What explains differential rates of ethnic violence in postcolonial Africa? I argue that ethnic groups organized as a precolonial state (PCS) exacerbated interethnic tensions in their postcolonial country. Insecure leaders in these countries traded off between inclusive coalitions that risked insider coups and excluding other ethnic groups at the possible expense of outsider rebellions. My main hypotheses posit that PCS groups should associate with coups because their historically rooted advantages often enabled accessing power at the center, whereas other ethnic groups in their countries—given strategic incentives for ethnopolitical exclusion—should fight civil wars more frequently than ethnic groups in countries without a PCS group. Analyzing originally compiled data on precolonial African states provides statistical evidence for these implications about civil wars and coups between independence and 2013 across various model specifications. Strikingly, through 1989, thirty of thirty-two ethnic group-level major civil war onsets occurred in countries with a PCS group.

Political scientists frequently use qualitative evidence to support or evaluate the empirical applicability of formal models. Despite this widespread practice, neither the qualitative methods literature nor research on empirically evaluating formal models systematically address the topic. This article makes three contributions to bridge this gap. First, it demonstrates that formal models and qualitative evidence are indeed frequently combined in current research. Second, it shows how process tracing can be as important a tool for empirically assessing models as statistical testing, because models and process tracing share a common focus on understanding causal mechanisms. Lastly, it provides new guidelines for using process tracing that focus on issues specific to the modeling enterprise, illustrated with examples from recent research. * We thank

A broad literature on how oil wealth affects civil war onset argues that oil production engenders violent contests to capture a valuable prize from vulnerable governments. By contrast, research linking oil wealth to durable authoritarian regimes argues that oil-rich governments deter societal challenges by strategically allocating enormous revenues to enhance military capacity and to provide patronage. This article presents a unified formal model that evaluates how these competing mechanisms affect overall incentives for center-seeking civil wars. The model yields two key implications. First, large oil-generated revenues strengthen the government and exert an overall effect that decreases center-seeking civil war propensity. Second, oil revenues are less effective at preventing center-seeking civil war relative to other revenue sources, which distinguishes overall and relative effects. Revised statistical results test overall rather than relative effects by omitting the conventional but posttreatment covariate of income per capita, and demonstrate a consistent negative association between oil wealth and center-seeking civil war onset.

Dictators face a power‐sharing dilemma: Broadening elite incorporation mitigates prospects for outsider rebellions (by either elites excluded from power or the masses), but it raises the risk of insider coups. This article rethinks the theoretical foundations of the power‐sharing dilemma and its consequences. My findings contrast with and provide conditionalities for a “conventional threat logic,” which argues that large outsider threats compel dictators to create broader‐based regimes, despite raising coup risk. Instead, I analyze a game‐theoretic model to explain why the magnitude of the elite outsider threat ambiguously affects power‐sharing incentives. Dictators with weak coup‐proofing institutions or who face deeply entrenched elites take the opposite actions predicted by the conventional logic. An additional outsider threat from the masses can either exacerbate or eliminate the power‐sharing dilemma with elites, depending on elite affinity toward mass rule. Examining the elite‐mass interaction also generates new implications for how mass threats affect the likelihood of coups and regime overthrow.

In contemporary democracies, backsliding typically occurs through legal machinations. Self-enforcing democracy requires that political parties refrain from exploiting legal opportunities to tilt electoral rules. Using a formal model, we argue that informal norms of mutual forbearance and formal constitutional rules are fundamentally intertwined via a logic of deterrence. By circumscribing how far each party can legally bend the rules, legal bounds create reversion points if mutual forbearance collapses. If legal bounds are symmetric between parties, they deter electoral tilting by making credible each party's threat to punish transgressions by the other. If legal bounds become sufficiently asymmetric, however, the foundations for forbearance crumble. Asymmetries emerge when some groups (a) are more vulnerable than others to legally permissible electoral distortions and (b) favored and disfavored groups sort heavily into parties. We apply this mechanism to explain gerrymandering and voting rights in the United States in the post-Civil Rights era.