Political Science ETDs

Publication Date



Using a mixed-method design, my dissertation examines three interrelated puzzles of state violence during counterinsurgency campaigns. First, why does state repressive violence effectively thwart rebellion is some cases while escalating it in others? Using simultaneous equation modeling on 139 cases, I find that collective (or indiscriminate) state violence is associated with a long-lasting backlash effect against the government, while selective (i.e., individual-level) state targeting is expected to deter rebellion effectively. The second puzzle is why are states so seldom selective and so frequently collective in their use of violence? I argue that the scope of state violence is in, in part, a function of the capacities of the state agencies in the security sector tasked with counterinsurgency. State agents with high coercive capacity (e.g., national militaries) tend to lack sufficient local sources to identify and target insurgents at the individual level. Most local security organizations (e.g., police) that would otherwise have access to local intelligence lack the organizational competence to exploit this knowledge and become an efficient counterinsurgency force. The result is to rely on information about collectives (locations, ethnic groups, etc.) and target accordingly, which is associated with an escalation of the threat. This argument is supported by the first two stages of the Punjab crisis in India, where initial police attempts at thwarting rising extremism were incompetent and the armys heavy-handed approach produced a backlash against the government. Selective targeting requires local intelligence capacity and competent organizational capacity. The third puzzle is how do states (or state agents) effectively overcome these obstacles to carry out selective targeting? Using qualitative process tracing supplemented by elite interviews in India, I explain how the Punjab Police built organizational and intelligence capacity over the ensuing years to identify insurgents effectively at the individual level, which is often credited for the eventual defeat of the Khalistan militancy in Punjab, India.

Degree Name

Political Science

Level of Degree


Department Name

Political Science

First Committee Member (Chair)

Stanley, William

Second Committee Member

Peceny, Mark

Third Committee Member

Mitchell, Neil




civil war, state repression, human rights, counterinsurgency

Document Type