The Politics of Unmixing: Riots, Segregation and Votes in India Open Access
Downloadable ContentDownload PDF
Do ethnic riots have long-term electoral consequences? While strategic political calculations are said to play a key role in the production of ethnic riots, existing studies illuminate only how riots shape electoral competition in the short-term. Drawing on a study of Hindu-Muslim violence in India, this dissertation argues that ethnic riots may also be thought of as a particularly brutal spatial strategy, designed by political actors to violently refashion social geography in ways that highlight and fix an ethnic divide in the long-term. Briefly, I advance a two-part hypothesis: (1) recurrent and severe riots shift the ethnic composition of electoral constituencies, constructing homogeneous constituencies where relative heterogeneity had been the norm - following Brubaker I name this process 'ethnic unmixing' (Brubaker 1995; 1998); and (2) greater ethnic homogeneity at the constituency level promotes lasting electoral support for ethnic parties. This dissertation articulates and tests several hypotheses about the efficacy of this spatial strategy in promoting long-term electoral support for matching ethnic parties, employing a mixed methods research design that combines large-N statistical analysis with extensive in- depth case study research. In doing so, the dissertation utilizes original quantitative data on religious demography at the local level in India, extensive interviews, and archival data gathered during 13 months of field research in India.