Abstract Why has the international community refrained from intervention in the Turkish-Kurdish conflict that has lasted for two decades and several interventions took place in the Iraqi-Kurdish conflict?  The conflict literature has not found an adequate answer to this question, mostly, I argue, due to the lack of domestic variables in intra-state conflict theories and data.  

This project builds on the selectorate theory (Bueno de Mesquita et al. 2003) to construct a formal model that derives testable hypotheses from the interactions between the rebels, the target government and the potential interveners in a civil conflict.  Regime type of both the target and the intervener shapes the preferences of all actors that may or may not lead to an intervention decision.  The conflict, on the other hand, is shaped by the lack of formal recognition of the minority identity and/or political access to the government.  An existing data set is improved through the concept of “politically relevant potential interveners.”  The objective is to show that domestic politics not only affect the civil war onset in the country, but also impinge on the decision of third party interveners.  The quantitative analysis tests six hypotheses on regime type derived from four games in the model.  The qualitative analysis provides supporting evidence for the argument through the comparative case studies of the Kurds in Iraq and Turkey.  The basis of this section is field research in both countries.  Finally, policy implications of the findings and the need of further scholarship on the effect of domestic factors in civil conflict onset and on intervention decisions in civil conflicts are discussed.