Nested Security as a Prerequisite for External Conflict Management: A Data Analysis of Civil Conflicts, 1946-2004

Co-authored with Juraj Medzihorsky and Levente Littvay.

The paper presents and tests a theory of nested security, which holds that third party mediation of sectarian conflicts is unlikely to succeed if the civil conflict is not first nested in a stable regional and hegemonic environment, not unlike Matryoshka nesting dolls. We test this argument using quantitative analysis of a Civil War Mediation (CWM) dataset, which includes data on civil conflicts from 1946 to 2004. Specifically, we test for whether civil conflicts are more likely to de-escalate when the wider regional environment has been stabilized—namely when there is no civil war in a neighboring state, there is no military intervention in the conflict, and when the host government is not neighboring a rival state. By conducting hierarchical analysis on these data, we test for whether there are strong associations between regional stability and conflict intensity (measured as the number of battle-related deaths). We find, among other things, that the presence of a neighbor state rivalry and military intervention are strong positive predictors of a conflict escalation, although we find little support for the civil war in the neighborhood. This provides initial support for the nested security model of conflict mediation that inter-state rivalry between the host and patron state and military intervention are associated with conflict escalation, although the lack of civil war in the neighborhood is not associated with de-escalation, suggesting that regional stability is only the first step toward successful third party mediation.