Other recent research: Nation-state formation | Ethnic boundary making | Boundaries in social networks

Ethnic exclusion and wars


From empire to nation-state: Explaining wars in the modern world, 1816-2001

In my previous comparative historical work I have tried to show how and why the transition to the modern nation-state politicizes ethnic dividing lines and transforms ethnicity and nationhood into major principles of inclusion and exclusion in various institutional fields. In this project I wanted to demonstrate that this institutional transformation has also generated many of the violent conflicts of the modern world. The existing quantitative literature on war takes the independent nation-state as the self-evident unit of analysis and largely excludes other political types from consideration. In contrast, I argue that the introduction of the nation-state form is itself a major cause for war. To test this hypothesis quantitatively, I have assembled, together with former political science graduate student Brian Min, a new data set that records the outbreak of war in fixed geographic territories from 1816 to 2001, independent of the political entity in control of a territory. The results of this study are published as an article in the American Sociological Review, which won the best article awards both of the Comparative Historical and the Political Sociology Sections of the American Sociological Association, and as chapters in two edited volumes. The war dataset as well as the replication dataset including documents describing coding rules are available from the data sets page. A separate article introducing the war dataset has been published by International Interactions. It is also co-authored with Brian Min.


Ethnic politics and armed conflict, 1946-2001

According to the approach developed above, the major causal mechanism linking nation-state formation to war is the delegitimization of ethnic hierarchies and the subsequent ethno-political mobilization against political exclusion. Unfortunately, no direct measurement of political exclusion is available. The existing indices measure ethno-demographic diversity without taking power relations into account, or they track the political status of only a subset of ethnic groups, such as in the Minorities at Risk dataset. Together with Lars-Erik Cederman from the ETH Zurich and UCLA's Brian Min, we have assembled a new dataset on ethnic power relations in all countries since the Second World War. The Ethnic Power Relations dataset is now available through the EPR website.

The analysis of this dataset have led me to move toward a more complex, disaggregated approach to understanding ethnic conflict. A first paper, co-authored with Lars-Erik Cederman and Brian Min, has been published by the American Sociological Review in 2009. It addresses the quantiative scholarship on civil wars, which has long debated whether or not ethnic diversity breeds armed conflict. We go beyond this debate and show that it is not ethno-demographic diversity as such but three ethno-political configurations of power that are associated with violent conflict: First, states that exclude large portions of the population on the basis of their ethnic background are likely to be challenged by armed rebellions. Second, segmented states where power is shared between a large number of competing ethnic elites risk violent infighting between them. Third, incohesive states with a short history of direct rule are more likely to experience secessionist conflicts. We test these hypotheses with EPR dataset which records ethnic power relations in all countries since 1945. Cross-national analysis shows that once properly conceived and measured, ethnic politics is as powerful and robust in predicting civil wars as is a country’s level of economic development. We then show through multinomial logit regressions that rebellion, infighting, and secession result from high degrees of exclusion, segmentation and incohesion respectively. More diverse countries, on the other hand, are not more likely to suffer from violent conflict. An appendix describing the dataset and offering further analysis is available here.

A second paper, first-authored by Lars-Erik Cederman, disaggregates the analysis further by taking politically relevant ethnic groups as units of analysis. Consistent with the above analysis, our findings indicate that representatives of ethnic groups are more likely to initiate conflict with the government (1) the more excluded from state power they are, especially if they have recently lost power, (2) the higher their mobilizational capacity, and (3) if they have experienced conflict in the past. This paper has appeared in World Politics and is available here.