Abstract
This article aims to explain why strategic violence against targeted groups emerges after some civil wars but not others. It argues that when one side has captured the coercive apparatus of the state, and the potentially hostile losing side is less vulnerable to predation, the leaders of the victorious group can reward their domestic constituents and conduct in-group policing to prevent opportunistic violence. But when an armed group fails to achieve state capture and the losing side remains in a vulnerable position relative to its former enemies, neither side can credibly guarantee their domestic allies a share of the resources of the state or conduct effective in-group policing of potential extremists. Using Kosovo and East Timor as case studies, this article shows that in these cases, strategic violence is less a function of a concerted attempt to spoil a settlement than of the internal bargaining of new splinter groups.