The subculture-of-violence theory holds that cultural values and social conditions rather than individual biological or psychological factors are significant causes of violent behavior. In the criminology and delinquency literature, a subculture-of-violence thesis has been used to explain social-structural causes of violence in urban areas. The general model of such a subculture is one characterized by "dense concentrations of socioeconomically disadvantaged persons with few legitimate avenues of social mobility, lucrative illegal markets for forbidden goods and services, a value system that rewards only survival and material success, and private enforcement of the informal rules of the game" (Gottesman and Brown, 1999, p. 297). Within this context, it has been well-documented that women sex workers on urban streets are typically at high risk for assault, rape, and other forms of physical violence. The perpetrators include muggers, serial predators, drug dealers, pimps, police, and customers. Because of the authors' long-term involvement in urban street studies in Miami, they were able to recruit six focus groups composed of eligible women sex workers, with an average of four participants per group. To be eligible for the study, the women had to be ages 16 to 49, to have traded sex for money or drugs at least three times a week in the past 30 days, and have used heroin and/or cocaine three or more times a week in the past 30 days. Nearly half of the respondents reported physical and/or sexual abuse as children, and over 40 percent had experienced violence from clients in the prior year; 24.9 percent were beaten, 12.9 percent were raped, and 13.8 percent were threatened with weapons. The consistent relationships between historical and current violent victimization suggest that female sex workers experience a continuing cycle of violence throughout their lives. Findings thus support the concept that female sex workers are enmeshed in a social milieu in which violence is common and victimization is expected. Policy and research implications of these findings are discussed. 5 tables and 59 references