Book project

“Publicity and the de-legitimation of lynching”

Lynching in the United States was once a violent public ritual that was widely praised and implicitly justified in public, but by the late 1920s lynching was universally portrayed as an abhorrent and deeply shameful act. How did the lynching come to be seen for the atrocity that it is? How does the public reception of violence change from applause to opprobrium?

Read more... Political scientists that who study violence seek to explain its occurrence, the forms it takes, and its political consequences. While research about violence spans many disciplines and encompasses questions ranging from war to policing and incarceration, and from organized crime to riots and pogroms, types of violence often are treated separately. All of these disparate forms of violence share a key trait: they are subjected to societal evaluations as either acceptable or unacceptable. Max Weber defined politics as contestation over the state as the "sole source of the 'right' to use violence." But how is the "right" to violence determined? The question of how violence becomes legitimate or illegitimate is one of the most fundamental questions in politics, yet it is little understood. Literature on violence addresses this the question only in passing. Research on communications and public opinion examines questions of public legitimacy, but rarely addresses evaluations of violence. And finally, international relations theories of norms and norm transformation provide insight on how legitimation takes place, but have not yet been applied to violence per se. This book places the legitimation and de-legitimation of violence squarely at the center of a new research agenda. After laying out the scope of this new problem, I propose a theory of `publicity shocks' to explain how non-state violence becomes publicly unacceptable. I argue that between the late 19th century and the eve of Second World War, the expanding geographic reach of publicity about lynching---due to revolutions in transportation and communication---and growing inclusion of African American voices in public debate brought about a dramatic reversal in public support for lynching. To test this argument, I compile and analyze new data on press coverage of lynching in over eight million newspaper issues from 1880 to 1940, railroad and telegraph networks, and campaigns of anti-lynching activists.


Do parties matter for ethnic violence? Evidence from India,” (Lead article, Volume 11, Issue 3, Quarterly Journal of Political Science) (supplementary appendix)

with Gareth Nellis and Steven Rosenzweig

Abstract Ethnic-group conflict is among the most serious threats facing young democracies. In this paper, we investigate whether the partisanship of incumbent politicians affects the incidence and severity of local ethnic violence. We theorize that incumbents from multiethnic parties with long-standing attachments to victimized minority groups face uniquely strong incentives to quell ethnic unrest. To test our argument, we use a novel application of the regression-discontinuity design and show that as-if random victory by candidates representing India's Congress party in close state assembly elections between 1962 and 2000 reduced Hindu-Muslim rioting. The effects are large. Simulations suggest that had Congress lost all close elections in this period, India would have experienced 10 percent more riots. Additional analyses corroborate the mechanisms underlying our theory. Our findings shed new light on parties' connection to ethnic conflict, the relevance of party labels in developing states, and the puzzle of democratic consolidation in ethnically-divided societies.

Reported in the Economic Times, Hindustani Times (Live Mint),, quartz, Navbharat Times (in Hindi), and Azhimukham (in Malayalam).

“A Two-Edged Sword: the risks and rewards of violence for ethnic parties,” (under review)

Abstract What are the political consequences of ethnic violence? Research over several decades suggests that ethnic political parties benefit from ethnic violence. While these claims are widespread and are the foundation for many theoretical explanations of ethnic politics, there are good reasons to doubt that ethnic violence necessarily causes support for ethnic political parties. Moreover, these claims have not been systematically tested. In this paper, I suggest that the effects of ethnic violence on support for ethnic political parties differ (and may be opposite) for individuals with and without an ethnonationalist ideology. I test these arguments using a novel research design and data from Northern Ireland. I find that the average effect of both inter- and intra-group violence on support for ethnic parties is nil. This is because violence causes people who hold an ethnonationalist ideology to support ethnic parties, while it causes people who don't to oppose them.

“Protesting and Policing Boundaries: The role of protest in changing ethnic boundaries during the Civil Rights Movement” (Under review)

Abstract How are ethnic boundaries altered in the wake of challenges to ethnic hierarchy? While ethnic boundaries may evolve in the longterm, I argue that in moments of rupture boundaries can change quickly. Mass incarceration and police stop-and-frisk policies evidence the fact that the security apparatus of the state can institutionalize racial and ethnic boundaries through the threat of and use of violence. In this paper, I examine how the 1966 Campaign by the Chicago Freedom Movement led by Martin Luther King, Jr. and the Southern Christian Leadership Conference altered the police behavior towards, and thus the racial boundary of, the black community in American cities. I use unique data, collected in 1966, on the details of nearly twenty thousand police-citizen interactions in Chicago, Boston, and DC. In the midst of this data collection, the SCLC began housing demonstrations in Chicago. I exploit this coincidence to test whether the protests led the policing of black communities and the application of state power at the racial boundary, to intensify or abate. By showing how the police responded to protest against the racial status quo, this paper furthers understanding of the intersection of race and criminal law. More generally, this paper employs a strong research design and unique data on ethnic practices at the micro-level to show that the contents of ethnic boundaries change quickly during social upheaval.

Working papers

Ongoing projects

“From Service to Suffrage: U.S. Civil War veterans, the experience of combat, and support for black suffrage.”

Abstract How does war shape post-conflict politics? One important source of change comes from returning veterans who have acquired new skills, experiences, and attitudes during the war. In the wake of the U.S. Civil War, the expansion of civil rights to African Americans was a central debate in American politics. In many states, the expansion of suffrage to black men was put to a popular referendum. I argue that returning veterans brought home new perspectives on expanding suffrage. The military service and combat experience of Union veterans increased their support for black suffrage for two reasons. First, the harrowing experiences and sacrifices made during combat motivated veterans to make the outcome of the war meaningful. This made them particularly responsive to Republican efforts to "wave the bloody shirt" when mobilizing support for equalizing rights. Second, many white veterans lived and fought alongside black soldiers during the war, which created an opportunity for them to break down racial stereotypes and develop respect for the equal humanity of their black comrades. Drawing on complete enlistment and service records of Union regiments and county referenda returns from Iowa, Wisconsin, and New York, I show in a difference-in-difference design that that counties where a greater proportion of military-aged males served and saw combat experience had much higher support for black suffrage after the war.

“Waging freedom: Black Civil War veterans and the fight to redeem the promise of rights”

Abstract Political and civil rights are not self-enforcing. When these rights are extended to previously excluded groups, attempts to exercise these rights are often stymied by violent suppression. Can the oppressed effectively obtain their rights by organizing in self-defense? While the efficacy of violence in denying people electoral and civil rights has been widely studied, the efficacy of organized resistance has not. During Reconstruction, Southern whites used an array of threats and violence to deny newly freed blacks access to their rights. But there are many anecdotes of black veterans used their military training to create local organizations that conducted military drills, escorted voters to the polls, and negotiated better labor contracts. Did the presence of black veterans with the new capacities to organize and use violence enable black citizens to exercise their rights? I match the nearly 200 thousand surviving veterans of the US Colored Troops to 1870 census records to identify counties with greater concentrations of veterans. I then test the effect of concentration of black veterans on black voter registration, turnout, land ownership, and equality before the criminal law. Because the settlement of veterans is not random, I condition on the presence of US military units, Freedmans Bureau agents, and pre-war black population. I also instrument for the presence of black veterans using distance from their place of mustering out. This project contributes not only to understanding the history of Reconstruction, but also adds to the literatures on ex-combatants and on electoral violence.