War and Rivalry among Great Powers

We combine analysis of rivalry with consideration of a possible bias. We discover that what makes great powers more likely to be rivals is statistically related to their propensity to experience war. A censored probit allows simultaneous estimation of the influence of six covariates on the probability great powers experience rivalry and war, while controlling for statistical linkage between the two dependent variables. Consequently, we recast some past research. For example, contiguity and parity actually reduce the probability of war between great power rivals once we control for their influence on the probability great powers are rivals in the first place. Our results compliment other research highlighting various problems of bias in world politics and suggest a new direction for future research on rivalry. -e draw together two literatures about processes in international politics. A growing literature concerned with various selectionrelated inferential threats recommends we pay as much attention to how cases get into our datasets as we do to what occurs in those cases. A large literature on rivalry analyzes pairs of states repeatedly experiencing conflict. Both literatures highlight the importance of process in international conflict. By process we mean linked instances sequentially and cumulatively resulting in outcomes. Although many conflict processes likely exist, we investigate linkages across a process of the occurrence of rivalry and the onset of war. We believe the existence of rivalry and the onset of war are linked expressions of one underlying process and argue that treating them in this way may teach us much. The literature on selection effects within international politics helps us produce more valid estimates of what makes rivals escalate their relations to war, while interstate rivalry provides a powerful example of where methodological concerns about can inform the development of theory in international politics.