Military Purges in Autocratic Sudan (1989–2019): Introducing a New Dataset

• In Event: Strategies of Authoritarian Regimes

Abstract

This paper investigates the political logic of military purges during Sudan’s former autocratic regime (1989-2019) using a new dataset built upon original primary sources collected in 2019 in Sudan.

Military purges are a common occurrence in authoritarian contexts and are often considered as “par for the course” for autocratic rule. Indeed, the military has been conceptualized as a major threats to such regimes since 1) it is usually the institution with the means to organize a coup and 2) coups are the most common way autocrats are overthrown (Svolik 2009). Recent events in Turkey (2016), Sudan (2019) and Mali (2021) appear to confirm the relevance of this finding.

Yet, systematic conceptualizations and investigations of autocratic purge processes remain rare. In qualitative area studies, the small amount of research on the topic tend to focus on the same cases: Stalin’s USSR (Whitewood 2015) and Franco’s Spain (Valero 1997; Balcells et Villamil 2020). Military purges in African contexts remain under researched, though some authors investigate the management of public servants in perspectives that dialog with the concept of purges (see notably Woldense 2018). African militaries have also been investigated in the extensive coup-proofing literature, which started to develop at the end of the 1990s. Yet, this strand of research rarely mentions purges explicitly. Scholars look mostly at practices of counterbalancing (Escribà-Folch, Böhmelt, et Pilster 2020; Belkin et Schofer 2005) and ethnic stacking (Morency-Laflamme 2018; Roessler 2011), seldom engaging with purges as a distinct strategy to be conceptualized to the same extent. However, an emerging strand of scholarship has started to fill this gap. It investigates the “strategic logic of purges” and their relationship to autocrats’ survival (Sudduth 2017; 2021; Wong et Chan 2021; Easton et Siverson 2018; Fruge 2019) and often uses quantitative methods and formal modelling.

This scholarship has raised new questions on processes of purges, notably on their timing, methods, and targets. This paper contributes to these current theoretical and empirical debates by presenting a new dataset on military purges in Sudan during Omar al-Bashir’s islamist and military regime (1989-2019). It is built upon primary sources that were collected during a field trip in Sudan in 2019. Those sources are documents collected by a Sudanese organization named the “Executive Committee of the Dismissed - All titles”, which has attempted to defend the rights of wrongfully terminated civil servants since 1997. In 2019, I was able to photograph the content on a binder about individuals dismissed from the military. It included two types of documents: forms filled out for the Committee by the dismissed soldiers and official forms of termination provided by the army. These documents contain data such as birthdates, hiring and termination dates, levels of education etc. In collaboration with a team of research assistants, I turned those 567 photographs into a quantitative dataset, first translating them then coding their content. This is, to the best of my knowledge, the first dataset on this topic on Sudan. The core attribute of the dataset is therefore that it is based on primary sources in the local language – most quantitative studies rely on English-speaking press as primary sources for datasets - and presents fine-grained data, at the level of the individual. Such data on purges are notoriously difficult to find and/or access (Balcells and Villamil 2020).

The paper is divided into three parts. The first one presents the state of the art regarding the topic of purges, highlighting how purges are conceptualized in relation to regime survival as part of a potential tradeoff between coup-proofing and military effectiveness that has paradoxical effects when protests or civil wars erupt. In this literature review, I highlight two limitations of current research: first, a tendency to define purges by their focus on high-level military officers, and second the bias of datasets based upon anglophone sources. The second part presents in more details the case of Sudan’s former autocratic regime and the data, describing their own limitations and how the dataset is being built. The last part is dedicated to present preliminary descriptive statistics and their analysis. To conclude, I highlight the potential theoretical contributions of this dataset to the quantitative and cross-national study of purges as well as to qualitative studies of authoritarianism and Sudan.

Author

• Anne-Laure Mahé, Institut de recherche stratégique de l'école militaire