A brief review of institutional riots and disturbances reveals not only a cyclical pattern of periodic flareups every decade or so, but also a steady increase in occurrences of disruptive behavior and a change in the seriousness of riots in terms of more severe personal injuries and loss of life. The cumulative evidence of empirical studies points to a combination of at least the following preconditions that may be necessary for riots to occur: (1) internal pressures on inmates resulting from racial and political tensions, the fragmentation of the inmate infrastructure, and power shifts among inmates; (2) staff conflict; and (3) external pressures brought forth by promises of reform, publicity, raised expectations, and official visits. The model for riot prevention and control presented here is based on the theory first formulated by Smelser (1962) and subsequently applied to the correctional setting by Glaser (1972). Smelser's six principal determinants of collective behavior have been modified, and indices for the preconditions of violence are developed by drawing from a wide body of knowledge in the fields of corrections, social control, and conflict management. The six determinants of collective violence are structural conduciveness, strain, generalized belief, precipitating factors, mobilization and organization for action, and the operation of mechanisms of social control. A method is delineated for identifying sources of strain, dissatisfaction, and arbitrariness underlying many of the problems of the prison system, and strategies are suggested for the impartial resolution of inmate grievances. About 90 references are included.