Abstract
Event data on the Israel-Lebanon and Israel-Palestinian conflicts in the Levant (1979-1999) and the Serbia-Croatia and Serbia-Bosnia conflicts in the Balkans (1991-1999) are used to test two sets of process-related hypotheses embedded in the theoretical and qualitative literatures on mediation. Cross-correlation analysis is used to examine the time delay in the effects of mediation on the level of violence over time. Results show that these effects vary somewhat, depending on the conflict, and differ substantially, depending on who is conducting the mediation. Whether conflict reduction is most effectively achieved with sanctions or rewards is tested. Results show that a reduction in violence is generally associated with mediation combined with conflictual action directed toward both of the antagonists and combined with cooperative action directed to the weaker antagonist.