Borders and the Feasibility of Rebel Conflict

Abstract

Contemporary spatial research on civil conflict in Sub-Saharan Africa has largely focused on border regions as spaces of limited political and economic opportunity. These studies largely adopt approaches that present borderlands as institutionally desolate regions lacking in governance, economic opportunity and political inclusion and giving rise to the feasibility of rebel conflict. While spatial analyses focus on territorially-based capabilities, such as state power projection, they typically overlook borderlands and their territorial distinctiveness with regards to rebel capabilities. This paper specifically explores the structural effects of borders on rebel capabilities and argues that Sub-Saharan Africa’s porous borders enhance the capabilities of rebels to operate in nearby territories. I empirically test this hypothesis with a zero-inflated negative binomial model and spatially disaggregated conflict events data from the Armed Conflict Location & Event Data Project dataset mapped to the PRIO-GRID 0.5-degree x 0.5-degree geographic data structure. In total, the analysis covers 14,120 georeferenced rebel conflict events in 37 countries between 1997-2019. The results provide strong evidence that territories nearer to borders are likely to experience more battle events relative to other territories, suggesting that borderlands may enable distinct conflict-related capabilities for rebels not found elsewhere. Additionally, the model also differentiates the effects that the border may have on conflict, testing the effect of rough terrain, resources, excluded groups, and towns at the border. Of the variables tested, the results suggest that territories with border towns significantly increase the capabilities of rebels to engage in conflict and suggest a more nuanced scholarly consideration of cross-border institutions that facilitate rebel conflict.