The most serious fragmentation of research on violence has occurred between explanations of violence that is socially defined as illegal and unacceptable (such as parents beating their children) and violence that is considered legal and acceptable (such as parents spanking a child). The popular definition of violence connotes evil, but a neutral definition used by conflict theorists would be a struggle to maintain, change, or protest asymmetric social relations governing the distribution of scarce resources by the threat or exertion of physical force. Most approaches suggest that violence is caused by malfunctioning or deficit states, but when it is characterized as goalseeking and rational behavior, normal social conditions can be explored as possible causes. Several studies on collective behavior are reviewed which debunk the idea that collective violence is outside the realm of normal behavior and therefore beyond the scope of traditional social science. These analysts assume that the incidence of violence decreases with interest in class position. This view is also compatible with sociobiological concepts of aggression. An explanation of social violence is presented through a chain of events that begins with unequal distribution of resources and proceeds through asymmetric social relations, nonviolent conflict, and violent conflict, first as normative violence with deviant violence as a last resort. Nonviolent forms of conflict usually precede violence and can serve as useful predictors. The behavior aspects of conflicts between different social levels are also discussed. The expression of violent behavior may be limited by social or legal proscriptions, insufficient organization and resources, and explicit intimidation. Responses to blocked expression of violence include pooling of resources, substituted violence, and scapegoating. These theories are illustrated by analyses of parental violence and differential rates of male and female violence. This approach challenges both class-based and sex-based explanations of violence, but tries to locate the common roots of social violence in the everyday experiences of inequality. Approximately 90 references and 15 footnotes accompany the paper.