J. N. Rosenau, ed., International Politics and Foreign Policy (2nd
ed., New York 1969), 573-82.
lu Martin C. McGuire, Secrecy and the Arms Race (Cambridge, Mass. 1965).
11 Robert C. North, "Research Pluralism and the International Elephant," in Klaus
Knorr and James Rosenau, eds., Contending Approaches to International Politics
(Princeton 1969), 218-42; Robert C. North, "The Behavior of Nation-States: Problems
of Conflict and Integration," in Morton Kaplan, ed., New Approaches to International
Relations (New York 1968), 203-356; Charles A. McClelland and Gary D. Hoggard,
"Conflict Patterns in the Interactions Among Nations," in Rosenau (fn. 9), 711-24.
12 Charles A. McClelland, "The Acute International Crisis," in Knorr and Verba
(fn. 3), 182-204; "Access to Berlin: The Quantity and Variety of Events, 1948-1963,"
in J. David Singer, ed., Quantitative International Politics (New York 1968), 159-86.
Event/interactions are international actions such as threats and promises (words) or
uses of force and offers of proposals (deeds). Event/interactions are different from
transactions such as trade and mail flows between nations. The present study deals
only with connective event/interactions since there were too few cooperative interactions
during the Berlin conflict of 1961 to perform statistical analysis.
13 See below, however, for a discussion of how recurring event sequences may be
subsumed under learning models and how such models explain limited rational search
behavior.
11 The mediated S-R model draws on internal attributes (perceptions) more than
the other models. Similarly, game theory models applied to world politics focus on
the rational intentions of decision-makers, which tap internal attributes of nations. A
major criticism of game theory models, however, is their treatment of an actor as a
black-box, ignoring psychological and behavioral attributes. See John C. Harsanyi,
"Rational-Choice Models of Political Behavior vs. Functionalist and Conformist Theories,"
World Politics, xxi (July 1969), 513-38; Michael Shapiro, "Rational Political
Man: A Synthesis of Economic and Social-Psychological Perspectives," American Po10
RAYMOND TANTER
the present study draws most from the event/interaction model. A hypothesis
derived from this model is that the current behavior of the
Warsaw Treaty Organization (WTO) in an East-West conflict is a consequence
of a prior pattern of North Atlantic Treaty Organization
(NATO) actions, and vice versa.
Recall the earlier suggestion that international system approaches suggest
interaction models while foreign policy approaches may imply
decision-making models. An early decision-making scheme is the one
pioneered by Richard Snyder and his associates.15 Although their original
decision-making scheme allows for international system determinants
of foreign policy behavior, the scheme mostly relies on the organizational
roles—communication, information, and personality variables,
especially motivation—which constitute the internal setting of decisions.
16 As with game theory, the decision-making scheme assumes
rationality, but rationality is a more limited concept than the comprehensive
version assumed in game theory. In game theory goals are
ranked, all alternatives are specified, consequences are calculated, and
rational choice consists of selecting the value-maximizing alternative.
In the decision-making scheme, however, men are bounded by: (i) the
lack of an explicit preference ordering; (2) incomplete information on
alternatives; and (3) inadequate computational skills to calculate the
consequences of each option. All three limitations violate the requirements
of comprehensive rationality.17
The Snyder scheme focuses on the attributes of individuals as well
titical Science Review, LXIH (December 1969), 1106-19. Simon modified game theory
by incorporating attributes of the actor and then inferring a new decision-rule—
"satisficing" (Simon, fn. 5, 241-60). Experimental gaming explicitly treats properties of
the actors such as competitiveness, risk, and temptation, as well as rewards and punishment.
Melvin Guyer, "A Review of the Literature on Zero-Sum and Non-Zero-Sum
Games in the Social Sciences," Mental Health Research Institute, University of Michigan,
Mimeo, n.d.
15 Richard C. Snyder and others, eds., Foreign Policy Decision-Making (New York
1962); James A. Robinson and Richard C. Snyder, "Decision-Making in International
Politics," in Herbert C. Kelman, ed., International Behavior (New York 1965), 433-63;
Glenn Paige, The Korean Decision (New York 1968); Charles F. Hermann, Crises in
Foreign Policy: A Simulation Analysis (Indianapolis 1969); J. A. Robinson and others,
"Search Under Crisis in Political Gaming and Simulation," in D. G. Pruitt and R. C.
Snyder, eds., Theory and Research on the Causes of War (Englewood Cliffs, N.J. 1969),
80-94.
16 Richard C. Snyder and Glenn D. Paige, "The United States Decision to Resist
Aggression in Korea: The Application of an Analytical Scheme," in f. N. Rosenau, ed.,
International Politics and Foreign Policy (New York 1961), 196.
17 Simon (fn. 5); James G. March and Herbert A. Simon, Organizations (New York
1958); Richard M. Cyert and James G. March, A Behavioral Theory of the Firm
(Englewood Cliffs, N.J. 1963).
INTERNATIONAL SYSTEM 11
as on their foreign policy organizations. The decision-making model
explicated by Graham Allison primarily stresses organizational processes.
18 Allison's model explains government behavior as the output of
large organizations functioning according to standard operating procedures
and search processes. Like Snyder's scheme, Allison's model assumes
limited rationality rather than the comprehensive rationality of
game theory models. Allison's organizational processes explanation
asserts the following principle: Stop searching with the first alternative
that is good enough—the "satisficing" rule.19 The present study draws
more on the Allison work than on Snyder's efforts. Consider Allison's
inference from an organizational processes model: "The best explanation
of an organization's behavior at [time] t is / — i; the best prediction
of what will happen at / -)- i is t."20 Following Allison's model, a
hypothesis is that the current behavior of WTO in an East-West conflict
is a consequence of its own prior pattern of actions, and similarly
for NATO.
The international system and foreign policy approaches may both
yield adequate explanations of international behavior. Similarly, event/
interaction and organizational processes models may apply to the same
situation. Thus, the study evaluates: (i) an event/interaction model;
(2) an organizational processes model; and (3) a combined interaction/
organizational model. Consider the following illustrations of these
three models. The event/interaction model assumes that WTO behavior
was a reaction to the prior pattern of NATO events. That is,
WTO countries decided to construct the Berlin Wall as a result of prior
NATO provocations, e.g., the encouragement of a mass refugee flow
from East Germany to West Germany via Berlin. Similarly, NATO
behavior was a reaction to prior WTO events. NATO countries increased
their defense budgets and sought alliance agreement on economic
sanctions in reaction to Soviet threats to sign a separate peace
treaty with the East Germans and to turn over control of Berlin access
routes.
An organizational processes model, on the other hand, might stress
such variables as standard operating procedures and the problem-solving
search processes of organizations as explanations for alliance actions.
Consider this explanation of an official U. S. reply to the Soviet aide
18 Allison (fn. 2). 19 Simon (fn. 5).
20 Allison (fn. 2), 87. Allison's "explanation" of present behavior as determined by
prior behavior is not an explanation in the sense of specifying why the present behavior
occurs. A learning model may be able to explain why organizations repeat or deviate
from prior patterns.
12 RAYMOND TANTER
memoire and subsequent U. S. actions during the Berlin conflict of
1961: For weeks President John F. Kennedy waited to reply to a Soviet
threat to Western access routes to Berlin which was implied by a Soviet
aide memoire. The Department of State drafted a reply; Kennedy rejected
it as stale and uninspired. He asked Theodore Sorensen to draft
a new reply. Then Kennedy discovered the new reply could not be released
without going through complicated allied and interdepartmental
clearances. He gave up the new attempt and issued the earlier State Department
reply.21 The organizational processes model anticipates standard
operating procedures and helps explain some of the foreign policy
output. Perhaps partly as a result of his dissatisfaction with the perfunctory
U. S. reply, Kennedy searched for more direct ways of answering
the Soviet aide memoire, e.g., by increasing the military budget.22
The interaction/organization model combines the reaction and organizational
process explanations into a single model. Prior studies suggest
that a combination may be more powerful as an explanatory device
than either the international system or foreign policy approach taken
separately. Consider the studies by Nazli Choucri and Robert North.
Although Choucri and North seek to explain international conflict behavior
over longer periods of time, their work is nevertheless relevant
here. Between 1870 and 1914, they find that a nation's role in international
conflict was less a consequence of changes in that nation's own
capabilities (i.e., the foreign policy approach) than of the changing
distances between itself and rival nations, particularly its closest rival
(i.e., the international system approach). They conclude, however, that
neither the foreign policy nor the international system approach alone
is adequate to explain the international conflict process.23 Thus, the present
study combines the international system and foreign policy type approaches
in creating an interaction/organization model. A specific
hypothesis based on the interaction/organization model is that WTO
behavior in an East-West conflict is a consequence of both its own prior
actions and prior NATO actions, and similarly for NATO.
The following three working hypotheses, thus, are: (1) an alliance's
behavior in conflict situations results from the prior pattern of actions
of its opponent (event/interaction); (2) an alliance's behavior in conflict
situations results from its own prior patterns of actions (organiza-
21 Theodore C. Sorensen, Kennedy (New York 1965), 587.
22 This interpretation of the organizational model seems to imply that Kennedy increased
the U.S. military budget because of his dissatisfaction with the State Department.
External factors such as the W T O threat clearly should be considered to explain
the increase in the military budget in this case.
23 See the essay by Nazli Choucri and R. C. North in this volume.
INTERNATIONAL SYSTEM 13
tional processes); (3) an alliance's behavior in conflict situations results
from both the opponent's prior pattern of behavior and its own prior
pattern of actions (interaction/organization).233
AN EVENT/INTERACTION MODEL
McClelland has laid the theoretical framework for the event/interaction
model in a series of essays. In the 1961 special issue of World
Politics, his essay on "The Acute International Crisis" explicates an
event/interaction model.24 He suggests that events in conflicts might
form a chain of interaction sequences, and the discovery of these sequences
would permit comparisons across cases. McClelland's model
describes the state of the international system in terms of its pattern
(process), structure, and performance. Needed data are of two types:
relationships to tap structure, and interactions as indicators of system
process.25 In a later article, McClelland evaluated several propositions
with interaction data concerning access to Berlin, 1948-1963.26 For example,
he evaluated one of the ideas put forward in the 1961 article: the
greater the number of intense conflicts between two actors, the more
likely each will develop routines for minimizing violence. These routines
develop as bureaucrats learn standard operating procedures to
process repetitive conflicts.27 Although the 1968 design does not provide
an explicit test of the learning idea, there is some evidence supporting
it in the Berlin case. Finally, an assumption of McClelland's event/interaction
model is that there are certain international processes, such as
arms races, which occur regularly with specific international situations
such as intense conflicts. The task of the analyst of the international
system is to discover the processes which accompany various situations
and to forecast future processes.28
23a The distinction between event/interaction and organizational processes is for t he
sake of convenience of presentation. In a sense, there is only one model that contains
interaction and organization parameters. Interaction parameters may be relatively more
important at times, while organizational factors may be more significant at other
times. See Tanter, 1972, for a more complete synthesis of interaction and organizational
parameters than given here.
"McClelland, in Knorr and Verba (fn. 3 ) .
25 McClelland (fn. 1), chapter 4.
26 McClelland, in Singer (fn. 12) 159-86.
27 McClelland, in Knorr and Verba (fn. 3 ) , 200-201. Note that one can explain
event/interaction processes with an organizational model, a partial synthesis of the
approaches of McClelland and Allison. Also, McClelland actually uses the term crisis
where the interpretation in the text above refers to conflicts. T h e word crisis refers
to the most intense phase of a conflict in the present study.
28 Robert A. Young, "Prediction and Forecasting in International Relations: An
Exploratory Analysis," unpub. Ph.D. diss., University of Southern California (June
1970).
14 RAYMOND TANTER
McClelland's event/interaction model is the least formal and the least
explicitly theoretical of the interaction models discussed above. It makes
the simple assumption that an interaction pattern will continue under
the conditions of a specific international situation and structure. Recall
Allison's inference from his organizational processes model: "The best
explanation of an organization's behavior at [time] t is / — i; the best
prediction of what will happen at t -f-1 is /." McClelland's model makes
a similar statement but it explains continuity of patterns by referring
to the international situation and structure. McClelland's model, however,
does not explain the continuation of a pattern by referring to
axiomatic assumptions regarding rationality or learning, assumptions
which would provide closure for either a deductive or inductive explanation.
For example, game theory draws upon rationality in a deductive
argument to explain rational choice. The power of game theory lies
in its elegant deductive explanation of a wide range of rational choice
behavior. When applied to the complexities of world politics, however,
game theory loses its elegance as well as its deductive power. In a model
of world politics, one cannot have deductive power without sacrificing
the empirical fit of the model. There are definite trade-offs between
logical closure on the one hand and empirical fit on the other hand.
One can gain some closure by assuming that event/interaction patterns
will continue as a consequence of prior reinforcement—a learning
model. The learning model explains inductively the continuity of specific
event patterns.
Regarding inductive and deductive explanations, Abraham Kaplan
asserts, ". . . we know the reason for something either when we can fit
it into a known pattern, or else when we can deduce it from known
truths."29 Kaplan states that the inductive pattern type of explanation
may be appropriate to a more mature science. Even in the early stages,
however, the generalizations explaining a continuing pattern can function
as general laws in a deductive argument. In addition, the patterned
behavior can be written as a tendency statement and then operate in
an inductive explanation.30
A learning model can explain why event/interaction patterns repeat.
In behavioral psychology, an individual's patterns result from prior
socialization. Kenneth Langton states that, " . . . the continuity of many
29 Abraham Kaplan, Conduct of Inquiry (San Francisco 1964), 332.
30 Carl G. Hempel, "Deductive-Nomological vs. Statistical Explanation," in H. Feigl
and G. Maxwell, eds., Minnesota Studies in the Philosophy of Science (Minneapolis
1962), 98-169.
INTERNATIONAL SYSTEM 15
patterns over time and place suggests that the individual has been modified
in the course of his development in such a way so that he often
exhibits persistent behavior apart from the momentary effect of his
immediate environment. This behavior results from the socialization
process: an individual's learning from others in his environment the
social patterns and values of his culture."31 Hence, socialization models
seem appropriate to explain why an event/interaction pattern will hold
in the future. One can classify learning and game models as similar
explanations of rational behavior. Simon asserts that, "Implicit in any
theory of learning is a motivational assumption—i.e., that learning consists
in the acquisition of a pattern of behavior appropriate to 'goal
achievement,' . . . In parallel fashion, game theory . . . (is) concerned
with discovering the course of action in a particular situation that will
'optimize' the attainment of some objective or 'payoff'."32
Since learning and game models both explain rational choice behavior,
it may be possible to subsume event/interaction patterns under
a more general model based on rationality.33 Thus, an event/interaction
sequence only appears to be non-rational. It may not be the least theoretical
of the interaction models discussed above. An event/interaction
analyst, however, need not pay attention to the implicit assumptions
concerning learning and/or rationality. For example, McClelland and
his associates identified recurring patterns in the flows of events with
little reference to assumptions about learning or rationality which
might have explained such patterns.34 Given their purpose of forecasting
from these patterns, it may be adequate just to know the existence
of patterns rather than why the pattern existed.
If one does not know why the pattern exists, he may have difficulty
anticipating changes in patterns. Learning models may explain why
international event patterns exist or change. In world politics, just as
in behavioral psychology, one may need to know prior reinforcement
and present behavior to forecast future behavior. Behavioral psychologists
initiate their investigations and/or therapy by establishing prior reinforcement
schedules. Thereafter, they monitor and reward present
behavior in relation to the prior schedules. McClelland and his associ-
31 Kenneth P. Langton, Political Socialization (New York 1969), 3.
32 Simon (fn. 5 ) , 274.
33 Learning models, unlike game theory, use a more bounded concept of rationality.
Goals may not be ranked, and search for an alternative which satisfies a goal replaces
choice of an optimal alternative.
34 McClelland, in Singer (fn. 12); McClelland and Hoggard, in Rosenau (fn. 9),
711-24.
16 RAYMOND TANTER
ates would be on more solid theoretical ground if they first attempted
to discover the prior reinforcement schedules of nations and then discovered
their performance records.35
The present study attempts to infer prior reinforcement from present
interaction patterns. For example, if WTO tends to respond to NATO
in the most intense phase of the Berlin conflict, this might reflect the
experience of prior situations when WTO leaders were rewarded for
responding to NATO actions during the intense phases of prior conflicts.
Indeed, an assumption in this regard is that alliance leaders are
more likely to recall learned behavior from the most intense phase of
a prior conflict than from less intense phases. Moreover, as conflictive
intensity increases, the greater may be the perception of interdependence
among the actors. Oran Young, furthermore, suggests that actual
interdependence increases during the most intense phase of conflict because
each actor is able to exercise less and less control over the interaction.
As a result, each actor increasingly considers both the actual and
potential actions of the other party.36
Nazli Choucri and Robert North also stress the interdependence of
interactions during periods of high conflict intensity. In their contribution
to this volume, Choucri and North discuss three models of international
conflict behavior that deal with national expansion, competition,
and crisis. The national expansion model assumes that a nation
generates its own dynamic of conflict behavior irrespective of its rivals.
The competitive model assumes that a nation's level of conflict may be
a consequence of the difference in power capability between itself and
its nearest rival. The crisis model assumes that a nation's involvement
in conflict is a response to the behavior of the opponent. The crisis
model anticipates reaction processes, as does the Richardson model. In
arguing for a mixed model, Choucri and North assert that the earlier
stages of a conflict are dominated by dynamics internal to the nation,
as explained by the national expansion model. During later stages,
processes of competition become more evident than the internal self-
35 Acknowledgments to Judith Tanter for assistance with the behavioral modification
analogy. Subsequently, McClelland and his associates have begun to use learning models
in their World Event/Interaction Survey. Thanks to Gary Hoggard and John Sigler
for bringing these learning models to the author's attention. See McClelland's "Verbal
and Physical Conflict in the Contemporary International System," Mimeo, August 1970,
especially 4-8.
36 Oran R. Young, The Politics of Force: Bargaining During International Crises
(Princeton 1968), 19, 28; Thomas C. Schelling, The Strategy of Conflict (Cambridge,
Mass, i960), 15-16. Note also that evidence suggests that perceptions become more
important the more intense the conflictive interactions. See Ole Holsti and others,
"Perception and Action in the 1914 Crisis," in Singer (fn. 12), 123-58.
INTERNATIONAL SYSTEM 17
generating forces. Even later come the interdependent interactions
characteristic of crises. Some of their most important discoveries are
the "breakpoints," where external dynamics begin to dominate internal
dynamics as determinants of conflictive interactions.
Following Choucri and North, the present study hypothesizes that
internal attributes are more important in pre- and post-crisis phases.37
The present study divides the Berlin conflict into three phases (precrisis,
crisis, and post-crisis) in order to consider whether interdependent
behavior between WTO and NATO increases during the crisis
phase in contrast to other phases. During the crisis phase, an event/
interaction model should explain alliance behavior more adequately
than an organizational processes model. In short, limited rational actors
learn patterns of interdependence from prior conflicts. They generalize
these patterns and, particularly at the most intense phase of an
ongoing conflict, tend to repeat the learned behavior.
AN ORGANIZATIONAL PROCESSES MODEL
Recall Charles McClelland's description of the international system.
He ignores the internal attributes of the actors and stresses prior interactions
as an explanation for current behavior. Graham Allison's foreign
policy approach, on the other hand, ignores prior interaction and emphasizes
standard operating procedures and the search behavior of complex
organizations within each actor.38 An event/interaction model can
employ the concept of learning to explain recurrent patterns between
actors; the organizational processes model can use learning to explain
organizational routines and search processes within actors.
One important set of organizational routines are standard operating
procedures (SOP's). The existence of standard operating procedures
implies that the actor is adaptively rational. Although the actors are
business firms, Richard Cyert and James March suggest that standard
operating procedures are the result of a long run adaptive process
through which a business firm learns.39 Standard operating procedures
are internal characteristics of the actor. If the actor has a need to behave
adaptively in the changing environment of a conflict, however, he has
37 T h e temporal domain of the present study differs from the Choucri-North study.
They base their study on observations covering the period 1870-1914, while the present
study concerns the eight-month period immediately prior, during, and after the intense
conflict over Berlin in 1961. While the important events in the Choucri-North study
unfold over a period of years or even decades, the theoretically meaningful unit of time
in the present study is a period of days.
38 Allison (fn. 2), explicitly acknowledges other models of foreign policy decisionmaking,
e.g., Allison's rational actor model explicitly includes interaction.
39 Cyert and March (fn. 17), 101 and 113.
18 RAYMOND TANTER
to take into account the dynamic nature of that environment. Standard
operating procedures are not tailored to specific environments. Rather,
they are generalized routines which have been applied previously to
similar problems.40
When a conflict occurs, standard operating procedures may not be
an adequate basis for decision-making. In routine situations, the explanation
of the output of an actor may depend heavily on standard operating
procedures. During a conflict, rational adaptation suggests that the
actor search for more innovative solutions than those provided by
standard operating procedures. As Julian Feldman and Herschel Kanter
assert: "The major variable affecting the initiation of search is dissatisfaction—
the organization will search for additional alternatives when
the consequences of the present alternatives do not satisfy its goals."41
The concept of search fits nicely with the idea of "satisficing"—an actor
searches until he finds an alternative which is satisfactory.42
During a conflict, the organizational standard operating procedures
tend to give way to search processes which are more likely to respond
particularly to the external environment. Even these search processes,
however, occur primarily in the neighborhood of prior or existing alternatives
because of the prominence of these options and the ease of
calculating their consequences. In this respect, search simply builds
incrementally on standard operating procedures relying on prior cases
to provide alternatives that may satisfy organizational goals.
Organizational processes models are to event/interaction models as
decision-making models of the firm are to some economic explanations
of firm behavior. That is, some economic explanations stress the environment
external to the firm as the basis of rational choice. Regarding
event/interaction models, the market-determined firm is equivalent to
the international system-determined nation. The external environment
in a market economy consists of all other competitive firms, e.g., all
firms are striving to maximize net revenue, given certain prices and a
technologically determined production function. Similarly, consider
nations as firms, where nations seek to maximize their national interest.
If the market determined each firm's behavior irrespective of internal
organizational processes, domestic attributes would be irrelevant to an
explanation of a nation's foreign policy decisions. Cyert and March provide
an alternative to the market-based ideas just as Allison provides
"Allison (fn. a ) , 85.
41 Julian Feldman and Herschel Kanter, "Organizational Decision-Making," in
James G. March, ed., Handbook of Organizations (Chicago 1965), 662.
42 Donald W. Taylor, "Decision-Making and Problem Solving," in March, ibid., 662.
INTERNATIONAL SYSTEM 19
an alternative to international system ideas. Cyert and March supplement
market analysis with an explanation of the internal operation of
the individual firm. Indeed, their analysis indicates that a firm's resource
allocation decisions are very dependent upon prior patterns of
allocation.43 In a related inquiry, Aaron Wildavsky finds that the most
important determinant of the size and content of a given year's budget
is the previous year's budget—a type of organizational incrementalism.44
Organizational processes models are to event/interaction models as
decision-making models of budgeting are to community power studies.
For example, John P. Crecine's study of municipal budgeting employs
a decision-making model that stresses organizational factors. His findings
provide empirical support to the organizational processes model of
Cyert and March. Crecine finds that the lack of adequate data on
agency performance leaves the decisionmakers with little choice. They
must use prior budgets as a reference for current budget decisions.
Crecine also discusses external citizen demand in the budgeting process.
This kind of external demand has a counterpart in the event/interaction
model of the present inquiry. Crecine acknowledges that external citizen
demand may determine the pattern of expenditure within certain
accounts. But he finds that there is no direct connection between political
pressure and departmental budget levels. Crecine does suggest,
however, that external pressures may have a cumulative, long run effect
on governmental problem-solving.45 In contrast, community power
studies assume a process of mutual interaction comparable to the event/
interaction model presented here. Community power studies do not
allow for organizational explanations of the process by which local
governments allocate values. The community power studies assume
that a business dominated elite, or multiple elites specializing in particular
issues, determine governmental resource allocation.46 In other
words the elitist and pluralist community power models both assume
that resource allocation in the polity is a consequence of external factors,
an assumption comparable to the logic of the event/interaction model.47
43 Cyert and March (fn. 17).
44 Aaron B. Wildavsky, The Politics of the Budgetary Process (Boston 1964), n ff.;
also cf. Charles E. Lindblom, "The Science of Muddling Through," Public Administration
Review, xxxvi (Spring 1959), 79-88; David Braybrooke and Charles E. Lindblom,
A Strategy of Decision: Policy Evaluation as a Social Process (New York 1963).
45 John P. Crecine, Governmental Problem-Solving: A Computer Simulation of
Municipal Budgeting (Chicago 1969), 219; "Defense Budgeting: Organizational
Adaptation to External Constraints," RAND Corporation (March 1970).
46 Floyd Hunter, Community Power Structure: A Study of Decision Makers (Chapel
Hill 1953).
4T Robert A. Dahl, Who Governs? Democracy and Power in an American City (New
Haven 1961).
20 RAYMOND TANTER
There are several implications from organizational studies which are
relevant to the present inquiry.48 One such inference is that rriost actions
taken by alliances may consist of the repetition or continuance of what
was done in the past. In the absence of some reason to change behavior,
alliances may simply continue doing what they have been doing.49 An
organizational processes model assumes that most present behavior is
a result of prior behavior and organizational routines. Explanation of
an action begins at the base line of prior behavior and routines, noting
incremental deviations.50 The incremental deviations may result from
the external environment. Thus, the organizational based studies also
suggest a combined interaction/organization model.
Recall the specific hypothesis emerging from a foreign policy decision-
making approach: an alliance's behavior during a conflict results
from its own pattern of actions. Given the discussion of conflict phases
above, consider the following expansion and modification of this hypothesis:
an alliance's behavior in pre- and post-crisis results from its
intra-organizational standard operating procedures and search processes.
Specifically, WTO should respond more to its own prior behavior
than to NATO during the pre- and post-crisis phases of the Berlin
conflict, and similarly for NATO. Finally, the interaction/organization
model simply combines the event/interaction and organizational processes
models.
DESIGN AND ANALYSIS DECISIONS
A fundamental assumption of the design is that indicators can tap
unmeasured concepts. That is, the data are the intensities of conflictive
interactions between the WTO and NATO alliances. No data are presented
here on such theoretically interesting concepts as learning, rationality,
standard operating procedures, or search processes. Nonetheless,
the design assumes that event/interaction patterns can be used as
indicators of these theoretically significant concepts.51
If an alliance's current actions are a response more to its own prior
behavior, the inference is that organizational processes are more important
than interaction patterns. Conversely, if an alliance's current
48 SOP's in bureaucracies imply long-term stability of behavior, while the present
analysis treats continuity of action over periods of several days. Nonetheless, the organizational
literature may provide useful analogies for the study of short-term conflict.
49 Morton H. Halperin, Bureaucratic Politics and Foreign Policy, The Brookings
Institute (March 1970).
50Ibid.; Allison (fn. 2).
51 Hubert M. Blalock, Jr., "The Measurement Problem: A Gap between the Language
of Theory and Research," in Hubert M. Blalock, Jr. and Ann B. Blalock, eds., Methodology
in Social Research (New York 1968), 5-27.
INTERNATIONAL SYSTEM 21
actions are a response more to the other alliance's prior behavior, then
the inference is that interaction patterns are more important than organizational
processes. In both cases, measured indicators (actions) tap
unmeasured concepts (e.g., event/interactions and organizational processes).
By no stretch of the imagination, then, does this design test
models or their implications. Rather, the design simply evaluates the
models which seem to be implied by certain patterns in the data. This
design is inductive in orientation, but it does more than search for regularities
in the data. The study uses patterns as a point of departure for
making inferences about models. In short, the design seeks to develop
an interface between strategies that stress logical closure via tight models
and those which search for empirical regularities.52
Specifically, the design allows for the evaluation of the following
hypotheses:
1. Prior WTO connective action intensities determine current
WTO action intensities.53
2. Prior NATO conflictive action intensities determine current
NATO action intensities.
3. Prior WTO conflictive action intensities determine current
NATO action intensities.
4. Prior NATO conflictive action intensities determine current
WTO action intensities.
5. Prior WTO and NATO conflictive action intensities determine
current WTO action intensities.
6. Prior WTO and NATO conflictive action intensities determine
current NATO action intensities.
The first four hypotheses correspond to the paths in Figure 1. Hypotheses
five and six combine paths one and four as well as paths two
and three respectively. Paths one and two are called vertical paths
while three and four are the diagonal paths in this study. If the diagonals
are greater than the verticals, this might indicate that an event/
interaction model is more valid than an organizational processes model.
If the verticals are greater than the diagonals, this might indicate that
an organizational processes model is more valid than an event/interaction
model. If both the diagonals and verticals are equally strong, this
might indicate that the interaction/organization model is the valid one
relative to its components. If neither the diagonals nor the verticals are
52 See the article by Oran R. Young in this volume regarding strategies that stress
logical closure and those that emphasize the search for empirical regularities.
53 The term action intensity includes both word and deed intensities.
22 RAYMOND TANTER
strong, this might indicate one or two things: ( i ) the models specified
here are invalid; (2) a significant amount of measurement error is
present in the data.
NATO
(2)
FIGURE 1
Prior Action Intensity
(3)
NATO
WTO
(4)
Current Action Intensity
(1)
WTO
With the six hypotheses diagrammed in Figure 1, the author hopes
to account for the systematic variance in the study. Other variance may
be due to error or is systematic variance which is extraneous here. The
design, therefore, seeks to minimize error variance and rule out extraneous
variance, e.g., rival hypotheses which might explain the dependent
variables. One plausible rival hypothesis, for example, is that
the actions of the Chinese People's Republic might determine the interactions
between WTO and NATO. There is some evidence of a close
connection between the long term connective actions of the C.P.R.,
U.S.S.R., and the U.S.54 An assumption of this study, however, is that
the relationship between the WTO and NATO countries in a given conflict
is not a result of their respective interactions with China.
A further design decision concerns the measurement of conflict intensity
and the identification of the distinct phases of the Berlin conflict.
Walter Corson made available his conflict intensity scale and coded
data from the Berlin conflict of 1961.55 Corson divides the Berlin con-
54 Walter H. Corson, "Conflict and Cooperation in East-West Relations: Measurement
and Explanation," paper delivered at the 66th Annual Meeting of the American Political
Science Association, Los Angeles, September, 1970. Also, see Allen S. Whiting,
"United States-Chinese Political Relations," The University of Michigan, Mimeo,
1970, 17.
55 Corson constructed the scale in two phases. He administered questionnaires to 53
citizens of 13 non-Western and Western countries. In the first phase, there were 54
conflictive actions arranged in irregular order. With each action printed on a separate
card, respondents arranged the actions in rank-order of increasing intensity. The responses
from these questionnaires constituted information to compute a mean rankorder
for each action, resulting in a 54-item rank-order conflict intensity scale. In the
second phase, respondents had 14 conflictive actions selected from the original group
of 54; these actions covered the full range of intensity. They were printed on separate
cards and presented to respondents in irregular order. Respondents assigned a number
to each action proportional to its intensity as they perceived it. Using the responses
INTERNATIONAL SYSTEM 23
flict into five phases on the basis of changes in the types and intensities
of both conflictive and cooperative behavior. Corson's second criterion
for disaggregating the total interaction process is. events which act as
obvious thresholds.
The present study draws partially on Corson's criteria to specify the
phases of the Berlin Conflict. In contrast to the Corson analysis, the
present study excludes cooperative interaction patterns.58 Instead, total
conflictive intensity scores are used for NATO and WTO by day from
i May 1961 to 31 December 1961.
The data show that conflictive intensity remains low until 25 July
when President Kennedy announced major U.S. military preparations.
Conflictive intensity peaked for WTO on 13 August when the East
Germans sealed the border, and for NATO on 17 August when France
and Britain strengthened their armed forces and NATO demanded an
end to the travel ban. The last high conflict peak occurred on 17 September
when the U.S.S.R. protested West German air intrusion over
Berlin. Beginning with the meetings between Soviet Premier Khrushchev
and Belgian Foreign Minister Spaak on 18 September over the
German treaty, events of moderate cooperative intensity occur with relative
frequency. The time from 25 July to 17 September is thus delineated
as the crisis phase for three reasons: (1) conflictive interaction
is more intense during this 56 day period than during any other; (2)
although this phase has several clear peaks, the intensity remained
high for several days; and (3) the crisis phase begins on 25 July with an
event of high conflictive intensity and ends with an event on 17 September
of high conflictive intensity. Figure 2 presents all three phases
of the 1961 Berlin conflict.57
The design evaluates the three models of conflict (event/interaction,
organizational processes, and a combination of both) and their corresponding
hypotheses by regressing each alliance's current conflictive
action intensity (dependent variable) on both its own prior conflictive
from these questionnaires, the geometric mean for each event reflected its intensity
across respondents. From these data, he developed a 14-item conflict intensity scale
and assigned intensity values by interpolation to the remaining 40 conflictive actions.
Details of the scaling project are given in Walter H. Corson, "Conflict and Cooperation
in East-West Crises: Dynamics of Crisis Interaction," unpublished Ph.D. thesis,
Harvard University, December, 1970.
56 T h e conflict phases outlined in this paper are based on empirical data from a
specific conflict and describe only that conflict. Work is under way by the author and
his colleagues on the development of a process model of conflict which will draw on
this analysis but not be limited to it.
57 Corson originally identified five conflict phases: pre-crisis, intensification, peak,
reduction, and post-crisis. For the present analysis, crisis includes intensification, peak,
and reduction. Corson (fn. 55).
24 RAYMOND TANTER
FIGURE 2
PHASES OF THE 1961 BERLIN CONFLICT
Phase Period No. of Days
All Phases 1 May 1961 - 31 December 1961 245
Pre-crisis 1 May - 24 July 84
Crisis 25 July -17 September 56
Post-crisis 18 September - 31 December 105
action intensity and the other alliance's prior action intensity for each
phase of the Berlin conflict. These operations yielded the path coefficients
reported in this analysis.
A crucial substantive and design problem confronted in this paper
is the meaning of time. As a variable and unit of analysis, time is usually
measured in terms of increments of solar time—minutes, hours,
days, weeks, months, years, and so on. Yet it is very likely that time
holds a different meaning for decisionmakers caught up in a crisis.
Time, thus, could be thought of as "diplomatic time" and measured in
a variety of ways including aggregating solar time to periods of specific
duration on the basis of explicit theoretical criteria, or abandoning solar
time units altogether. A variety of studies of crisis58 converge in their
identification of two criteria integral to the nature of crises: (1) action
intensity, and (2) elapsed time between actions.
Corson finds that the elapsed time between actions varies inversely
with total conflictive intensity in his study of the 1961 Berlin conflict.59
This implies that as events increase in conflictive intensity, they also
become more frequent. However, a day is the unit of time in this study
for three reasons: (1) the author knows of no theory relating the frequency
of events with the intensity of conflict in a continuous fashion;
58 Holsti and others, in Singer (fn. 12); Hermann (fn. 15); Corson (fn. 55). Thanks
to Paul Smoker for his thoughts on the study of time.
59 Corson (fn. 55), 186. Corson speculated that time period should be an aggregation
of days rather than a single day. T h e criteria he employed are three: (1) if total conflictive
intensity for N A T O and W T O on a given day was less than 30 on the Corson
scale, the intensity of conflictive actions on that day and the preceding six days predicted
the intensity of conflictive actions for the next three days; (2) if total given
intensity on a given day was between 30 and 150, action intensity on that day and the
preceding four days predicted action intensity for the next three days; and (3) if total
intensity on a given day was greater than 150, action intensity on that day and the
preceding two days predicted deed intensity for the next two days. There are at least
three difficulties with this method. First, it is difficult to implement this aggregation
scheme without overlapping time periods. Second, the method is discontinuous when it
should most properly be of the form
t = all
where / is the aggregation period, /, the conflictive intensity, and, a, the proportionality
constant. As the intensity gets large the aggregation period gets small. Third,
the method only represents more intuition than empirical finding.
INTERNATIONAL SYSTEM 25
(2) if time periods were aggregated, stronger relationships would be
found according to the "ecological fallacy;" (3) the multi-lagged
models tested did not contribute any additional information over and
above the single-lagged models.00 Thus, the study predicts current action
intensity by prior action intensity across each day of the conflict.
A further decision was to aggregate data to the alliance level of analysis.
61 An initial decision was to study only Soviet-American behavior in
the Berlin conflict. It became apparent, however, that East and West
Germany would have to be included. Then what does one do with
relevant actions by other countries during the conflict? These actions
should also be taken into account. Hence, the alliance became the unit
of aggregation. The alliance unit of aggregation may be more valid
for a case such as Berlin than for a case such as the Cuban Missile Crisis
of 1962. There, alliance participation was secondary to the Soviet-American
confrontation.02
Regarding the data, there are 337 events for the Berlin conflict from 1
May 1961 through 31 December 1961—245 days. Primary data sources
included the New Yor\ Times front page, Deadline Data on World Affairs,
as well as The World Almanac and Boo\ of Facts, 1961, 1962. The
present study does not use events per se in analysis. Rather, the daily
intensities aggregated across events for each alliance comprise the data
for analysis. The coding and aggregation design decisions prepared the
data for analysis. The method used is path analysis, which consists of regression
analyses of theoretically specified relationships using standardized
data.03 Path analysis is appropriate for determining the relative
contribution of competing paths in explaining a dependent variable.
The assumptions of the method compare nicely with the measurement
60 Multi-lagged models were run under two hypotheses: (1) the connective intensity
on any given day would be some linear combination of the conflictive intensities of
the previous six days; (2) conflictive intensity would have a decreasing effect as time
from the present increased. Neither of these two hypotheses were supported by the
models or the data. The only different model which arose out of this analysis is found
in footnote 73.
61 Aggregating to the alliance level as in the present study may result in a lack of fit
between an organizational model and the alliance. The study assumes, however, that
organizational models are equally valid irrespective of the level of analysis.
62 In the Berlin conflict of 1961, 2 8% of W T O actions recorded involved other W T O
members acting with or without the U.S.S.R.; 48% of all N A T O actions recorded
involved other NATO members acting with or without the United States. See ibid.
63 Here is a summary of the methodology: The independent variables are prior
WTO and/or prior NATO action intensities. Both word and deed intensity comprise
the action category. The author standardized action intensity within the three conflict
phases for each alliance, e.g., action intensity had a mean of zero and a standard deviation
of unity, pre-conditions for path analysis. Standardized NATO and WTO action
intensities were each regressed on standardized prior NATO and WTO action intensities,
resulting in the path coefficients.
26 RAYMOND TANTER
system and theoretical specification of this study. For example, path
analysis assumes interval scale data and specification of some of the
paths. The Corson scale probably meets the interval level assumption,
and the present study specifies most of the paths explicitly.
ANALYSIS AND RESULTS
Here is a very brief historical overview of key events in the Berlin
conflict from i May 1961 through 31 December 1961, followed by the
path analysis whose purpose is to evaluate the three proposed models.64
During May of 1961 (pre-crisis) WTO countries began to intensify
their demands that the West terminate its presence in Berlin. There
was concern with the problem of the flow of refugees fleeing East Germany—
almost 200,000 in i960. The refugee problem was a major
motivating factor in precipitating the conflict. Recall the inference from
the Choucri-North study that during the pre-crisis phase it is likely
that the focus would be on internal attributes of the actor rather than
on the opponent's actions. Intra-alliance factors such as the refugee
problem and potential unrest in East Germany appear to be more important
than NATO actions as determinants of WTO conflict intensification.
There followed a slow but steady intensification of conflict
which, although self-generated, was modified by Western actions occasionally.
The WTO "ultimatum" of June, the threat to sign a separate
peace treaty with East Germany and end the legal basis for the Western
presence in Berlin, illustrates a key event in the intensification.
In the crisis phase there seemed to be a greater amount of competitive
action and reaction than in the pre-crisis phase. For example, the WTO
actions of 13 August 1961 to erect the Wall may have resulted from
WTO dissatisfaction with Western response to the demand for a separate
peace treaty with East Germany. The Western response consisted
partly of a reiteration of three essentials: (1) continued allied presence
in Berlin; (2) unrestricted access routes to and from Berlin; and (3)
64 For a historical overview of the Berlin crisis, see: George Bailey, "The Gentle
Erosion of Berlin," The Reporter (April 26, 1962); Arnold L. Horelick and Myron
Rush, "The Political Offensive Against Berlin," Strategic Power and Soviet Foreign
Policy (Chicago 1965), chap. 10; John W. Keller, Germany, the Wall and Berlin:
Internal Policies During an International Crisis (New York 1964); Jean Edward Smith,
"Berlin: The Erosion of a Principle," The Reporter (November 21, 1963); Jean Edward
Smith, The Defense of Berlin (Baltimore 1963); Hans Speier, Divided Berlin (New
York 1961); Jack M. Schick, "The Berlin Crisis of 1961 and U.S. Military Strategy,"
Orbis, vm (Winter 1965); Oran R. Young, The Politics of Force (Princeton 1966);
Charles McClelland, "Access to Berlin: The Quantity and Variety of Events, 1948-
1963," in Singer (fn. 12), 159-86.
INTERNATIONAL SYSTEM 27
freedom for West Berliners to choose their own form of government.
The Western response consisted of concrete acts which strengthened
NATO military forces and reinforced NATO troops in Berlin. (One
could select an event from the post-crisis phase to illustrate the deemphasis
on interaction and the consequent reassertion of domestic
factors, but it is not necessary to illustrate the point.) One problem with
selecting historical incidents as illustrations is that it is generally easy
to find an event which demonstrates the idea! Systematic comparative
inquiry seeks to avoid such biased sampling "to prove" one's ideas. A
comparison of action intensities across time, based on a universe of
events, is more valid than the selective sampling of events in a verbal
descriptive account, although both are necessary.
Another way of analyzing the Berlin conflict is to look at the level
of conflictive intensities over time. For example, from i May through
24 July, total conflictive intensity was low for both alliances.65 Disaggregating
conflictive action into its components for a moment, consider
the period between 25 July through 12 August. WTO threats were
much higher in intensity than WTO disapproval, demands, or deeds.
In contrast, NATO conflictive deeds were much higher in intensity
than its words: disapproval, demands, or threats.66
From 13 August to 17 September, conflictive intensities were at their
highest levels. Disaggregating the conflictive actions from 13 to 26 August
shows that WTO conflictive actions were comprised of low demand,
high threat, and low to moderately intense deeds. In contrast, NATO's
conflictive actions in this period had moderately intense deeds (including
troop movements), high demand, and low threat intensity (including
frequent protests of the border closing but few threats of action
which would counter the closing). Between 27 August and 17 September
the nature of WTO and NATO conflictive intensity levels are similar:
threats and deeds were relatively high, disapproval and demands
were relatively low.07 In the post-crisis phase, 18 September through 31
December, events of cooperative intensity were more frequent than
those of conflictive intensity.68 In summary, total conflictive intensity for
65 For the purposes of this study, 1 May 1961 is the beginning of the Berlin conflict.
This establishes a base line period several weeks prior to the WTO ultimatum in early
June.
6 6Corson (fn. 55).
67 Corson, ibid.
68 T h e meetings between Soviet Premier Khrushchev and Belgian Foreign Minister
Spaak on 18-19 September mark the transition to the post-crisis phase. The analysis
ends on 31 December 1961 because the frequency and intensity of actions began to
approach the pre-crisis level of June.
28 RAYMOND TANTER
WTO and NATO averaged lowest in the pre-crisis phase (daily average
= 14 points on the Corson scale), moderate in the post-crisis phase
(29 points on the Corson scale), and highest in the crisis phase (96
points on the Corson scale).
Given this brief historical overview and the description of intensities
of conflictive behavior, Figure 3 contains the results of the quantitative
analysis. Recall the general proposition that the organizational processes
model should explain alliance behavior in the pre- and post-crisis phases
FIGURE 3
RESULTS FOR THE 1961 BERLIN CONFLICT1'9
(N = 245)
Pre-Crisis NATO,.,
.27
NATO WTO.
Conflict Days 1-84
N =83
Crisis NATO,
.12
NATO WTO,
Conflict Days 85-140
N =55
Post-Crisis NATO
.13
NATO
WTO
7-1
.14
WTO,
Conflict Days 141-245
N=104
69 Note that when variables are lagged, you lose one degree of freedom. Thus, N is
always smaller than the number of conflict days in each phase.
INTERNATIONAL SYSTEM 29
while the event/interaction model should explain such behavior during
the crisis phase.'0
The values in Figure 3 are path coefficients, which generally range
from —1.0 to -f-i-o- They indicate the relative magnitude of each path
in determining current alliance action intensity. High vertical path
coefficients relative to the diagonals are consistent with an organizational
processes model. Large diagonal coefficients relative to the verticals
are compatible with an event/interaction model.71
How are the numbers to be interpreted in light of the hypotheses of
the study? Paths may be thought of as flows of influence between variables,
indicating both the direction of the "flow" and the strength of
the dependence of one variable on another. Donald Stokes likens paths
to a system of interlocking waterways with the flow of water through
the paths directed by gates and the amount of water flowing through
each gate determined by the magnitudes of the path coefficients.72
According to the organizational processes model for the pre-crisis
phase, the vertical paths for both WTO and NATO should be
stronger than the diagonals, and, indeed, this is generally the case. However,
for WTO, the vertical path coefficient is only .06 while the diagonal
is just •—-.ii. These results suggest that the model is inadequate
and/or there is large measurement error in the data. On the other
hand, the vertical path coefficient for NATO is .27 while the diagonal
is .21. Given the small difference between these two values, it would
seem that both the organizational process model and the event/interaction
model apply to NATO activity in this period, but one should not
draw strong inferences because of the small magnitude of the coefficients.
The fit does not improve for the crisis phase, although, as one would
expect, the event/interaction model has a slight edge in predictive
power. For NATO, the diagonal path coefficient, .23, is twice as large
as the vertical, .12, but given the small sample size the difference would
70 Besides the organizational processes model, there are several other foreign policy
type models that might explain incremental outputs during the pre- and post-crisis
phases (cf. Allison and Halperin in this volume).
71 The path coefficients for the entire period can be seen in the following diagram:
22 ^~~^~^^^ 14 Entire P«riod
* ^ ^ " ^ - ^ * N = 244
NATO,
72 See Donald E. Stokes, "Compound Paths in Political Analysis," The University of
Michigan, Mimeo, n.d.
30 RAYMOND TANTER
not be statistically significant.73 The fit is even worse for WTO; the
diagonal path coefficient, .11, is about the same as the vertical path coefficient,
.08. Thus, one should be cautious in drawing any inferences.
In the post-crisis phase, the vertical coefficients are larger than the
diagonals, but not of sufficient magnitude to suggest a reversion to
standard operating procedures and the organizational processes model.
For NATO the vertical coefficient is .13 while the diagonal coefficient
is .01. For WTO the vertical coefficient is .14 while the diagonal coefficient
is —x>8.74
Given the inconclusive results of the data analysis, the author cannot
select between the models. The organizational processes model may or
may not be adequate for the pre-crisis and post-crisis phases; the organizational
processes model might be as relevant or irrelevant to the crisis
phase as the event/interaction model.75
There are at least three possible explanations for the inconclusive
results. First, the models may be mis-specified; that is, not all of the
predictive variables were included in the analysis. Recall that the refugee
problem was a major motivating factor in precipitating the conflict.
Thus, at least one major causal variable was left out of the model. Indeed,
internal conflictive behavior was left out of the model. There is
justification for expecting little relationship between domestic and
foreign conflictive behavior, but now may be the time to re-examine the
73 This study does not use statistical inference procedures in evaluating the models.
Here is an alternate model for NATO in the crisis phase, the only instance where the
author felt the multiple lags contributed new information:
= 55
Here is an indication of the strength of the interaction relationships. NATO reacts
strongest to WTO actions lagged by four days.,
74 Results of an earlier analysis using the aggregation periods defined in footnote 59
supported the hypotheses that organizational processes were more important in the preand
post-crisis phases, and that organizational processes may have been important in
the crisis phase as well. Because of the reasons stated in the text, most especially because
of the ecological fallacy, these results are not presented here. Although the ecological
fallacy demonstrates little effect on the functional relationship between variables,
the regression coefficient, it has profound effects on the strength of that association,
the beta weight. In path analysis only the beta weight is presented, which would be
inflated because of the aggregation of days. Thus, it may be misleading to draw inferences,
based as they might be on an artifact of aggregation. For a reference, see
Hubert M. Blalock, Jr., Causal Inferences in Nonexperimental Research (Chapel Hill
1964), 97-114.
73 Given the alternative model in footnote 73, it appears that the event/interaction
model has more explanatory power for NATO during the crisis phase.
INTERNATIONAL SYSTEM 31
internal-external relationship, given the nature of the Berlin conflict.76
Second, the Corson scale as used here may be inappropriate. Summing
conflictive intensities about interactions may not be sufficient
to tap such theoretically interesting concepts as standard operating procedures.
The Corson scale should be re-examined for its assumptions
and its applicability to models of relevance to the current study.
Third, there is some question as to whether action intensity, which is
the aggregate of both words and deeds, should be used as the indicator
in the models. Action intensity was used initially as a means of tapping
the total behavior of the actors, but this may be unsatisfactory for three
reasons: (i) deeds have a longer preparation time than words; (2)
international politics words can be disregarded with greater frequency
than deeds—it is very difficult to ignore the Berlin Wall; (3) words are
subject to greater misinterpretation than deeds. The author is at present
addressing himself to the above problems in his forthcoming book on
the 1948-49 and 1961 Berlin conflicts."
IMPLICATIONS FOR CONFLICT MODELLING AND MANAGEMENT
The twin goals of this paper are to make a tentative evaluation of
models based on an international system approach, a foreign policy
approach, and a combination of these two approaches; and to infer
from the evaluation of these models some implications for conflict
modelling and management. The inconclusive nature of the above
analysis only points up the problems facing the conflict manager, and
in this section the author attempts to address the implications of the
study of the Berlin conflict of 1961 for the more general problem of conflict
management.
If one is to generalize about conflicts, it would make sense to have
information on as many cases as possible. An analyst hopes to draw
inferences from a limited number of cases which are representative of
the larger universe of all conflicts. The Berlin conflict of 1961 may not
be at all representative. As McClelland indicates, the Berlin conflicts of
1948, 1958, and 1961 may have been increasingly routinized as a consequence
of a bureaucratic processing that became almost self-generating.
That is, conflict over Berlin occurred so frequently that organizational
processes assumed greater importance over time. Quite possibly, standard
operating procedures grew up around the conflicts as a result of this
76 Raymond Tanter, "Dimensions of Conflict Behavior Within and Between Nations,
1958-1960," Journal of Conflict Resolution, x (March 1966), 41-64.
77 Raymond Tanter, The Berlin Crises: Modelling and Managing International Conflicts
(forthcoming, 1972).
32 RAYMOND TANTER
repetitive pattern.78 Yet, in a case such as the Cuban Missile Crisis of
1962, the event/interaction model might be more valid within the
crisis phase. Thus, it is important to create a universe of cases for the
comparative inquiry of conflicts before drawing firm inferences from
any one case. (The new conflict data should include information on the
interactions and on organizational processes if possible.)
In addition to obtaining data on more cases, it is necessary to explicate
further the present models and to develop additional models to explain
conflictive interactions. The present models allow one to make little
sense of patterns in the data. Certainly, this result of the data analysis
indicates a need to develop further process models that would describe
and explain the evolution of conflict situations.'0 In the interest of preventing
the explosion of conflicts into crisis, it is extremely important to
discern the connection, if any, between apparently dissimilar conflicts.
This might be accomplished through the development of models and
through the long and tedious process of making, rejecting, and accepting
hypotheses based on these models.
The results of this study have other tentative implications for an effort
at model-building. For instance, the evidence does not indicate that
the author should ally himself with "disillusioned interaction analysts"
and join the growing number of organizational analysts. Such a decision
would be premature, especially since the times call for a synthesis
of the two approaches. Perhaps the Thomas Schellings and Charles
McClellands overemphasize the role of interaction processes; perhaps
Graham Allison and Morton Halperin overemphasize organizational
processes in relation to interaction notions. It is not for the author to
say at this time; the jury is still out.
How would the organizational theorists view the Berlin conflict?
Halperin, for example, might claim that "In periods viewed by senior
players as crises . . . , organizations will calculate how alternative policies
and patterns of action will affect future definitions of roles and
missions. . . . [Organizations] will press for policies which they believe
78 This study, however, does not compare intensities for the three Berlin conflicts;
rather, it only has data on the Berlin conflict of 1961. Thus, there are no hard data
presented here on the routinization of conflict decision-making.
79 Process modelling is a research strategy designed to disaggregate a complex set
of interrelated events and behaviors into stages representing discrete actions or distinct
choice points. Process models serve several useful purposes. First, they direct our attention
to processes such as learning, forgetting, or precedent search which underlie
highly complex patterns of behavior. Thus, process models reduce complex situations
to their basic elements, permitting an economy of description and explanation. Finally,
process modelling could explain the breakpoints in a conflict—those points where the
internal dynamics give way to external factors.
INTERNATIONAL SYSTEM 33
will maintain or extend their roles and missions, even if at some cost
to the immediate objectives of the President. . . ."so Regarding the present
study, Halperin's explanation suggests that alliances should respond
more to intra-alliance than to inter-alliance considerations. Halperin's
explanation also poses the question whether alliances are useful units
of analysis to tap organizational processes. (See footnote 61.) If bureaucracies
respond, as Halperin contends, to their roles as defined within a
particular country, there is no reason to suppose that this response is
consistent with other countries in the alliance. Indeed, one might suspect
the contrary. The alliance problem may account for the relatively
weak organizational process link found in the analysis of the Berlin
conflict of 1961. To determine the effects of organizational processes, it
might be better to examine individual countries and, especially, the
various bureaucracies in those countries.
One of the more interesting aspects of this study comes from the
examination of the plots of conflictive intensity. For the pre-crisis and
post-crisis phases, activity is relatively minor; many of the days register
no activity at all. This might conform to Halperin's statement that,
". . . most of the actions taken by bureaucrats . . . involve doing again
or continuing to do what was done in the past. In the absence of some
reason to change their behavior, organizations keep doing what they
have been doing."81 This notion of "bureaucratic incrementalism," explaining
the performance of foreign service personnel around the
world, is certainly intuitively appealing. Evidence from the budgeting
studies, moreover, suggests that municipal politicians may have something
in common with their statesmen counterparts in the foreign
service.
There is a problem, however, with the incrementalist thesis. How
can the incrementalist thesis account for an innovative sequence of interactions
such as WTO's ultimatum to NATO, NATO's response increasing
its conventional military capabilities, the Berlin Wall, and
negotiations ? Although these events are measured, the present quantitative
analysis fails to account for such innovative sequences. Similarly,
the budgeting studies which stress quantitative budget totals may overlook
the quality of the programs. Thus, quantitative analysis needs to
be supplemented by a study of the qualitative aspects. The latter may be
more apt to yield event/interaction sequences.82
80 Halperin (fn. 49), 50.
81 Ibid., 9.
82 As stated previously, however, one must be careful to avoid selecting historical
events in order "to prove" one's hypothesis. Thanks to Alexander George for the
critique of the incrementalist thesis regarding the quality of programs.
34 RAYMOND TANTER
In summary, this study implies that in modelling conflict an analyst
should: (i) specify a universe of cases for comparative inquiry across
conflicts; (2) further explicate the event/interaction and organizational
processes models, emphasizing their formal axioms and data requirements;
(3) develop process models that describe and explain the evolution
of conflict in general—emphasizing breakpoints where internal
dynamics give way to external factors; and (4) integrate qualitative
evaluation of events with quantitative analysis, to ensure that one takes
into account the nature of events.
A project underway by the author and his colleagues seeks to implement
those modelling implications with the construction of a
Computer-Aided Conflict Information System (CACIS). Coders are
classifying major power conflicts since World War II in terms of environmental
factors, policy options, national interests and involvement,
goals, intentions, resources employed (military, economic diplomatic),
and outcomes. CACIS will also include a capability for specifying event/
interaction and organizational models, among others, within the general
framework of a process model of conflict. An important aspect of
the process model will be its formal status. Rather than using the relatively
loose verbal models of the present study, CACIS will emphasize
tight, deductively oriented formal models.
One principal attribute of CACIS is that it is being built around four
separate but interrelated modules:
1. The memory module which stores information about prior
conflicts.
2. The experience module which stores evaluations of strategies
used in prior conflicts, and the number of successes, failures, or
indeterminate outcomes.
3. The involvement module which estimates the type and magnitude
of interests (or values) of conflict participants.
4. The operational environment module which includes external
events and domestic political factors. This module could serve
as the basis for the evaluation of the relative potencies of internal
processes vs. external events on the policy-making process,
as well as provide parameters for an all-machine simulation
of conflict decision-making.
A second major characteristic of CACIS is its reliance on the process
of precedent search.83 That is, a party to a conflict, in seeking a solution
83Hayward R. Alker, Jr. and Cheryl Christensen, "From Causal Modelling to Artificial
Intelligence: The Evolution of a U.N. Peace-Making Simulation," Massachusetts
Institute of Technology, Mimeo, n.d.
INTERNATIONAL SYSTEM 35
commensurate with its goals, will search for prior conflicts similar to
the current conflict as policy guides. Precedent search behavior assumes
the existence of rules or "precedent logics"84—i.e., criteria guiding precedent
search—as well as the identification of dimensions of similarity
and differences along which conflicts may be located.
CACIS supplements the Computer-Aided System for Handling Information
on Local Conflicts (CASCON), developed by Lincoln
Bloomfield and Robert Beattie.85 CASCON focuses on local conflicts
between small powers or between a small power and one major power,
while CACIS will include mainly the CASCON cases and those conflicts
involving more than one major power. Some overlap, however, is
expected in the sample of cases selected. CACIS will offer more options
to the analyst through the programming of multiple models rather than
the single model of local conflict of Bloomfield and Amelia Leiss in
CASCON.86 Finally, unlike CASCON, CACIS is expected to have a
machine simulation capability enabling the user to look at "what might
have been" by recalling prior relevant cases, applying alternative policy
options, and examining the simulated outcomes in relation to a current
conflict.
Implications of the present study for conflict management are less
certain. Glenn Paige faced a similar problem in deciding whether to
draw implications for conflict management from a single case—Korea,
1950. He wondered ". . . whether it is not premature and irresponsible
for the student of decision-making analysis to venture suggestions of an
applied nature on the basis of a single case. . . ." Paige concluded that
international crises are such important phenomena that it is well worth
the risk to venture suggestions.8' Following Paige's lead, the present
study will also make inferences regarding conflict management, with
similar caveats about over-generalizing.
The idea of conflict management assumes that conflicts are similar
enough to plan for in advance. Some national security policy planners
argue that the element of surprise places great constraints upon planning.
For example, G. A. Morgan asserts: "The number of theoretically
possible crises in the years ahead is virtually infinite. Even to try to plan
systematically for all that are moderately likely would be a questionable
silbid., 21.
85 Lincoln Bloomfield and Robert Beattie, "Computers and Policy-Making: The
CASCON Experiment," Journal of Conflict Resolution, xi (March 1971); Robert Beattie,
and Lincoln Bloomfield, CASCON: Computer-Aided System for Handling Information
on Local Conflicts (Cambridge, Mass. 1969); also cf. Fisher Howe, The Computer and
Foreign Affairs (Washington 1967).
86 Lincoln Bloomfield and Amelia Leiss, Controlling Small Wars: A Strategy for
the igjo's (New York 1969).
"Paige (fn. 15).
36 RAYMOND TANTER
expenditure of resources."88 Klaus Knorr and Oskar Morgenstern agree
with this, concluding that planning is difficult because intense conflicts
are " . . . essentially unpredictable.. . ."80
The notion that conflict planning is virtually impossible because of
unpredictability overlooks the fact that contingency planning takes
place in several areas where phenomena are not easily predicted. For
example, earthquakes are rarely predictable in advance. Nonetheless,
areas where they frequently occur have developed standard operating
procedures for processing the injured, alleviating congestion, and communicating
in the absence of normal channels. Similarly, in international
security planning, conflict need not be fully predictable for management
plans to be written and used as general guides.
Social scientists should not feel uncomfortable at being unable to
make point predictions of specific events. Physicists often do not forecast
individual events, but they are able to explain and forecast processes
and general classes of events. Social scientists also should seek to explain
and forecast processes and classes of events. Process models are promising
ways of developing explanatory and predictive theory both for
processes and general event-classes. The development of conflict intensity
scales is a way of constructing more general event-classes.90 Computer
based models and the acquisition of comparable data on a series
of historical cases promise to improve the generality of event concepts.
The creation of computer based models such as CACIS should
facilitate conflict management in several ways. For example, the results
of the coming inquiry might serve as a basis for specifying models in
CACIS. Suppose then, that these analyses found that an organizational
processes model explained WTO and NATO alliance behavior better
than an event/interaction model, especially in the pre- and post-crisis
phases. In such a case, a foreign policy decision-making approach may
yield more than an international system approach for the conflict. If a
new Berlin conflict were to erupt, an analyst might expect the predominance
of intra- as opposed to inter-alliance factors. CACIS would allow
the analyst to compare recurring conflict over Berlin with what oc-
88 G. A. Morgan, "Planning in Foreign Affairs: The State of the Art," Foreign
Affairs, xxxix (January 1961), 278. T h e thrust of Morgan's argument is for selective
planning. However, some authors advocate more planning—J. C. Ausland and J. F.
Richardson, "Crisis Management: Berlin, Cyprus, Laos," Foreign Affairs, XLIV (January
1966), 291-303.
89 Klaus Knorr and Oskar Morgenstern, Political Conjecture in Military Planning,
Princeton University, Center of International Studies, Policy Memorandum No. 35
(1968), 10-15.
90 A conflict intensity scale produces more general classes than raw event data. That
is, the scales allow an analyst to aggregate across a variety of events to calculate a
general intensity score for the actor.
INTERNATIONAL SYSTEM 37
curred in 1948, 1958, and 1961, especially regarding the organizational
processes of the actors. If such a comparison proved useful, the analyst
might expect the bureaucratic patterns of the past to repeat themselves.
As a result, the analyst can develop his plans anticipating standard
operating procedures and search processes.
Another way that CACIS might facilitate conflict management is
as an aid to memory in the form of an information retrieval system.
The information would describe prior conflicts, the policy measures
used, and their consequences. The institutionalization of prior crisis
patterns, and the policy measures employed, is important for several
reasons. First, the memory of complex organizations too often resides
in now departed personnel who were instrumental in prior conflict
problem-solving. CACIS would thus be an aid to memory in immediately
accessible form. As an aid to memory, CACIS would facilitate the
search for alternative options. Recall the search style of limited rational
actors—they learn to search for alternatives until they find the one that
satisfies goal achievement.91
It is also very important to institutionalize alternatives. During a conflict
there is a higher probability that stress may cause the replacement
of complex problem solving habits by more basic forms. That is, if
stress is intense and persistent, there is a tendency for more recent and
usually more complex behavior to disappear and for simpler and more
basic forms of behavior to reappear.92 Thus, there might be a tendency
to revert to the standard operating procedures and other familiar organizational
routines during periods of highest conflictive intensity.
Rather than bringing about a greater sensitivity to the external environment,
crisis induced stress may result in increased reliance upon standard
operating procedures in the intense crisis phase.
Finally, institutionalization of alternatives would permit the examination
of the consequences of conflict management attempts in prior
cases. For example, Alexander George specifies seven principles of crisis
management, some of which relate nicely to the present inquiry. He
asserts that there should be: (1) high level political control of military
options; (2) pauses in military operations; (3) clear and appropriate
demonstrations to show resolution; (4) military action coordinated
with political-diplomatic action; (5) confidence in the effectiveness and
discriminating character of military options; (6) military options that
91 James G. March, "Some Recent Substantive and Methodological Developments in
the Theory of Organizational Decision-Making," in Austin Ranney, ed., Essays on the
Behavioral Study of Politics (Urbana 1962), 191-208.
92 Thomas W. Milburn, "The Management of Crisis," Mimeo, 1970.
38 RAYMOND TANTER
avoid motivating the opponent to escalate; and (7) avoidance of the
impression of a resort to large scale warfare.93 CACIS may aid the control
over military options by specifying alternatives (emphasizing political
ones?) and estimating consequences. CACIS could be used to
evaluate the effects of timely pauses in military operations in a current
conflict by suggesting what the implications were for such pauses in prior
conflicts. CACIS may help develop clear and appropriate demonstrations
of resolution, as well as help discriminate among options based
upon such intensity scaling as developed by Corson. In addition, an improved
Corson scale might allow for a more subtle selection of politicomilitary
options and decrease the probability of escalation.
SUMMARY
The present study evaluates an international system and a foreign
policy decision-making approach via their corresponding models:
event/interaction, organizational processes, and interaction/organizational
models. The design used actions between East and West in the
Berlin conflict of 1961 to infer the unmeasured models. The Corson
scale of conflict intensity provided a discriminator of politico-military
options, even though there may be problems with the scale and the
coding.94 The Berlin conflict of 1961 provided a laboratory for the exploration
of the three models. The resulting path coefficients did not
support the original hypotheses. The magnitude of the coefficients is so
low that the results are inconclusive.
The implications of this study for conflict modelling and management
are tentative but potentially promising. Regarding modelling, the
study concludes that analysts should: (1) specify a universe of cases for
comparative inquiry across conflicts; (2) explicate the event/interaction
and organizational processes models, emphasizing formal axioms and
data requirements; (3) develop process models that describe and explain
the evolution of conflict, emphasizing breakpoints where internal
dynamics give way to external factors; and (4) integrate qualitative
evaluation of events with their quantitative analysis to make sure that
93 Alexander George and others, The Limits of Coercive Diplomacy (Boston 1971),
8-15.
94 Cf. Edward Azar, "Analysis of International Events," Peace Research Reviews, iv
(November 1970), 83. Azar asserts that, "We code events and measure their violence
content with the 13 point interval scale. Although we realize that participants to a
conflict situation do not use such an objective instrument, we maintain that they employ
an implicit (or possibly explicit) scale which ranks signals by their violence
content." Also see William A. Garrison and Andre Modigliani, Untangling the Cold
War: A Strategy for Testing Rival Theories (Boston 1971), for an attempt to quantify
and scale East-West interactions.
INTERNATIONAL SYSTEM 39
the quality of the policies is taken into account. Regarding conflict management,
the study concludes that: (i) the results of the present inquiry
could help specify models for a Computer-Aided Conflict Information
System, which could be used to compare a current conflict with prior
relevant cases; and (2) CACIS might institutionalize prior alternatives
and estimate their consequences in similar cases. Such institutionalization
should expand the political options short of military force available
to decision-makers. Finally, CACIS should not be used to freeze options
on the basis of historical precedents. Rather, CACIS should provide
a fresh set of alternatives for the adaptively rational actor.95
95 Also, see Sidney Verba, "Assumptions of Rationality and Non-Rationality in
Models of the International System," in Knorr and Verba (fn. 3), 93-117. Acknowledgments
to Dennis Doolin for calling attention to the danger of freezing options on
the basis of historical precedents with a system such as CACIS. There is a great need
for what Doolin calls ". . . creative politics—which is really the essence and true
genius of politics—and there seems to be a danger in an approach that could view
routinization as a rule of action." Letter from Dennis Doolin, 28 June 1971. CACIS
attempts to address itself to Doolin's perceptive critique and to facilitate "creative
politics."