Free beta access. 50 free AI tokens each week. Optional top-ups available. See what is included
Opening Word document…

That’s descriptivist fallacy – deterrence theory doesn’t have to prove that the decision maker’s mental choices perfectly align, just that the consequences are aligned with the theory

Achen and Snidal 89 (Christopher H. and Duncan, professor of politics at Princeton University, professor of international relations at Oxford University, “Rational Deterrence Theory and Comparative Case Studies,” 1989, https://www.cambridge.org/core/journals/world-politics/article/abs/rational-deterrence-theory-and-comparative-case-studies/05B3C8641D5EF6F9A36067EEA0210CCE#:~:text=Several%20recent%20books%20have%20argued%20that%20comparative%20case,This%20paper%20shows%20that%20such%20contentions%20are%20unwarranted.)

There is a second objection to the empirical testing of rational deterrence in comparative case studies—namely, that the verification procedure fails to engage the rational deterrence theory of how states behave. In particular, rational deterrence is implicitly misconstrued as a theory of how decision makers think. Case study analysts have often expressed the opinion that, if decision makers do not really carry out the appropriate mental calculations, rational deterrence theory does not apply. This argument is the "descriptivist fallacy." As a glance at any appropriate text will show, the axioms and conclusions of utility theory refer only to choices. Mental calculations are never mentioned: the theory makes no reference to them. Indeed, a major reason for the various axiomatizations of expected-utility theory is to show that decision makers need not calculate. If they simply respond to incentives in certain natural ways, their behavior will be describable by utility functions.65 Rapoport puts it this way: Now, in being asked to choose between the two [lottery] tickets, the man is not asked to calculate anything. He is asked simply to choose between the two. It is from his choices that the theoretician will construct the man's util- ity scale on which all the outcomes will be assigned numbers (utilities).66 To our knowledge, no one who does rational-choice theorizing disputes this point, but it seems not to be widely understood. Rational deterrence theory does contain some minimal psychological content: for example, the initiator must realize that the defender exists and threatens to defend. But rational deterrence is agnostic about the actual calculations decision makers undertake. It holds that they will act as if they solved certain mathematical problems, whether or not they actually solve them. Just as Steffi Graf plays tennis as if she did rapid computations in Newtonian physics (and in game theory, too—at least against Navratilova), so rational deterrence theory predicts that decision makers will decide whether to go to war as if they did expected-utility calculations. But they need not actually perform them.67 To avoid misunderstandings, we want to point out that our point here has nothing to do with Friedman's famous argument that the truth of assumptions need not matter.68 Friedman maintained that a theory might sometimes work well for certain purposes, even though its assumptions were known to be false. We have our doubts; but whatever its merits, this line of reasoning has no bearing on our argument. Our point is that even if decision makers do not actually calculate, or if they rationalize their actions after the fact with foolish calculations, the assumptions of rational choice theory may yet remain true. Understanding what rational deterrence demands of decision makers is important because it sharply limits what counts against the theory. A proper understanding of the theory eliminates most of the arguments in the historical deterrence literature that are supposed to overturn it. Consider, for example, the fact that many decision makers do not seem to think in terms of probabilities at all. This point is made by Steinbruner with respect to Kennedy's remark that he faced certain impeachment if he did nothing about the Cuban missiles;69 Stein has made it in connection with Anwar Sadat's forecasts of various foreign policy consequences, for Egypt in the early 1970s.70 Both authors maintain that, because each decision maker failed to discuss probabilities in situations where rational deterrence theory implies he should, rational deterrence is inaccurate in describing what happened. Related arguments have been advanced by Jervis and by Betts.7 ' These conclusions do not follow. Let us repeat that rational deterrence theory deals with choices, not mental calculations. It makes no predictions about what decision makers say influenced them, only about what actually did so. The distinction is particularly important in the case of postdecision reconstructions. As diplomatic historians have long been aware, the historical record often differs sharply from decision makers' memories, even memories about their own thoughts at the time.72 Since the critics of deterrence wish to adduce evidence that is not only dubious but apparently irrelevant to the theory under discussion, they bear the burden of proof. They must show that no plausible psychological mechanisms could have introduced a gap between the reality and the reconstruction, and they must demonstrate that what decision makers say influenced them is identical with what actually happened. Neither Stein nor Steinbruner have made such a case. Indeed, several plausible psychological mechanisms would render their evidence irrelevant—mechanisms of the sort that they and others have often put forward as key factors in international decision making. Suppose, then, that decision makers actually behave according to rational-choice theory. Indeed, no matter what he said, it is hard to believe that Kennedy's Cubanmissile decisions were unaffected by the difference between a.possibility of impeachment, a livelihood of impeachment, and a certainty of impeachment. But suppose that, like the rest of us, Kennedy subsequently bolstered his decision by thinking and saying that any other choice would have brought on unpleasant political consequences with certainty. If so, he would speak in the apparently nonrational language that he actually used, but his behavior would be predicted perfectly by rationality assumptions. Obvious competing interpretations like this one have not been excluded by the critics of rational deterrence. It is not easy to do so. Indeed, no matter how detailed the historical records, disagreements break out routinely over interpretations of decision makers' thoughts and intentions.73 The degree of uncertainty in historical interpretations of motive and intention is often substantial, and conclusions deriving from such interpretations must be discounted appropriately. Yet case-study analysts usually provide no assessment of the reliability of their historical judgments. In particular, without a detailed showing that the questionable memories of decision makers represent the historical process accurately, no amount of evidence about their apparent calculation errors is relevant to the issue of whether rational deterrence theory predicts decision making.^ In the absence of such a showing, rational deterrence theory stands unrebutted.75 It is not our central point here to set out one particular list of inferential slips by case-study analysts. After all, one can imagine case studies that select cases in a nonprejudicial fashion, and that take proper account of the fallibility in assessments of decision makers' intentions. Well-designed case-study tests may not be decisive, but they can be highly enlightening and strongly persuasive.?6 They are certainly an indispensable first step before proceeding to statistical methods—a point quantitative researchers have too often ignored at their peril. In principle, case studies are capable of doing everything statistical analysis can do; they may simply replace symbols with words and replicate the random samples, precise definitions, and rigorous inferences of statistical methods. But in practice, inferential rigor is not the best tool for what case studies should accomplish, and the comparative method works best when it enforces no such discipline. As Jack Snyder notes, "The best analysts in the [case study] field have always used a rough-and-ready version of the scientific method," which they have identified, not with the powerful counterintuitive logic of the real McCoy, but rather with the substantially tamer "self-conscious, systematic application of commonsense rules of inference."77 When the purposes of comparative case studies are properly understood, informality is the right choice. Creativity is enhanced when historical cases can be chosen at liberty and analyzed in accord with the investigator's intuitions. The great many degrees of freedom in these studies are inordinately helpful in finding useful variables and producing empirical generalizations. Like all good things, however, this free play for unaided common sense has a price. Used in isolation from scientific methods, it creates inevitable inferential errors and makes decisive theory verification well-nigh impossible. In case studies of rational deterrence, the consequences have been both predictable and devastating for the credibility of their conclusions

Sign in to recut, update, or save this card.