When do nonviolent revolutions result in stable democratic regimes? Existing scholarship is optimistic about the positive relationship between nonviolent revolutionary transitions and post-revolutionary democracy. However, several regimes established following nonviolent revolutions fail to consolidate into democracy, often being derailed by coups and civil wars. Nevertheless, we know comparatively little about when and why nonviolent revolutions lead to unstable or undemocratic regimes. I theorize that security force defection during nonviolent revolutions undermines stability and democratic consolidation in the regimes that follow. Security force defectors possess distinctive motives, opportunities, and coercive capabilities that equip them to destabilize democracy in the post-revolutionary era. Empirical analysis using global data capturing nonviolent revolutions from 1945 to 2013, reveals that security force defection both significantly increases the likelihood of post-revolutionary coups and civil wars and diminishes the democratic character of post-revolutionary regimes. This article thereby revises the scholarly consensus regarding the benefits of security force defection during nonviolent revolutionary movements.

Why do some winning rebel groups build loyal postwar militaries, while others do not? Surprisingly, though victorious rebel leaders cooperate successfully with their military commanders to seize sovereign power, they often struggle to share it.  In roughly one-third of rebel regimes, military elites attempt to depose their leaders through military coups. I contend that postwar coup risk is primarily shaped by the wartime roles of rebel leaders. When rebel leaders occupy a frontline role, exposing themselves to the hazards of the conflict theater, they are less likely to suffer postwar coup threats. Frontline leaders win greater personal prestige, and stronger legitimacy among military audiences. Additionally, frontline leaders’ greater proximity to their military commanders allows them to develop institutional structures that more effectively reduce military autonomy. However, remote leaders, who are absent from the conflict theater, exercise weaker postwar control over the armed forces and therefore face greater coup risk. I find support for this theory using original cross-national data on all rebel regimes and the wartime roles of victorious rebel leaders between 1945 and 2010. A controlled case comparison of civil-military relations in two rebel regimes - Guinea-Bissau and Uganda - further illustrates the mechanisms tying wartime roles to postwar coups.

The complex nature of civil conflict makes it inherently difficult to study, which is exacerbated by the challenges of conducting field research in volatile conflict zones. How do these challenges affect the data gathered and inferences drawn about armed groups? This article examines the epistemological, practical, and ethical issues of researching armed organisations across varying power relations, space, and time. We highlight the divergent perspectives that emerge within armed group hierarchies, the varying realities across conflict geographies, and shifts in armed group trajectories and narratives that arise over time. Based on research in Africa and Latin America, we identify the pitfalls of capturing isolated ‘snapshots’ related to with whom, where, and when research was conducted, and we advocate for a ‘panoramic’ approach to the study of armed groups, recognising both limits to available data and the evolving nature of armed groups and their behaviour. Ultimately, we offer a conceptual roadmap for understanding the main axes along which variation in armed group organisation and behaviour unfolds, provide recommendations on how to navigate them, and call for caution in the generalisation of research findings and policy prescriptions, given the limitations and liminal nature of knowledge about armed groups.

An influential line of scholarship argues that regimes originating in violent rebellion are extremely durable. However, the legacies of violent founding struggles vary markedly. Some rebel victories give rise to powerful regimes enjoying prolonged rule, yet others engender comparatively weak regimes that rapidly succumb to armed threats. Why? I argue that how victorious rebellions are organized critically shapes regime durability. Specifically, I argue that victors who engaged in extensive rebel governance during wartime establish the most durable regimes. Rebel governance demands that rebels establish robust institutions dedicated to reining in fighters and governing civilian masses. The skills, structures, and social ties developed in the course of wartime governance persist into the postwar era. Governing victors establish postwar regimes with more reliable armies, skilled administrators, and socially penetrative state infrastructures - critical resources for maintaining power over the long term. By contrast, victors with a negligible history of governance made comparatively limited wartime investments in ruling civilians and regulating fighters. The ensuing rebel regimes lack the military discipline, administrative wherewithal, and social control with which to parry existential threats to rule. I find strong empirical support for this argument using global data on all rebel victors and subsequent regimes from 1945-2015. Leveraging fieldwork materials, primary and secondary sources, I further illustrate the theory through a controlled case pairing of regimes established by victorious rebellions in Uganda. These cases highlight how the distinctive organizational residues of rebel governance enhance regime durability, and how the organizational deficits of nongoverning rebels undercut lasting rule.