¹ Singer, J. David and Small, Melvin, The Wages of War, 1816–1965: A Statistical Handbook (New York: Wiley, 1972)Google Scholar. This handbook has been updated, and expanded to include new data on civil wars, in Small and Singer, 1982.

² See Singer, and Small, , “Formal Alliances, 1815–1939: A Quantitative Description,” Journal of Peace Research, No. 3 (1966), 1–32CrossRefGoogle Scholar; Small, and Singer, , “Formal Alliances, 1816–1965: An Extension of the Basic Data,” Journal of Peace Research, No. 3 (1969), 257–82CrossRefGoogle Scholar; Small, and Singer, , “The Diplomatic Importance of States, 1816–1970: An Extension and Refinement of the Indicator,” World Politics 25 (July 1973), 577–99CrossRefGoogle Scholar, reprinted in Singer, 1979, pp. 199–222.

³ Other important analysts who entered the project while they were graduate students at Michigan include: John Stuckey, Yoshinobu Yamamoto, Hugh Wheeler, Cynthia Cannizzo, Michael Mihalka, Thomas Cusack, Richard Stoll, Michael Champion, and Paul Diehl.

⁴ Waltz, Kenneth, “Theory of International Relations,” in Greenstein, Fred I. and Polsby, Nelson W., eds., Handbook of Political Science, International Politics, Vol. 8 (Reading, MA: Addison-Wesley, 1975), 9Google Scholar; cf. Waltz, , Theory of International Politics (Reading, MA: AddisonWesley, 1979)Google Scholar, 5, 8.

⁵ Ibid. (1979), 12–13.

⁶ Ibid., 8–10.

⁷ Ibid., 4.

⁸ Ibid., 8. In an earlier version (fn. 4, 1975, p. 9), the term “theoretical” did not appear. This new qualification makes Waltz's argument true by definition, since the real question is not whether facts produce explanation, but whether the documentation of patterns through research is an essential step in producing knowledge.

⁹ Waltz (fn. 4, 1979), 8.

¹⁰ Ibid., 7.

¹¹ Waltz (fn. 4, 1975), 7, 14.

¹² See Vasquez, John A., “Colouring It Morgenthau: New Evidence for an Old Thesis on Quantitative International Politics,” British Journal of International Studies 5 (October 1979) 210–28CrossRefGoogle Scholar; Vasquez, John A., The Power of Power Politics: A Critique (New Brunswick, NJ: Rutgers University Press, 1983)Google Scholar, chap. 5.

¹³ Waltz (fn. 4, 1975), 2.

¹⁴ Singer, , “The ‘Correlates of War’ Project: Interim Report and Rationale,” World Politics 24 (January 1972), 243–70CrossRefGoogle Scholar, at 251.

¹⁵ Brown, Robert, Explanation in Social Science (Chicago: Aldine, 1963)Google Scholar, chap.2

¹⁶ This approach follows the spirit of Jervis's call for the improvement of cumulation through the development of richer theories by asking questions such as: If this explanation is true (or inaccurate), then what else should follow? What are the implications? What other events should or should not occur? What does the theory say about cases or data not yet analyzed? See Jervis, Robert, “Cumulation, Correlations, and Woozles,” in Rosenau, James, ed., In Search of Global Patterns (New York: Free Press, 1976), 183.Google Scholar

¹⁷ Vasquez (fn. 12, 1983), chap. 8.

¹⁸ On misspecification, see Zinnes, Dina, “An Analytical Study of the Balance of Power Theories,” Journal of Peace Research, No. 3 (1967), 270–88CrossRefGoogle Scholar; Waltz (fn. 4, 1975), 12–14; Jervis, Robert, “The Costs of the Quantitative Study of International Relations,” in Knorr, Klaus and Rosenau, James, eds. Contending Approaches to International Politics (Princeton: Princeton University Press, 1969), 177–217Google Scholar. On research designs, see the essays on the Correlates of War project in Hoole, Francis and Zinnes, Dina, eds., Quantitative International Politics: An Appraisal (New York: Praeger, 1976), 21–145Google Scholar; Most, Benjamin and Starr, Harvey, “Conceptualizing 'War': Consequences for Theory and Research,” Journal of Conflict Resolution 27 (March 1983)CrossRefGoogle Scholar, 137–59; Siverson, Randolph and Sullivan, Michael P., “The Distribution of Power and the Onset ofWar,” Journal of Conflict Resolution 27 (September 1983), 473–94CrossRefGoogle Scholar. On measurement, see Altfeld, Michael, “Arms Races?–and Escalation?: A Comment on Wallace,” Inter- national Studies Quarterly 27 (June 1983), 225–31CrossRefGoogle Scholar, and Michael Wallace's reply in the same issue; Goertz, Gary and Diehl, Paul, “Measuring Military Allocations,” Journal of Conflict Resolution 30 (September 1986), 553–81CrossRefGoogle Scholar.

¹⁹ Wallace, Michael D., “Polarization: Towards a Scientific Conception,” in Sabrosky, 1985, pp. 95–113Google Scholar, at 112.

²⁰ It should be pointed out that the Correlates of War project has not yet produced an integrated explanation. For Singer's view of what, in principle, would constitute an adequate explanation of war, see his “Peace Research and Foreign Policy Prediction” (Presidential address to the Peace Research Society) Peace Science Society (International) Papers 21 (1973)Google Scholar, 13, reprinted in Singer, 1979, pp. 155–71, at 161–63. For his own review and assessment of Correlates of War findings, see Singer, J. David, “Accounting for International War: The State of the Discipline,” Journal of Peace Research 18 (No. 1, 1981), 1–18.CrossRefGoogle Scholar For an early survey of a variety of studies on war, see Deutsch, Karl and Senghaas, Dieter, “The Steps to War: A Survey of System Levels, Decision Stages, and Research Results,” in Patrick J. McGowan, ed., Sage International Yearbook of Foreign Policy Studies 1 (1973), 275–329.Google Scholar

²¹ Singer has suggested that conflict be seen as an unfolding escalatory process involving different stages; see Singer, 1980, pp. xxiv, xxix–xxx.

²² Singer, David, “Confrontational Behavior and Escalation to War 1816–1980: A Research Plan,” Journal of Peace Research 19 (No. 1, 1982), 40.CrossRefGoogle Scholar

²³ See Rummel, R. J., War, Power, Peace, Understanding Conflict and War, Vol. 4 (Beverl Hills, CA: Sage, 1979), 186Google Scholar on testing; Maoz, Zeev, “Resolve, Capabilities, and the Outcomes of Interstate Disputes, 1816–1976,” Journal of Conflict Resolution 27 (June 1983), 195–229.CrossRefGoogle Scholar

²⁴ See Wallace, Michael D., “Status, Formal Organization, and Arms Levels as Factors Leading to the Onset of War, 1820–1964,” in Russett, Bruce M., ed., Peace, War, and Numbers (Beverly Hills, CA: Sage, 1972), 49–69Google Scholar; Diehl, Paul F., “Arms Races to War: Testing Some Empirical Linkages,” The Sociological Quarterly 26, (No. 3, 1985), 331–32CrossRefGoogle Scholar.

²⁵ Leng, Russell J., “When Will They Ever Learn? Coercive Bargaining in Recurrent Crises,” Journal of Conflict Resolution 27 (September 1983), 379–419.CrossRefGoogle Scholar

²⁶ Wallace, Michael D., “Arms Races and Escalation: Some New Evidence,” Journal ofConflict Resolution 23 (March 1979), 3–16CrossRefGoogle Scholar, reprinted in J. David Singer and Associates, 1979, pp. 240–252, and in Small and Singer, 1985, pp. 196–206.

²⁷ Gochman, Charles S. and Leng, Russell J., “Realpolitik and the Road to War: An Analysis of Attributes and Behavior,” International Studies Quarterly 27 (March 1983), 97–120.CrossRefGoogle Scholar

²⁸ Leng (fn. 25).

²⁹ Holsti, Ole R., North, Robert C., and Brddy, Richard A., “Perception and Action in the 1914 Crisis,” in Singer, J. David, ed., Quantitative International Politics (New York: Free Press, 1968), 123–58Google Scholar.

³⁰ For a justification of this assumption, see Vasquez, John A., “Foreign Policy, Learning, and War,” in Hermann, Charles F., Kegley, Charles Jr., and Rosenau, James N., eds., New Directions in the Study of Foreign Policy (Winchester, MA: Allen & Unwin, 1987), 366–83Google Scholar.

³¹ de Mesquita, Bruce Bueno, The War Trap (New Haven: Yale University Press, 1981), 4–5.Google Scholar

³² Organski, A.F.K. and Kugler, Jacek, The War Ledger (Chicago: University of Chicago Press, 1980)Google Scholar; Gilpin, Robert, War and Change in World Politics (Cambridge: Cambridge University Press, 1981)CrossRefGoogle Scholar; Modelski, George, “The Long Cycle of Global Politics and the Nation State,” Comparative Studies in Society and History 20 (April 1978), 214–35CrossRefGoogle Scholar; Thompson, William R., “Succession Crises in the Global Political System: A Test of the Transition Model,” in Bergesen, Albert, ed., Crises in the World System (Beverly Hills, CA: Sage, 1983), 93–116.Google Scholar For an attempt to get at some of the details, see Doran, Charles F. and Parsons, Wes, “War and the Cycle of Relative Power,” American Political Science Review 74 (December 1980), 947–65CrossRefGoogle Scholar. For a thoughtful review of this type of explanation, see Levy, Jack S., “Theories of General War,” World Politics 37 (April 1985), 344–74CrossRefGoogle Scholar.

³³ War is a relatively rare event. Only 67 inter-state and 51 other wars (involving at least one nation-state against a nonrecognized state) were fought between 1816 and 1980 (Small and Singer, 1982, pp. 59–60). According to these data, war is even a rarer event in the history of specific inter-state relations. Peter Wallensteen shows that some periods of history—namely 1816–1848, 1871–1895, 1919–1932, 1963–1976—are devoid of major-power wars and have few military confrontations; see Wallensteen, , “Universalism vs. Particularism: On the Limits of Major Power Order,” Journal of Peace Research 21 (No. 3, 1984), 243–57CrossRefGoogle Scholar, at 245–46.

³⁴ For evidence, see Maoz (fn. 23); Leng, Russell J., “Influence Strategies and Interstate Conflict,” in Singer, 1980Google Scholar, pp. 124–57; Leng (fn. 25).

³⁵ Within the project, the work most relevant to these questions is by Michael Wallace (fn. 24). For the factors that lead actors to become involved in disputes, see Gochman, Charles S., “Status, Capabilities, and Major Power Conflict,” in Singer, 1980Google Scholar, pp. 83–123; and Maoz, Zeev, Paths to Conflict: International Dispute Initiation, 1816–1976 (Boulder, CO: Westview Press, 1982)Google Scholar, chaps. 4 and 5.

³⁶ Levy, Jack, “Alliance Formation and War Behavior: An Analysis of the Great Powers, 1495–1975,” Journal of Conflict Resolution 25 (December 1981), 581–613CrossRefGoogle Scholar. A similar conclusion had been reached earlier by Singer and Small on the basis of more limited tests; see “National Alliance Commitments and War Involvement, 1815–1945,” Peace Research Society (International) Papers 5 (1966), 109–40Google Scholar, and Singer, and Small, , “Foreign Policy Indicators: Predictors of War in History and in the State of the World Message,” Policy Sciences 5 (September 1974), 271–96CrossRefGoogle Scholar, reprinted in Singer, 1979, pp. 298–329, esp. 325–26.

³⁷ Levy (fn. 36), 597–98, esp. Table 7.

³⁸ But note that 60% of the wars of the 20th century were preceded by alliances, compared to 18%, 14%, 35%, and 25% of the wars in the 16th, 17th, 18th, and 19th centuries, respectively. Levy (fn. 36), 599, Table 8.

³⁹ Ostrom, Charles W. Jr., and Hoole, Francis W., “Alliances and War Revisited: A Research Note,” International Studies Quarterly 22 (June 1978), 215–36CrossRefGoogle Scholar; see pp. 228–29 for the findings reported here. On the earlier tests, see Singer and Small, “Alliance Aggregation and the Onset of War, 1815–1945,” in Singer 1968 (fn. 29), 247–86 (reprinted in Singer, 1979, pp. 225–64), and Singer and Small (fn. 36, 1966).

⁴⁰ Sabrosky, Alan Ned, “Interstate Alliances: Their Reliability and the Expansion of War,” in Singer, 1980Google Scholar, pp. 161–98, shows that even at the most fundamental level—whether an ally will fight on your side in a war—expectations are not fulfilled. More often than not, allies remain neutral (p. 176). In addition, some alliances, like neutrality pacts (especially those between major and minor powers) are more likely to be violated than honored; i.e., if one party (especially a major power) enters the war, it will enter against its ally (particularly if the ally is a minor power) (p. 196).

⁴¹ Siverson and King, “Alliances and the Expansion of War,” in Singer and Wallace, 1979, pp. 37–49, esp. 45. See also Siverson, and King, , “Attributes of National Alliance Membership and War Participation, 1815–1965,” American Journal of Political Science 24 (February 1980), 1–15.CrossRefGoogle Scholar

⁴² Siverson and King (fn. 41, 1979), 48.

⁴³ Yamamoto, Yoshinobu and Bremer, Stuart A., “Wider Wars and Restless Nights: Major Power Intervention in Ongoing War,” in Singer, 1980, pp. 199–229Google Scholar; see pp. 216–17 for findings reported here.

⁴⁴ Although the two-way conditional model holds for both the 19th and 20th centuries, the finding is stronger for the 20th. Major powers in the 20th century were more likely to intervene in ongoing wars, and were more affected by what others did. Ibid., 218–21. While this inference is based on a very small number of cases, it suggests that alliances in the 20th century were more dangerous than in the 19th.

⁴⁵ Siverson and King (fn. 41, 1979), 46–47.

⁴⁶ Most, Benjamin A. and Starr, Harvey, “Diffusion, Reinforcement, Geopolitics, and the Spread of War,” American Political Science Review 74 (December 1980), 941–44.CrossRefGoogle Scholar

⁴⁷ Yamamoto and Bremer (fn. 43, p. 226) find that, when only one side is joined by major powers, the war is usually brought to a rapid close (the chief exception being the Crimean War); but when major powers join opposite sides, as has often happened in the 20th century, the war is more severe.

⁴⁸ See Singer (fn. 20, 1981), 8–9.

⁴⁹ Wayman, Frank Whelon, “Bipolarity, Multipolarity and the Threat of War,” in Sabrosky, 1985, pp. 128–31Google Scholar, shows that polarity based on the number of blocs and polarity based on the distribution of power are two uncorrelated and empirically separate dimensions (.01 for 1816—1965). See also his extended review of the various measures in “Bipolarity and War,” Journal of Peace Research 21 (No. 1, 1984), 74–75Google Scholar. For Wallace's argument in favor of combining the two dimensions, see his “Polarization” in Sabrosky, 1985.

⁵⁰ de Mesquita, Bruce Bueno, “Systemic Polarization and the Occurrence and Duration of War,” Journal of Conflict Resolution 22 (June 1978), 241–67CrossRefGoogle Scholar, reprinted in Singer and Associates, 1979, 113–38; Wayman (fn. 49, both articles); Levy, Jack S., “The Polarity of the System and International Stability: An Empirical Analysis,” in Sabrosky, 1985, pp. 41–66.Google Scholar

⁵¹ Deutsch, Karl W. and Singer, J. David, “Multipolar Power Systems and International Stability,” World Politics 16 (April 1964), 390–406CrossRefGoogle Scholar, reprinted in Small and Singer, 1985, pp. 182– 95. The first test of their propositions was in Singer and Small (fn. 39, 1968).

⁵² Wallace, Michael D., “Alliance Polarization, Cross–Cutting, and International War, 1815–1964,” Journal of Conflict Resolution 17 (December 1973), 575–604CrossRefGoogle Scholar, reprinted in Singer and Associates, 1979, pp. 83–111, at 105–07. Kegley, Charles W. Jr., and Raymond, Gregory A., in “Alliance Norms and War: A New Piece in an Old Puzzle,” International Studies Quarterly 26 (December 1982), 585–89CrossRefGoogle Scholar, find that when alliance structures are either extremely flexible or extremely rigid, the magnitude and severity of war are high.

⁵³ The correlation between an increase in interaction opportunities and the occurrence of war is probably a function of the rise of new actors rather than of a reduction in blocs. Wayman, in “Bipolarity, Multipolarity, and the Threat of War” (fn. 49), 137, makes a similar point against Bueno de Mesquita (fn. 50). The sheer increase in interaction opportunities increases conflict and war—contrary to what Deutsch and Singer expected—because new states often raise fundamental issues that challenge a given order; they must also create a (territorial) place for themselves in the system. For evidence on the latter, see Zeev Maoz, “Joining the Club of Nations: Independence, Revolution, and Conflict in the Interstate System, 1816–1976,” paper presented to the World Congress of the International Political Science Association, Paris, July 16, 1986. Evidence that an increase in the number of states is associated with an increase in wars and serious disputes is provided, respectively, in Small and Singer, 1982, pp. 130, 135, 141, and in Gochman, Charles S. and Maoz, Zeev, “Militarized Interstate Disputes, 1816–1976,” Journal of Conflict Resolution 28 (December 1984), 591–94CrossRefGoogle Scholar, excerpted in Small and Singer, 1985, pp. 27–36; see 34–35.

⁵⁴ Bueno de Mesquita (fn. 50); see 130–32 for the findings reported here.

⁵⁵ Since creation of alliances leads to a reduction of interaction opportunities, one would expect a negative correlation between change in interaction opportunities and war. This re– lationship may not have been uncovered by Bueno de Mesquita because such a reduction would have been wiped out by the rise of new actors, especially in the post-1945 period. It is suggestive, therefore, that he does find a slight (but statistically insignificant — .25) relationship between a reduction of interaction opportunities and the onset of wars involving major powers in the 19th century (see fn. 50, Table 2, p. 130). The positive .33 for the 20th century may be explained away as a function of system size (see fn. 53).

⁵⁶ Ostrom and Hoole (fn. 39), 229.

⁵⁷ See Bueno de Mesquita (fn. 50), 134–35, for the idea that an increase in poles encourages nations to prepare for war, and for his findings on discreteness and duration of war. On the importance of cross-cutting (per se) for reducing conflict, see Deutsch and Singer (fn. 51); Dean, P. Dale and Vasquez, John A., “From Power Politics to Issue Politics,” Western Political Quarterly 29 (March 1976), 24–27CrossRefGoogle Scholar; and Levy (fn. 36), 608.

⁵⁸ Steven Rosen, “War Power and the Willingness to Suffer,” in Bruce M. Russett (fn. 24), 167–83. Wayman, Frank W., Singer, J. David, and Goertz, Gary, in “Capabilities, Allocations, and Success in Militarized Disputes and Wars, 1816–1976,” International Studies Quarterly 27 (December 1983), 506–11CrossRefGoogle Scholar, show that the single best capability predictor of an initiator's prospects for winning a war is its industrial capability rather than its military or demographic capability.

⁵⁹ Kennedy, Paul, in “The First World War and the International Power System,” International Security 9 (Summer 1984), 7–40CrossRefGoogle Scholar, gives a graphic description of how allies prolong wars by providing additional resources, particular economic staying power, at critical junctures.

⁶⁰ Bueno de Mesquita (fn. 50), 131–32.

⁶¹ Sabrosky, Alan Ned, “Alliance Aggregation, Capability Distribution, and the Expansion of Interstate War,” in Sabrosky, 1985, pp. 148Google Scholar, 151, 181.

⁶² For an elaboration of these distinctions, see Vasquez, John A., “Capability, Types of War, Peace,” Western Political Quarterly 38 (June 1986), 313–27Google Scholar, where wars are classified on the basis of three dimensions: capability (wars between equals vs. wars between unequals), the number of participants (dyadic vs. complex), and the goals and means (limited vs. total war).

⁶³ On the relationship between norms justifying violence and the use of force, see Gurr, Ted Robert, Why Men Rebel (Princeton: Princeton University Press, 1970), 170Google Scholar; Maoz (fn. 35), 96, 98–100.

⁶⁴ J. David Singer, Stuart A. Bremer, and John Stuckey, “Capability Distribution, Uncertainty, and Major Power War, 1820–1965,” in Russett (fn. 24), 19–48; reprinted in Singer, 1979, an pp. 265–97, and in Singer and Associates, 1979, 159–88.

⁶⁵ Vasquez (fn. 62).

⁶⁶ de Mesquita, Bruce Bueno, “Risk, Power Distributions, and the Likelihood of War,” International Studies Quarterly 25 (December 1981), 541–68CrossRefGoogle Scholar. Likewise, Wayman et al. (fn. 58), 506, 509 show that, among major powers, the stronger are as likely as the comparatively weaker to initiate wars and disputes; compare Tables 3 and 5.

⁶⁷ Bueno de Mesquita (fn. 50), Table 4, p. 132.

⁶⁸ Levy (fn. 50), esp. 50–59.

⁶⁹ For example, Wayman measures shifts in polarity within the 19th and 20th centuries, while Levy codes the entire 1815–1945 period as multipolar. Compare Levy (fn. 50), Table 4.1, p. 49, with Wayman (fn. 49, 1985), Table 7.2, p. 127. Nevertheless, Levy can make a plausible argument that the 1815–1945 period is relatively multipolar when compared to other periods between 1495 and 1975. Levy's classification of unipolar periods is more controversial; Michael Wallace, in the same volume, takes sharp exception to it. See Wallace, , “Polarization,” in Sabrosky, 1985, pp. 108–9Google Scholar.

⁷⁰ See Wayman (fn. 49, 1985), at 126–27, 131–33 for the findings discussed below.

⁷¹ Rosecrance, Richard, “Bipolarity, Multipolarity and the Future,” Journal of Conflict Resolution 10 (September 1966), 317–27CrossRefGoogle Scholar; Haas, Michael, “International Subsystems: Stability and Polarity,” American Political Science Review 64 (March 1970), 98–123CrossRefGoogle Scholar.

⁷² Wayman (fn. 49, 1985), 133–35; Bueno de Mesquita (fn. 50), 135.

⁷³ Wayman (fn. 49, 1985), 129.

⁷⁴ Sabrosky (fn. 61), 181. Further evidence of the importance of minor powers in the occurrence of world wars is provided by Manus Midlarsky, who is outside the project. He finds that such systemic wars occur when serious disputes in the system begin to accumulate and are no longer stable (i.e., they are no longer randomly distributed in the sense that there is a balance in the number of disputes that are begun and ended in a given period). He finds that if only the disputes between major powers are examined, the periods prior to the two World Wars, 1893–1914 and 1919–1939, are stable. However, if the disputes involving “mid-range powers” (like Serbia, Rumania, or Belgium) are included, the system becomes very unstable. Midlarsky, , “Preventing Systemic War,” Journal ofConflict Resolution 28 (December 1984), 563–84CrossRefGoogle Scholar, and “A Hierarchical Equilibrium Theory of Systemic War,” International Studies Quarterly 30 (March 1986), 77–105CrossRefGoogle Scholar.

⁷⁵ Wallace, , “Polarization,” in Sabrosky, 1985, pp. 110–11Google Scholar. Also see Singer and Small, “Foreign Policy Indicators,” in Singer, 1979, pp. 322–23, for evidence of the different effects of alliances in the two centuries.

⁷⁶ In my view, it is not the balance-of-power aspect that is the significant difference: in the 19th century, alliances more frequently aimed to prevent war by coming to an understanding about how to deal with major issues; failing that, they aimed to keep any war that did occur limited. Bismarck was particularly adept at using alliances in this manner. After his departure, the approach of Kaiser Wilhelm II made alliances more threatening and therefore more likely to give rise to counter-alliances and arms races. The notion that a balance of power in and of itself can prevent war appears to be incorrect; see Levy (fn. 36), 608, esp. n. 19, and Vasquez (fn. 62). On the differences in alliances in different centuries, see Levy (fn. 36), 604–07.

⁷⁷ Wallace (fn. 24), 64.

⁷⁸ Wallace, , “Polarization,” in Sabrosky, 1985, pp. 110–11Google Scholar.

⁷⁹ Testing a complex interaction model originally offered by Mansbach, Richard W. and Vasquez, John A., In Search of Theory: A New Paradigm for Global Politics (New York: Columbia University Press, 1981)Google Scholar, chap. 7, Wayman finds considerable support for this hypothesis explaining up to 88% of the variance; see Wayman, Frank Whelon, “Voices Prophesying War: Events and Perceptions as Indicators of Conflict Potential in the Middle East,” in Singer and Stoll, 1984, pp. 180–82Google Scholar.

⁸⁰ Richardson, Lewis Fry, Arms and Insecurity (Pittsburgh: Boxwood Press, 1960)Google Scholar; Singer, J. David, “Threat-Perception and the Armament-Tension Dilemma,” Journal of Conflict Resolution 2 (March 1958), 90–105CrossRefGoogle Scholar, and Singer, , “The Outcome of Arms Races: A Policy Problem and a Research Approach,” Proceedings of the International Peace Research Association 2 (1970), 137–46Google Scholar, both reprinted in Singer, 1979, pp. 27–47, 145–54, respectively; also see Singer, 1984, chap. 7.

⁸¹ See Singer, “Threat Perception” (fn. 80), 33–34.

⁸² Singer, , “Escalation and Control in International Conflict: A Simple Feedback Model,” General Systems Yearbook 15 (1970), 163–73Google Scholar, reprinted in Singer, 1979, p. 73. Also see Singer, 'Threat Perception” (fn. 80), 36–37; Lowi, Theodore J., “Making Democracy Safe for the World,” in Rosenau, James, ed., Domestic Sources of Foreign Policy (New York: Free Press, 1967), 320–23Google Scholar; Wayman et al. (fn. 58), 498.

⁸³ For Singer's views on the role of the military and other bureaucratic and domestic groups in sustaining arms races, see his “Escalation and Control,” in Singer, 1979, pp. 77–78; “Outcome of Arms Races” (fn. 80), 151–52; Singer, 1984, pp. 259–60; and Singer, “The Responsibilities of Competence in the Global Village” (presidential address to International Studies Association), International Studies Quarterly 29 (September, 1985), 247–48, 254, 257–59Google Scholar.

⁸⁴ Wallace (fn. 26), in Singer and Associates, 1979, esp. 251–52.

⁸⁵ Weede, Erich, in “Arms Races and Escalation: Some Persisting Doubts,” Journal of Conflict Resolution 24 (June 1980), 285–87CrossRefGoogle Scholar, has criticized Wallace for treating each arms race and serious dispute as a dyadic case rather than counting all disputes and wars involving the same set of participants as a single case, regardless of the number of dyads. This is particularly important for World War I and World War II, both of which Wallace treats as several discrete cases. Wallace has re-analyzed the data and eliminated all cases in which two or more allies simultaneously entered the war against a common foe, thereby reducing the 28 cases of serious disputes having arms races to 15. Nevertheless, the relationship still holds; II of the 15 serious disputes escalate to war while only 4 do not. Conversely, only 2 of the 65 serious disputes escalate when there is no ongoing arms race. See Wallace, Michael D., “Armaments and Escalation,” International Studies Quarterly 26 (March 1982), 37–56, at 46CrossRefGoogle Scholar.

Diehl, Paul F., in “Arms Races and Escalation: A Closer Look,” Journal of Peace Research 20 (No. 3, 1983), 205–12CrossRefGoogle Scholar, manages to eliminate most of the relationship Wallace establishes; but to do so, he must remove 17 cases from the sample. Ten cases are eliminated because they are connected with ongoing wars and 7 more are deleted because they are connected with the First and Second World Wars (p. 210). In addition, some new cases are added and changes are made in the arms-race index. These manipulations are too extensive to support a conclusion that arms races are unrelated to the escalation of serious disputes. Diehl's main contribution is to suggest that much of Wallace's finding depends on the presence of ongoing wars and on World War I and World War II. (On the latter, see Houweling, Henk W. and Siccama, Jan, “The Arms Race-War Relationship: Why Serious Disputes Matter,” Arms Control 2, September 1981, p. 161, n. 13.CrossRefGoogle Scholar) This suggests that, while arms races may escalate disputes between rivals (relative equals), they may not be a factor in other types of wars. This hypothesis would also help to account for Diehl's finding (p. 209) that 77% of the wars that occur are not preceded by an arms race, since many of these wars are wars between unequals. Finally, the connection between arms races, escalation of disputes, and ongoing wars may provide a clue about how wars spread to become world wars. The presence of a series of disputes that might escalate to war may encourage others to arm because of the threatening environment, and thereby encourage war, when it comes, to enlarge.

⁸⁶ Organski, A.F.K., World Politics (New York: Knopf, 1958), 325–33Google Scholar.

⁸⁷ See Wallace, “Armaments and Escalation” (fn. 85), at 42–45, 47–51, for the findings reported here.

⁸⁸ Also relevant to evaluating the peace-through-strength argument are two studies by Stuart Bremer which show that stronger nations tend to be involved in wars more often; see his “The Trials of Nations: An Improbable Application of Probability Theory” and “National Capabilities and War Proneness,” both in Singer 1980, pp. 3–35 and 57–82. In a related study, Wolf-Dieter Eberwein finds that the more powerful a nation, the more likely it is to initiate or join a(n ongoing) serious dispute; see “The Seduction of Power: Serious International Disputes and the Power of Nations, 1900–1976,” International Interactions 9 (No. 1, 1982), 57–74CrossRefGoogle Scholar.

Paul Diehl (fn. 24), 343, also finds the peace-through-strength hypothesis deficient. Although Diehl questions the importance of arms races, per se, in the escalation of disputes, he does show that both arms races and military build-ups can be critical indirect factors in the escalation of disputes between states that have become rivals (i.e., have already experienced two disputes). His reservations about Wallace's finding on arms races, however, stem primarily from the fact that he has removed a number of cases associated with the two World Wars and has included 42 (out of 104) cases involving rivalries where at least one party has nuclear capacity (Table 1, p. 335). Among the latter are a large number of cases that have not resulted in the escalation of disputes to war; this should not be interpreted as undercutting Wallace's finding, which is primarily a product of the prenuclear era. On the whole, the Correlates of War project has not treated the post-1945 period in a way that permits an assessment of the impact of nuclear weapons; for an initial attempt by someone outside the project, see Altgeld, Michael, “Nuclear Weapons and War-Choice Decisions,” in Sabrosky, 1985, pp. 191–207Google Scholar.

⁸⁹ See Singer, “The Outcome of Arms Races” (fn. 80), esp. 147, 153. What is known is that many wars grow out of serious disputes and that it is difficult to avoid war when there is a pattern of repeated military confrontations. Wallensteen, Peter, in “Incompatibility, Confrontation, and War: Four Models and Three Historical Systems, 1816–1976,” Journal of Peace Research, 18 (No. 1, 1981), 74–75, 84CrossRefGoogle Scholar, a study of major power pairs, found that 75% (12 of 16) of the pairs of nations that had repeated confrontations also experienced war. See also Houweling and Siccama (fn. 85), 157–97.

⁹⁰ Leng, Russel J. and Wheeler, Hugh B., “Influence Strategies, Success and War,” Journal of Conflict Resolution 23 (December 1979), 655–84CrossRefGoogle Scholar. See also Leng, Russell J. with Goodsell, Robert A., “Behavioral Indicators of War Proneness in Bilateral Conflicts,” in Patrick J. McGowan, ed., Sage International Yearbook of Foreign Policy Studies 2 (1974), 191–226Google Scholar, reprinted in Singer and Associates, 1979, pp. 208–39.

⁹¹ Leng, in Singer, 1980, pp. 143–56. For an attempt to create a model that predicts which disputes will escalate on the basis of interstate interactions, see Maoz, Zeev, “A Behavioral Model of Dispute Escalation: The Major Powers, 1816–1976,” International Interactions 10 (Nos. 3–4, 1984), 373–99CrossRefGoogle Scholar.

⁹² Leng, “When Will They Ever Learn?” (fn. 25), esp. 398–99, 412–15; Leng, , “Reagan and the Russians: Crisis Bargaining Beliefs and the Historical Record,” American Political Science Review 78 (September 1984), 338–55CrossRefGoogle Scholar, reprinted in Small and Singer 1985, pp. 207–29.

⁹³ Gochman and Leng (fn. 27).

⁹⁴ The hypothesis that crises initiated by physical threats escalate to war is also supported by James, Patrick and Wilkenfeld, Jonathan, “Structural Factors and International Crisis Behavior,” Conflict Management and Peace Science 7 (Spring 1984), 33–53CrossRefGoogle Scholar, employing an independent data base.

⁹⁵ It is probably the territorial aspect of issues that is particularly explosive. See Most and Starr (fn. 46); Singer, , “Confrontational Behavior and Escalation to War, 1816–1980: A Research Plan,” Journal of Peace Research 19 (No. 1, 1982), 40CrossRefGoogle Scholar; and Diehl, Paul F., “Contiguity and Military Escalation in Major Power Rivalries, 1816–1980,” Journal of Politics 47 (No. 4, 1985), 1203–11CrossRefGoogle Scholar. However, it seems that issues that involve several parties are also explosive, since an intervention by a third party into an ongoing dyadic dispute will increase the likelihood of escalation to war. See Cusack, Thomas and Eberwein, Wolf-Dieter, “Prelude to War: Incidence, Escalation and Intervention in International Disputes, 1900–1976,” International Interactions 9 (No. 1 1982), 9–28CrossRefGoogle Scholar.

⁹⁶ Barringer, , War: Patterns of Conflict (Cambridge: MIT Press, 1972)Google Scholar; see pp. 102–12 for the findings reported here.

⁹⁷ Singer (fn. 95), 40, hypothesizes that the more equal the protagonists are militarily, the more likely an escalation to war.

⁹⁸ Little quantitative work has been conducted on the effect of the domestic political context, and especially the role of hard-liners, on the decision of leaders to go to war or use force. For a set of hypotheses that seek to identify general patterns, see Vasquez (fn. 30).

⁹⁹ Singer, “Escalation and Control” (fn. 82), 72–78, and Singer (fn. 95), 40.

¹⁰⁰ Vasquez (fn. 30).

¹⁰¹ Holsti et al. (fn. 29), 146, 148, 152–57.

¹⁰² For an attempt to model different hostile spirals so as to distinguish those that will escalate to war from those that will not, see Zinnes, Dina A. and Muncaster, Robert G., “The Dynamics of Hostile Activity and the Prediction of War,” Journal of Conflict Resolution 28 (June 1984), 187–229CrossRefGoogle Scholar.

¹⁰³ Singer would probably find this statement too strong; but see his “Confrontational Behavior” (fn. 95), 42–43. It should also be kept in mind that this statement does not mean that those who have advocated such realist practices are necessarily in favor of war. No one familiar with the political life of Hans Morgenthau or Reinhold Niebuhr would make such an assertion.

¹⁰⁴ Vasquez (fn. 12, 1983), 215–23.

¹⁰⁵ See in particular the essays reprinted in Singer, 1979, chaps. 2, 4, 6, 7, 15, and Singer, 1984, epilogue (also reprinted as “Nuclear Strategies and Peace,” in Small and Singer, 1985, pp. 377–9).

¹⁰⁶ See Singer, 1984, p. 299, and his two edited books on quantitative indicators: Singer and Wallace, 1979, and Singer and Stoll, 1984.

¹⁰⁷ See Singer, 1979, pp. 131–32; Singer, , The Scientific Study of Politics: An Approach to Foreign Policy Analysis (Morristown, NJ: General Learning Corporation, 1972)Google Scholar, excerpted in Singer, 1979, pp. 133–44, see esp. 136–37, 143; Singer and Small, “Foreign Policy Indicators,” in Singer, 1979, 328–29.

¹⁰⁸ Wallensteen (fn. 33), especially Table 2, p. 246.

¹⁰⁹ Specifically, they find that war occurs in every half-decade (from 1820–1914) among great powers when commitments are not considered binding (clausa rebus sic stantibus) in the international legal culture, but only in 50% of the half-decades when they are considered binding (pacta sunt servanda). See Kegley and Raymond (fn. 52), 586.

¹¹⁰ Kegley, Charles W. Jr., and Raymond, Gregory A., “Alliance Norms and the Management of Interstate Disputes,” in Singer and Stoll, 1984, 199–220Google Scholar.

¹¹¹ ” Because serious disputes (including wars) also decrease when there is a moderate degree of alliance flexibility (as opposed to rigid, polarized, or extremely fluid alliance structure), these findings (fn. no, esp. p. 210) can also be explained by a deterrence logic. The deterrence explanation seems less compelling than the one offered here because the explanation that the creation of a consensus on the rules of the game is an independent factor producing peace is consistent not only with the evidence presented by Kegley and Raymond, but also with the other evidence presented in this section. In addition, Kegley, and Raymond, , in “Normative Constraints on the Use of Force Short of War,” Journal of Peace Research 23 (No. 3, 1986), 213–27, esp. pp. 217–24CrossRefGoogle Scholar, use the notion of rules to account in detail for actual historical practices.

¹¹² An allocation mechanism is a set of rules and procedures, no matter how informal, for making and implementing political decisions, thereby providing a way for the disposition of stakes. See Mansbach and Vasquez (fn. 79), 282–87.

¹¹³ Doran, Charles F., The Politics of Assimilation (Baltimore: The Johns Hopkins University Press, 1971)Google Scholar.

¹¹⁴ A stable peace also reduces the number of major-minor wars and confrontations (see Wallensteen, fn. 33, p. 246); but such a peace does not imply justice, since it may be used by major states to divide or control minor states or other territories, as was done in the 19th century.

¹¹⁵ Singer, J. David and Wallace, Michael D., “Intergovernmental Organization and the Preservation of Peace, 1816–1965: Some Bivariate Relationships, International Organization 24 (Summer 1970), 520–47CrossRefGoogle Scholar.

¹¹⁶ Wallace (fn. 24), 65–66. finding, Kjell Skjelsbaek's, in “Shared Membership in Intergovernmental Organizations and Dyadic War, 1865–1964,” in Fedder, Edward H., ed., The United Nations: Problems and Prospects (St. Louis: Center for International Studies, Universitof Missouri, 1971), 31–61Google Scholar, that shared membership in intergovernmental organizations decreases prior to the outbreak of war is consistent with Wallace's findings. So, too, is the elaboration and re-test of the Singer and Wallace study (fn. 115) by Faber, J. and Weaver, R., “Participation in Conferences, Treaties, and Warfare in the European System, 1816–1915,” Journal of Conflict Resolution 28 (September 1984), 522–34CrossRefGoogle Scholar, which provides evidence for the notion that the presence of effective rules of the game reduces the need to go to war. For an interpretation of Wallace's findings along these lines, see Mansbach and Vasquez (fn. 79), 304–13. This evidence is also consistent with much of what Gilpin (fn. 32) has to say about the outcome of hegemonic struggles.

¹¹⁷ Singer, “Accounting for International War” (fn. 20), 14.