Dissertation: Veto Players and Civil War Duration

 

My dissertation explains why some civil wars are resolved quickly and others are not. I argue that the number of actors involved in a civil war directly affects the ability of all parties to reach a negotiated settlement, and therefore how long the war lasts. Multi-party negotiations are harder for four reasons. First, in multi-party conflicts there is a smaller range of agreements that all parties prefer to continued warfare. Second, when multiple actors are involved it is more likely that at least one will overestimate its probability of victory and therefore block agreement. Third, in multiparty negotiations each actor has an incentive to hold out at the negotiating table to try to get the best deal. Fourth, shifting alliances between parties prevent the emergence of negotiating blocks that can make negotiating easier.

 

The dissertation contains two empirical tests of this veto player approach. Statistical analysis using a new dataset reveals a strong correlation between a higher number of parties and longer civil wars. Comparative case studies based on a most similar case design of civil war negotiations in Rwanda and Burundi show that multi-party negotiations in Burundi were more prone to breakdown due to factions walking out and refusing to participate. This dissertation demonstrates that the assumption in the existing literature that civil wars are two-party hinders our ability to explain the duration of these conflicts.

 

Committee: Barbara Walter (chair), Kristian Gleditsch, David Lake, Clark Gibson and Robert Horowitz

 

Link to my dissertation and précis.