Non-State Conflict Management and Civil War Duration: Do NGO Interventions Contribute to Shorter Civil Wars?

Outsiders often try to influence outcomes of internal conflicts. A growing body of literature explores the effects of third-party intervention, focusing primarily, though not exclusively, on states’ military and, more recently, diplomatic interventions. Non-state actors have become central players both in the conduct of war – e.g. rebel groups, terrorist networks or mercenaries – and in efforts of conflict resolution. Regional organization, churches and professional conflict management organizations are examples of groups that try to end wars. Such conflict management initiatives conducted by non-state actors are proliferating – but how relevant are they in an international system still dominated by states? Can actors without the means of state power affect the behavior of conflict parties and help end conflicts? Considering the growing number of NGOs and individuals getting involved in conflict resolution, it is worthwhile to analyze their effectiveness. I focus my research on track-two diplomacy in ongoing internal conflicts, where track two is defined as any conflict management initiative that is led by an individual or organization not directly tied to a government or an intergovernmental organization and targets the leadership level of parties in a conflict. The focus on internal conflicts is practical: Non-state actors find more political space to insert themselves into negotiations in civil wars than in interstate conflicts. Thus, if non-state conflict management has the effect it aims for – contributing to conflict resolution – then this effect should be most pronounced in intrastate wars. Using a new data set of track two diplomacy, I develop a duration model of African civil conflicts in the time period 1990-2010, hypothesizing that civil conflicts that experience non-state conflict management efforts will end earlier than those in which no non-state conflict management takes place, all else being equal. Early models find statistical significance for the main explanatory variable, but not in the expected direction. More work is needed to understand the relationship between non-state conflict management and conflict dynamics. This paper and the larger project it is part of contribute to the literature on civil war, mediation and conflict resolution and non-state actors in international relations. Findings will be interesting for states and non-state actors engaged in mediation and conflict management.