World Politics

Research Article

The Enemy Within: Personal Rule, Coups, and Civil War in Africa

Philip Roesslera1*

a1 University of Oxford, Email: philip.roessler@politics.ox.ac.uk.

Abstract

Why do rulers employ ethnic exclusion at the risk of civil war? Focusing on the region of sub-Saharan Africa, the author attributes this costly strategy to the commitment problem that arises in personalist regimes between elites who hold joint control of the state's coercive apparatus. As no faction can be sure that others will not exploit their violent capabilities to usurp power, elites maneuver to protect their privileged position and safeguard against others' first-a rising internal threat, rulers move to eliminate their rivals to guarantee their personal and political survival. But the cost of such a strategy, especially when carried out along ethnic lines, is that it increases the risk of a future civil war. To test this argument, the author employs the Ethnic Power Relations data set combined with original data on the ethnicity of conspirators of coups and rebellions in Africa. He finds that in Africa ethnic exclusion substitutes civil war risk for coup risk. And rulers are significantly more likely to exclude their coconspirators—the very friends and allies who helped them come to power—than other included groups, but at the cost of increasing the risk of a future civil war with their former allies. In the first three years after being purged from the central government, coconspirators and their coethnics are sixteen times more likely to rebel than when they were represented at the apex of the regime.

(Online publication April 07 2011)

Philip Roessler is a senior research fellow in the Department of Politics and International Relations, University of Oxford. He is currently finishing a book on the internal politics of civil war onset in sub-Saharan Africa.

* I thank Andreas Wimmer, Lars-Erik Cederman, and Brian Min for sharing the Ethnic Power Relations data set and Patrick McGowan for sharing his African Military Intervention Events codebook and data file. I am also grateful for helpful comments from Leo Arriola, Nic Cheeseman, Donald Horowitz, Adrienne LeBas, Nic van de Walle, Harry Verhoeven, Andreas Wimmer, and four anonymous reviewers at World Politics. An earlier version of this paper was presented at the 2009 meeting of the American Political science Association in Toronto.