Lynx users, please scroll past our index to get to major content.

Katerina Bezrukova

Home Page

Curriculum Vitae

 Statistics 40
 Diversity 156
 I-O Psychology 157
 Advanced I-O Psyc 161

Research Projects
 Group Faultlines



Katerina Bezrukova, Psychology Department
Santa Clara University

The Impact of Violent and Nonviolent Action on the Evolution of Group Faultlines

Research Team:

Katerina Bezrukova, Chester Spell, & Spencer Graves

When people are killed and property destroyed, the apparent perpetrators often make enemies. This seemingly obvious observation is commonly forgotten as individuals in one group move to defend themselves and people with whom they identify from apparently inappropriate violence from others. The violence of each side tends to strengthen with-group identification and the salience and stability of existing faultlines.1 Nonviolent action generally seems to have the opposite effect in destabilizing faultlines, reducing cohesion within the groups on each side, allowing other issues and group identities to become salient and fostering movement of faultines.2

If people understand these phenomena better, they will tend to make better choices about when violence can be used productively and when it is more likely to be counterproductive by strengthening their opponents. We plan to use text mining and content analysis of data on Twitter, Facebook, YouTube and elsewhere to trace changes in the level of commitment of individuals to alternative group identities in response to violent and nonviolent events. We hope thereby to produce model the probability distribution of changes in the strength and location of faultlines resulting from alternative actions. Collateral damage may be the single greatest contributor to the outcome of most wars. If so, an empirically validated quantitative theory of this nature could help reduce the counterproductive activities of all sides in conflict.

1Lau, D., & Murnighan, J. K. (1998) "Demographic diversity and faultlines: The compositional dynamics of organizational groups", Academy of Management Review, 23, 325-340.

2Sharp said that violence tends to concentrate power and nonviolent action tends to diffuse it. Gene Sharp, Politics of Nonviolent Action, esp. vol. 3, pp. 799-806. See also Seymour Martin Lipset, Continental Divide: The Values and Institutions of the United States and Canada (NY: Routledge, 1990, p. 109). Lipset said, “The emphasis on due process, free speech, and other rights of individuals as superseding the needs of the state for the maintenance of order is severely challenged during wartime. Threats to the very existence of the nation lead governments everywhere to suspend or ignore legal rights. America and Canada have not been exceptions. In the United States, habeas corpus was suspended during the Civil War; free speech was denied opponents of the conflict in World War I, and many, particularly socialists and other radicals, were imprisoned; citizens of Japanese ancestry were placed in concentration camps on both sides of the 40th parallel for no other offense than their national origin; and a wave of intolerance against Communists and other leftists accused of complicity with them, which entered history under the generic name of McCarthyism, swept the United States during the Korean War, a conflict with two Communists states. The courts did little to restrain the authorities, thus revealing the fragility of legal guarantees in the face of wartime hysteria.” See also Michel Gourd, “Quand l’Arbitraire Policier s’impose au Canada: Au Nom de la Lutte contre le Terrorisme”, Le Monde Diplomatique (Feb. 2005): 12-13. See also "The Relative Effectiveness of Violence & Nonviolence" at "".

A Field Study in Ethnopolitical Conflict Areas

We plan to conduct a field study to test how the basic faultline model maps on to actual situations of ethnopolitical conflict. This study will include three natural experiments carried out in Crimea, Bosnia, Sri Lanka, and Burundi. The field study will be a conservative test of our model in that we will measure ethnic faultlines on two dimensions, race and national heritage, to first parallel the lab study while controlling for class, religion, gender and other potentially influential demographic variables (retaining the data for future analysis of a more complex construct of faultlines including more demographic alignment variables).  In all cases we will measure the levels of conflict as our dependent variable of interest.

Experimental Design Summary Table


Exp 1.

High Group Identity


Low Group Identity

Exp 2.

Strong Dual Leadership


No Strong Dual Leadership

Exp 3.

Strong Coalition Formation


Weak Coalition Formation


Sri Lanka


Sri Lanka


Sri Lanka




Bosnia, Burundi