BEFORE GENOCIDE: EVALUATING THE ATROCITY RISK ASSESSMENT FRAMEWORK

Abstract

This thesis examines the effectiveness of the Atrocity Risk Assessment Framework (ARAF) in predicting the risk of mass identity-based killing, including genocide. Recognizing the persistent yet infrequent nature of such atrocities, it investigates how accurately this model assesses risk in complex sociopolitical contexts. The primary research question was as follows: How effectively does the ARAF predict the risk of mass identity-based killing? Following an initial literature review on atrocity prevention, this study evaluated both qualitative and quantitative risk assessment methods, with a focus on this qualitative framework. Then, the ARAF was applied to the Rwandan genocide, analyzing pre-1994 events across 21 risk factors. The results revealed that 18 factors were manifest in Rwanda, though five factors were redundant, and the economic assessment component was weak. The thesis concludes that while the framework identifies significant indicators of atrocity risk, its complexity may limit practical application. Recommendations include refining the framework to improve efficiency and adaptability across varied risk contexts, ultimately enhancing early warning capabilities.