Conflict management of enduring rivalries: The frequency, timing, and short‐term impact of mediation

Abstract
In this study, we explore the frequency, timing, and impact of a particular form of conflict management—namely, mediation—and we do so in the dangerous context of international militarized rivalries over the period 1816–1992. We wish to explore how often does mediation actually occur in the context of enduring rivalries and understand at what phase mediation efforts are undertaken (if indeed they are made at all). Are mediation efforts only made in the most severe rivalries, or are they attempted in rivalries before they become enduring? At what stage in the rivalry is mediation attempted, and how does this fit in with the prescriptions derived from the extensive literature on timing and mediation success? Beyond a description of mediation in the context of rivalries, we wish to assess its impact on the short‐term dynamics of rivalries. Do mediation efforts make a difference? Do they help to postpone the onset of violence, lessen conflict severity, or prevent a war? The results indicated that enduring rivalries experience more mediation efforts than other conflicts. Compared to other less intense rivalries, enduring ones were up to ten times more likely to involve a third party (more than a majority of enduring rivalries actually had at least one mediation effort), and the average number of mediation attempts was significantly greater in the enduring rivalry context. Contrary to some prescriptions, we did not find mediation efforts to occur necessarily later in rivalries; mediations generally occurred at various stages of the rivalry process, regardless of whether it was a isolated, proto, or enduring rivalry. Overall, we found mediation attempts to have relatively little impact on the behavior of states in rivalries. They did not apparently influence the likelihood of subsequent war between rivals nor the level of severity for conflict that fell short of the war threshold. The most notable effects of mediation were found in their relationship to dispute “waiting times” or the interval from one dispute to the next. We found that positive mediation outcomes could, in some rivalries, lead to a delay in the onset of new militarized conflict, but the effect was modest. As anticipated, mediation attempts occurring in the latter stages of a rivalry did increase waiting times, but we judge this relationship to be spurious given the basic trend in rivalries toward greater time between disputes as rivalries mature.