Catalysts of conflict : how refugee crises lead to the spread of civil war

Abstract

The spread of civil war due to refugee crises has occurred repeatedly throughout history. In some refugee crises, the sending state pursues the refugees, subjecting them to military attack. In other cases, militant exiles use the refugee camps as rear bases in their attacks on the sending state. As the cross-border attacks escalate, the risk of international war grows. The refugee crisis spawned by the 1994 genocide in Rwanda provides the most extreme example of this phenomenon. The militant actions of the Rwandan Hutu refugees in Zaire eventually sparked two international wars that led to further massive population displacement in the region. The recurring pattern of violent refugee crises, prompts the following three questions: 1) How widespread is the phenomenon of political violence involving refugees? 2) Under what conditions do refugee crises cause a civil war to spread across borders? 3) What role can international actors, such as the United Nations or the United States government, play in preventing the spread of violence? To answer the above questions, this dissertation presents new time series data on refugee-related political violence and also systematically compares violent and non-violent crises involving Rwandan, Bosnian, and Afghan refugees. This dissertation advances a political explanation for the spread of civil war in refugee crises and tests it against the prevailing socioeconomic explanation. According to the political explanation, three factors combine to cause the spread of civil war. 1) Strong political cohesion among the group before exile determines initial refugee militancy.

(cont.) 2) A refugee hosting state that lacks the capability and/or willingness to secure borders and demilitarize refugees facilitates the spread of war. 3) Third party states and non-state actors that intentionally or inadvertently contribute resources to combatants expand the conflict. The humanitarian assistance literature and the policy community routinely offer socioeconomic explanations that ignore the political context of the crisis. According to those explanations, camps near the border, large populations in camps, the presence of bored young men, and poor living conditions cause cross-border violence. This dissertation finds that none of those four socioeconomic propositions satisfactorily explain the spread of civil war.