¹ For an assessment of the relevant evidence from outside international politics considered within the frameworks developed by studies of international deterrence, see Stern, Paul et al., eds., Perspectives on Deterrence (New York: Oxford University PressGoogle Scholar, forthcoming).

² For an extended discussion, see Jervis, Robert, Perception and Misperception in Politics (Princeton: Princeton University Press, 1976)Google Scholar, chap. 3. Also see Wildavsky, Aaron, “Practical Consequences of the Theoretical Study of Defense Policy,” Public Administration Review 25 (March 1965), 90–103CrossRefGoogle Scholar. For a recent summary of the evidence, see Patchen, Martin, Resolving Disputes Between Nations: Coercion or Conciliation? (Durham, NC: Duke Press, 1988)Google Scholar. A complicating factor is that many abstract and seemingly historical or scientific arguments are driven by political preferences because debates about deterrence are central to American foreign policy. Thus many disagreements about the origins of World Wars I and II or Khrushchev's rationality in putting missiles in Cuba are not purely academic.

³ These cases are more complicated than the stereotypes commonly used by political scientists. See Robert Jervis, “War and Misperception,” Journal of Interdisciplinary History (Spring 1988), 685–88; Glynn, Patrick, “The Sarajevo Fallacy,” National Interest, No. 9 (Fall 1987), 3–32Google Scholar; Lebow, Richard Ned and Stein, Janice Gross, “Beyond Deterrence,” Journal of Social Issues 43, No. 4 (1988), 33–35Google Scholar.

⁴ See, for example, Young, Oran R., “Professor Russett: Industrious Tailor to a Naked Emperor,” World Politics 21 (April 1969), 486–511CrossRefGoogle Scholar; Russett, Bruce, “The Young Science of Inter national Politics,” World Politics 22 (October 1969), 87–94CrossRefGoogle Scholar; Knorr, Klaus and Rosenau, James, eds., Contending Approaches to International Politics (Princeton: Princeton University 1969)Google Scholar, esp. Marion Levy, Jr., “ ‘Does It Matter If He's Naked?’ Bawled the Child.”

⁵ Good reviews of these kinds of theories are Paul Schoemaker, “The Expected Utility Model: Its Variants, Purposes, Evidence and Limitations,” Journal of Economic Literature (June 1982), 529–63; and Machina, Mark, “Choice Under Uncertainty: Problems Solved and Unsolved,” Economic Perspectives 1 (Summer 1987), 121–54CrossRefGoogle Scholar. Debates about this approach not unique to political science. In anthropology the argument is joined, for example, by Sahlins, Marshall, Stone Age Economics (New York: Aldine, 1972)Google Scholar and Harris, Marvin, Cultural Materialism: The Strugglefor a Science of Culture (New York: Random House, 1979)Google Scholar. A summary of the arguments between psychologists and economists in the latter's own field can be found in the special issue of The Journal of Business 59 (October 1986). For the arguments in sociology, see Hirsch, Paul, Michaels, Stuart, and Friedman, Ray, “ ‘Dirty Hands’ Versus ‘Clean Models’: Is Sociology in Danger of Being Seduced by Economics?” Theory and Society 16 (May 1987), 317–36CrossRefGoogle Scholar. Also, see Gravovetter, Mark, “Economic Action and Social Structure: The Problem of Indebtedness,” American Journal of Sociology 91 (November 1985), 481–510CrossRefGoogle Scholar. Also see Barry, Brian, Sociologists, Economists, and Democracy (Chicago: University of cago Press, 1978)Google Scholar. For the purposes of this article, I will equate rationality with probability-utility calculus, as most authors do, but this glosses over the question of whether it is always rational to choose by this criterion. Doing so, for example, could mean following a policy that entailed some chance of enormous loss, and we might not want to say that a person who chose a safer policy with a lower expected utility was violating the tenets of all rational models.

⁶ Achen and Snidal, “Rational Deterrence Theory and Comparative Case Studies,” World Politics 41 (January 1989), 156–59Google Scholar.

⁷ See Almond, Gabriel and Genco, Stephen, “Clouds, Clocks, and the Study of Politics,” World Politics 29 (July 1977), 489–522CrossRefGoogle Scholar, and Hirschman, Albert, “The Search for Paradigms as a Hindrance to Political Understanding,” World Politics 22 (April 1970), 329–43CrossRefGoogle Scholar.

⁸ For good discussions, see Eckstein, Harry, “Case Study and Theory in Political Science,” in Greenstein, Fred and Polsby, Nelson, eds., Handbook of Political Science, Vol. 7 (Reading, MA: Addison-Wesley, 1975), 79–137Google Scholar; George, Alexander L. and McKeown, Timothy, “Case Studies and Theories of Organizational Decision-Making,” in Coulam, Robert and Smith, Richard, eds., Advances in Information Processing in Organizations, Vol. 2 (Greenwich, CT: JAI Press, 1985), 21–58Google Scholar; Campbell, Donald, “ ‘Degrees of Freedom’ and the Case Study,” Comparative Political Studies 8 (July 1975), 178–93CrossRefGoogle Scholar.

⁹ See, for example, the path-breaking study by Abelson, Robert and Rosenberg, Milton, “Symbolic Psycho-logic,” Behavioral Science 3 (January 1958), 1–13CrossRefGoogle Scholar. Artificial intelligence models are in this tradition.

¹⁰ See Jervis, Robert, “Deterrence Theory Revisited,” World Politics 31 (January 1979), 301CrossRefGoogle Scholar, and The Illogic of American Nuclear Strategy (Ithaca, NY: Cornell University Press, 1984)Google Scholar.

¹¹ George, Alexander L. and Smoke, Richard, Deterrence in American Foreign Policy (New York: Columbia University Press, 1974), 519–33Google Scholar. As Achen and Snidal (fn. 6) note, however, these propositions were reached inductively.

¹² For further discussion, see Achen and Snidal (fn. 6), 152; Richard Ned Lebow, “Deterrence: A Political and Psychological Critique,” in Stern (fn. 1); George Quester, “Some Thoughts on ‘Deterrence Failures,’ ” ibid.

¹³ Jervis (fn. 3), 677–79.

¹⁴ For further discussion, see George and Smoke, “Deterrence and Foreign Policy,” World Politics 41 (January 1989), 170–82CrossRefGoogle Scholar.

¹⁵ Achen and Snidal's assertion that RDT has influenced policy then implies that it could not previously explain it: Achen and Snidal (fn. 6), 153. For a discussion of the alternative purposes of SEU models, see Schoemaker (fn. 5), 538–41.

¹⁶ But see Waltz, Kenneth, Theory of International Politics (Reading, MA: Addison-Wesley, 1979)Google Scholar. The application of the idea of evolutionary dynamics is not entirely straightforward. Traits with survival value do not automatically appear; they can survive even if they lack such value. A great deal depends on the harshness of the environment and the competition. For good discussions of evolution and SEU models, see Nelson, Richard and Winter, Sydney, An Evolutionary Theory of Economic Change (Cambridge: Harvard University Press, 1982)Google Scholar, and McKeown, Timothy, “The Limitations of'Structural' Theories of Commercial Policy,” International Organization 40 (Winter 1986), 52–55CrossRefGoogle Scholar.

¹⁷ Whether prescriptive, descriptive, or both, it is often unclear whether the claim is that the policy is an appropriate way to reach the person's goals or the best way to do so. The latter and stronger claim is obviously extremely difficult to verify since doing so requires comparing the results of the policy with those imputed to alternatives—in one version those policies the person thought of and in another version all those that could have been adopted. A strict interpretation of SEU would imply that the policy is the best one; most tests in social science more modestly seek to demonstrate that the policy is adequate for the goals.

¹⁸ A good summary of relevant literature is Robert Wilson, “Deterrence in Oligopolistic Competition,” in Stern et al., eds. (fn. 1).

¹⁹ Achen and Snidal (fn. 6), 153.

²⁰ See, for example, Wohlstetter, Albert, “Swords Without Shields,” National Interest, No. 8 (Summer 1987), 31–57Google Scholar; Gray, Colin, Nuclear Strategy and National Style (Lanham, MD: Hamilton Press, 1986)Google Scholar. Part of the explanation for this dispute is a disagreement on whether decision makers focus on relative or absolute gains and losses. See Jervis (fn. 10, 1984), 59–63; Arthur Stein, Dilemmas of Interdependence: Logics of International Conflict and Cooperation, forthcoming.

²¹ For convenience, I will treat these empirical studies as though the methods and results were entirely compatible with each other, which is not the case.

²² For a discussion of the second wave, see Jervis (fn. 10, 1979), 291–301. It should also be noted that, while these writings are generally called a theory, they do not meet strict criteria for this designation.

²³ See Lebow, Richard Ned, Between Peace and War (Baltimore, MD: The Johns Hopkins University Press, 1981)Google Scholar; Jervis, Robert, Lebow, Richard Ned, and Stein, Janice Gross, Psychology and Deterrence (Baltimore, MD: The Johns Hopkins University Press, 1985)Google Scholar; Lebow and Stein (fn. 3). I am using the Nehru case, drawn from Lebow, Richard Ned and Maxwell, Neville, India's China War (New York: Pantheon, 1970)Google Scholar, as a stylized illustration and am ignoring the complexities and ambiguities.

²⁴ For a similar argument, see Tetlock, Philip, “Testing Deterrence Theory: Some Conceptual and Methodological Issues,” Journal of Social Issues 43, No. 4 (1987), 85–92CrossRefGoogle Scholar.

²⁵ For this reason, Lebow and Stein argue that the resulting theory would not be one of deterrence: “Rational Deterrence Theory: I Think, Therefore I Deter,” World Politics 41 (January 1989), 212–14Google Scholar. Also see George and Smoke (fn. 14), 181–82.

²⁶ The classic statements are Wohlstetter, Albert, “The Delicate Balance of Terror,” Foreign Affairs 37 (January 1959), 211–34CrossRefGoogle Scholar, and Schelling, Thomas, “The Reciprocal Fear of Surprise Attack,” in Schelling, Strategy of Conflict (Cambridge: Harvard University Press, 1960), chap. 9Google Scholar. Many previous wars developed because decision makers believed that the choice was between “war now and war later”: Schilling, Warner et al. ,. American Arms and a Changing Europe (New York: Columbia University Press, 1973), 172–74Google Scholar. For an argument that the main danger of such dynamics today is largely psychological, see Jervis, Robert, Implications of the Nuclear Revolution (Ithaca, NY: Cornell University Press, forthcoming), chap. 5Google Scholar.

²⁷ Indeed these points follow from a full SEU model: see Mueller, John, Retreatfrom Doomsday: The Obsolescence of Major War (New York: Basic Books, forthcoming)Google Scholar.

²⁸ For a further discussion, see George and Smoke (fn. n), 77–78; George and Smoke (fn. 14), 181–82; Jervis (fn. 10, 1979). The use of the term “diagnostic” in George and Smoke (fn. 14), 180 refers to the utility of deterrence theory in pointing out the situations in which deterrence is likely to be challenged, but the theory cannot tell statesmen whether they are facing such a situation.

²⁹ This also was the main point of earlier critics of the second wave: see, for example, Osgood, Charles, An Alternative to War or Surrender (Urbana: University of Illinois Press, 1962)Google Scholar.

³⁰ Thus George has stressed the need to develop differentiated theories because cases are of different types: George and Smoke (fn. 11); George, Alexander L., Farley, Philip, and Dallin, Alexander, eds., U.S.-Soviet Security Cooperation (New York: Oxford University 1988)Google Scholar, chaps. 1 and 29; George and Smoke (fn. 14), 170–73. (A parallel argument from another field is that “no single theory is capable of explaining international trade in all commodities and at all times”: El-Agraa, Ali, The Theory ofInternational Trade [London: Helm, 1983], 85Google Scholar; also see Gilpin, Robert, The Political Economy of International [Princeton: Princeton University Press, 1987], 174–80Google Scholar.) For further discussion of why the theory would no longer be one of deterrence, see Lebow and Stein (fn. 25).

³¹ Achen and Snidal (fn. 6), 160–63; also see Lebow and Stein (fn. 3), 7–8; George and Smoke (fn. 11), 516–17; Jervis, Lebow, and Stein (fn. 23), 13; Jervis (fn. 2), 368–70; Levite, Ariel, Intelligence and Strategic Surprises (New York: Columbia University Press, 1987), 23–24Google Scholar. In The War Trap (New Haven: Yale University Press, 1981)Google Scholar, Bruce Bueno de Mesquita argues that social science should seek only necessary conditions, in which case the problem does not arise. The limitations of this approach are obvious, however.

³² Morgan, , Deterrence: A Conceptual Analysis (Beverly Hills, CA: Sage, 1977), 31–43Google Scholar.

³³ Huth, Paul and Russett, Bruce, “Deterrence Failure and Crisis Escalation,” International Studies Quarterly 32 (March 1988)CrossRefGoogle Scholar, at 30, acknowledge this, but some of their analysis then proceeds as though this were not the case, as the students in my graduate seminar pointed out to me.

³⁴ Lebow and Stein (fn. 3), 33–36.

³⁵ See George and Smoke (fn. 14), 179.

³⁶ See George and Smoke (fn. 14), 173–74 and George, Farley, and Dallm (fn. 30), 5, 11–12.

³⁷ Supporting evidence and detailed arguments can be found in Snyder, Glenn and Diesing, Paul, Conflict Among Nations (Princeton: Princeton University Press, 1977)Google Scholar; George and Smoke (fn. 11); Morgan (fn. 32); Lebow (fn. 23); Jervis (fn. 2); Jervis, Lebow, and Stein (fn. 23); George, Alexander L., Hall, David, and Simons, William, The Limits of Coercive Diplomacy (Boston: Little, Brown, 1971)Google Scholar; Janis, Irving and Mann, Leon, Decision Maying (New York: Free Press, 1977)Google Scholar; and White, Ralph, Fearful Warriors (New York: Free Press, 1984)Google Scholar.

³⁸ Simon, , Models of Man (New York: Wiley, 1957)Google Scholar.

³⁹ The best treatment is Jack Snyder, Myths of Empire: Domestic Politics and Strategic Ideology (Ithaca, NY: Cornell University Press, forthcoming).

⁴⁰ Recent studies of the origins and termination of the World War II conflict between America and Japan show in fascinating detail the opportunities that were missed when one country failed to perceive the nature of the coalitional struggles within its adversary: see Barnhart, Michael, Japan Prepares for Total War (Ithaca, NY: Cornell University Press, 1987)Google Scholar, and Sigal, Leon, Fighting to a Finish: The Politics of War Termination in the United States and Japan, 1945 (Ithaca, NY: Cornell University Press, 1988)Google Scholar.

⁴¹ See, for example, Allison, Graham, Essence of Decision (Boston: Little, Brown, 1971)Google Scholar; Lebow, Richard Ned, Nuclear Crisis Management (Ithaca, NY: Cornell University Press, 1987)Google Scholar; Sagan, Scott, “Nuclear Alerts and Crisis Management,” International Security 9 (Spring 1985), 99–139CrossRefGoogle Scholar; Jervis (fn. 26), chap. 3.

⁴² For conflicting arguments about the relative frequency of the errors of mirror-imaging and lack of empathy, see Laqueur, Walter, A World of Secrets (New York: Basic Books, 1985), 190–94Google Scholar, 272–77; White (fn. 37), 160–67; Jervis, Robert, “Beyond What the Facts Will Bear,” International Journal of Intelligence and Countenntelligence 1 (Spring 1986), 146–48Google Scholar.

⁴³ For a nice example, see Thies, Wallace, When Governments Collide (Berkeley: University of California Press, 1980)Google Scholar.

⁴⁴ See George, Hall, and Simons (fn. 37); Snyder and Diesing (fn. 37); Jervis (fn. 10, 1979); Stein, Janice, “Deterrence and Reassurance,” in Tetlock, Philip et al., eds., Behavior, Society, and Nuclear War, Vol. 2 (New York: Oxford University Press, forthcoming)Google Scholar.

⁴⁵ For a parallel argument, see George and Smoke (fn. 14), 182.

⁴⁶ For discussions of this issue, see Elster, Jon, “Introduction,” in Elster, ed., Rational Choice (New York: New York University Press, 1986), 4–5Google Scholar, 13–16; Barry O'Neill, “Game Theory and the Study of the Deterrence of War,” in Stern (fn. 1); Lebow and Stein (fn. 25), 216–18.

⁴⁷ The familiar analogy to athletes, put forward by Friedman, Milton and Savage, Leonard, “The Utility Analysis of Choices Involving Risk.” Journal of Political Economy 56 (August 1948)CrossRefGoogle Scholar and Achen and Snidal (fn. 6), 164, does not really help. Statesmen act in a realm of much greater complexity, deal with many fewer instances, and only rarely receive unambiguous feedback. For a discussion of the more interesting issue of whether information-processing biases and fallacies really can be considered deviations from rationality, see Cohen, L. Jonathan, “Can Human Irrationality be Experimentally Demonstrated?” Behavioral and Brain Sciences 4 (September 1981), 317–31CrossRefGoogle Scholar, and the comments, ibid., 331–70. For an SEU model that takes information-processing costs into account, see Anderson, Paul and McKeown, Timothy, “Changing Aspirations, Limited Attention, and War,” World Politics 40 (October 1987), 1–29CrossRefGoogle Scholar.

⁴⁸ Riker, William and Ordeshook, Peter, An Introduction to Positive Political Theory (Engle-wood Cliffs, NJ: Prentice-Hall, 1973)Google Scholar; Downs, Anthony, An Economic Theory ofDemocracy (New York: Harper & Row, 1957)Google Scholar. Also see Jervis (fn. 2), chap. 4.

⁴⁹ For some relevant findings, see Sears, David and Freedman, Jonathan, “Selective Exposure to Information: A Critical Review,” Public Opinion Quarterly 31 (Summer 1967), 194–213CrossRefGoogle Scholar. McGuire, William, “Selective Exposure: A Summing Up,” in Robert Abelson et al., eds., Theories of Cognitive Consistency (Chicago: Rand McNally, 1968), 797–800Google Scholar; Tetlock, Philip, “Ac countability: The Neglected Social Context of Judgment and Choice,” in Staw, Barry M. and Cummings, L. L., eds., Research in Organizational Behavior, Vol. 7 (Greenwich, CT: JAI Press, 1985)Google Scholar.

⁵⁰ Robert Jervis, “Deterrence and Perception,” International Security 7 (Winter 1982/83), 14-17.

⁵¹ See, for example, the work summarized in Kahneman, Daniel, Slovic, Paul, and Tversky, Amos, eds., Judgment under Uncertainty: Heuristics and Biases (Cambridge: Cambridge University Press, 1982)CrossRefGoogle Scholar, and Tversky and Kahneman, “Rational Choice and the Framing of Decisions,” in Journal ofBusiness (fn. 5), S251–78.

⁵² The classic account is Smith, M. Brewster, Bruner, Jerome, and White, Robert, Opinions and Personality (New York: Wiley, 1956)Google Scholar.

⁵³ For further discussion, see Jervis (fn. 2), 374–78; and Stein, Arthur, “When Misperception Matters,” World Politics 34 (July 1982), 505–26CrossRefGoogle Scholar.

⁵⁴ Schelling (fn. 26), 36–43.

⁵⁵ Elster (fn. 46), 10–12.

⁵⁶ For related arguments, see Elster, Jon, Sour Grapes: Studies in the Subversion of Rationality (Cambridge: Cambridge University Press, 1983)CrossRefGoogle Scholar, chaps. 3 and 4, and Larson, Deborah, Origins of Containment (Princeton: Princeton University Press, 1985)Google Scholar.

⁵⁷ Warner Schilling, “The Politics of National Defense: Fiscal 1950,” in Schilling, Warner, Hammond, Paul, and Snyder, Glenn, Strategy, Politics, and Defense Budgets (New York: Co lumbia University Press, 1962), 1–266Google Scholar; Thies (fn. 43).

⁵⁸ Tokushiro, Ohata, “The Anti-Comintern Pact, 1935–1939,” in Morley, James, ed., Deterrent Diplomacy: Japan, Germany, and the USSR, 1935–1940 (New York: Columbia University Press, 1976), 47–111Google Scholar.

⁵⁹ See Gaddis, John, Strategies of Containment (New York: Oxford University Press, 1982)Google Scholar.

⁶⁰ Holsti, Ole and Rosenau, James, American Leadership in World Affairs: Vietnam and the Breakdown of Consensus (Boston: Allen & Unwin, 1984)Google Scholar.

⁶¹ Modigliani, Andre, “Hawks and Doves, Isolationism and Political Distrust,” American Political Science Review 66 (September 1972), 960–78CrossRefGoogle Scholar. For rational choice discussions of how institutions can provide solutions to cyclical majorities, see Shepsle, Kenneth and Weingast, Barry, “Political Solutions to Market Problems,” American Political Science Review 78 (March 1984), 417–34CrossRefGoogle Scholar; and Shepsle, Kenneth and Weingast, Barry, “Structure Induced Equilibrium and Legislative Choice,” Public Choice 37, No. 3 (1982), 503–19CrossRefGoogle Scholar.

⁶² Japanese and U.S. policies at the end of World War II provide several examples; see Sigal (fn. 40).

⁶³ See Jervis, Robert, “Realism, Game Theory, and Cooperation,” World Politics 40 (April 1988), 324–28, 340–44CrossRefGoogle Scholar.

⁶⁴ The wine example comes from Thaler, Richard, “Toward a Positive Theory of Consumer Choice” Journal of Economic Behavior and Organization 1 (March 1980), 43CrossRefGoogle Scholar; the other examples are from Tversky, Amos and Kahneman, Daniel, “The Framing of Decisions and the Psychology of Choice,” Science 211 (January 1981), 453–58CrossRefGoogle Scholar. Also see Grether, David and Plott, Charles, “Economic Theory of Choice and the Preference Reversal Phenomenon,” American Economic Review 69 (September 1979), 623–38Google Scholar; Tversky and Kahneman (fn. 51); Jack Knetsch, Richard Thaler, and Daniel Kahneman, “Experimental Tests of the Endowment Effect and the Coase Theorem” (unpub.); MacCrimmon, Kenneth and Wehrung, Donald, Taking Risfy (New York: Free Press, 1986)Google Scholar. For applications to foreign policy, see Jervis (fn. 26), chap. 5, and Nancy Kanwisher, “Cognitive Heuristics and American Security Policy,” Journal of Conflict Resolution (forthcoming).

⁶⁵ See Simon, Herbert, “Human Nature in Politics: The Dialogue of Psychology and Political Science,” American Political Science Review 79 (June 1985), 292–303CrossRefGoogle Scholar; Jervis (fn. 63), 324–29.