Understanding conflict at the highly local level, analyzing the behavior of combatants and perpetrators, and evaluating the specific mechanisms that generate conflict and other forms of political violence — these three aims motivate my research. To achieve them, I study domestic conflict and political violence, with a specific focus on using political, economic, and environmental factors to explain and predict violence inside and outside of conflict settings. As a researcher I devote considerable energy into developing “localized conflict” approaches to understanding manifestations of violence traditionally analyzed at the country level.
International relations: political violence; environmental conflict; civil war; the geography of violence
Methodology: computational social science; mixed distribution models; causal inference; mixed methods; event data
Mukherjee, Bumba and Ore Koren. Winning and Losing Civil Wars. Under contract at Oxford University Press.
When are paramilitaries and militias more likely to demobilize, disarm and re-integrate into the political spheres? If they demobilize, do these groups join existing political parties or form their own in order to participate in the electoral process? Conversely, under what conditions do paramilitaries or militias opt to not disarm and integrate into the political process, choosing instead to continue their violent and criminal activities and undermine the state? Finally, when will these activities lead to the renewal of civil war? Answering these questions is more than just an academic exercise; the answers can provide new, important explanations for when peace-building works, and when civil war recurs. These, in turn, can inform our current understanding of post-war politics that emphasize transitional justice, peacekeeping activities, and democratic and market integration.
Paramilitaries, Militias, and Political Demobilization after Civil War provides a thorough, meticulous analysis of the role that paramilitaries and militias play after the war ends. Whereas ongoing research into these groups focuses primarily on their behavior during conflict, this book explains the strategic choices made by these groups – to disarm and join the political process, continue their criminal activities, or actively attack the state – and shows that these choices have a crucial impact on the success of peace building processes in post-civil war situations.
Peer-Reviewed Conference Proceedings
Bagozzi, Benjamin E., and Ore Koren. 2017. “Using Machine Learning Methods to Identify Atrocity Perpetrators.” Proceedings of the IEEE Big Data 2017 Conference.
“Big data” on atrocities events are now widely analyzed in the social sciences. Unfortunately, these data often contain incomplete information on the identities of atrocity perpetrators. This study addresses this deficiency by developing a machine learning approach for the accurate recovery of unknown perpetrator identities within existent atrocities datasets. In doing so, it demonstrates how to transform and standardize a large number of auxiliary variables into text-compatible data. It next shows how to leverage this information to train a series of classifiers on observed atrocities data. After identifying the ideal set of machine learning algorithms and evaluating their performance in this context, this study then uses an ensemble of the best performing algorithms to classify all unknown atrocity perpetrators included within a prominent atrocities dataset, validating the results with external data from the Iraq conflict.
Other Conference Proceedings
Koren, Ore. “Conflict Prevention in Isolated World Ship Societies.” Proceedings of the Tennessee Valley Interstellar Workshop.
Scholars have long argued that so-called “closed societies” frequently experience severe violence perpetrated for political motivations. Living in socioeconomic and political vacuum aboard large world ships – without inter-societal migration, external penetration of ideas, conflict management and peacebuilding by external parties, or even regular communications with the outside world – interstellar societies will be especially vulnerable to internal conflict. In such contexts, enmities can easily fester, nonviolent quarrels can escalate into violent conflict and political disagreement can deteriorate into mass killing of opposition members. Building on recent research, this paper identifies specific factors that could engender conflict on interstellar world ships to offer strategies of mitigation and prevention. The paper begins by discussing four types of conflict on Earth that are especially likely in the context of world ships and listing their relevant socioeconomic and political causes. These pathways are then articulated to operationalize empirical models of violence. The strongest predictors are identified empirically by statistically analyzing large datasets of different conflict types on Earth over the last half century. To do so, computer simulations are first applied to out-of-sample data, i.e., data not used to compile the original models, to cross-validate the relevant strength of the predictors identified in the previous stage. This process is repeated twice, first on a sample of all countries globally for the years 1961–2011 and then again on a sample consisting solely of countries with very high population densities – societies that resemble those most likely to exist on world ships. Finally, literature on conflict mitigation is applied to pinpoint strategies to address causes of conflict identified as having an especially strong impact as societal density increases, including political restrictions, economic inequalities, ethnic divisions and limitations on access to food. By applying lessons from earth societies to interstellar travel, this paper will inform the creation of a sustainable, peaceful governance system for future on-board colonies.
Koren, Ore, and Bumba Mukherjee. “Violent Repression as a Commitment Problem: Urbanization, Food Shortages, and Civilian Killings under Authoritarian Regimes.” HiCN Working Paper 296.
Authoritarian regimes frequently commit systematic killings of their own subjects, yet the mechanisms governing this behavioral shift remain unclear. We address this puzzle by developing a formal model that shows authoritarian elites perpetrate systematic killing campaigns preemptively in response to an exogenous shock where urban development levels are sufficiently high. In these contexts, the civilians cannot commit not to mobilize and pose a credible threat to the regime, which often preempts these efforts using systematic killings. Statistical analyses of a global high-resolution sample within all authoritarian states between 1996 and 2008 confirm the model’s predictions. This study thus explicates when elites would resort to systematic killing as a rationalist strategy, and identifies an important dynamic that explains geographical and temporal variations in systematic killings within authoritarian states.
Koren, Ore, and Laura Mann. “Nighttime Light, Superlinear Growth, and Economic Inequalities at the Country Level.”
Research has highlighted relationships between size and scaled growth across a large variety of biological and social organisms, ranging from bacteria, through animals and plants, to cities an companies. Yet, heretofore, identifying a similar relationship at the country level has proven challenging. One reason is that, unlike the former, countries have predefined borders, which limit their ability to grow “organically.” This paper addresses this issue by identifying and validating an effective measure of organic growth at the country level: nighttime light emissions, which serve as a proxy of energy allocations where more productive activity takes place. This indicator is compared to population size to illustrate that while nighttime light emissions are associated with superlinear growth, population size at the country level is associated with sublinear growth. These relationships and their implications for economic inequalities are then explored using high-resolution geospatial datasets spanning the last three decades.
Koren, Ore. “Hunger Games: Food Security and Strategic Preemptive Conflict.” HiCN Working Paper 253.
A growing number of studies draw linkages between violent conflict and food scarcities. Yet, evidence suggests that at the subnational level conflict is likely to revolve around food resources abundance. In focusing on conflict waged by groups to prevent their rivals from securing food resources, this paper offers a theory to understand the relationship between food security and violent conflict. I develop a formal model that incorporates three actors: civilian producers who grow crops, raiders, and defense forces. Equilibrium and comparative static results show that violent conflict is more likely in regions with an abundance of food resources. The model is validated at the subnational level using new high specificity spatial data on staple crop production for the years 1998-2008, and used to forecast conflict for 2009-2010. In line with theoretical expectations, food resources have a positive and statistically significant effect on the strategic behaviors of different actors.
Koren, Ore. “Reparations for State Crimes as a Diffusive Norm.”
This study applies extant theories on the diffusion of human rights to reparations for mass killing, and tests these theories using modeling techniques that account for the diffusive nature of reparations. I argue that reparations for state crimes are a rare, diffusive event, and that in order to understand their spread one must account for the conditional relationship between where and when diffusive processes are more likely. Drawing on extant theories of international policy diffusion and international law, I derive testable hypotheses regarding the diffusion of reparations for state crimes. These hypotheses are then tested on newly available data on reparations for the years 1971-2011 using Bayesian Weibull models that account for the underlying differences affecting the baseline likelihood and baseline hazard, respectively, of reparations, based on theoretical expectations. Primarily, I find that regimes with less political rights are significantly less likely to provide reparations. In addition, evidence from these theory specific hierarchical models suggests that more international reparations precedents are associated with increased time until reparations, whereas GATT/WTO membership and perpetrating regimes are associated with decreased time until reparations. These models also suggest that reparations become more likely over time, and can be used to identify countries that are inherently more likely to adopt them.