Risking War: Regime Crises, Political Exclusion and Indiscriminate Violence in Africa

Thumbnail Image
umi-umd-4752.pdf(689.43 KB)
No. of downloads: 1233
Publication or External Link
Roessler, Philip
Lichbach, Mark I.
Between 1956 and 1999 one-third of the civil wars in the world occurred in sub-Saharan Africa. The prevailing explanation given to account for this fact is the economic weakness of African states. While low income is a robust determinant of civil war onset in global models, it is not as precise a predictor within sub-Saharan Africa. Instead, I argue that civil war is often a consequence of how African rulers respond to threats to regime survival, such as failed coups d'etat and other regime crises. In the wake of regime crises, rulers, concerned by their tenuous hold on power, seek to reduce the risk of future coups by eliminating disloyal agents from within the government and increasing spoils for more trusted clients to try to guarantee their support should another coup or threat materialize. The problem for the ruler is distinguishing loyal agents from traitors. To overcome this information problem rulers often use ethnicity as a cue to restructure their ruling networks, excluding perceived 'ethnic enemies' from spoils. The consequence of such ethnic exclusion is that, due to the weakness of formal state structures, the ruler forfeits his leverage over and information about such societal groups, undermining the government's ability to effectively prevent and contain violent mobilization and increasing the risk of civil war. To test this hypothesis, I employ a nested research design. The first part quantitatively tests the causal logic on a sample of 40 African countries between independence and 1999. I find that in the five years after a regime crisis there is a significant increase in the risk of civil war onset, often when the government resorts to indiscriminate violence to regulate the opposition. Part two examines this argument at the micro-level by examining two cases in Sudan based on hundreds of interviews during more than 14-months of fieldwork between 2005 and 2006. The second Sudan case illustrates that the civil war in Darfur in 2003 was a consequence of how the central government responded to a crisis within the Islamic movement in 1999 and 2000.