¹ Allison, Graham T., Essence of Decision: Explaining the Cuban Missile Crisis (Boston: Little, Brown 1971)Google Scholar. See also Allison, Graham T. and Halperin, Morton, “Bureaucratic Politics: A Paradigm and Some Policy Implications,” in Tanter, Raymond and Ullman, Richard H., eds., Theory and Policy in International Relations (Princeton: Princeton University Press 1972), 40–70Google Scholar.

² Halperin, Morton, Bureaucratic Politics and Foreign Policy (Washington: Brookings Institution Press 1974)Google Scholar.

³ Downs, Anthony, Inside Bureaucracy (Boston: Little, Brown 1967), 266, 272–73CrossRefGoogle Scholar.

⁴ Steinbruner, John D., The Cybernetic Theory of Decision (Princeton: Princeton University Press 1974), 123Google Scholar.

⁵ Holsti, Ole R., “The 1914 Case,” American Political Science Review, Vol. 59 (June 1965), 365CrossRefGoogle Scholar. See also Holsti, , Crisis Escalation War (London: McGill-Queens University Press 1972), 13CrossRefGoogle Scholar. In this study, crises are defined as situations which threaten important national goals, allow a short time for decision, and occur as surprises to policy makers. See Hermann, Charles F., “Time, Threat, and Surprise: A Simulation of International Crisis,” in Hermann, , ed., International Crises: Insights from Behavioral Research (New York: Free Press 1972), 187Google Scholar. The terms “policy makers” and “decision makers” are used loosely to include middle-level government officials.

⁶ Moffitt, J. Weldon and Stagner, Ross, “Perceptual Rigidity and Closure as a Function of Anxiety,” Journal of Abnormal and Social Psychology, Vol. 52 (May 1956), 354–57CrossRefGoogle Scholar.

⁷ Janis, Irving L., Victims of Groupthink (Boston: Houghton Mifflin 1972), 198Google Scholar.

⁸ Hermann, Charles F., Crises in Foreign Policy: A Simulation Analysis (Indianapolis: Bobbs-Merrill 1969), 155Google Scholar.

⁹ Pruitt, Dean G., “Problem Solving in the Department of State” (Northwestern University 1961), 60Google Scholar, as quoted in Hermann, Charles F., “Some Consequences of Crisis which Limit the Viability of Organizations,” Administrative Science Quarterly, VIII (June 1963), 69Google Scholar.

¹⁰ Howard H. Lentner, “The Concept of Crisis as Viewed by the United States Department of State,” in Hermann (fn. 5), 130.

¹¹ Boulding, Kenneth E., “National Images and International Systems” Journal of Conflict Resolution, in (June 1959), 124–25.Google Scholar

¹² Jervis, Robert, “Hypotheses on Misperception,” World Politics, XX (April 1968), 454–79CrossRefGoogle Scholar; and Jervis, Robert, Perception and Misperception in International Politics (Princeton: Princeton University Press 1976)Google Scholar.

¹³ White, Ralph K., Nobody Wanted War: Misperception in Vietnam and Other Wars (Garden City, N.Y.: Doubleday 1970)Google Scholar.

¹⁴ LeVine, Robert A. and Campbell, Donald L., Ethnocentrism: Theories of Conflict, Ethnic Attitudes, and Group Behavior (New York: Wiley 1972), 212–23.Google Scholar

¹⁵ Finlay, David J., Holsti, Ole R., and Fagen, Richard R., Enemies in Politics (Chicago: Rand McNally 1967), 8Google Scholar.

¹⁶ Jervis (fn. 12), “Hypotheses on Misperception,” and Perception and Misperception, chap. IX.

¹⁷ White (fn. 13), 16.

¹⁸ LeVine and Campbell (fn. 14).

¹⁹ Jervis (fn. 12), “Hypotheses on Misperception,” and Perception and Misperception, 354–55.

²⁰ Rosenblatt, Paul C., “Origins and Effects of Group Ethnocentrism and Nationalism,” Journal of Conflict Resolution, VIII (June 1964), 131–46.CrossRefGoogle Scholar

²¹ Postman, Leo and Bruner, Jerome S., “Perception Under Stress,” Psychological Review, Vol. 55 (November 1948), 314–23CrossRefGoogle Scholar.

²² Political games are to be distinguished from international relations simulations like the Inter-Nation Simulation, which involves imaginary nations, more structure, and quantification.

²³ Goldhamer, Herbert and Speier, Hans, “Some Observations on Political Gaming” World Politics, XII (October 1959), 71–83CrossRefGoogle Scholar.

²⁴ Bloomfield, Lincoln P. and Gearin, Cornelius J., “Games Foreign Policy Experts Play: The Political Exercise Comes of Age,” Orbis, XVI (Winter 1973), 1012–13.Google Scholar

²⁵ Letter from William M. Jones, RAND Corporation, February 26, 1975. A 1964 series of Pentagon games on a projected American bombing of North Vietnam, involving Curtis LeMay, Earle Wheeler, John McNaughton, and McGeorge Bundy, had the startling (at the time) resolution that the bombing would not have much effect. These Vietnam games are described in Halberstam, David, The Best and the Brightest (New York: Random House 1972), 460–62.Google Scholar

²⁶ While I was a summer intern at the C.I.A., I conducted a game on the Kurd-Iraq-Iran dispute in 1974 and one on a China-Vietnam dispute in 1975. Other C.I.A. games were on the Chile-Peru border dispute and on internal factions in Brazil. These games had a political rather than military emphasis and involved only the roles of foreign actors in foreign crises.

²⁷ The POLEX, POLEX-DAIS, DETEX, and CONEX series are reviewed here. (POLEX I-which had n o government personnel-is excluded.) The games covered many potential crises, including civil strife in Angola, a pro-Castro regime in Venezuela, Chinese infiltration of Burma, and insurgency in India.

²⁸ Reviewed are the SUPER CRISIS, INTERMEDIATE, and L OW LEVEL series on (respectively) a Soviet invasion of China, Soviet attacks in the Middle East, and tensions between India and Pakistan. Although there are many other RAND political games, the group examined here consists of the unclassified ones RAND deems most worthy of postgame analysis. Most of the games not included are informal “dry runs” with little planning or documentation and with no Defense Department personnel involved.

²⁹ The Yale games were designed to replicate the C.I.A. game of July 1975 on island disputes between China and Vietnam; a C.I.A. game participant gave a talk on the crisis. The Yale games were conducted to see whether, and how, student games with rigid experimental controls would be similar to professional-level games.

³⁰ One evaluation of benefits from the early M.I.T. political games is Barringer, Richard E. and Whaley, Barton, “The MIT Political-Military Gaming Experience” Orbis, IX (Summer 1965), 437–58Google Scholar. A recent review of five Pentagon games is Lt. Colonel Joel Snyder, “Measuring the Benefits of Politico-Military Simulations” (unpub., 1974). But these studies and the few attempts to use political gaming to test hypotheses are often flawed by misuse of participants' reactions and demand characteristics, misunderstanding of gaming effects, and ambiguity of indicators.

³¹ Examples of role-playing studies are in Elms, Alan C., ed., Role Playing, Reward, and Attitude Change (New York: Van Nostrand Reinhold 1969)Google Scholar. Examples of group dynamics studies are in Maier, Norman R. F., Problem Solving and Creativity in Individuals and Groups (Belmont, Calif.: Brooks/Cole Publishing Co. 1970Google Scholar).

³² For limitations on analyzing perceptions, see Cook, Stuart and Selltiz, Claire, “A Multiple-Indicator Approach to Attitude Measurement” Psychological Bulletin, Vol. 62 (July 1964), 36–55CrossRefGoogle Scholar; and Holsti, Ole R. and others, “Perception and Action in the 1914 Crisis,” in Singer, J. David, ed., Quantitative International Politics (New York: Free Press 1968), 129Google Scholar. Standard methods of attitude measurement (personal interviews, questionnaires, or direct observation), other types of simulation (INS, TEMPER, or PRINCE), and other forms of psychological intervention (such as devil' advocates or T-Groups), have rarely been effective here.

³³ The underlying assumption in advancing these two different sets of hypotheses is that distortions in internal decision making can be reduced through political gaming because they are primarily the result of the particular structural environment in which foreign policy makers normally operate during crises; but distortions in international images can merely be reflected through political gaming because they are the result of deep-seated belief structures that are relatively independent of the decision-making process and the organizational context.

³⁴ A reliability test was used to check the clarity of the rules for counting the decision-making factors. To assess the significance of the change in these factors across game periods, the F-test was used because both the period and the quantity of each factor are interval variables.

³⁵ Bloomfield, Lincoln P. and others, User's Manual for CASCON II: Computer-Aided System for Handling Information on Local Conflicts (Cambridge: M.I.T. C.I.S. monograph #C/72–14, September 1972), 12Google Scholar.

³⁶ These two perspectives are referred to as the “actor” and “observer” perspectives. See Jones, Edward E. and Nisbett, Richard E., “The Actor and the Observer: Divergent Perceptions of the Causes of Behavior,” in Jones, Edward E. and others, Attribution: Perceiving the Causes of Behavior (Morristown, N.J.: General Learning Press 1972), 93–94Google Scholar.

³⁷ Three reliability tests were used to check the clarity of the rules for rating the perceptions. To assess the significance of the difference between the perspectives of actors and observers, the Mann-Whitney U-test (standardized by the Z-score) was used because one variable (the team) is nominal and the other (the rating) is ordinal.

³⁸ A special hypothesis tested in the Yale games showed that results were similar among expert and non-expert student players.

³⁹ Postperiod questionnaires were administered to DETEX II participants by Lincoln Bloomfield. The roles played by these participants corresponded to their agency backgrounds and were separable into political (State-type) and military (Defense-type) groups.

⁴⁰ Perceptions were compared by studying the results of two series of games; the SUPER CRISIS series was compared to the LOW LEVEL series.

⁴¹ The similarity of expert and non-expert results in the Yale games and the similarity of the results of the Yale games to those of the M.I.T. and RAND games indicate that players at the professional level may not always be needed to draw valid conclusions from the games.

⁴² See Raser, John, “Theories of Deterrence,” Peace Research Reviews, in (February 1969), 14Google Scholar.

⁴³ George, Alexander and Smoke, Richard, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press 1974), 44–45Google Scholar.

⁴⁴ Several government analysts mentioned that more research on international perceptions would be useful in their own work.

⁴⁵ A number of government analysts who participated in the C.I.A. games commented spontaneously on this improvement.

⁴⁶ The insights in this section are drawn from the author' discussions with middle-level government analysts and from his observation of their interaction and modes of behavior during their routine working conditions.

⁴⁷ Interview, January 8, 1976.

⁴⁸ The highly classified transcripts of the Pentagon games were examined by the author while he was an intern at the C.I.A.

⁴⁹ Bloomfield and Gearin (fn. 23), 1026.

⁵⁰ For a discussion of the state of art, see DeLeon, Peter, Scenario Designs: An Overview (Santa Monica, Calif.: RAND Paper Series, #R-1218-ARPA, June 1973), V.Google Scholar