1. Steinberg, J., “The Copenhagen Complex,” Journal of Contemporary History, vol. 1, no. 3, 1966, pp. 23–24CrossRefGoogle Scholar. On the lessons of Munich and Korea, see May, Ernest R., “Lessons” of the Past (New York: Oxford University Press, 1973)Google Scholar; Jervis, Robert, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976), chap. 6Google Scholar; and Khong, Yuen Foong, Analogies at War: Korea, Munich, Dien Bien Phu, and the Vietnam Decisions of 1965 (Princeton, N.J.: Princeton University Press, 1992)Google Scholar. On Vietnam, see Ravenal, Earl C., Never Again: Learning from America's Foreign Policy Failures (Philadelphia: Temple University Press, 1978)Google Scholar; and Holsti, Ole R. and Rossnau, James N., American Leadership in World Affairs (Boston: Allen and Unwin, 1984)Google Scholar. On 1815 and 1919, see Butterfield, Herbert, History and Human Relations (London: Collins, 1951), pp. 176–77Google Scholar. On the Great Depression, see Kindleberger, Charles P., The World in Depression 1919–1939 (Berkeley: University of California Press, 1973)Google Scholar; and Odell, John S., “From London to Bretton Woods,” Journal of Public Policy 8 (07–12 1988), pp. 287–315CrossRefGoogle Scholar.

2. See May, “Lessons” of the Past; Jervis, Perception and Misperception in International Politics; Etheridge, Lloyd S., “Government Learning,” in Long, Samuel L., ed., The Handbook of Political Behavior, vol. 2 (New York: Plenum, 1981), pp. 73–161CrossRefGoogle Scholar; Etheridge, Lloyd S., Can Governments Learn? (New York: Pergamon, 1985)Google Scholar; and Neustadt, Richard and May, Ernest, Thinking in Time (New York: Free Press, 1986)Google Scholar. Interest in learning goes beyond international relations. Hall argues that “The notion of social learning is on the verge of becoming a key element in contemporary theories of the state and of policymaking more generally.” See Hall, Peter A., “Policy Paradigms, Social Learning, and the State,” Comparative Politics 25 (04 1993), p. 276CrossRefGoogle Scholar.

3. See Breslauer, George W., “Ideology and Learning in Soviet Third World Policy,” World Politics 39 (04 1987), pp. 429–48CrossRefGoogle Scholar; Breslauer, George W., “Explaining Soviet Policy Changes: Politics, Ideology, and Learning,” in Breslauer, George W., ed., Soviet Policy in Africa: From the Old to the New Thinking (Berkeley: University of California Press, 1992), pp. 196–216Google Scholar; Breslauer, George W. and Tetlock, Philip E., eds., Learning in U.S. and Soviet Foreign Policy (Boulder, Colo.: Westview, 1991)Google Scholar; Bennett, Andrew Owen, “Theories of Individual, Organizational, and Governmental Learning and the Rise and Fall of Soviet Military Interventionism 1973–1983,” Ph.D. diss., Harvard University, 1990)Google Scholar; Bennett, Andrew Owen, “Patterns of Soviet Military Interventionism 1975–1990,” in Zimmerman, William, ed., Beyond the Soviet Threat: American Security Policy in a New Era (Ann Arbor: University of Michigan Press, 1992)Google Scholar; Evangelista, Matthew, “Sources of Moderation in Soviet Security Policy,” in Tetlock, Philip E. et al. , eds., Behavior, Society, and Nuclear War, 3 vols. (New York: Oxford University Press, 1991), pp. 2:254–354Google Scholar; Wallander, Celeste, “Opportunity, Incrementalism, and Learning in the Extension and Retraction of Soviet Global Commitments,” Security Studies 1 (Spring 1992), pp. 514–42CrossRefGoogle Scholar; Mendelson, Sarah E., “Internal Battles and External Wars: Politics, Learning, and the Soviet Withdrawal from Afghanistan,” World Politics 45 (04 1993), pp. 327–60CrossRefGoogle Scholar; and Janice Gross Stein, “Political Learning by Doing: Gorbachev as Uncommitted Thinker and Motivated Learner,” this issue of International Organization.

4. See Snyder, Jack, Myths of Empire: Domestic Politics and International Ambition (Ithaca, N.Y.: Cornell University Press, 1991)Google Scholar; Khong, Analogies at War; Nye, Joseph S. Jr, “Nuclear Learning and U.S.–Soviet Security Regimes,” International Organization 41 (Summer 1987), pp. 371–402CrossRefGoogle Scholar; George, Alexander L., Farley, Philip J., and Dallin, Alexander, eds., U.S.–Soviet Security Cooperation (New York: Oxford University Press, 1988)Google Scholar; Leng, Russell J., “When Will They Ever Learn? Coercive Bargaining in Recurrent Crises,” Journal of Conflict Resolution 27 (09 1983), pp. 379–419CrossRefGoogle Scholar; Leng, Russell J., “Crisis Learning Games,” American Political Science Review 82 (03 1986), pp. 179–94CrossRefGoogle Scholar; Huth, Paul and Russett, Bruce, “What Makes Deterrence Work? Cases from 1900 to 1980,” World Politics 36 (07 1984), pp. 496–526CrossRefGoogle Scholar; and Dan Reiter, “Learning, Realism, and Alliances,” World Politics, forthcoming.

5. See Goldstein, Judith and Keohane, Robert O., eds., Ideas and Foreign Policy: Beliefs, Institutions, and Political Change (Ithaca, N.Y.: Cornell University Press, 1993)Google Scholar; Haas, Ernst B., When Knowledge Is Power (Berkeley: University of California Press, 1990)Google Scholar; Haas, Peter M., ed., “Knowledge, Power, and International Policy Coordination,” special issue of International Organization 46 (Winter 1992)Google Scholar, and in particular the contribution by Adler, Emanuel, “The Emergence of Cooperation: National Epistemic Communities and the International Evolution of the Idea of Nuclear Arms Control,” pp. 101–45Google Scholar.

6. See Tamashiro, Howard, “Algorithms, Heuristics, and the Artificial Intelligence Modeling of Strategic Statecraft,” in Sylvan, Donald A. and Chan, Steve, eds., Foreign Policy Decision Making (New York: Praeger, 1984), pp. 197–226Google Scholar; Khong, Analogies at War; Vertzberger, Yaacov Y.I., The World in Their Minds (Stanford, Calif.: Stanford University Press, 1990)Google Scholar; and Hybel, Alex Roberto, How Leaders Reason (Cambridge: Basil Blackwell, 1990)Google Scholar.

7. See Axelrod, Robert, The Evolution of Cooperation (New York: Basic Books, 1984)Google Scholar; Wagner, R. Harrison, “Uncertainty, Rational Learning, and Bargaining in the Cuban Missile Crisis,” in Ordeshook, Peter C., ed., Models of Strategic Choice in Politics (Ann Arbor: University of Michigan Press, 1989), pp. 177–205Google Scholar; Powell, Robert, Nuclear Deterrence Theory: The Search for Credibility (Cambridge: Cambridge University Press, 1990)CrossRefGoogle Scholar; and Farkas, Andrew, “State Learning and International Change,” Ph.D. diss., Princeton University, 1994Google Scholar.

8. I have dealt briefly with some of these issues in Levy, Jack S., “Learning from Experience in U.S. and Soviet Foreign Policy,” in Midlarsky, Manus I., Vasquez, John A., and Gladkov, Peter, eds., From Rivalry to Cooperation (New York: HarperCollins, 1994), pp. 56–86Google Scholar.

9. On learning in foreign economic policy, see Odell, John S., U.S. International Monetary Policy (Princeton, N.J.: Princeton University Press, 1982)CrossRefGoogle Scholar.

10. On social learning theory, see Bandura, Albert, Social Learning Theory (Englewood Cliffs, N.J.: Prentice-Hall, 1977)Google Scholar. See also Lott, Bernice and Lott, Albert J., “Learning Theory in Contemporary Social Psychology,” in Lindzey, Gardner and Aronson, Elliot, eds., Handbook of Social Psychology, vol. 1 (New York: Random House, 1985), pp. 109–35Google Scholar; and McGuire, William J., “Attitudes and Attitude Change,” in Lindzey, and Aronson, , The Handbook of Social Psychology, vol. 2, pp. 233–346Google Scholar. On socialization, see Greenstein, Fred I., Children and Politics (New Haven, Conn.: Yale University Press, 1965)Google Scholar; and Sears, David O., “Political Socialization,” in Greenstein, Fred I. and Polsby, Nelson W., eds., Handbook of Political Science, vol. 2 (Reading, Mass.: Addison-Wesley, 1975), pp. 93–153Google Scholar.

11. For excellent reviews of alternative conceptions of political learning, see Etheridge, “Government Learning”; and the following essays in Breslauer and Tetlock, Learning in U.S. and Soviet Foreign Policy: Breslauer, George W. and Tetlock, Philip E., “Introduction,” pp. 1–19Google Scholar; Tetlock, Philip E., “Learning in U.S. and Soviet Foreign Policy: In Search of an Elusive Concept,” pp. 20–61Google Scholar; Haas, Ernst B., “Collective Learning: Some Theoretical Speculations,” pp. 62–99Google Scholar; and Breslauer, George W., “What Have We Learned About Learning?” pp. 825–56Google Scholar.

12. See p. 324 of Levitt, Barbara and March, James G., “Organizational Learning,” Annual Review of Sociology 14 (1988), pp. 324–34CrossRefGoogle Scholar. Similarly, most research in social psychology concludes that interpretation of reality tends to be more theory-driven than data-driven. See Nisbett, Richard and Ross, Lee, Human Inference (Englewood Cliffs, N.J.: Prentice-Hall, 1980)Google Scholar.

13. Organizational search is a central concept in organizational theory, and information search is central to rational choice theories of politics.

14. Wildavsky writes (of hazardous technologies) that through “small-scale trial and error, we develop skills for dealing with whatever may come our way from the world of unknown risks.” See Wildavsky, Aaron, Searching for Safety (New Brunswick, N.J.: Transaction Books, 1988), p. 37Google Scholar. A deliberate “strategy of small losses” to maximize learning from failure is proposed by Sitkin, Sam B., “Learning Through Failure: The Strategy of Small Losses,” Research in Organizational Behavior, vol. 14 (JAI Press 1992), pp. 231–66Google Scholar. Such a strategy is more feasible, however, for organizational managers than for statesmen confronting national security threats.

15. See Sagan, Scott D., The Limits of Safety: Organizations, Accidents, and Nuclear Weapons (Princeton, N.J.: Princeton University Press, 1993)Google Scholar; Levitt, and March, , “Organizational Learning,” p. 334Google Scholar; Tamuz, Michael, “Monitoring Dangers in the Air: Studies in Ambiguity and Information,” Ph.D. diss., Stanford University, 1988Google Scholar.

16. See Evangelista, , “Sources of Moderation in Soviet Security Policy,” pp. 275–77Google Scholar; Mueller, John, “The Marketing of Ideas,” manuscript, 1992Google Scholar; and Jervis, Robert, The Logic of Images in International Relations (Princeton, N.J.: Princeton University Press, 1970)Google Scholar.

17. Erickson, John, “Threat Identification and Strategic Appraisal by the Soviet Union, 1930–1941,” in May, Ernest R., ed., Knowing One's Enemies (Princeton, N.J.: Princeton University Press, 1984), pp. 375–423Google Scholar, cited in Jarosz, William W. with Nye, Joseph S. Jr, “The Shadow of the Past: Learning from History in National Security Decision Making,” in Tetlock, Philip E. et al. , Behavior, Society, and International Conflict, p. 3:150Google Scholar.

18. Freedman, Lawrence and Karsh, Efraim, The Gulf Conflict, 1990–1991 (Princeton, N.J.: Princeton University Press, 1993), pp. 279Google Scholar and 435–36.

19. Causal learning may be universal or restricted to certain categories of actors or spatial or temporal domains.

20. I thank Richard Hermann for emphasizing this point to me.

21. See Edwards, Ward, Lindman, H., and Savage, L.J., “Bayesian Statistical Inference for Psychological Research,” Psychological Review, vol. 70, 1963, pp. 193–242CrossRefGoogle Scholar; and Iverson, Gudmund R., Bayesian Statistical Inference (Beverly Hills, Calif.: Sage, 1984)CrossRefGoogle Scholar.

22. On heuristics and biases, see Kahneman, Daniel, Slovic, Paul, and Tversky, Amos, eds., Judgment Under Uncertainty: Heuristics and Biases (Cambridge: Cambridge University Press, 1982)CrossRefGoogle Scholar; and Khong, Analogies at War, chap. 2.

23. Bayesian learning involves an interesting conceptual problem. Probability updating follows directly from environmental stimuli, so that mediating cognitive variables have no independent causal impact. In this sense Bayesian updating is more like structural adjustment or adaptive learning, which I discuss later.

24. On sequential games with incomplete information, see Alt, James, Calvert, Randall, and Humes, Brian, “Reputation and Hegemonic Stability: A Game-Theoretic Analysis,” American Political Science Review 82 (06 1988), pp. 445–66CrossRefGoogle Scholar; Powell, Nuclear Deterrence Theory; and Wagner, “Uncertainty, Rational Learning, and Bargaining in the Cuban Missile Crisis.” For a non-Bayesian analysis of interactive learning, see Weber, Steven, “Interactive Learning in U.S.–Soviet Arms Control,” in Breslauer, and Tetlock, , Learning in U.S. and Soviet Foreign Policy, pp. 784–824Google Scholar. For an attempt to integrate rational and psychological models of learning, see Larson, Deborah Welch, “Experiential Learning in Soviet–American Interaction,” presented at the meeting of the American Political Science Association, Chicago, 3–6 09 1992Google Scholar.

25. See Deutsch, Karl, The Nerves of Government (London: Free Press, 1963), p. 92Google Scholar; Nye, , “Nuclear Learning and U.S.–Soviet Security Regimes,” p. 380Google Scholar. This parallels the distinction between “single-loop” and “double-loop” learning suggested by Argyris, Chris and Schon, Donald A., Organizational Learning (Reading, Mass.: Addison-Wesley, 1980), pp. 20–26Google Scholar.

26. Tetlock, , “Learning in U.S. and Soviet Foreign Policy,” pp. 27–31Google Scholar.

27. See p. 9 of Goldstein, Judith and Keohane, Robert O., “Ideas and Foreign Policy: An Analytic Framework,” in Goldstein, and Keohane, , Ideas and Foreign Policy, pp. 3–30Google Scholar. On normative change, see Keohane, Robert O. and Nye, Joseph S. Jr, “Power and Interdependence Revisited,” International Organization 41 (Autumn 1987), pp. 725–53CrossRefGoogle Scholar.

28. Argyris, and Schon, , Organizational Learning, pp. 26–28Google Scholar.

29. Bennett, , “Theories of Individual, Organizational, and Governmental Learning and the Rise and Fall of Soviet Military Interventionism 1973–1983,” p. 106Google Scholar.

30. Tversky, Amos and Kahneman, Daniel, “Availability: A Heuristic for Judging Frequency and Probability,” in Slovic, Kahneman, and Tversky, , eds., Judgment Under Uncertainty, pp. 163–78Google Scholar.

31. There is an extensive literature on organizational learning. See March, James G. and Olsen, Johan P., “The Uncertainty of the Past: Organizational Learning Under Ambiguity,” in March, James G., Decisions and Organizations (New York: Basil Blackwell, 1988), pp. 335–58Google Scholar; Levitt and March, “Organizational Learning”; Hedberg, Bo, “How Organizations Learn and Unlearn,” in Nystrom, Paul C. and Starbuck, William H., eds., Handbook of Organizational Design, vol. 1 (Oxford: Oxford University Press, 1981), pp. 3–27Google Scholar; Huber, George P., “Organizational Learning: The Contributing Processes and the Literatures,” Organization Science 2 (02 1991), pp. 88–115CrossRefGoogle Scholar. The literature on learning in foreign policy has been slow to incorporate the insights of organizational theorists, but that is beginning to change.

32. Routines refer to the forms, rules, procedures, conventions, strategies, and technologies around which organizations are constructed and through which they operate, as well as the organizational culture and paradigms through which they are interpreted. See Levitt, and March, , “Organizational Learning,” p. 320Google Scholar.

33. The quotations are drawn from Argyris, and Schon, , Organizational Learning, pp. 9–11, 20, and 28Google Scholar; from Heclo, Hugh, Modern Social Politics in Britain and Sweden (New Haven, Conn.: Yale University Press, 1974), p. 306Google Scholar; and from Hedberg, , “How Organizations Learn and Unlearn,” p. 3Google Scholar, respectively.

34. Ravenal, , Never Again, pp. 27–28Google Scholar.

35. Lovell, John P., “‘Lessons’ of U.S. Military Involvement: Preliminary Conceptualization,” in Sylvan, and Chan, , eds., Foreign Policy Decision Making, p. 135Google Scholar.

36. This builds on the conception of a closed cycle of organizational learning elucidated by March, James G. and Olsen, Johan P., Ambiguity and Choice in Organizations (Bergen: Universitetsforlaget, 1976)Google Scholar.

37. Etheridge argues that one of the reasons the U.S. government did not learn from the Bay of Pigs fiasco was that “subordinates were at personal risk if they told the truth”; see Can Governments Learn? p. 100.

38. My conception of organizational learning involves intendedly rational action by individuals to improve organizational routines and behavior. It differs from models of cybernetic learning, which emphasize preprogrammed responses rather than outcome calculations and the evolutionary selection of routines that work. See Steinbrunner, John, The Cybernetic Theory of Decision (Princeton, N.J.: Princeton University Press, 1974), pp. 78–79Google Scholar.

39. Haas and Evangelista each give less emphasis to the institutionalization of learning than I do. See E. Haas, “Collective Learning”; and Evangelista, , “Sources of Moderation in Soviet Security Policy,” p. 272Google Scholar.

40. Hermann, Charles F., “Changing Course: When Governments Choose to Redirect Foreign Policy,” International Studies Quarterly 34 (03 1990), pp. 3–31CrossRefGoogle Scholar.

41. Mendelson, , “Internal Battles and External Wars,” pp. 341–46Google Scholar.

42. See Jervis, Perception and Misperception in International Politics, chap. 4; and Nisbett and Ross, Human Inference.

43. The quotations are drawn from Jarosz, with Nye, , “The Shadow of the Past,” pp. 130 and 180Google Scholar, respectively.

44. The quotations are drawn from Heclo, , Modern Social Politics in Britain and Sweden, p. 306Google Scholar; Hall, , “Policy Paradigms, Social Learning, and the State,” p. 278Google Scholar; and Farkas, , “State Learning and International Change,” p. 20Google Scholar, respectively.

45. In fact, Farkas constructs an evolutionary model of foreign policy change in which individual learning based on cognitive change does not necessarily play a central role.

46. Jervis, , Perception and Misperception in International Politics, p. 222Google Scholar.

47. The quotation is from Tetlock, , “Learning in U.S. Foreign Policy,” p. 22Google Scholar; see also pp. 27–38. Tetlock classifies the first element as the “belief system” (or “cognitive psychological”) approach to learning. Etheridge also defines learning in terms of accuracy and effectiveness. See Etheridge, , Can Governments Learn, p. 66Google Scholar. Note that the efficient matching of means and ends can involve either the adoption of more effective strategies for pursuing one's original goals or the redefinition of one's goals in more realistic ways.

48. Breslauer, , “What Have We Learned about Learning?” in Breslauer, and Tetlock, , Learning in U.S. and Soviet Foreign Policy, p. 825Google Scholar.

49. Stein, , “Political Learning by Doing,” p. 171Google Scholar.

50. Tetlock, , “Learning in U.S. and Soviet Foreign Policy,” pp. 35–36Google Scholar.

51. For a discussion of ill-structured contexts, see Voss, James F. and Post, Timothy A., “On the Solving of Ill-structured Problems,” in Chi, Michelene, Glaser, Robert, and Farr, Marshall J., eds., The Nature of Expertise (Hillsdale, N.J.: Erlbaum, 1988), pp. 261–85Google Scholar.

52. Holsti and Rosenau, American Leadership in World Affairs; and Zimmerman, William and Axelrod, Robert, “The Lessons of Vietnam and Soviet Foreign Policy,” World Politics 34 (10 1981), pp. 1–24CrossRefGoogle Scholar.

53. Similarly, Haas argues that “Notions of affect, imitation, intelligence, effectiveness, and therapy must be banished from our discussion of learning”; see Haas, E., “Collective Learning,” p. 75Google Scholar. Keohane and Nye suggest that we “need not identify [learning] with morally improved action”; see “Power and Interdependence Revisited,” p. 749.

54. Moreover, our ultimate aim is to explain policy change, and incorrect learning is an important source of policy change.

55. See p. 108 of Lounamaa, Pertti H. and March, James G., “Adaptive Coordination of a Learning Team,” Management Science 33 (01 1987), pp. 107–123CrossRefGoogle Scholar.

56. May, , “Lessons” of the Past, p. xiGoogle Scholar.

57. Jervis, , Perception and Misperception in International Politics, p. 228Google Scholar.

58. See Khong, Analogies at War, chap. 2; and Vertzberger, The World in Their Minds, chap. 6.

59. See Tetlock, , “Learning in U.S. and Soviet Foreign Policy,” pp. 22 and 32–35Google Scholar; and Etheridge, , “Government Learning,” pp. 76–79Google Scholar.

60. There are also important methodological problems. The operationalization and measurement of cognitive structure can be very difficult, time-consuming, and data-intensive.

61. Tetlock, concedes this point in “Learning in U.S. and Soviet Foreign Policy,” pp. 34–39Google Scholar.

62. Rosati, Jerel A., The Carter Administration's Quest for Global Community: Beliefs and Their Impact on Behavior (Columbia: University of South Carolina Press, 1991), p. 105Google Scholar.

63. See Tetlock, , “Learning in U.S. and Soviet Foreign Policy,” pp. 33–35Google Scholar; Abelson, Robert P. et al. , eds., Theories of Cognitive Consistency (Chicago: Rand McNally, 1969)Google Scholar; Jervis, Perception and Misperception in International Politics, chap. 4; Stein, “Political Learning by Doing”; and Larson, Deborah Welch, Origins of Containment: A Psychological Explanation (Princeton, N.J.: Princeton University Press, 1985), p. 49Google Scholar.

64. Tetlock, Philip E., “Accountability and Complexity of Thought,” Journal of Personality and Social Psychology 45 (07 1983), pp. 74–83CrossRefGoogle Scholar.

65. Gronich, Lori Helene, “Expertise, Naivete, and Decision-making: The Cognitive Processing Theory of Foreign Policy Choice,” Ph.D. diss., University of California, Los Angeles, 1991)Google Scholar, and Chi et al., The Nature of Expertise.

66. Tetlock, , “Learning in U.S. and Soviet Foreign Policy,” p. 40Google Scholar.

67. I prefer “structural adjustment” to “adaptation” to avoid any confusion created by the equating of adaptation and learned behavior in evolutionary biology or in cybernetics.

68. This is a refinement of Bennett's, conceptualization in “Theories of Individual, Organizational, and Governmental Learning and the Rise and Fall of Soviet Interventionism, 1973–1983,” p. 102Google Scholar.

69. We could distinguish the two by referring to learning through structural adjustment as “adaptive learning.”

70. See Nye, , “Nuclear Learning and U.S.–Soviet Security Regimes,” p. 372Google Scholar; and Tetlock, “Learning in U.S. and Soviet Foreign Policy.”

71. On neorealism, see Waltz, Kenneth N., Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979)Google Scholar; and Keohane, Robert O., Neorealism and Its Critics (New York: Columbia University Press, 1986)Google Scholar.

72. See Waltz, , Theory of International Politics, pp. 74–79Google Scholar; and Ikenberry, G. John and Kupchan, Charles A., “Socialization and Hegemonic Power” International Organization 44 (Summer 1990), pp. 283–315CrossRefGoogle Scholar.

73. The death rates of regimes and individual leaders are much higher than for states. Waltz makes no attempt to explain this variation in learning, which reduces somewhat his ability to explain the patterns of interactions among states.

74. This follows March and Olsen's classification of six theories of organizational change: variation and selection, intendedly rational problem solving, experiential learning, bargaining and negotiation, contagion or imitation, and regeneration and turnover. See March, James G. and Olsen, Johan P., Rediscovering Institutions: The Organizational Basis of Politics (New York: Free Press, 1989), pp. 58–89Google Scholar.

75. See ibid., pp. 58–59; Campbell, Donald, “Variation and Selective Retention in Sociocultural Evolution,” General Systems, vol. 16, 1969, pp. 69–85Google Scholar; Nelson, Richard R. and Winter, Sidney G., An Evolutionary Theory of Economic Change (Cambridge, Mass.: Harvard University Press, 1982)Google Scholar; and Hirschman, Albert O., Exit, Voice, and Loyalty (Cambridge, Mass.: Harvard University Press, 1970)Google Scholar.

76. Axelrod, The Evolution of Cooperation. The second round of the tournament, wherein some players constructed new strategies after observing the results of the first round, involves learning.

77. See Asprey, Robert B., Frederick the Great (New York: Ticknor and Fields, 1986), pp. 551–59Google Scholar; Levy, Jack S., “Preferences, Constraints, and Choices in July 1914,” International Security 15 (Winter 1990–1991), pp. 151–86CrossRefGoogle Scholar; and Levy, Jack S., “Correspondence: Mobilization and Inadvertence in the July Crisis,” International Security 16 (Summer 1991), pp. 189–94CrossRefGoogle Scholar.

78. Nye, , “Nuclear Learning and U.S.–Soviet Security Regimes,” p. 381Google Scholar.

79. See Hall, “Policy, Paradigms, Social Learning, and the State”; and Mendelson, “Internal Battles and External Wars.”

80. Mendelson, , “Internal Battles and External Wars.” Mendelson argues that even before Gorbachev, Andropov had learned from policy failures and concluded that a withdrawal was necessary but politically infeasible (p. 347)Google Scholar. See also Breslauer, , “Explaining Soviet Foreign Policy Changes,” p. 211Google Scholar. On the 1914 case, see Levy, , “Preferences, Constraints, and Choices in July 1914,” pp. 174–78Google Scholar.

81. See Bennett, “Patterns of Soviet Military Interventionism 1975–1990”; Evangelista, , “Sources of Moderation in Soviet Security Policy,” pp. 275–79Google Scholar; Kingdon, John W., Agendas, Alternatives, and Public Policies (Boston: Little, Brown, 1984)Google Scholar; and Mueller, “The Marketing of Ideas.”

82. Differences among technical experts give political leaders the opportunity to rely on those experts whose ideas are compatible with their own and who can be used to legitimate their own preexisting policy preferences or enhance their domestic power base. This point is often underemphasized in the theoretical literature on epistemic communities.

83. See Mendelson, “Internal Battles and External Wars”; Checkel, Jeff, “Ideas, Institutions, and the Gorbachev Foreign Policy Revolution,” World Politics 45 (01 1993), pp. 271–300CrossRefGoogle Scholar; and Stein, “Political Learning by Doing.” This constitutes an improvement on some of the early theoretical work on epistemic communities, which gave primary emphasis to the one-directional flow of information and influence from specialists to political leaders. See E. Haas, When Knowledge Is Power.

84. See Mendelson, “Internal Battles and External Wars”; and Moltz, James Clay, “Divergent Learning and the Failed Politics of Soviet Economic Reform,” World Politics 45 (01 1993), pp. 301–25CrossRefGoogle Scholar; and Wallander, “Opportunity, Incrementalism, and Learning in the Extension and Retraction of Soviet Global Commitments,” respectively.

85. See Evangelista, , “Sources of Moderation in Soviet Security Policy,” p. 328Google Scholar; and Lynch, Allen, Gorbachev's International Outlook: Intellectual Origins and Political Consequences, Institute of East-West Security Studies Occasional Paper no. 9 (Boulder, Colo.: Westview, 1989)Google Scholar, respectively.

86. The quotations are drawn from Breslauer, , “Ideology and Learning in Soviet Third World Policy,” p. 443Google Scholar; and Breslauer, , “Explaining Soviet Policy Changes,” p. 209Google Scholar. See also Wallander, “Opportunity, Incrementalism, and Learning in the Extension and Retraction of Soviet Global Commitments”; and Bennett, “Theories of Individual, Organizational, and Governmental Learning and the Rise and Fall of Soviet Military Interventionism 1975–1990.”

87. See Nisbett and Ross, Human Inference; and Khong, Analogies at War, chap. 2.

88. Stein, “Political Learning by Doing.”

89. See Hough, Jerry F., Russia and the West: Gorbachev and the Politics of Reform (New York: Simon and Schuster, 1988)Google Scholar; and Cohen, Stephen F. and Heuvel, Katrina Vanden, Voices of Glasnost: Interviews with Gorbachev's Reformers (New York: W.W. Norton, 1989)Google Scholar.

90. Evangelista, , “Sources of Moderation in Soviet Security Policy,” pp. 264–65Google Scholar.

91. Jervis, , Perception and Misperception in International Politics, pp. 249–57Google Scholar.

92. It is also necessary to show that these shared beliefs, and not other variables, shaped policy change. For criticisms of the generational change explanation of Soviet foreign policy change under Gorbachev, see Meyer, Stephen M., “The Sources and Prospects of Gorbachev's New Political Thinking on Security,” International Security 13 (Fall 1988), pp. 124–63CrossRefGoogle Scholar; and Stein, “Political Learning by Doing.”

93. The availability heuristic is important here, as are changing paradigms that make certain phenomena more salient. See Tversky and Kahneman, “Availability.” See also Jervis's discussion of the evoked set in Perception and Misperception in International Politics, chap. 11.

94. See May, “Lessons” of the Past; Jervis, Perception and Misperception in International Politics; and Levy, “Learning from Experience in U.S. and Soviet Foreign Policy.” Thus I disagree with Snyder, who argues that Truman's references to Munich were instrumental rather than manifestations of genuine learning; see Snyder, , Myths of Empire, pp. 255–304Google Scholar.

95. Individuals may also draw foreign policy lessons from domestic events. Ronald Reagan's firing of domestic air traffic controllers may have influenced others' images of his resolve in international politics.

96. See Hedberg, , “How Organizations Learn and Unlearn,” p. 16Google Scholar; Wong, Paul T.P. and Weiner, Bernard, “When People Ask ‘Why’ Questions, and the Heuristics of Attributional Search,” Journal of Personality and Social Psychology vol. 40, no. 4, 1981, pp. 650–63CrossRefGoogle Scholar; and Sitkin, “Learning through Failure.”

97. Reiter, “Learning, Realism, and Alliances.”

98. Jervis, , Perception and Misperception in International Politics, pp. 275–79Google Scholar; the quotation is from p. 275.

99. Hackworth, David H., “The Lessons of the Gulf War,” Newsweek, 24 06 1991, pp. 22–24Google Scholar.

100. See Lant, Theresa K. and Montgomery, David B., “Learning from Strategic Success and Failure,” Journal of Business Research, vol. 15, 1987, pp. 503–17CrossRefGoogle Scholar; Wong and Weiner, “When People Ask ‘Why’ Questions and the Heuristics of Attributional Search”; Sitkin, , “Learning Through Failure,” p. 236Google Scholar; and Levitt, and March, , “Organizational Learning,” p. 325Google Scholar.

101. See Tversky, Amos and Kahneman, Daniel, “Rational Choice and the Framing of Decisions,” Journal of Business, vol. 59, no. 4, part 2, 1986, pp. S251–78CrossRefGoogle Scholar; and Levy, Jack S., “An Introduction to Prospect Theory,” and “Prospect Theory and International Relations: Theoretical Applications and Analytical Problems,” Political Psychology 13 (06 1992), pp. 171–86CrossRefGoogle Scholar and 283–310, respectively.

102. Bismarck suggested a more nuanced proposition: fools learn by experience while wise men learn by other peoples' experience; noted in Jervis, , Perception and Misperception in International Politics, pp. 239–43Google Scholar. On vicarious learning, see Huber, , “Organizational Learning,” pp. 96–97Google Scholar.

103. See Leng, “When Will They Ever Learn?”; Huth and Russett, “What Makes Deterrence Work?”; and Jentleson, Bruce W., Levite, Ariel E., and Berman, Larry, “Foreign Military Intervention in Perspective,” in Levite, Ariel E., Jentleson, Bruce W., and Berman, Larry, eds., Foreign Military Intervention (New York: Columbia University Press, 1992), pp. 303–25Google Scholar.

104. May, , “Lessons” of the Past, p. 81Google Scholar.

105. See Vasquez, John A., The War Puzzle (Cambridge: Cambridge University Press, 1993), pp. 207–10CrossRefGoogle Scholar; and Levy, Jack S., “The Diversionary Theory of War,” in Midlarsky, Manus I., ed., Handbook of War Studies (London: Unwin-Hyman, 1989), pp. 259–88Google Scholar.

106. Farkas incorporates this into his dynamic model of evolutionary state learning. Change in influence is a function of the distance of one's policy preferences from successful policies or from a new policy that emerges from the search and reevaluation following unsuccessful policies. See Farkas, “State Learning and International Change.”

107. Snyder, , Myths of Empire, pp. 253–54Google Scholar.

108. When political leaders use history instrumentally is an interesting question. Robertson argues that, “The use of lessons as leverage in political conflict pervades policy areas where facts are contested, values are complex, and partisan differences are sharp.” See p. 55 of Robertson, David Brian, “Political Conflict and Lesson-drawing,” Journal of Public Policy 11 (01–03 1991), part 1, pp. 55–78CrossRefGoogle Scholar. One testable implication is that the instrumental use of history is more common in security policy than in foreign economic policy. See Moltz, , “Divergent Learning,” pp. 304–6Google Scholar.

109. Taylor is cited in Jervis, , Perception and Misperception in International Politics, p. 217Google Scholar; Fairbank is cited in Hoffmann, Stanley, Gulliver's Troubles (New York: McGraw-Hill, 1968), p. 135Google Scholar. See also Khong, , Analogies at War, pp. 8–9Google Scholar. On domestic interests, see Snyder, Myths of Empire.

110. The Vietnam War probably generated more learning about the conduct of war and the importance of domestic support than about the proper conditions (if any) for U.S. intervention.

111. Jarosz, with Nye, , “The Shadow of the Past,” pp. 164–78Google Scholar.

112. For a good discussion of hypotheses on the sources of learning see Jervis, , Perception and Misperception in International Politics, pp. 239–71Google Scholar.

113. These comparisons might include people who are in and out of positions of political power or who represent different functional and area ministries in the government, different states or districts with different economic interests, or different firms with different exposure to the global economy.

114. Khong, Analogies at War.

115. Snyder, Myths of Empire. Snyder's primary theoretical task is to construct a domestic political model of imperial expansion, and consequently his learning model is not as well-developed as Khong's.

116. In terms of Khong's study, it would be useful to extend the domain of comparison and explore whether hypotheses on learning can explain which U.S. decision makers were the first to shift away from a military solution as the war continued.

117. See Leng, “When Will They Ever Learn?”; Huth and Russett, “What Makes Deterrence Work?”; and Reiter, “Learning, Realism, and Alliances.”

118. Gochman, Charles S. and Maoz, Zeev, “Militarized Interstate Disputes, 1816–1976,” Journal of Conflict Resolution 28 (12 1984), pp. 585–616CrossRefGoogle Scholar.

119. There are other uses of quantitative methods that might be better able to analyze intervening learning processes. It might be possible to operationalize cognitive structure and beliefs and measure them through the content analysis of statements or documents. See Tetlock, Philip E., “Integrative Complexity of American and Soviet Foreign Policy Rhetoric: A Time Series Analysis,” Journal of Personality and Social Psychology vol. 49, no. 6, 1985, pp. 1565–85CrossRefGoogle Scholar; and Maoz, Zeev and Astorino, Allison, “The Cognitive Structure of Peacemaking: Egypt and Israel, 1970–1978,” Political Psychology 13 (12 1992), pp. 647–62CrossRefGoogle Scholar. Cognitive mapping techniques might also be useful. See Johnston, Alastair Iain, “An Inquiry into Strategic Culture: Chinese Strategic Thought, the Parabellum Paradigm, and Grand Strategic Choice in Ming China,” Ph.D. diss., University of Michigan, 1993Google Scholar.

120. On process tracing see George, Alexander L., “Case Studies and Theory Development,” paper presented at the Second Annual Symposium on Information Processing in Organizations, Carnegie Mellon University, 15–1610 1982Google Scholar.

121. See Kuklinski, James H., Luskin, Robert C., and Bollard, John, “Where Is the Schema? Going Beyond the ‘S’ Word in Political Psychology,” American Political Science Review 85 (12 1991), pp. 1341–56CrossRefGoogle Scholar; and Stein, “Political Learning by Doing.”