Three Essays on the Dynamics of Conflict in Civil Wars

Abstract

Civil wars in the last three decades have produced staggering death tolls, unleashed huge waves of human migration through refugee flows, and generated incalculable human suffering. Understanding the dynamics of civil conflicts -- how they are fought, how they end, and their legacies on the societies that survive them -- is of critical importance, perhaps now more than ever. In this dissertation I explore three central dimensions of civil war dynamics, using the case study of the Colombian civil war as an empirical context with which to evaluate my theory-building. Chapter 2 analyzes how the electoral process shapes patterns of violence in countries experiencing conflict. I combine statistical models with fine-grained data on the timing of local elections and the prevalence of violence during three decades of Colombian history to show that insurgents respond to the electoral process and wield violence to achieve electoral goals. The results raise caution about the prospect of democratization as a palliative to conflict. Chapter 3 explores how attempts to mitigate conflict by promoting economic growth can backfire. I argue that in contexts where the state is weak, the expansion of land-intensive industries can incentivize land-grabbing and the displacement of civilians. I collect original data on the rapid expansion of the palm-oil industry in early 21st century Colombia to show that growth in this industry was associated with mass civilian displacement. The findings warn against intuition that economic growth necessarily reduces violence and instead suggests that actors can take advantage of ongoing conflict for private gain. Finally, Chapter 4 focuses on the challenge of generating public support for conflict-termination in deeply divided societies. I conceptualize the broad points over which state and insurgent actors have to agree to reach settlement, and draw testable hypotheses for how different kinds of settlements will move public opinion. I use novel survey experiments fielded during the 2016 Colombian peace process to demonstrate that normative questions bearing on punishment deeply divided citizens. I derive implications for policymakers seeking to construct peace agreements with broad bases of public support.