Dissertation:
The Politics of Violence: Empathy and action amidst conflict
My dissertation focuses on the consequences of violence for post-conflict citizen politics. Although war is popularly associated with destruction, recent research suggests war can have generative effects, particularly as it relates to citizen civic engagement. Despite expectations that victims of violence will disengage from civic life out of fear or trauma, the positive association between exposure to violence and future civic action is notably persistent across regions and political contexts. Why this relationship persists, however, is less clear. Predominant explanations emphasizing the role of victimization have neglected the experiences of those who are not direct victims and favor large-n statistical analysis, which tends to obscure processes and mechanisms. What explains the positive relationship between exposure to violence and civic engagement? My dissertation posits a new theory that highlights the crucial role of empathy in politics - an important but understudied phenomena. In doing so, my study opens new avenues for research into the affective elements of political life, the intersection of rational and emotion-based theories of decisionmaking, and civilian agency during civil war.
I theorize and analyze the mechanism driving the relationship between violence and civic engagement via a multimethod research design that integrates rich, ethnographic evidence with large-n statistical analysis and experimental methods. I leverage evidence from one key case – Mexico’s criminal conflict – to inductively generate a theory of civic engagement amidst violence. In the theory-generating stage, I fielded in-depth interviews with victims and non-victims of violence, and conducted ethnographic participant observation in violent and nonviolent communities in one of Mexico’s most violence-plagued cities, Monterrey. I find that the relationship between violence and civic engagement is motivated not by the trauma experienced by direct victims of violence, but instead by the empathy experienced by bystanders to violence. When citizens living amidst violence can imagine themselves in the shoes of a victim, they are more likely to perceive themselves as vulnerable to victimization. This sense of vulnerability motivates citizens to take preventative action, often in the form of civic activities that address violence. I test and demonstrate the external validity of these arguments outside the context of criminal violence in Mexico via statistical analysis of large-n survey data from two cases of civil war in Africa: Uganda and Liberia. A novel survey experiment conducted with a nationally representative sample of Mexican citizens provides further evidence of my theory. This research is grounded in over 16 months of field research conducted between 2013 - 2018.
Unlike many mixed methods projects, my work does not pair methods - it integrates them. This means that each element of my research - ethnography, statistical analysis, and experiments - grows out of and reflects processes of inquiry from the others. I developed the "Ethnographically-Embedded Experimental Design" approach for this project, which grounds the design of experiments in rich, ethnographic data.
I theorize and analyze the mechanism driving the relationship between violence and civic engagement via a multimethod research design that integrates rich, ethnographic evidence with large-n statistical analysis and experimental methods. I leverage evidence from one key case – Mexico’s criminal conflict – to inductively generate a theory of civic engagement amidst violence. In the theory-generating stage, I fielded in-depth interviews with victims and non-victims of violence, and conducted ethnographic participant observation in violent and nonviolent communities in one of Mexico’s most violence-plagued cities, Monterrey. I find that the relationship between violence and civic engagement is motivated not by the trauma experienced by direct victims of violence, but instead by the empathy experienced by bystanders to violence. When citizens living amidst violence can imagine themselves in the shoes of a victim, they are more likely to perceive themselves as vulnerable to victimization. This sense of vulnerability motivates citizens to take preventative action, often in the form of civic activities that address violence. I test and demonstrate the external validity of these arguments outside the context of criminal violence in Mexico via statistical analysis of large-n survey data from two cases of civil war in Africa: Uganda and Liberia. A novel survey experiment conducted with a nationally representative sample of Mexican citizens provides further evidence of my theory. This research is grounded in over 16 months of field research conducted between 2013 - 2018.
Unlike many mixed methods projects, my work does not pair methods - it integrates them. This means that each element of my research - ethnography, statistical analysis, and experiments - grows out of and reflects processes of inquiry from the others. I developed the "Ethnographically-Embedded Experimental Design" approach for this project, which grounds the design of experiments in rich, ethnographic data.
Violence and Conflict Processes
"Overkill: Understanding excessive violence during conflict" (working paper) | with Jerome Marston
This article advances a new conceptual category of violence, “overkill,” which describes acts of violence whose aim seems to surpass the goal of killing alone. We use quantitative data from Colombia and Mexico to show that, as a subcategory of conflict violence, overkill has distinct drivers and consequences for conflict trajectories.
“Threat Perception and Deception” (working paper)
How do different manifestations of violence shape citizen perceptions of threat? In this article, I demonstrate how the presence or absence of overkill-style violence shapes citizens’ perceptions of insecurity.
"Hazarding Data Collection in the Field: Confronting the dangers of measurement error and selection bias" (working paper) | with Jerome Marston
The methodological strategies we employ affect our research findings. This is particularly true for field researchers of violence and human rights, who work in high-risk but low-information environments and face distinct logistical challenges. There, the probability of methodological misstep is heightened, as are the accompanying empirical and normative consequences. In this article, we interrogate two methodological hazards particularly common to research in violent contexts - measurement error and selection bias - and draw on our field experience to lay out clear strategies for minimizing these hazards.
This article advances a new conceptual category of violence, “overkill,” which describes acts of violence whose aim seems to surpass the goal of killing alone. We use quantitative data from Colombia and Mexico to show that, as a subcategory of conflict violence, overkill has distinct drivers and consequences for conflict trajectories.
“Threat Perception and Deception” (working paper)
How do different manifestations of violence shape citizen perceptions of threat? In this article, I demonstrate how the presence or absence of overkill-style violence shapes citizens’ perceptions of insecurity.
"Hazarding Data Collection in the Field: Confronting the dangers of measurement error and selection bias" (working paper) | with Jerome Marston
The methodological strategies we employ affect our research findings. This is particularly true for field researchers of violence and human rights, who work in high-risk but low-information environments and face distinct logistical challenges. There, the probability of methodological misstep is heightened, as are the accompanying empirical and normative consequences. In this article, we interrogate two methodological hazards particularly common to research in violent contexts - measurement error and selection bias - and draw on our field experience to lay out clear strategies for minimizing these hazards.
Contentious Politics
“Vigilantism, politics, and violent civic responsibility: Evidence from Mexico" (working paper)
What explains participation in vigilante movements? In this article, I leverage qualitative evidence with statistical analysis of large-n survey data to explore this question. Unlike existing theories of vigilantism that emphasize self-help behaviors under weak institutional environments and/or legacies of violence, I show that individual factors associated with support for vigilantism are strikingly similar to individual factors associated with nonviolent civic engagement. This study thus makes an important revision to how we understand vigilante movements by emphasizing the political underpinnings of vigilantism and re-framing vigilantism as a political choice, rather than a survival strategy.
What explains participation in vigilante movements? In this article, I leverage qualitative evidence with statistical analysis of large-n survey data to explore this question. Unlike existing theories of vigilantism that emphasize self-help behaviors under weak institutional environments and/or legacies of violence, I show that individual factors associated with support for vigilantism are strikingly similar to individual factors associated with nonviolent civic engagement. This study thus makes an important revision to how we understand vigilante movements by emphasizing the political underpinnings of vigilantism and re-framing vigilantism as a political choice, rather than a survival strategy.
The Politics of Public Security
"Who is Safe? Varieties of Security in the American City" (working paper) | with Richard O. Snyder
The public good of local safety can be provided by a host of public, private and community agents combining in a multiplicity of different ways. Even within a single city, we observe striking variation in the provision of local security across neighborhoods. To describe and explain these local varieties of security, we propose a multilateral, multilevel and subnational framework. The framework offers a far stronger understanding not only of who is safe, but also how and why. We leverage evidence gathered via subnational comparative analysis of neighborhoods within one single American city, Providence, Rhode Island.
The public good of local safety can be provided by a host of public, private and community agents combining in a multiplicity of different ways. Even within a single city, we observe striking variation in the provision of local security across neighborhoods. To describe and explain these local varieties of security, we propose a multilateral, multilevel and subnational framework. The framework offers a far stronger understanding not only of who is safe, but also how and why. We leverage evidence gathered via subnational comparative analysis of neighborhoods within one single American city, Providence, Rhode Island.
