Proxy Wars : The Effects of External Support on Multiparty Conflicts

Sammanfattning: How does external support affect the number of rebel groups active in a conflict? The aim of this thesis is to investigate whether there is a systematic relationship between external support and civil wars with more than one rebel group active, referred to as multiparty conflicts. Drawing on previous literature on the causes of such conflicts, I argue that external support increases the probability of multiparty conflicts by providing resources to nascent rebel groups while also causing polarization and inhibiting rebel alliances. Additionally, I argue governments involved in interstate rivalries are more likely to be involved in multipart conflicts, as the government’s rivals are more likely than other states to support the rebels. I test these theories using global data on the number of actors in intrastate armed conflicts 1975-2009 using logistic regression and Random Forest machine learning. I find that conflicts with external supporters have an increased probability of being multiparty conflicts. This holds true for both support from states to the government and the rebels, as well as support from non-state groups to the rebels. I also find that governments involved in interstate rivalries are more likely to be involved in multiparty conflicts.