Over the last fifteen years, the number of civil conflicts worldwide has declined and negotiated settlements have increased. The spread of democracy and greater international concern about domestic conflicts have encouraged states to adopt a negotiations-based approach to addressing minority grievances. In many conflicts, international intervention has played a significant role in facilitating dialogue and peace settlements. The complexities of cases of ethno political conflict suggest that a twin track approach, which looks at the domestic and the international levels of analysis, is critical. Yet, the existing literature on conflict management tends to study either international intervention or domestic institutions. Intersections between the two are ignored. Combining a cross-national analysis using the Minorities at Risk (MAR) dataset with a case study of Sri Lanka, this research project examines the relative and combined impact of domestic institutions and international intervention on the management and de-escalation of conflict. Uncertainty and mistrust between the state and minority groups drives political violence. Mitigating this uncertainty and building trust become essential for building peace. The extant literature fails, however, to recognize that the pathways to building trust and reducing uncertainty vary according to domestic political capacity. When the conflict-affected state is facing domestic institutional anarchy, coercive forms of international intervention, such as offering security guarantees through peacekeeping troops, become necessary. In contrast, where conflict co-exists with relative political stability and some measure of democracy, non-coercive intervention, such as mediation, becomes critical. In contrast to collapsed states, the challenge in such countries is to reform, rather than replace or create, institutions. Where the state is not facing a collapse of authority, facilitative intervention, such as mediation, can be a more cost-effective conflict management tool than high-cost, high-impact actions. This class of cases, which includes Sri Lanka, India, Indonesia and the Philippines, has not been adequately investigated in the existing scholarly literature. The arguments and findings presented here make an important contribution by focusing on the interactive role of domestic and international variables, particularly in relatively stable states.