Terminated or Just Interrupted? How the End of a Rivalry Plants the Seeds for Future Conflict


  • Paul F. Diehl will share all data and coding for replication purposes. An earlier version was presented at the Annual Meeting of the International Studies Association-Midwest, St. Louis, MO, November 5–7, 2010. The authors would like to thank Amanda Murdie, Gary Goertz, and the journal's reviewers for their comments and suggestions.

Direct correspondence to Paul F. Diehl, Department of Political Science, University of Illinois, 420 DKH, 1407 West Gregory Drive, Urbana, IL 61801 〈pdiehl@illinois.edu〉.



The objectives of the study were to explain why some interstate rivalries recur while others remain terminated. We argue that the recurrence is largely contingent on the congruence between the first rivalry's military outcome and the resulting political settlement as well as the intensity of the original rivalry.


Our hypotheses are tested using a data set of interstate rivalries from 1816 to 2001. We use a logistic regression to examine which rivalries recur. As a robustness check, we use an event history model to test the effect of factors that change after the initial rivalry.


Our results indicated that a rivalry with an incongruent termination is nearly three times more likely to recur than one where the settlement reflects the preceding outcome, and more intense rivalries recur more than twice as frequently as their less intense counterparts.


Whether a rivalry recurs is strongly conditioned by how it concludes, and not by later events after the initial termination.