Many insurgents engage in acts of violence against civilians in rural areas during civil war. In the absence of protection from the state, some villages within those areas organize armed groups to defend themselves against such violence. This project investigates the formation of these armed self-defense groups.

The formation of armed self-defense groups is under-studied and yet it is also significant and not uncommon phenomenon in civil war. Examples of conflicts in which armed self-defense groups have emerged include the civil wars in Algeria, Kenya, Peru, Sierra Leone and Sudan. Recent analyses have documented the prevalence of insurgent violence against rural villages. These studies have concentrated on explaining such violence but have neglected variation in civilian response.

This project conducts an in-depth comparative analysis of four rural villages, located close to one another, in southwest Sudan. The ethnic composition and level of insurgent threat was similar across the four villages during the early years of the civil war in south Sudan (1983-2005). Yet only two communities created armed self-defense groups. The project combines in-depth interviews, formal models, and network analysis to identify the process--events, actions and decisions--underlying the formation and non-formation of armed self-defense groups.

Armed self-defense groups can have two important effects on the evolution of civil wars. First, armed self-defense groups hinder insurgent action. Insurgents often need civilian support--forced or voluntary--for food, new recruits and intelligence. Armed self-defense groups usually form an obstacle to such support. Second, armed self-defense groups often come to play an important role in the state's counterinsurgency campaign. In the absence of weapons and military know-how, communities under threat typically resort to the state or to local army commanders to receive the material support and training that are necessary to build and sustain their armed self-defense groups. The state predictably demands some level of control in exchange for its support. As these exchanges continue, armed self-defense groups become co-opted by the state, developing into government sponsored militias that conduct offensive operations against the insurgents for the state.

This project has implications for general policy debates around disarmament, demobilization and reintegration (DDR) of former combatants, and for more specific policy debates around the future of south Sudan. There is a trend in debates over DDR towards the development of programs that are sensitive to local power dynamics. This project's identification of reasons and conditions underlying the initial armament of civilians fits within this trend and could inform DDR efforts. As for the future of south Sudan, many policy makers are concerned about the resurrection of government sponsored militias in the south. This project?s capacity to shed light on the origin of those militias should facilitate the production of policy responses to resurrection threats.

Project Report

In many civil wars, insurgents use violence against civilians. In some cases, civilians organize armed self-defense groups in response. Armed self-defense groups are armed groups, organized from within the village to protect the village against threats of insurgent violence. Armed self-defense groups are under-theorized yet consequential and not uncommon in civil war. They have emerged in various conflicts, including the civil wars in Algeria, Burundi, Chad, Mozambique, Peru, Sierra Leone and Sudan. Armed self-defense groups often influence the course of civil wars. They tend to become more offensive over time, turning into government sponsored counter-insurgent militias. They finally create long-term polarization between social groups affiliated with insurgents and social groups related to militias. In this project, PI and Co-PI investigated variation in formation of armed self-defense groups. Researchers focused on the second civil war in Sudan (1983-2005). They concentrated specifically on 24 tribal areas, asking why armed self-defense groups emerged in some areas but not in others. Researchers conducted interviews with tribal leaders (N=78), past insurgent commanders, past army commanders, past political leaders and religious leaders (N=27). Based on their interviews, researchers found that armed self-defense groups emerged in a particular tribal area when tribal leaders interpreted insurgent violence in local terms, i.e. they perceived insurgent violence as violence from the neighboring tribe. When tribal leaders did not interpret insurgent violence in local terms, armed self-defense group did not emerge. Researchers found that the interpretation of insurgent violence in local terms led tribal leaders to believe that future threats of violence were proximate, constant and manageable. Based on this belief, the formation of armed self-defense groups became both necessary and effective in mitigating threats of future insurgent violence.

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Standard Grant (Standard)
Application #
Program Officer
Brian D. Humes
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Wisconsin Madison
United States
Zip Code