Universität KonstanzExzellenzcluster: Kulturelle Grundlagen von Integration

Inequality, Identity and Conflict

The Ethnic and Socioeconomic Dimensions of Internal Political Violence

Prof. Dr. Gerald Schneider, Dr. Margit Bussmann, Lilli Banholzer


This project will examine how the internal social potential of violence is influenced by the mutual interaction of ethnic and socioeconomic lines of conflict. In the process, we will also be testing the empirical content of classical models of political sociology as well as that of newer approaches. Our comparative study touches upon three explanatory levels:

  1. Using macro-quantitative approaches, we will evaluate the extent to which the interaction between group identity, income inequality within a group and overall social inequality have an effect on the potential for violence.
  2. An examination of the World Value Surveys should clarify the interaction of the effects of socioeconomic status and group identification on political extremism, using comparisons between countries as well as individuals surveyed.
  3. Using field studies of selected Rwandan municipalities, we wish to explain which income levels were overproportionately represented among the various parties involved in the civil war.

Once we have worked out the precise relations between these effects, the project consists more concretely of three partial projects:

  1. A macro-quantitative study of the relationship between group identification, income inequality and the use of political violence: Studies which compare countries with respect to the influence of inequality on the risk of civil war or other forms of violence have been incomplete primarily. This is due to the fact that data has been collected on income inequality in nations but not in groups. The basis for these new measurements of income inequality will be surveys such as the Latino and Euro Barometers or the World Value Surveys. To measure the potential for violence we consider a variety of indicators: civil wars, revolution, state collapse, strikes, political assassinations, etc. According to an instrumentalized understanding of political violence, the forms of violence are for the most part substitutes that actors choose between in order to achieve their political goals. While using the influences on internal political conflicts identified in the research literature as a control, we will use a panel-analysis to assess the effects of economic inequality and ethnic factionalism as separate factors and above all as interactive effects.
  2. Individual data analysis of the individual propensity to violence: With assistance from survey data, it can be measured to what extent an individual identifies with a group, how legitimate he/she views political violence to be, and to what extent he/she is critically inclined vis-a-vis other groups. We want to conduct such a parallel study employing the usual approaches to data. On the one hand, there is the well-known World Value Survey; on the other, there are similarly regional surveys such as the Euro Barometer, the Latino Barometer, the Asia Barometer, and the Africa Barometer. In order to test the hypothesis of the influence of income inequality and group identification on the individual propensity for non-parliamentary politics, we wish to evaluate multilevel models that permit a parallel assessment of individual and national effects.
  3. Comparison of Rwandan Municipalities: Since the survey data allow one to measure attitudes but not actual behavior, we wish to attempt to do so by means of field studies in one conflict region—here Rwanda—on the basis of surveys on location as well as making determinations of the socioeconomic and ethnic backgrounds of victims and perpetrators.