Interstate Security Networks and State Support for Rebel Groups

• In Event: Bringing IR Back in: The International in Conflict Processes

Abstract

With the observed decline of interstate war, states have increasingly adopted “gray zone” strategies short of militarization to prosecute international disputes. This paper explains one particular gray zone strategy: state sponsorship of nonstate armed groups. Existing research highlights the characteristics of the sponsoring state, the rebels, and/or the target adversary. But interstate conflicts rarely occur in a vacuum. This project advances a network approach to understanding state sponsorship. I argue that the structure of interstate competition and alignment in the international system shapes states' incentives to sponsor. Security networks in which multiple state challengers each express competing demands for concessions from the same target state are characterized by higher demand for the services of opposition groups operating within the target’s territory, as the supply of labor and resources in organized resistance against the target is limited. This pressures challengers to rush to forge sponsorship relationships in order to maintain bargaining position with the adversary relative to other challengers, increasing the sponsorship in the multilateral interstate conflict. Furthermore, scholars have focused primarily on explaining sponsorship in the context of interstate rivalry or in “politically relevant” dyads, defined with an eye to the contexts in which conventional militarized disputes are possible. But, only about 31% of sponsorship-years in San-Akca’s (2016) dataset occur within 5 years of an interstate rivalry period; implying that explaining state sponsorship and proxy war behavior requires examining a broader array of interstate relationships beyond contiguity and major power status. Empirically, I construct a time-variant network of states in the international system leveraging existing data on interstate rivalry (Thompson and Dreyer 2011), security competition (Maoz 2010 and Markowitz and Fariss 2017), and coded event data from ICEWS and TERRIER datasets to specify a latent variable model of interstate hostility for each dyad. The measure takes into account the dyad’s position in the broader interstate security network. I test the theory in a regression framework, measuring state sponsorship of rebel groups using San-Akca’s (2016) data and the level of interstate competition in a given target's security environment using network analysis techniques to summarize the structure of overlapping interstate hostility.

Author

• Michael A. Rubin, University of California, San Diego