“The Moderating Effect of Democracy on Climate-Induced Social Conflict: Evidence from Indian Districts.” (with Ore Koren). Political Research Quarterly. 75(3):892–905. View paper »
Do political institutions moderate the effect of environmental stress on social conflict? We posit that while the frequency of social conflict in developing agrarian states can increase during drought, democratic competition reduces conflict and can facilitate cooperation, reversing this effect. This hypothesis is tested on a sample of all districts in India over a period from 2001 to 2014. The dependent variable captures the number of crimes perpetrated against scheduled castes—so-called “untouchables”—and scheduled tribes—India’s Indigenous groups — during a given district-year. When the effect of drought is moderated using a local electoral competition index, findings show that although droughts increase the frequency of social conflicts where political institutions are weak, they reduce it where political institutions are strong. The results are robust to alternative operationalization choices. Our findings, thus, have relevance both to scholars of the climate–conflict nexus and to policymakers working to address climate change’s effects.
“Structure of Religion, Ethnicity & Insurgent Mobilization: Evidence from India.” World Politics. 73(1):82–127. View paper »
Abstract: Colonialism is associated with the expansion of organized religion among ethnic groups across the world. These religious interventions came in varied forms and brought about socio-structural changes in targeted societies, which had long-term political implications. This paper relates such changes to post-colonial rebel mobilization. It leverages micro-level variations in changes associated with the spread of organized religion, and unique data on rebel recruitment providing information on the entire universe of recruits in an ethnic insurgency in India, to examine the effect of the socio-structural context on rebel recruitment. Drawing upon a combination of archival research, semi-structured interviews and statistical analysis, this paper demonstrates that enhanced structural connectivity of the population resulting from a network of highly centralized churches and civil society organizations under the Welsh Presbyterian Mission significantly bolstered insurgent recruitment. The findings are remarkably robust to various alternative measures of structural connectivity, a range of specifications and controls and are unlikely to be explained by unobserved variables. Relying on ethnographic evidence, this paper demonstrates that the findings are not unique and extend to other cases.
“State Capacity, Insurgency and Civil War: A Disaggregated Analysis.” (with Ore Koren). International Studies Quarterly. 62(2):274–288. View paper »
Abstract: Studies on civil war usually approximate state capacity using broad indicators such as gross domestic product, bureaucratic quality, military spending and mountainous terrain. While these factors can produce observable effect on the onset of civil war, they do not adequately account for the actual degree to which states can exercise control over their peripheries, or correctly proximate the distribution of this capacity within the state’s boundaries. We provide an alternative highly-disaggregated measure of state capacity, nighttime light, which we believe better accounts for the degree of intrastate control and the variations therein. We postulate theoretical mechanisms linking the provision of electricity with the expansion of state capacity. We establish the validity of this measure against a variety of previously-used measures of state capacity at different levels of disaggregation and across multiple contexts. In global analyses of civil war onset, we find that, counterintuitively, civil war is more likely to erupt in locations where the state exercises higher degrees of control. These results hold even after the effect of cells that are practically immune to civil war is taken into account, and suggest that more attention should be given to the manners in which scholars conceptualize state capacity in the study of civil war. We articulate three mechanisms accounting for the onset of civil wars in areas with higher state capacity.
“Impact of Drone Strikes on Terrorism in Pakistan.” (with Patrick Johnston). International Studies Quarterly. 60(2):203–219. View paper »
Abstract: This study analyzes the effects of US drone strikes on terrorism in Pakistan. We find that drone strikes are associated with decreases in the incidence and lethality of terrorist attacks, as well as decreases in selective targeting of tribal elders. This matters for key ongoing debates. Some suggest that drone strikes anger Muslim populations and that consequent blowback facilitates recruitment and incites Islamist terrorism. Others argue that drone strikes disrupt and degrade terrorist organizations, reducing their ability to conduct attacks. We use detailed data on US drone strikes and terrorism in Pakistan from 2007–2011 to test each theory’s implications. The available data do not enable us to evaluate if drone strikes resulted in increased recruitment, but the data do allow us to examine if these strikes resulted in changes in terrorist activities. While our findings do not suggest long-term effects, the results still lend some credence to the argument that drone strikes, while unpopular, bolster US counterterrorism efforts in Pakistan.
“Insurgent-Population Ties and the Variation in the Trajectory of Peripheral Civil Wars.” Comparative Political Studies. 47(10):1470–1500. View paper »
Abstract: This paper seeks to account for the variation in the trajectory of conflict in the context of peripheral civil wars by emphasizing the importance of rebel-population ties. In examining the trajectory of peripheral civil wars, it expands the range of outcomes traditionally examined in the literature. The empirical validity of the theoretical argument is tested using a new dataset of 166 rebel groups involved in 58 peripheral insurgencies between 1960 and 2010. I trace social embeddedness of rebel groups to pre-war political processes and institutions. Rebel groups are coded on six characteristics to create a seven-point rebel-type scale as a measure of their embeddedness in the population. The results provide support to the argument that the social embeddedness of a rebel group exercises a significant effect on the trajectory of peripheral civil wars. However, the substantive effect of embeddedness on some outcomes is more pronounced than on others.