Working Papers
The Representation Trap: How and Why Muslims Struggle to Maintain Power in India [Abstract]
This paper is adapted from my dissertation project which you can learn about at the Dissertation tab.
Why do democracies often fail to represent marginalized groups in politics? I argue that one reason for this is that political gains for members of a marginalized group can create the very conditions for their political demise. When a marginalized group ascends to power without institutional protections it can result in a representation trap, where political incentives and identity are restructured in such a way that divides the marginalized group and unifies the dominant group. These shifts are likely to harm the political prospects of the marginalized group and increase majoritarianism. I provide evidence for the existence of a representation trap by studying one of the largest marginalized groups in the world's largest democracy, Indian Muslims. Since Indian independence Muslims have remained the only marginalized group to have no substantial progress toward political inclusion; in fact their representation has worsened over time. Using a regression discontinuity design, I reveal the representation trap: a Muslim political victory leads to a lower likelihood of future Muslim success. I document the causal logic driving this effect through additional election analysis, an original face-to-face survey of about 5000 Muslim and Hindu voters, and qualitative evidence drawing on about 150 elite and voter interviews in North India. Taken together, the theory and findings challenge the perspective that representation generally serves as a catalyst for the political empowerment of marginalized groups.
States worldwide use facial recognition technology (FRT) to assist in policing citizens, monitoring public goods, and even running elections. This article asks how FRT in polling stations affects voter turnout. Existing research on technology in elections offer ambiguous predictions for the direction and magnitude of the effect. I leverage a state-run randomized pilot of FRT in local elections in a municipality in Telangana, India to show that polling stations with FRT have lower turnout compared to those without. I discuss how three possible mechanisms might explain this effect: logistical issues, shifts in fraudulent activity, and apprehension about government surveillance particularly among marginalized citizens. Given the small sample of this pilot, the findings should be viewed as suggestive but indicative of the need for future research on the consequences that new technologies in governance can have on citizens in democracies.
Sacred Time and Religious Violence: Evidence from Hindu-Muslim Riots Conditionally Accepted at Journal of Conflict Resolution [Abstract]
How and when can religious times become focal points for communal violence? In the context of Hindu-Muslim riots in India, I argue that incompatible ritual holidays where one religion’s rituals are at odds with the other religion (e.g. sacrificing cows or engaging in processions with idolatry) explains the positive effect of sacred time on religious rioting. Holidays with incompatible rituals provide doctrinal differences that make riots more likely. I provide support for this argument by (1) analyzing riot data across 100 years of Hindu-Muslim riots (2) exploring individual-level surveys responses on holidays and (3) describing the way different holiday rituals play a role in violence. By focusing on the content of religion, this paper demonstrates how particular religious holidays can provide the underlying conditions that riot entrepreneurs use to incite religious violence.
Marriage, Conversion, and Religious Violence: Evidence from Colonial Burma [Abstract]
Inter-group marriages and conversions remain a constant source of agitation for majority-minority relations, yet how this phenomenon can contribute to violence remains unexplored. I theorize that elites can incite anti-minority violence by using marriage and conversion as a signal of a majority group’s loss of dominance. I demonstrate this dynamic in the context of the 1938 anti-Indian riots in the understudied context of Burma. I use colonial riot reports to assemble unique data on the victims of violence by religion. I find that townships where inter-religious marriage was salient were associated with increased rioting against Muslims—whose marriages involved conversion—but not Hindus. This effect is stronger when religious elites engaged in rioting. Qualitative evidence from archival documents demonstrates how elites framed marriage and conversion as a threat to the economic, political, and cultural dominance of Burmese Buddhists. These findings introduce an important but often overlooked driver of anti-minority hostility.
Works in Progress
Religious Identity Choice: Evidence from Indian Names (with Tanushree Goyal and Saad Gulzar)
Telegraphs and the Technology of Conflict: Evidence from the 1857 Indian Rebellion (with Zuhad Hai)
Other Writings
Using IOM Flow Monitoring Data to Describe Migration in West and Central Africa (with Darin Christensen, Guy Grossman, and Jeremy M. Weinstein) [Report] [Coverage in The Economist]
Book Review of Mass Religious Ritual and Intergroup Tolerance: The Muslim Pilgrims’ Paradox. Nationalities Papers. 2019. [Link]