Repression and refuge

This article asks why ethnic exclusion from executive-level state power leads to armed conflict in some cases but not in others. To resolve this puzzle, it focuses on the possible role of five additional, qualitatively coded factors that have been considered by either grievance or opportunity theories of civil war but for which quantitative data are not readily available. To assess the combined relevance of these factors, the authors use qualitative comparative analysis (QCA) to explore the diverging conflict trajectories of 58 ‘most similar’ ethnic groups. These groups have a uniformly high conflict propensity because they are politically excluded, situated in poor countries, live geographically concentrated, and comprise substantial parts of the population; yet, only 25 of them actually experienced violent conflict. The results show that the resentment created by ethno-political exclusion translates into violent conflict if the state reacts against initial protests and mobilization with indiscriminate violence, and if there is a refuge area either within or outside the country that allows regime opponents to organize armed resistance. Moreover, a more processual analysis of conflict dynamics reveals that the conditions conducive to ethnic rebellion appear in a particular temporal sequence with a clear and universal escalation pattern.