KANTIAN PEACE 9 resolving disputes grow,even nonliberal states may be impelled to use regional or international organizations to help settle their disputes rather than accept the political,military,and economic costs imposed by the liberal community as penalties for using military force.Thus in- creases in the Kantian influences at the system level may have beneficial effects on the behavior of dyads that are not particularly democratic, economically interdependent,or involved in international organizations. This is not an ecological fallacy.21 We do not make inferences about dyadic conflict from information about conflict at the systemic level.In all our analyses we address the incidence of militarized disputes among pairs of states.We investigate the consequences of purely dyadic char- acteristics for dyadic behavior,but we also do consider the effects of evolutionary changes in the international system.To capture the effects of such systemic changes,we use the annual mean scores of democracy, bilateral trade as a proportion of gross domestic product(GDP),and joint memberships in IGOs graphed in Figure 1.They effectively gauge the pervasiveness of changes in international norms and institutions and document the example of the success of liberal states in the com- petition among nations.We also consider the influence of the leading state,the hegemon,on interstate relations.We investigate this aspect of leading realist theories with measures of the relative power of the hege- mon,states'satisfaction with the status quo,and the hegemon's sense of its own security. HISTORICAL DOMAIN,KEY VARIABLES,AND SOURCES OF DATA As our analysis spans the years 1885 to 1992,it enables us to examine the effects of democracy,economic interdependence,and international organizations over a long period before the cold war and for a few years after.Realists often contrast the dynamics of bipolar and multipolar systems,though there is disagreement over their consequences for in- terstate relations.By Waltz's criteria,the international system was mul- tipolar for the centuries preceding 1945 but bipolar during the cold war.22 And the current,post-Soviet world is neither bipolar nor multi- polar but perhaps is best understood as unipolar,at least as measured by 2 Identified by W.S.Robinson,"Ecological Correlations and the Behavior of Individuals,"Ameri- can Sociological Review 15,no.3(1950).On how some inferences can be made,see Gary King,A Solu- tion to the Ecological Inference Problem(Princeton:Princeton University Press,1997). 22 Kenneth Waltz says that it is the power of the units(states)themselves that defines polarity and not the number or power of the alliances they lead;see Waltz,Theory of International Politics(Reading, Mass.:Addison-Wesley,1979),98-99.Thus the formation of two opposing alliance systems prior to World War I did not change the structure of the multipolar system.Waltz's emphasis on the systemicKANTIAN PEACE 9 resolving disputes grow, even nonliberal states may be impelled to use regional or international organizations to help settle their disputes rather than accept the political, military, and economic costs imposed by the liberal community as penalties for using military force. Thus in creases in the Kantian influences at the system level may have beneficial effects on the behavior of dyads that are not particularly democratic, economically interdependent, or involved in international organizations. This is not an ecological fallacy.21 We do not make inferences about dyadic conflict from information about conflict at the systemic level. In all our analyses we address the incidence of militarized disputes among pairs of states. We investigate the consequences of purely dyadic char acteristics for dyadic behavior, but we also do consider the effects of evolutionary changes in the international system. To capture the effects of such systemic changes, we use the annual mean scores of democracy, bilateral trade as a proportion of gross domestic product (GDP), and joint memberships in IGOs graphed in Figure 1. They effectively gauge the pervasiveness of changes in international norms and institutions and document the example of the success of liberal states in the com petition among nations. We also consider the influence of the leading state, the hegemon, on interstate relations. We investigate this aspect of leading realist theories with measures of the relative power of the hege mon, states' satisfaction with the status quo, and the hegemon's sense of its own security. Historical Domain, Key Variables, and Sources of Data As our analysis spans the years 1885 to 1992, it enables us to examine the effects of democracy, economic interdependence, and international organizations over a long period before the cold war and for a few years after. Realists often contrast the dynamics of bipolar and multipolar systems, though there is disagreement over their consequences for in terstate relations. By Waltz's criteria, the international system was mul tipolar for the centuries preceding 1945 but bipolar during the cold war.22 And the current, post-Soviet world is neither bipolar nor multi polar but perhaps is best understood as unipolar, at least as measured by 21 Identified by W. S. Robinson, "Ecological Correlations and the Behavior of Individuals, "Ameri can Sociological Review 15, no. 3 (1950). On how some inferences can be made, see Gary King, A Solu tion to the Ecological Inference Problem (Princeton: Princeton University Press, 1997). 22 Kenneth Waltz says that it is the power of the units (states) themselves that defines polarity and not the number or power of the alliances they lead; see Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979), 98-99. Thus the formation of two opposing alliance systems prior to World War I did not change the structure of the multipolar system. Waltz's emphasis on the systemic