Date of Award

Spring 5-15-2017

Author's School

Graduate School of Arts and Sciences

Author's Department

Political Science

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Hamas, the LTTE, FARC, the Zapatistas, and ETA all waged violent dissent for decades to no avail. Conversely, Anti-Fujimori protests, the Cedar Revolution, and the Jasmine Revolution overthrew governments through civil dissent within months. Despite the efficiency and efficacy of nonviolence, dissidents seldom change strategies.Extant studies on civil war onset and dissident strategies use panel datasets to determine what causes annual tactical decisions, overemphasizing the possibility of a change. Other works analyze strategic choice by collapsing tactics into a single category (violence or nonviolence) for dissidents' entire lifespans, ignoring tactical change. This dissertation parts ways from these studies by using a two-stage approach. First, I address which factors lead to initial dissident strategic choice. Second, I analyze the conditions responsible for the longevity of a given strategy. I also combine these analyses within a Heckman duration model to address the causes of persistent, violent dissent.Before these analyses, I revisit the issue of strategic change by unpacking the binary distinction between violent and nonviolent resistance. I revise the Nonviolent and Violent Campaigns and Outcomes (NAVCO) dataset to account for the empirical reality that many dissident groups use both strategies at once (dual strategies). In addition, I expand the dataset to include quiescent periods (phases after a government commits a grievance against a group, but before active resistance begins.) Altogether, I code 4,692 dissident-years for 232 groups. Contrary to the assertion that nonviolence is more effective than violence, I find that dual strategies are just as effective as nonviolent dissent.Using this new dataset, I test whether conditions during quiescent periods affect initial strategy choices. Along with violent ideologies and the necessity to ensure survival before success, I argue that failed nonviolent actions by a resistance group's precursor organization lead dissidents to select violence. Additionally, limited time to amass the social capital or organizational resources necessary for nonviolence increases the likelihood of violent dissent. My analyses support these claims.Thereafter, I test whether certain state policies designed to thwart violence actually prolong it. I argue that terrorist labels, restriction of movement, and state controlled media erode the mechanisms through which nonviolent dissent succeeds. Using survival analyses that measure the lifespan of a dissident group, I find that terrorist labels and state controlled media prolong dissent.


English (en)

Chair and Committee

Sunita Parikh

Committee Members

Justin Fox, Matthew Gabel, John Patty, Betsy Sinclair,


Permanent URL:

Available for download on Saturday, May 15, 2117