We mind and we care but have things changed? Assessment of progress in the reporting, investigating and prosecution of allegations of rape : Two steps forward and two steps back: Rape investigation and prosecution in the wake of the Stern Review

This paper discusses three themes in relation to assessing progress in the Criminal Justice System's (CJS) ability to investigate and successfully prosecute rape. The themes are business as usual, the justice gap and implementation problems. They arose as a consequence of two discussion papers in this volume by Kate Cook and Jan Jordan, who take positions of optimism and pessimism with respect to improvements for rape victim survivors after a series of reforms and changes in police and prosecution procedures. The stimulus for these papers was the latest review in England and Wales into the CJS's approach to rape conducted by Baroness Stern (2010). The present paper observes that the lack of systematic evaluative research makes it difficult to assess whether innovations in the CJS with respect to rape have been effective or to partial out the relative contribution law reform and changing police or prosecution practice may have made to reporting and conviction rates. Finally, it is suggested that intensification of effort to fully implement review recommendations may achieve limited improvement because of the characteristics of an adversarial legal model.

Keywords: rape; rape conviction rates; Attrition

Eeyore, the old grey Donkey, stood by the side of the stream, and looked at himself in the water. "Pathetic," he said. "That's what it is. Pathetic." He turned and walked slowly down the stream for twenty yards, splashed across it, and walked slowly back on the other side. Then he looked at himself in the water again. "As I thought," he said. "No better from this side. But nobody minds. Nobody cares. Pathetic, that's what it is". (Milne, [26], p. 76)

A pessimistic view about the Criminal Justice System's (CJS) approach towards rape would say that, on one hand, despite many efforts to reform the law, reformulate police and prosecuting practices and provide more support for the victim–survivors of rape, nothing changes, and the efforts are pathetic when attempting to deal with a problem that has been called pandemic (Westmarland & Gangoli, [35]). On the other hand, a more optimistic stance lauds that reforms have been made, more support is being given to victim–survivors, reviews are undertaken and academic journals carry papers discussing and debating sexual violence as testament that people both mind and care.

Baroness Vivien Stern was recently commissioned to undertake an independent review into how rape complaints are handled by public authorities in England and Wales (Stern, [32]). Her report is another examination of rape investigation and prosecution, following a familiar pattern to previous similar undertakings: some cause célèbre provoking public disquiet, a media brouhaha and the commissioning of an enquiry. Thus, in 1982 as a consequence of Roger Graef's fly-on-the-wall documentary about Thames Valley Police, which broadcast the oppressive interviewing of a rape victim-survivor, the ensuing outcry resulted in a Sexual Offences Steering Committee being established in 1983 (Lea, Lanvers, & Shaw, [19]) which, together with recommendations from a study by the Women's National Commission, resulted in the issuing of a number of Home Office circulars (25/83 and 69/86) requiring police to revise their procedures. In 1987 the first Sexual Assault Referral Centre (SARC) was established at St Mary's in Manchester, England. Changes were made in English/Welsh law, such as including rape in marriage as an offence and restrictions placed on sexual history evidence. Police training was re-thought, with the introduction of specially trained sexual offence investigation officers (SOITs; see Horvath & Yexley, [14]). Concerns re-emerged at the turn of the century; this time, Her Majesty's Inspectorates of Constabulary (HMIC) and Crown Prosecution Service (HMICPS) undertook a joint review (HMCPSI/HMIC, [12]). The law was further changed (Sexual Offences Act, 2003) and a raft of recommendations included closer cooperation between police and prosecuting authorities and the need for specially trained prosecutors. Baroness Stern's review was provoked by two high-profile investigative failures in the apprehension of serial rapists, John Worboys and Kirk Reid.

Such a cycle is not unique to England and Wales. In her paper, Jordan (this issue) describes the Louise Nicholas case, in which the acquittal of New Zealand police officers accused of rape resulted in government-funded research and the creation of a taskforce for action on sexual violence. Maurice Punch ([28]) details the Dutroux scandal in Belgium, where there were significant police and judicial failures resulting in the prolonged abduction and death of several young children. Two official enquiries followed, calling for reform of the police service.

This cycle of discernible shortcomings in the investigation and unsuccessful prosecution of rape, review of policy and practice and recommendations for reform stimulated the two preceding papers by Kate Cook (this issue) and Jan Jordan (this issue) to ask the question of whether things are actually getting better or worse, using the latest review, i.e. Stern's, as a marker. The present paper draws on the observations made by Cook and Jordan and offers some further commentary about the state of rape investigation, prosecution and convictions. While both preceding authors detect some movement in policy and practice, both are somewhat sanguine about actual change wrought by reforms in law or practice innovation. Jordan takes the more pessimistic, even depressive, Eeyore position that such movement as there has been has possibly resulted in raised expectations but dashed hopes in complainants receiving justice. Thus, the banks of river Jordan are bleak whichever side she is on. Cook is more cautiously optimistic, in that she identifies a "new strand" in Stern, that of victim rights to services that will help them recover and rebuild their lives. This supposes there are other outcomes of value apart from a guilty verdict. What is evident from Stern herself and Cook's paper is that there are improved services and dedicated professionals who care and whose efforts are appreciated by those caught up in the trauma of sexual victimisation; but what they, and Jordan, also say is that these efforts are not dispersed evenly throughout the criminal justice system; indeed they are subject to a postcode lottery in terms of effectiveness of provision which may well be under further threat from the present fiscal crisis besetting the United Kingdom (and other countries).

Both Jordan and Cook discuss the cycle alluded to above whereby some scandal provokes a public reaction and by way of response government or official bodies initiate an enquiry accompanied by recommendations often to reform the law, make changes in investigative or prosecution practice or innovate support services to victim survivors. There are three themes that emerge from Cook and Jordan's papers I wish to draw out and comment upon in this paper. The first is the idea that "it's business as usual" following enquiries or reviews, i.e. professionals are still failing and rates of prosecution remain low. This has been referred to as the "attrition problem" (Lees, [20]). The second is the role and limits of the criminal justice system in securing convictions. This has been referred to as the "justice gap" (Temkin & Krahé, [34]). The third is the notion that it is not the policies that are the problem; rather, the failures are of implementation (Stern, [32]). In addressing whether progress has been made Jordan describes her "frustration at what often appears to be faith-based claims and confidence that any previous faults with the justice system had been fixed". I would agree that there is a fundamental limitation in charting progress by the general lack of systematic evaluative evidence that goes beyond description to provide detailed causative mechanisms to account for success or failure. There are few, if any, randomised control trials (RCTs) held to be the gold standard of evaluation (Farrington & Welsh, [8]); rather, most reviews utilise mixed-methods designs employing a temporal frame of reporting and conviction rates before and after some reform has been implemented (e.g. HMCPSI, [13]; Success Works, [33]). While it is perhaps contentious whether RCTs are appropriate or even possible to evaluate reforms in criminal justice practice, nevertheless there is a shortage of systematic evaluations. Bryant and Loader ([3]), for example, undertook a review of dedicated rape investigation units on behalf of the UK Association of Chief Police Officers (ACPO) but were unable to establish statistically and robustly the reasons for different organisational outcomes as a consequence of the different ways of managing resources to investigate rape. In broad terms, without a reliable evidence base, it is unclear whether any improvements are coincidental to the introduced reforms or due to some other extraneous factors, or if failures can be attributed to poor and patchy implementation or subversion by professionals. As well as this, Cook's point about Baroness Stern's staring position that "the policies are right" is itself problematic, as these were not within her terms of reference and so not subject to evaluation in her review.

Both Jordan and Cook spend some time examining the attrition issue, i.e. dropping-out of cases from the criminal justice system whether this be at the stage of reporting, investigating, prosecuting or convicting. Both make the point that the dropout is large at each stage and does not appear to improve despite the raft of legal reform and changes in investigative and prosecution practices. Reviewing a time trend in numbers of rapes reported, recorded, forwarded to prosecuting authorities and resulting in a guilty verdict, Daly and Bouhours ([5]) find some improvement in rates for England and Wales. The latest available data for England and Wales show that the Home Office recording of sexual crime counted 43,579 serious sexual offences in 2009/10 (Flatley et al., [10]). This represents a 7% increase compared with the 40,748 recorded in 2008/09 and a change in the downturn of previous reporting. More recently, the Metropolitan Police Service (MPS; in London, UK) published data to show an increase in the last 12 months to October 2010 of 3123 recorded rapes (up 32%), with the greatest increase being within intimate and acquaintance relationships. The Sapphire unit responsible for investigating rapes did so in 87% of recorded offences [Metropolitan Police Authority (MPA), [25]]. The increases in reporting were thought to be due in part to tighter compliance with national recording standards. When the Australian State of Victoria undertook a major reform programme to improve the criminal justice effectiveness in its approach to rape, a mixed-method evaluation of the impact of these reforms concluded: "across the reform period, there has been relatively little change in the number of sexual assault victim survivors reporting to the police" (Success Works, [33], p. 40). While it is difficult to compare these data because of scale and methodological differences, and as Cook points out it depends upon what is included and excluded within definitions of reported and recorded cases, nevertheless results are suggesting either no change or some increase in reporting and recording of rape. What we cannot account for fully in these data are whether static or increased rates of reported and recorded rape are due to different definitions of offence, recording practices or changes in procedures. Of course, not all rapes are reported to the police. Kelly, Lovett, and Regan ([16]) compiled a data set of users of Sexual Assault Referral Centres and police data from the United Kingdom. They report a steady increase in younger (under 20 years of age) referrals to just over 40%, with 5% of over-45-year-olds remaining static. While the research evidence does not uniformly report a decrease in attrition at its various key points, where change has been identified in the reporting of rapes, numbers are increasing but we have no clear handle on why this might be or what accounts for fluctuations in the reporting statistics.

If it is the case that more cases are being reported, how does the CJS fare with respect to detecting offenders? For England and Wales between 1997 and 2003/4, Feist et al. ([9]) indicate that detection rates for rape (i.e. charging a suspect) dropped from 76% to 39%, although there was a 97% increase in reported rapes during that same period. Bryant and Loader's ([3]) analysis of all 42 forces in England and Wales found that detection rates ranged from just over 40% to fewer than 20%. Only two forces achieved more than 40% detections and out-performed the lowest performing force by a factor of three. In total, only seven of 42 forces achieved a detection rate of more that 30%. In 2011, the MPS reported that 20% of rapes were detected in the capital (MPA, [25]), whereas the picture from Victoria, Australia was more optimistic, with a reported increase in the number of cases received by the prosecuting authority and the listing of cases, i.e. time taken to complete the investigation, improved considerably (Success Works, [33]). So once again, the evidence from different jurisdictions is not clear-cut. Through her citations, Jordan says that causative factors are "inflated levels of police belief in the prevalence of false rape complaints, combined with high evidential requirements". However, there are other research findings to suggest that police officer attitudes towards rape myths are no different to a sample of students (Brown & King, 1998) and that police officers may now accept fewer rape myths than has been found in previous studies (Page, [27]). Brown, Hamilton, and O'Neill ([1]) report that police officers were not unaware of or indifferent to the trauma experienced by complainants, but that they were trained to assess the evidential quality of the victim's account and tend to "second"-guess probable prosecutor response to case files. The MPA ([25]) report has an interesting footnote: "The MPA and MPS recognise that the use of the word 'survivor' can better reflect the courage and determination of individuals who have experienced violence and abuse than the word 'victim'. For the purpose of this report we use the term victim to reflect someone who has experienced a crime for which the perpetrator is responsible, to reflect the policing context" (p. 6). On one hand, this shows a level of increased awareness and sensitivity, but on the other hand retaining the word "victim" as the preferred term is indicative of the pre-eminence of the police's power in their relationship with the complainant and place rape within a crime discourse.

These research findings do not indicate whether any shift in attitude is associated with greater compliance to procedural innovations. The Success Works evaluation of the Victoria police showed that victim-survivors with previous experience of the system noticed an improvement in their treatment at police stations and that police officers are now more likely to "believe" victim-survivors. The MPA ([25]) also report increased satisfaction rates: 87% of complainants were satisfied or very satisfied with the way they were initially treated, 92% were satisfied or very satisfied with their specialist SOIT officer and 93% were satisfied or very satisfied with the staff at the SARC. Conversely, other researchers find bias and prejudicial decision-making among police officers that responds to stereotypic thinking (McMillan & Thomas, [24]; Temkin & Krahé, [34]).

While the research evidence about the outcomes of changing practices is equivocal, there is no doubt that those practices have developed since the oppressive style of policing rape cases exposed by Roger Graef in 1982. Specialist officers, advances in DNA and SARCs have been actively embraced by the police (Horvath & Yexley, [14]). They cite the example of success in cold-case review work when DNA matches forensic evidence from historic cases. Hits have enabled previously unsolved rapes to be brought to trial and guilty verdicts achieved. The MPA ([25]) indicates that, in all, 1209 cold cases have been reviewed, resulting in 127 full DNA profiles of unknown offenders being loaded into the national DNA data base. Since 2009, 15 men have been charged with rape. Against these successes are the systemic failures pointed out by the Independent Police Complaints Commission when investigating the Metropolitan Police Service's protracted efforts to catch serial rapist John Warboys (IPCC, [15]).

Jordan accepts that the merits of specialisation have been recognised but that training roll-out has been slow and limited, and police cynically "tick the box" to demonstrate fulfillment of government directives. This seems a harsh judgement, especially when there is research evidence to suggest improvements in complainants' satisfaction in how they are dealt with by police (MPA, [25]; Success Works, [33]). However, I would agree with Jordan's conclusion that in the current fiscal climate specialist units are likely to struggle to preserve their existence and funding. I would add a further point. If more victims of sexual assault and rape are to be encouraged to report to the police, then there is likely to be a capacity problem. Could the police cope with an upsurge in reporting under more normal circumstances, let alone when they are facing real cuts in budget? The MPA ([25]) stated that: "It would be misleading to suggest that the financial crisis has brought nothing but positive new ways of tackling violence against women. The reduction in funding for boroughs means services will close. Posts dedicated to domestic violence are already being lost; IDVA/ ISVA and outreach service provision reduced and even refuges closing" (p. 50).

Stern, Jordan and Cook spend some time discussing the relative merits of the percentage calculation used to indicate conviction rate success once cases do get to court. Stern notes that 58% of court cases result in guilty verdicts, a percentage that has been steadily increasing. In the Victoria case, conviction rates remain relatively unchanged. Cook opines that Stern's attack on conviction rates is "silly", but nevertheless the fact that two-thirds of cases getting to court in England and Wales have a chance of success is to be welcomed. Guilty verdicts achieved in the Victoria courts are even higher (69%). Jordan concludes: "we cannot fully honour victims and respect their rights while our so-called justice system fails to hold many perpetrators accountable for their sexual offending". She also suggests that alleged offenders are likely to be acquitted, and further that the justice system is weighted to guard against the risk of wrongful conviction and it is often the victim who is his/her own witness is put under scrutiny. Temkin and Krahé's ([34]) notion of the justice gap is the discrepancy between reports of rape and those that result in convictions. They argue (Temkin & Krahé, [34]) that "victims of sexual assault can expect little by way of redress and assistance ... Despite all the efforts and undoubted improvements over the last thirty years, the rape trial as it is configured in the common law world is frequently not up to the task of delivering justice for rape victims" (p. 209). Jordan reviews the various special measures designed to reduce the negative impact of the trial experience on victim-survivors, but concludes that these have had limited success. She states that reform of legal procedure "will always be compromised in an adversarial system" [emphasis added]. Cook draws attention to the idea that rape complainants should be represented by a lawyer in part to offset the often hostile style of cross-examination. Burman ([4]) summarises the strategies and tactics used by defence counsel within an adversarial system which are designed to discredit witnesses (who are often the victim): asking the same question repeatedly in an increasingly louder voice, with little opportunity for the victim–survivor to respond; use of a hostile tone, suggesting the allegation was false and calling the victim–survivor a "timewaster" and "liar"; testing the victim–survivor's credibility and consistency as a witness. In a further paper, Krahé and Temkin ([17]) conclude that the justice gap in rape cases not only requires improvement but the proposals for doing so need to be evidence-based. They suggest strategies to inform jurors about rape myth prior to trial, and more radical solutions such as abolishing juries and having judge-only trials and preferring female judges to hear cases. They acknowledge these suggestions are outwith the remit of common law legal systems, and in any event await evaluation as to their efficacy. Cook, too, is doubtful that while formal representation of rape victim survivors in court has an instinctive appeal, within the current legal system this is unlikely to be achieved.

Probably the most contentious issue with respect to radical suggestions has been raised by Mary Koss in the United States (Koss, [18]) and Kathleen Daly (Daly & Stubbs, [7]) in Australia. After three decades of research on rape, Koss advocates "expanded justice alternatives", arguing that tightly defined forms of restorative justice should complement criminal prosecution, as it can deliver some of what some victim–survivors want, especially control and voice. Daly and Bouhours ([5]) conclude, in their review of practices in five common law jurisdictions (Australia, Canada, England, Wales, Scotland and the United States), that restorative justice conferences as a supplement to court sentencing or as a pre-sentence activity, post-guilty plea or part of a guilty plea could give victims a greater opportunity to describe the impact of an offence and for others to check the denials of wrongdoing. Again, many people do not like this idea, believing that it decriminalises rape, is justice on the cheap and that it is impractical (Daly, 2011).

Cutting the Gordian knot that binds rape into a criminal justice process, which may have inherent limitations because of its adversarial nature and presumptions of innocence, will necessarily have to be undertaken by the bold. It seems highly unlikely that there is the political will to wield the scissors for the innovation suggested by restorative justice alternatives or non jury trials.

Both previous authors, particularly Jordan, suggest that implementation of recommendations is both limited and incomplete. While there has been a great deal of problem description and analysis, as witnessed by the plethora of reports and policy innovation in the form of law reform and changes in practice, sexual violence remains under-reported, under-investigated and under-prosecuted, with patchy service provision in the form of SARCs or health care. Implementation failure is one of three suggested reasons why reforms fail (Lewis & Greene, [21]), the other two being programme over expectation and conceptual failure. The latter occurs when a reform project is based on inaccurate or incomplete theory of causation. Cook's position argues "that [to say] policies are right is fairly stunning". The preceding argument demonstrates that research evaluations are limited in their explanatory and causative powers, and findings are equivocal. In short, we do not know whether the policies are right or have been operationalised appropriately. Rein ([29]) suggests that recommendations frequently call for better coordination between the relevant constituent parties in a change programme. He points out that different stakeholders have different reasons for improved communication: efficiency, accessibility, advocacy and/or greater participation. Often, in attempting to operationalise the recommended communication improvement, there is a lack of clarity about purpose and outcome contributing to what Lewis and Greene ([21]) identify as reform over expectation. It is this heightened expectation from victim-survivors that Jordan complains about. Rein ([29]) adds that implementation requires practitioners to change their daily behaviour and Lewis and Greene ([21]) further propose that implementation requires changes in the internal working environment of professionals. The HMIC/CPS review referred to above and the work of the IPCC into failed police enquiries discussed earlier show that there is evidence for a mindset among some criminal justice professionals that still disbelieves women when reporting allegations of sexual violence under circumstances they, police officers and prosecutors find sufficiently unconvincing, such that they decline to proceed with an investigation or prosecution. In her discussion, Jordan poses the question as to whether failure to take reports of rape seriously is attributable to a systemic culture of scepticism within the police or an individual "rotten apple". Punch ([28]) suspects that the rotten apple metaphor is part of a rationalisation by "beleaguered" institutions when they are defending themselves and that the problem is more probably a diseased orchard. In a similar vein, Shiner ([30]) discusses the MPS's response to the charge of institutional racism made by Lord MacPherson in his inquiry when conducting a review of the failed murder investigation of the black teenager Stephen Lawrence. Shiner suggests that police officers were personally affronted by the charge, not seeing the wider institutional processes that contributed to discrimination. As a consequence, he suggests, they engaged in a series of denial or displacement strategies to avoid serious critical analysis of systems and processes or how to remedy the problem of racism. Such an analysis might be applied to the police's response to recommendations concerning sexual violence. Examination of the IPCC's findings in relation to failed investigations reveals examples of the police declining to believe women's allegations in preference to the accounts provided by male suspects. Recent publicity surrounding the behaviour of the MPS's Sapphire officers (a unit dedicated to the investigation of sexual crime) shows some interference with women complainants' statements and a subversion of process (London Evening Standard, 2010). The MPS has since re-organised the central department responsible for the investigation of sexual offences (MPA, [25]).

Both Jordan and Cook suggest that the police resist change. However, as documented by Horvath and Yexley ([14]), the MPS has initiated change in rape investigative practice such as SOITs, more women forensic examiners and specialist rape suites. Newer policies emphasise that in the present constrained financial landscape more partnership working and pooling of resources will be necessary and the offering more holistic resources (MPA, [25]). Certainly there is evidence for patchy implementation, documented in Stern's review, but we simply lack the evaluative research to say whether it is individual resistance, systemic problems or incomplete implementation that results in either specific scandalous failures or limited increases in reporting or conviction rates.

It is certainly reasonable to ask the question as to whether progress has been achieved in preventing sexual violence, permitting victim-survivors to report rape, have their allegations investigated and presented at court and the securing of guilty verdicts. Both Cook and Jordan observe movement by the criminal justice system in being more responsive to complainants. Cook particularly welcomes the shift of emphasis discernable in Stern, that rebuilding a life shattered by sexual violence is a legitimate focus of CJS activity. Jordan is more pessimistic about the reality and sustainability of changed outcomes. In particular, she notes that offenders are more likely to be acquitted once a case gets to court. While it is true that rape remains a largely unreported crime and many cases drop out of the criminal justice process, certainly in England and Wales (and evidence from other jurisdictions such as Victoria, Australia), if cases get to court there is a greater than 50% chance of a guilty verdict. It is evident that there are cycles of outrage, inquiry and reform when a particularly horrific rape becomes the subject of media comment. Stern notes that there is no lack of policies, but she locates the limited improvements in reform as implementation failures. In the absence of well-designed evaluative studies, we lack the evidence base to conclude whether either the policies are right, their operationalisation appropriate or wayward CJS professionals subvert efforts to change. It is also virtually impossible to identify the mechanisms to account for success or failures—so it is difficult to answer the question definitively.

Cook draws our attention to the feminist project of ending rape as the ultimate reform of the CJS in making its procedures redundant. This as an aspiration is a noble cause, but as Westmarland and Gangoli ([35]) state: "it is a depressing reality that rape still remains internationally prevalent in the 21st century. Despite many advances in gender equality, male violence against women continues at a level described by WHO as pandemic" (p. 11). The history of crime prevention suggests that the idea of eliminating crime through reducing opportunity (e.g. target hardening) was unrealistic, and political realism changed tack to a more modest crime reduction programme. This, too, became a political hot potato as politicians of various hues were criticised if crime rates increased. There then came the move to focus on the potential criminal, especially the young, and the idea of criminality interruption through diverting young people into more productive activity. Hallsworth and Lea ([11]) discuss pre-emptive criminality whereby stop-and-search, CCTV and intrusive surveillance is designed to inhibit criminal activity through heightening the likelihood of detection. None of this addresses the deeper structural and cultural causes of crime, and these are especially pertinent to the underlying inequalities between men and women and the commodification of sex. Westmarland and Gangali ([35]) say we have little choice but to continue the project in order to free the world from sexual violence; but the debate engaged between Cook and Jordan does not resolve where to prioritise or target our best efforts. Punch ([28], p. 194) concludes that "the more systemic [CJS] deviance [is] the more profound and far reaching the changes have to be". He says that the Dutroux scandal was of such far-reaching proportions in Belgium that it required change at societal and political levels, but given the evaporation of public outrage driving reform he was of the view that such fundamental changes were unlikely. Both Cook and Jordan also call for more profound change if justice professionals are to close the justice gap.

Is it reasonable to continue trying to alter or tighten procedures or changing the law to achieve higher reporting and conviction rates? There is an argument that an adversarial system with its presumption of innocence and beyond reasonable doubt test of evidence is too high a bar and too blunt an instrument to deal with the complexities of the range and circumstances in which sexual violence occurs and between whom in order to assign guilt in a court of law with its attendant checks and balances. Spohn and Horney ([31]) are of the view that it is not so much changing an individual officer that will reform practice, but change is required at the level of the organisation's ethos. If this is so, then simply to go on amending law, tightening adherence to new practices or resource full implementation of recommendations is unlikely to close the justice gap or achieve the feminist project of eliminating rape. Other, and possibly better, outcomes may be possible through controversial radical alternatives such as restorative justice conferencing but which itself is hotly contested. In the meantime, as Westmarland and Gangoli ([35]) argue, the problems of rape are intractable, but that should not stop us trying to cut the Gordian knot.

Thanks to Jan Jordan and Kate Cook for engaging so fully in this debate, Miranda Horvath for her editorial direction and Kathy Daly for reading and commenting on an earlier draft of this paper.