Essays on intraethnic and interethnic violence.

Abstract

The three essays that make up this dissertation examine the stability and volatility of interethnic violence using agent-based models and historical case studies. The first essay examines how interethnic engagement and intraethnic policing---two mechanisms that respectively promote interethnic cooperation---interact with each other. The first mechanism involves local networks of civic engagement, whereas the second involves a common expectation that transgressions by members of other ethnic groups will be observed and punished by their compatriots. My results indicate that both mechanisms work together in a curious way: if groups police their members, violence can ensue despite civic ties that cut across ethnic groups, and peace is possible in the absence of these ties. The framework I develop begins to explain why some group conflicts ultimately result in more extreme levels of violence than others, or why peaceful relations between groups are maintained in some instances but break down in others. The second essay examines the conditions under which <italic>intra</italic>-group norms---rules that shape the behavior of group members toward nominal ethnic rivals---are instituted and enforced by entrepreneurs to enlarge the set of participants and increase the scale of intergroup violence. The focus, therefore, is on intra-group mechanisms that generate intergroup violence. My results indicate that ethnic norms are not equally likely to form in groups with similar aggregate preferences; that violence promoting norms can emerge in groups initially opposed to violence; that ethnic norms are more likely to emerge with strong punishments, unrestricted interaction among group members, and ethnic entrepreneurs; and that the probability of norm creation exceeds the probability of norm shift. The third essay examines how members of ethnic groups learn from <italic> external</italic> events---such as the persecution of ethnic kin in neighboring states---with implications for ethnic domination and intergroup violence at home. I combine a historical account of group relations with a simple learning model to trace the diffusion of events across countries. Building on the historical account, the model I develop generates patterns that bear a close resemblance to the trajectories of domination in Rwanda and Burundi, reveals that variation in the level of interethnic tension in dyads characterized by cross-border ethnic ties is tied to power transitions, and supports the notion that interethnic tension is likely to be higher under minority than majority rule.