Building a Science of World Politics

The contributions to this special issue provide a vehicle for moving the scientific study of conflict and war forward. These articles reflect the results of a workshop on Emerging Methodologies in the Quantitative Study of Conflict held at Binghamton University in March 2001. The objective was to represent recent efforts to apply emerging methodologies to some of the more knotty empirical problems facing scholars of conflict and its management. As this community develops more nuanced theoretical arguments, we need to be able to test our ideas empirically in ways consistent with our theoretical specifications. The collection of articles in this issue clearly shows that the scientific study of world politics has reached a high degree of methodological maturity. They reveal not only the achievement of a high level of sophistication but also an ability to adopt and adapt methods designed to deal with research problems unique to world politics. Some of our recent advancements involve more refined arguments to account for the processes that lead states into militarized conflict, driven in part-but not exclusivelyby understanding the effects of processes that operate at different levels of analysis. The dyad and/or the event, for example, pose methodological issues regarding strategic choice, opportunity, and the stationarity of time-series data. Furthermore, the entire notion of how we can best test our models, or hypotheses derived from them, has important methodological implications. That is, competing model testing has developed as a viable alternative to testing the null hypothesis (see Bennett 1997) and purports to be an important approach to addressing some of the contemporary debates in