Prosecuting Domestic Violence without Victim Participation Louise Ellison* This article explores the evidential challenges victim withdrawal presents in domestic violence cases. More specifically it examines innovative measures taken in the United States to overcome problems of proof typically associated with domestic violence prosecutions. These evidentiary initiatives have facilitated a shift towards so-called `victimless' prosecution in the context of domestic violence which dispenses with victim participation. Drawing upon a `freedom model' of criminal justice, this article examines whether recent developments in the United States might be emulated as a means of addressing the high rate of attrition in domestic violence cases in England and Wales. As long as the investigation and the prosecution focus on the victim, cases will be inadequately investigated and prosecutors will fail to win convictions in large numbers of cases. 1 Complainant withdrawal presents prosecutors in England and Wales with formi- dable challenges in domestic violence cases. In one recent study as many as 46 per cent of victims withdrew their support for a prosecution following an initial complaint. 2 Others estimate a significantly higher withdrawal rate. In England and Wales, complainant withdrawal in the context of domestic violence appears to have an almost singular effect; namely, discontinuance. Research to date suggests that police and prosecutors rarely proceed with cases without a co-operative complainant. With regards to police decision-making, Hoyle, for example, found that the wishes of the victim not to proceed were crucial in the decision not to charge, even in very serious cases. She reports that it had become an almost inviolable working rule that victim withdrawal marked the end of a case. 3 Similarly, the main reason for discontinuance by Crown Prosecutors was a victim’s with- drawal of support after a case had been sent for prosecution. Cases without a co- operative victim were classified as ‘unprosecutable’ and ‘unwinnable’. 4 Research by Cretney and Davis paints a similar picture. The researchers found that prosecutors would generally only proceed if the complainant was committed to the prosecution process, even in respect of very serious assaults. 5 They conclude: ‘As far as ‘‘domestic’’ violence . . . [is] concerned, the low conviction rate is a direct result of victims’ inability or unwillingness to make a formal complaint or to give evidence when called upon to do so’. 6 This has remained true following the ß The Modern Law Review Limited 2002 (MLR 65:6, November). Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF and 350 Main Street, Malden, MA 02148, USA. 834 * Lecturer in Law, University of Manchester. I would like to thank Andrew Sanders and Elaine Baker for their valuable comments on an early draft of this article. 1 C. Gwinn, Toward Effective Intervention: Trends in the Criminal Prosecution of Domestic Violence (San Diego City Attorney’s Domestic Violence Unit) <http://www.sandiegodvunit.org/article1.htm>. 2 Crown Prosecution Service Inspectorate, The Inspectorate’s Report on Cases Involving Domestic Violence (London: Crown Prosecution Service Inspectorate, 1998). 3 C. Hoyle, Negotiating Domestic Violence (Oxford: Clarendon Press, 1998) 180. 4 ibid 168. 5 Cretney and Davis report a 52 per cent discontinuance rate in their sample of 205 domestic violence cases. A. Cretney and G. Davis, ‘Prosecuting ‘‘Domestic’’ Assault’ [1996] Criminal Law Review 162 166. 6 A. Cretney and G. Davis, Punishing Violence (London: Routledge, 1995) 144.