We address the role of victim cooperation in the prosecution of domestic violence cases in a specialized court in Toronto, Canada. We first examine what factors predict whether a case will proceed to prosecution. We find that, even in a court designed to minimize reliance on victim cooperation through the use of other types of evidence, when prosecutors perceive a victim to be cooperative, the odds that a case will be prosecuted are seven times higher than if a victim is not perceived to be cooperative. In the second part of our analysis, where we seek to determine the correlates of victim cooperation, we find that the two most important determinants of victim cooperation are the availability of videotaped testimony and meetings between victims and victim/witness assistance workers. We discuss the implications of these findings for future research and policy.The last several decades have been marked by careful scrutiny of criminal justice responses to domestic violence. Considerable research has examined both the role of police discretion in responding to domestic violence calls and the effect of arrest on recidivism rates (Fagan 1995;Sherman and Berk 1984; for review of arrest studies, see Fagan et al. 1995;Garner, Fagan, and Maxwell 1995). This attention to police practices, however, has not been paralleled by * We thank the three anonymous Justice Quarterly reviewers for their helpful comments,