|Donald Black's (1993, 1998) theory of conflict management suggests that the social structure of a conflict, comprised of the social statuses and ties of the principal parties and third parties, predicts whether it will be handled violently. Focusing on the principal parties, Black argues that five conditions increase the probability of violence: relational distance, functional independence, immobility, equality, and cultural distance. Although Black's theory is built on a wealth of evidence, it has not been tested. To test Black's theory, I conducted interviews with 102 men incarcerated for violent conflict management (assault and homicide) committed against another man. Each interview addressed a pair of conflicts: the crime and a similar conflict the respondent handled non-violently. If, for example, a drug debt led to the crime then the respondent also describes a drug debt he handled non-violently. Because the respondent must choose a matching non-violent conflict that occurred during the same time period, the design controls for individual and ecological influences outside the theoretical model; if two conflicts occur around the same time then the respondent tends to have the same individual characteristics and live in the same neighborhood, meaning such influences are a constant for the pair. The interview produces quantitative and qualitative data. Conditional logistic regression is used to examine the quantitative data (CLR is appropriate for matched case-control studies because it avoids the problem of autocorrelation). The qualitative data offer context and nuance. In general, the results offer support for the theory.
(Return to Program Resources)