Despite widespread depiction of civil war as a pathway to autocracy or state failure, the empirical record shows significant variation in post-civil war states' regime trajectories. While some states settled into durable authoritarianism, others went on to enter the ranks of electoral democracies shortly after belligerents laid down their arms. What explains this variation? In the extreme, how is it that a state that is staunchly autocratic at the war's outbreak can emerge from it a nascent democracy? This study proposes that post-civil war regime outcomes have wartime origins. Differences in the nature of rebel governance of civilians generate different social and institutional legacies across civil wars. These legacies can endure into peacetime politics, affecting the latter in often unintended ways. The theory centers on two wartime transformations that result from different forms of rebel governance. First, where rebels depend heavily on civilian material support, civilians become mobilized as a political force.

Widespread social mobilization can in turn create political pressures on postwar elites to respond with a democratization strategy. Second, where rebel groups engage in extensive wartime "statebuilding," they create formal and informal institutions of governance which they can carry over into postwar politics should they prevail in the war. Because institutions are sticky, how they govern civilians in times of war can affect how they will govern in times of peace. These arguments are tested using both quantitative and qualitative methods. An original cross-national dataset on rebel governance for all civil wars ending between 1950 and 2006 serves as the basis, first, for a novel empirical analysis of rebel governance in civil war, then for statistical tests of the theory. To further probe the theory's causal claims, the study engages in an in-depth analysis of the Nepalese civil war and its political aftermath based on field interviews. The theory is further tested in a comparative analysis of the Ugandan, Tajik, and Mozambican civil wars.

Together, empirical findings show that rebel governance in civil war can catalyze significant social and political change, with enduring impacts on postwar political regimes. The study offers theoretical and practical implications for our understanding of, and response to, the politics of violent rebellion and its effects on regime development.