Rebel-Civilian Social Ties and Their Effects on Violence Against Civilians in Civil War

Abstract

Why do some rebel groups target civilians extensively while others exercise restraint? This dissertation argues that prewar social ties between rebel combatants and the civilian communities they control are a critical and undertheorized determinant of rebel conduct. Social ties are bonds of personal familiarity, shared local identity and community solidarity that arise from cohabitation prior to conflict. They reduce rebel violence against civilians through three mechanisms. First, social ties increase the mutual legibility of rebels and civilians, reducing the information problems that drive violence. Second, they allow civilians to impose effective constraints on rebel behavior through informal institutions and reputational pressure. Third, social ties encourage rebels to delegate governance functions to existing civilian informal institutions. The dissertation also argues that these effects erode as social ties decay over the course of a conflict and that groups without ties use violence in particular to identify loyal civilians after territorial takeover. The empirical analysis uses a mixed-methods design. The quantitative analysis draws on original cross-national data on rebel-civilian social ties for a global sample of rebel groups, combined with a novel measurement model that captures both lethal and nonlethal forms of civilian abuse. The results show that rebel groups with social ties use substantially less violence against civilians than groups without such ties. They are also less likely to establish formal justice institutions and begin targeting civilians significantly later in conflict. A qualitative analysis of rebel-civilian ties and violence in the Syrian civil war, based on original fieldwork with conflict survivors, provides further support for the legibility and constraints mechanisms. Under groups with social ties, civilians reported greater certainty about rebel behavior and leveraged community institutions to challenge fighter conduct. Under groups without ties, these institutions were actively dismantled. This dissertation makes three contributions. It identifies prewar social ties as an independent determinant of rebel violence and centers civilian agency in the analysis of civilian targeting. It advances the measurement of violence against civilians beyond fatality counts to capture broader patterns of abuse. Finally, it carries direct implications for policymakers. The social origins of a rebel group are observable at rebellion onset and are a meaningful predictor of violence against civilians. Where social ties are present, strengthening existing community institutions enables civilians to hold rebels to account and limit the scope of rebel violence against them.