Dismantling the Conflict Trap : Essays on Civil War Resolution and Relapse

Abstract: Countries that have experienced civil war suffer a greater risk for new conflict than countries with no prior history of civil war. This empirical finding has been called a conflict trap where the legacy of previous war - unsolved issues, indecisive outcomes, and destruction – leads to renewed fighting. Yet, countries like Cambodia, El Salvador, Indonesia, and Mozambique have managed to overcome decade-long conflicts without relapse. This dissertation addresses this empirical puzzle by seeking to dismantle the conflict trap and look at microlevel explanations for civil war resolution and relapse. It adds to existing scholarship in three ways: first, by using disaggregated empirics on war termination and how fighting resumes; second, by exploring government agency in conflict processes; and third, by disaggregating rebel organizations. Essay I present original data on the start and end dates and means of termination for all armed conflicts, 1946-2005. Contrary to previous work, this data reveal that wars does not always end through victory or peace agreement, but commonly end under unclear circumstances. Essay II addresses how developments exogenous to the conflict influence governments’ decision to engage in a peace process. The results show that after natural disasters when state resources need to be allocated towards disaster relief, governments are more willing to negotiate and conclude ceasefires with insurgents. Essay III focuses on the post-conflict society, and posits that security concerns among former war participants will push them towards remobilizing into rebellion. The findings indicate that if ex-belligerent elite’s security is compromised, the parties of the previous war will resume fighting, while insecurity among former rank-and-file leads to the formation of violent splinter rebel groups. Finally, Essay IV seeks to explain why governments sometimes launch offensives on former rebels in post-conflict countries. The results show that internal power struggles provide leaders with incentives to use force against domestic third parties to strengthen their position against intra-government rivals. Taken together, this dissertation demonstrates that there is analytical leverage to be had by disaggregating the processes of violence in civil war and post-conflict societies, as well as the actors involved – both the government and rebel sides.