Title: Ending rebellion early: The initial stages of insurgency and counterinsurgency
Name: Janet I. Lewis

Political Science Department, U.S. Naval Academy
589 McNair Road
Annapolis, MD 21402

Year: 2011
Type: Dissertation Fellowship

How do armed rebellions begin? Scholars often probe the origins and onset of internal conflict, but rarely scrutinize how violence initially emerges. This study does so by examining the inception of all rebel groups that formed in Uganda since 1986, when the current ruling party gained power. It focuses on understanding why only some nascent groups become viable, while others fail too early to be captured in news media and thus remain omitted from the historical record and existing conflict analyses. Understanding these early stages of conflict is particularly important for policymakers who hope to end conflicts before substantial humanitarian costs accrue. By comparing the initial stages of rebellion for groups that become viable with those of groups that fail early, this project offers a rare view of how armed conflict begins  and how it sometimes ends before large- scale violence occurs.

The projects central findings highlight the importance of information in the initial trajectories of aspiring rebels. I show that most insurgencies begin as small, vulnerable, clandestine groups whose primary challenge is to avoid information leaks to the government. Several arguments supported by original qualitative and quantitative evidence, most of which was collected via fieldwork throughout Uganda, follow from this conceptualization of incipient rebellion.

First, in weak states  those with minimal institutional penetration and thus minimal monitoring in territory beyond the capital  barriers to entry for clandestine groups are low and therefore rebel formation occurs more commonly and with less spatial predictability than several dominant theories of conflict initiation suggest. Second, the decisions of civilians who live near newly-formed rebel groups, many of who could provide information about nascent rebels to the government, are critical in determining whether nascent groups survive. Civilians make decisions about whether to provide information to the government based primarily on what they believe other civilians will do; thus, variation in the structure of civilian information networks importantly influences incipient rebels chances for becoming viable. By showing a link between ethnicity and networks in rural Africa, the project advances a new understanding of how ethnicity can influence conflict onset. A third finding demonstrates the importance of domestic intelligence institutions in allowing states to access local information networks, deterring the initiation of new rebel groups.


Janet I. Lewis, 2012. How Rebellion Begins: Insurgent Group Formation and Viability in Uganda. Dissertation, Harvard University, Government.