Threats of violence, physical assault, stalking, and danger were measured by interview questionnaires at the time of the filing of charges and at 3 and 6 months later. Among the 90 women, 48 percent had insufficient evidence for charges; of the 52 percent making charges, 11 percent eventually dropped the charges; 37 percent of the perpetrators were arrested; and 4 percent remained fugitives. With the exception of danger at 3 months for one group of women, levels of violence did not differ by charges accepted or perpetrator arrested as measured at 3 and 6 months. The act of women attempting to file assault charges, whether the charges were accepted or the perpetrator arrested, resulted in equally lower levels of future violence. These findings support the experiments conducted in Omaha, Neb. (Dunford, Huizinga, and Elliott, 1990); Charlotte, N.C. (Hirschel, Huthinson, and Dean, 1992); Milwaukee, Wis. (Sherman et al., 1992); and Colorado Springs, Col. (Berk et al., 1992). These studies found no significant reduction in the prevalence of new violence following on-scene arrest of offenders. The findings in the current study do show that for this cohort of 90 women, the act of filing assault charges against an intimate partner was a powerful deterrent to future violence, whether or not the charges were accepted or the offender was arrested. Perhaps abused women acting in partnership with service agencies presents the greatest deterrent to future violence against them. 3 tables and 29 references