Chapter 9 
Sometimes, irrationality 
doesnÕt make sense
Some might find our effort to get people to act more rationally in discussion, argumentation, and negotiation a bit too idealistic. 
People donÕt act rationally, they say. They choose their positions based on Ògut feelingÓ, emotion, social relations and power relations; not logical calculation and reasoned consideration. Logic, reasoning and argumentation are just used as tools or tricks to convince (or fool) others, in service of emotional or selfish goals. 
ItÕs remarkable how different professions seem to have radically different views of human nature. Classical mathematical game theory, argumentation theory, and mathematical logic view people as always acting to maximize self-interest. They posit a Òutility functionÓ that puts a specific number on every situation that says whether a specific individual will find that situation good or bad. This is a completely ÒrationalÓ view. 
Economists, also, view people as an idealized Òhomo economicusÓÑalways acting rationally to maximize economic gain. But they donÕt have any idea of what actually provides value for people, except to quantify it in terms of money. Economists go so far as to work backwards, assuming that value is anything that people are willing to pay money for. If people are irrational and pay money for worthless products such as Òpet rocksÓ, economists assume that it is because they must provide some sort of value. That kind of reasoning is circular. 
Mathematical theory completely ignores social, emotional, and aesthetic motivations ([Goleman 1995], [Mason 2008]) and others show that we have considerable Òemotional intelligenceÓ that is essential to human problem solving. Modern social network theories which analyze the graphs of social connections and ßow of information between them show the importance of social network connections and patterns of collaboration in problem solving ([Hidalgo 2015], [Watts 2004]). Nobody yet has any credible theory of the role of aesthetics in problem solving. 
Advertisers, marketers, media executives, and cynical politicians and business people treat people as being almost completely irrational (weÕll refer to these people as irrationalists). They acknowledge people as being motivated only by a few primal animal impulses: need for food, shelter, clothing, sex, and power. They dismiss with disdain any suggestion that people form beliefs by conscious thought, or are capable of rational reasoning or morality. They seek to manipulate the public with emotional and subconscious appeals, and project a false sense of urgency in the hope that they can sell them a product before any rational thought has a chance to kick in. 
Oddly enough, though, they view themselves as being rational in the economic sense described in the previous paragraph, using their pseudoscientific psychology to maximize their own economic gain. ThereÕs a galling sense of superiority amongst these people: they believe they are the shepherds, and we, the public, are the sheep. 
The irrationalists will tell you that the proof of their claim is that it works. Advertising, bad as it is, sells products, they say. P.T. Barnum, perhaps the best exponent of this theory, said ÒNobody ever lost money by underestimating the intelligence of the American publicÓ. He said some other things, too.
But be suspicious of any claims by the irrationalists. Remember who itÕs coming from and what they actually think of you. ÒMarketing scienceÓ will tell you that, in a lab experiment, given a choice between a row of no-name detergents and the heavily advertised Tidetm detergent, people will pay more for Tide and choose it more often, without any rational evidence that itÕs any better. Maybe. 
But that doesnÕt mean that one more TV ad for Coketm or Pepsitmin the already ad-saturated real world will have any real effect on its market share. Remember, the job of the advertiser is not to sell the product to the consumer, it is to sell advertising to the vendor. The advertiser attempts to hoodwink the vendor, the same way they attempt to hoodwink you, the customer. 
Marketing studies that claim the efficacy of subliminal and emotional advertising are like the old toothpaste ads that had a white-coated ÒscientistÓ shill tell you their study proved that the toothpaste led to 44% fewer cavities. We predict that the fraudulent nature of most advertising will eventually catch up with it. 
We predict that the advertising industry will experience a catastrophic meltdown very similar to what is happening with the so-called Òmusic industryÓ. The music industryÕs job was supposed to be to both increase listenersÕ access to music they love, and to provide a living for musicians. But decades of the music industry exploiting their ÒleverageÓ, with low royalty rates for musicians and contractual restrictions, and exploiting the listeners with artificially high prices and limited availability, meant that the music industry was mainly working for itself. Once distribution alternatives were available, nobody had much sympathy for the music industry. It wasnÕt adding value to the product. Thus it collapsed. Similarly, advertisers exploit consumers and vendors alike. It doesnÕt add value to the product. Their turn is next. 
Advertising is an example of an arms race. In the PrisonerÕs Dilemma chapter we discuss advertising as an example, and show how an arms race requires cooperation from people other than those who perpetrate it [Thompson 2014]. The continuance of war depends on people electing warmongering politicians, people paying war taxes, people joining the army, etc. If they stop cooperating with the arms race, it collapses. 
To the extent that people are rational, our rationality certainly has natural limits. We only have so much time and energy. Several authors ([Ariely 2008], [Kahneman 2011], [Gigerenzer 1999], [Mullainathan 2013]) have detailed the extent to which our rationality can fail, usually due to one or more of the following causes: scarcity of time, effort, and resources; hidden assumptions; heuristics that are usually reliable but fail in particular circumstances; deliberation/action tradeoffs, etc. This is studied in AI as cognitive biases and bounded rationality ([Simon 1957], [Minsky 1988], [Manoogian 2016]).
ThereÕs been a recent movement to turn the deceptive tricks of advertisers and marketers around for good instead of evil [Thaler 2009]. This movement recognizes that people will inevitably have cognitive biases and limited rationality, but seeks to harness these techniques in service of positive instead of negative principles: tricking people into saving rather than spending; nudging people into exercising instead of watching TV; promoting choosing healthy foods instead of junk foods. They use gamification, incentive design, nudging and other techniques. If it works, and people donÕt mind or even like it, hard to argue with that. 
Trouble is, even though these may have some effect, itÕs usually limited to extrinsic motivation in the short term (which is suitably convenient to be verified by short term experiments). WeÕll discuss more about intrinsic vs. extrinsic motivation, and short-term vs. long-term, later. This movement shares with the advertisers the basic disrespectful view of the public as sheep to be manipulated by the elite gamifiers and incentive designers. 
Despite what the marketers will tell you, irrationality has its limits too. This was best expressed by Abraham Lincoln: ÒYou can fool some of the people all of the time, and all of the people some of the time, but you canÕt fool all of the people all of the timeÓ. After a while, peopleÕs rationality will indeed kick in. People will begin to think for themselves rather than blindly accept what theyÕve been told. Word will get around. The short term will turn into the long term. Push will come to shove. Something will hit the fan. 
But this isnÕt assured. A problem with the Lincoln quote is, as advertisers have discovered, that itÕs possible to fool enough of the people enough of the time to make a profit. But if people are better educated, through media literacy, to think more critically about what advertisers are saying and why theyÕre saying it, they may become less gullible. Once, ads were the only way you had of discovering the availability of products. Now, review web sites, recommender systems, online forums and other tools give consumers alternative ways of evaluating products. If people come to rely more on these, advertising will lose its bite. 
Actually, weÕre all a mix of rational reasoning and irrational impulses. But the mix is complicated, itÕs different for different people, and it shifts dynamically and according to context. Neither psychology, nor any other science has yet figured out exactly what that balance is, nor how it is affected by external conditions. Given that neither the total rationalists nor the total irrationalists are correct, where are we? Where should we be? 
We argue that the balance in contemporary society is far too much in the direction of the irrational. While acknowledging that people are not always rational, public discourse should probably put greater emphasis on appeals to peopleÕs rationality, than try to optimize utilization of peopleÕs irrational impulses. Though being rational is more work, and people are sometimes lazy, we ought to encourage people to do that work when appropriate, rather than appeal to their laziness. 
A second difficulty is that being rational, or understanding a rational argument, requires basic principles of math and science. Here we run up against the poor state of math and science education in this country. Appeals to simple emotional drives donÕt require any of this. 
Efforts to improve math and science education will reap enormous long-term benefits in improving peoplesÕ ability to make better decisions across a wide variety of situations in their lives. They will be less susceptible to exploitation and trickery, to short-term impulsive decisions that will have deleterious long-term consequences. ItÕs always better to give people the respect they deserve, and hope theyÕll live up to it, than disrespect them at the outset on the grounds they canÕt possibly merit respectful treatment. 
Education now spends much time on rote memorization of facts. Instead, these topics need to be explicitly taught in schools:
Making a scientific case. What counts as evidence. How to infer from data and specific examples to general principles, and vice versa. Scientific arguments are usually presented in science classes, but the process of scientific investigation itself receives little attention. Learning that is more important than any specific scientific topic. 
Argumention and debate. How to understand other peoplesÕ points of view. How to convince somebody of something. How to recognize and deßect ad hominem attacks and other Òdebate tricksÓ. Some of this is taught to debating teams, but it needs to be divorced from the counterproductive competitive stance of such teams. We also think that the technology of decision-support systems should be taught routinely, just as itÕs now unthinkinable to train an accountant without using spreadsheets. 
Conßict resolution techniques. Finding common values and win-win solutions. How to handle anger and the emotional aspects of conßicts. How to disagree respectfully. Negotiation techniques. How to de-escalate conßicts.
Cognitive biases. Advertisers, marketers, and politicians have learned to selfishly profit from the bounded rationality and unavoidable limits on rationality that Ariely, Kahnemann [Ariely 2008] and others point out. Students need to recognize and learn such tricks and understand how to think in spite of them. When you see the $3.99 product, immediately round up to $4 (maybe even $4.50 with tax or other add-ons youÕre supposed to forget about!). 
Trying to work on improving rationality at least holds the possibility that we could work together to come to the best solutions possible in a given circumstance. All the irrationalists can promise is an endless war between competing impulses, competing nudges, competing incentives, and subliminal suggestions. Dan Ariely has argued that people are Òpredictably irrationalÓ. We say, theyÕre both unpredictably irrational and unpredictably rational. Given that unpredictability, letÕs give them the benefit of the doubt.