CQ No. 25 SA Crime Quarterly no 32 • June 2010 19 This article describes a summary of key findings from a study undertaken in Gauteng Province that tracked the processing of rape complaints, as well as their outcome, from the time such complaints were reported in 2003, to the point at which they were disposed of, either by the police or the courts. This exploration of case attrition (the dropping or filtering of cases from the criminal justice system prior to a trial's conclusion) not only establishes the measure of justice afforded rape complainants, but also provides an insight into how and why justice may be eroded. STUDY METHODS The study was undertaken in Gauteng where, in 2003, a total of 11 926 rapes were reported at the province's 128 police stations. A two-stage procedure was developed to draw the sample. The first stage drew a sample of 70 police stations, using probability proportional to size, where size was based on the number of rape cases reported to the police in 2003. Within each police station all the closed rape cases for the year were identified by their Crime Administration System (CAS) numbers, and a random sample of 30 dockets was selected (or all cases if fewer than 30 cases were reported in that year to the sampled police station). The dockets were then located either at the police station or at the specialist Family Violence Child Protection and Sexual Offences (FCS) Units, and data extracted. There was no replacement of dockets that were unavailable. This procedure provided a sample of 2 068 c ases for the study. Records for those cases that went to court were obtained from both High Courts in the province, as well as all 30 magistrates' courts. In 2003, the year chosen for the study, the common law definition of rape was still in effect (it has now been replaced by the statutory definition contained in the new Sexual Offences Act of 2007). In terms of the common law definition, only women and girls could be victims of rape and only men and boys be accused of the crime. Consequently, in this report we refer to victims as female and perpetrators as male. Worth their while? Pursuing a rape complaint through the criminal justice system lisa@tlac.org.za If the criminal justice system were to advertise itself to victims of sexual violence, would its tag line be L'Oreal's 'Because you're worth it'? Or would the Foundry Premium Cider comment on the building of a pyramid: 'Twelve years in the making and it still doesn't taste very good' be more apposite? Certainly, a lack of faith in, or fear of, legal processes, is one of the reasons why eight out of nine women who are raped do not report the matter to the police.1 This article explores the extent to which victims' lack of faith in legal processes is warranted and asks if laying a complaint is worth the effort. Lisa Vetten, Rachel Jewkes, Romi Sigsworth, Nicola Christofides, Lizle Loots and Olivia Dunseith* * Lisa Vetten and Olivia Dunseith, Tshwaranang Legal Advocacy Centre to End Violence Against Women. Rachel Jewkes, Lizle Loots and Nicola Christofides, Gender and Health Research Unit, Medical Research Council. Romi Sigsworth, Gender Programme, Centre for the Study of Violence and Reconciliation. SELECTED CHARACTERISTICS OF REPORTED RAPES A total of 2 068 individual victims' dockets was included in our sample. Four of these cases however, were little more than shells and consisted only of a CAS number and an indication that the charge was one of rape. They were excluded from the analysis, which focused on the remaining 2 064 c ases. Twenty-eight of the rape incidents involved two or more victims and accounted for 60 (or 2,9%) of the women, teenagers and girls in our sample.2 Overall, 94,3% (n = 1186) o f our sample experienced a completed rape involving an act of penetration. The remaining 5,7% (n = 114 c ases) reported an attempted rape, or a suspected sexual assault where it was unclear from witness statements or other evidence whether or not a sexual assault had taken place. In the remaining 68 cases, there was insufficient information in the docket to determine the nature of the sexual violation that had occurred. The role of age in r ape victimisation Almost two thirds of rapes (60,2%) reported in Gauteng in 2003 involved adult victims. Teenage girls (defined as girls between the ages of 12-17) comprised one in four victims (25,2%) and girls aged between 0-11 years, one in seven victims (14,6%). Age significantly affected many characteristics of reported rape, with adults three times more likely to be raped by strangers than were girls (48,1% vs 14,6%). A lmost one in five women (18,8%) was raped by a current or former intimate male partner. Adults were also twice as likely as young girls to be the victims of gang rape (20,0% vs 8,2%). Half (49,5%) of the rapes perpetrated against adults involved an abduction where the perpetrator encountered the woman in one place and then forcibly took her elsewhere. Adult women were the group most likely to be attacked outdoors with more than one in four rapes occurring in an open space (24,9%) and a further 7,8% occurring in an alleyway or by a road. A sizeable proportion of women were also attacked opportunistically in the course of their routine daily activities such as walking (51,9%), or while simply being at home (19,7%). Rapes directed against adult women were the most likely to involve weapons, force, threat and injury. Perpetrators were ten times more likely to be armed with some sort of weapon when they raped adult women (40,9% of rapes) than when they raped girls (4,7%). In one in five cases (19,3%) this weapon was a gun. S ome form of bodily force was used against 70,1% of adult women and more than one in three (38,3%) was threatened with death or injury. Injuries to the body were 20 times more likely in women as in young girls and found in 39% of adult survivors. One in two adults sustained some form of genital injury. Repeat and chronic abuse was most evident amongst girls aged 0-11, of whom 16,4% stated that they had been raped at least once before by the same accused. This percentage, however, under-reports chronic abuse in this age category. While some medico-legal examinations found both old injuries and scarring to the girls' hymens, as well as evidence of prior anal penetration, only one charge of rape had been recorded, even though it was clear that multiple incidents had taken place. Repeat abuse was also evident amongst teenage girls, of whom 12,5% had been raped before by the suspect, but least likely amongst adult women, of whom 2,6% had been previously raped by the same suspect. Rape and disability A very small number of victims (41 or 1,9%) were recorded as having some form of disability. These figures fall below the prevalence of disability in Gauteng, calculated as affecting 3,8% of the female population in the province in the 2001 Census.3 It is impossible to know whether our figures reflect under-recording of disability on the J88s (the form recording the doctor's findings from the forensic examination) and dockets; 20 Institute for Security Studies SA Crime Quarterly no 32 • June 2010 21 under-reporting of rape of disabled victims; or a lower vulnerability of rape amongst disabled people. The last explanation seems unlikely. Research internationally has found the incidence of sexual victimisation experienced by women with disabilities to be either similar to, or greater than that reported by non-disabled women.4 The next section describes what became of all these charges. THE POLICE INVESTIGATION Of the 2 055 c ases for which information was available, 918 (44,7%) were withdrawn by the police. This proportion increased where adult women were concerned, with half of their cases (50,2%) withdrawn in comparison to only one in three (34,9%) cases involving girls under the age of 12. This proportion is somewhat better than that reported by the South African Law Commission (SALC) in 2000.5 Their study6 found that 68% of rape cases involving adult victims and 58% of those involving children did not progress beyond the police investigation. As we too found, failure to identify a suspect was the most common reason why cases did not proceed beyond the police investigation. The majority of cases (52,3%) were closed because a perpetrator could not be traced. Within this category of untraceable perpetrators we included cases where a suspect could not be identified by name or appearance, as well as those cases where the suspect was known but subsequently disappeared. More than half of adult women (55,4%) and teenage girls' (51,8%) cases were closed because the perpetrator could not be identified or found. The high number of withdrawals resulting from the failure to locate the perpetrator can, in part, be attributed to the higher number of stranger rapes perpetrated upon these two groups of victims. By contrast, just over one in three (36,3%) girls' cases were closed because the perpetrator could not be found. This is to be expected, given that over 80% of girls were raped by family members, friends or neighbours, whose identities and addresses were more likely to be known. It is worth pointing out how police practice contributed to suspects remaining 'undetected.' Descriptions of the perpetrator were absent from more than three-quarters of victims' statements (78,4%). Additionally, in more than half the dockets (52,7%), an instruction to arrest the suspect had to be issued twice or more before the investigating officer complied with it. In 30,2% of the cases where an instruction was issued on two or more occasions to arrest the suspect, he had disappeared. The 30% of victims who became untraceable could potentially also have been reduced. The victim's residential address was not recorded in the docket in 2,5% of cases and a work address not captured in three-quarters (75,2%) of the dockets. In only 17,8% of cases were the details of a contact person (in addition to the victim) captured. The median number of attempts made to contact untraceable victims was three, ranging from no effort whatsoever, to a maximum number of 15 attempts. In 25% of cases where the Table 1: Reasons for complaints being withdrawn by the police, by victim age Perpetrator untraceable 52,3% 36,3% 51,8% 55,4% Victim untraceable 30,1% 32,3% 27,7% 30,2% No evidence of rape 2,6% 17,6% 1,5% 0,5% False accusation 4,4% 5,9% 9,7% 2,3% Victim too traumatised 0,1% 0,2% Pressure from victim's guardian/caregiver/ parent 0,4% 2,9% 0,5% Families of victim and accused have resolved matters 0,9% 1,5% 0,8% Victim and accused have resolved matters 2,3% 0,5% 3,3% Victim wants to get on with her life or other- wise uncooperative 5,3% 2,0% 5,6% 5,9% Docket lost 0,2% 0,3% Other 1,3% 2,9% 1,0% 1,1% Total n=918 0-11 years n=102 12-17 years n=196 18+ years n=617 victim disappeared, as few as four days had passed between the investigating officer's first and last attempt to trace her. These particular examples of inadequate policing point to where attrition could potentially be reduced at the investigation stage. Table 2 sets out further indicators we used to assess the quality of docket completion. Some of these indicators (such as the absence of a first witness statement) will influence the prosecutor's decision whether or not to place a case on the court roll to enable it to go to trial. The medico-legal evidence is another crucial factor influencing whether or not an individual rape case proceeds to court. Here too we identified problems with the quality of the forensic examination and handling of evidence. Medico-legal evidence The J88 was available for our scrutiny in 77% of cases. According to the dockets, a Sexual Assault Evidence Collection Kit (SAECK) was completed in 67% of cases and in 51% of cases it was sent to the police's forensic laboratory. However, in only 16,4% of cases was the suspect's blood taken. Collecting DNA evidence from the victim is a meaningless exercise if it cannot be tested against a suspect's DNA. This practice certainly contributed to the fact that a report from the laboratory containing the results of DNA testing was available in a s cant 2% of dockets. Fewer children (39%) had a kit completed as compared to 61% of teenagers and 77% of adults. The SAECKs of girls were significantly more likely to be analysed and a report made available after being sent to the FSL, than for adults. The forensic examiner's concluding statements on the J88 are particularly relevant to case progression. The examiner is required to draw two conclusions: one based on the gynaecological examination and the other on the general examination. Some one in six (16,2%) J88s did n ot have a concluding statement related to the gynaecological examination. A concluding statement was missing in a higher proportion (18,9%) of adult cases than girls' cases, where 11,9% did not have a concluding statement. A concluding statement was missing from the general examination section in 41,5% of cases – which could be due to clinicians focusing their examination on the gynaecological examination alone. Clinicians often wrote 'Alleged Rape' in the conclusion section of either the general examination or gynaecological examination. This phrase does not qualify as a conclusion about the examination. In 13,4% of J88s, both concluding 22 Institute for Security Studies Address not recorded 2,5% 3,4% 2,5% 2,0% Work address not recorded7 75,2% 75,0% 72,9% 76,1% Contact person not recorded 82,2% 76,2% 78,3% 85,3% Address where rape occurred not recorded 13,7% 19,2% 13,3% 12,3% Handwritten victim statement not in docket 7,5% 30,2% 6,6% 2,1% Victim not asked to describe perpetrator 78,4% 87,4% 82,0% 74,7% First witness statement not taken 45,8% 26,0% 38,1% 53,6% Other witnesses but no statements taken 41,6% 35,2% 37,1% 46,7% Instruction issued to make arrest before it was made 41,2% 36,9% 38,8% 43,3% Instruction issued 2+ times to make arrest before it was made 52,7% 55,7% 54,5% 52,6% Table 2: Indicators of problems with the quality of the docket completion, by victim age Total n=2 042 0-11 years n=292 12-17 years n=514 18+ years n=1 227 Opening case 2 064 298 514 1230 Perpetrator arrested or 1 036 164 291 575 asked to appear in (50,5%) (55,0%) (56,7%) (46,8%) court Charged at court8 885 138 247 495 (42,8%) (46,3%) (48,1%) (40,2%) Trial commenced 358 66 109 181 (17,3%) (22,1%) (21,2%) (14,7%) SA Crime Quarterly no 32 • June 2010 23 Those who raped young girls were twice as likely to be convicted of a crime (10,1%) as those who perpetrated their crimes against adults (4,7%). Differential case outcomes on the basis of age have been found elsewhere. In one study 9,1% of children's cases resulted in convictions, as opposed to 5% of adults' cases,9 while data for the year 2000 identified 8,9% of children's cases and 6,8% of adults' cases to have led to convictions. This rate varied across provinces.10 Just over one in 20 (6,2%) c ases resulted in conviction. However, some of these convictions were for lesser charges such as indecent assault; so overall only 4,1% of cases reported as rape resulted in convictions for rape. Some one in seven (15,6%) convicted rapists received less than the mandated ten years minimum sentence. The other prescribed sentence for rape, life imprisonment, was very rarely applied. Thirty- four (or 41%) of the men convicted of rape were eligible for life imprisonment. This sentence was handed down in only three cases, all of which involved girls under 16. None of the cases involving gang rape, assault with intent to cause grievous bodily harm, or a victim with a disability resulted in life sentences – even though this is mandated in such cases. SURVIVORS' ENGAGEMENT WITH THE CRIMINAL JUSTICE SYSTEM Up to this point, the article has examined the contribution of the police, forensic examiners and statements were missing or the clinician had recorded 'Alleged Rape'. This was more likely to have occurred when the survivor was an adult than a child aged 0-11 (16,1% and 6,3%). Analysis of the J88 forms also showed that penetration may actually be more forceful in rape in South Africa, with 57,5% of cases resulting in an ano-genital injury. This is a higher proportion of such injury than has been found in studies conducted in the developed world. Nonetheless, a key finding of this study is that in 42% of cases no genital injuries were recorded while in 72% of cases no injuries to the remainder of the body were documented. This confirms findings from other countries that vigorously argue that absence of injuries should not be interpreted as indicating that no rape took place. Yet, as shown below, in relation to adult women the presence of injury clearly played a determining role in judicial officers' decisions. THE COURTS Courts disposed of cases in three ways: nolle prosequi decisions (16,1%), the withdrawal of matters (20,1%) and striking of matters from the court roll (2,2%). The greatest proportion of cases were 'nollied', on the basis that there was too little evidence to warrant a prosecution (25,7%) while the greatest proportion of cases were withdrawn due to the victim becoming untraceable (33,5%). One in four cases was withdrawn within five weeks (35 days) of the accused being charged. On average however, three months (92 days) elapsed between the accused being charged and the case being withdrawn. Three-quarters of cases were withdrawn within eight months (235 days). One case spent three years and four months (1 029 days) on the court roll before it was withdrawn. Summarised in the next column are the outcomes of the 2 064 c ases in our study. As this table shows, while half of cases resulted in arrests (50,5%), a s omewhat lower percentage (42,8%) were charged in court. Trials commenced in less than one in five cases (17,3%). Adult women fared worst at the hands of the criminal justice system relative to girls and teenagers. Table 3: Attrition of cases of rape at each stage of the criminal justice system, by victim age Total 0-11 years 12-17 years 18+ 24 Institute for Security Studies courts to case attrition. This section takes a closer look at the 37% of survivors who opted out of criminal justice system proceedings. Contrary to police and prosecutors' perceptions, only one in 20 victims withdrew their complaints following reconciliation with the accused or his family. The same proportion of victims withdrew because they found proceedings too upsetting or disruptive of their lives. What requires further investigation however, are the circumstances leading to the disappearance of almost one in four victims (23%). When victims became untraceable, they were most likely to do so after reporting the rape and during the course of the police investigation. Two thirds of those who disappeared (67,2%) did s o at this stage in criminal justice proceedings. It is possible that at least some of these survivors may have wanted to remain in contact with the police, but were unsure of their rights and the extent to which the police would welcome their inquiries. They may well have lacked the confidence to demand information from the police and waited instead to be contacted by them. Such follow-up contact would never have happened in cases where the police took down inadequate contact details from victims. In some instances, victims becoming untraceable could be ascribed to the police's failure to confirm victims' addresses and contact details at the time the reports were made. Some of the statements also suggested that victims belonged to a mobile section of the population that migrates across provinces, as well as within the province, either looking for work or moving between family and friends. They therefore did not have fixed, permanent addresses. Survivors' continuous cell phone ownership and airtime availability were also not guaranteed, given that the majority of women in the study were unemployed. Also, some women may not have wished to be found, only wanting to register that a crime had been done to them without having to engage with the criminal justice system on a longer-term basis. Indeed, engagement with the system often prompts traumatic memories that survivors may wish to put behind them. Intimidation by the perpetrator or his friends and family may also have contributed to survivors' disappearance. Finally, it is impossible to know how many of these disappearances are the result of police corruption or, in the face of case overload, the selective prioritisation of cases based on an assessment of the victim's perceived cooperativeness. Earlier it was reported that some police officers made no or only one attempt to locate the victim before filing the case. In another suspicious example, the investigating officer arrived at court claiming the victim could not be found and suggesting the matter be closed. The victim had, of her own accord, arrived at court and so the matter was postponed instead. However, by the next court appearance the victim had indeed become 'untraceable' and so the matter was withdrawn. IN CONCLUSION A certain amount of attrition in rape cases is inevitable. Some perpetrators may never be identified and some victims, for a variety of reasons, may well decide that pursuing a complaint is not in their best interests. In still other cases, there may genuinely be insufficient evidence to prove a matter beyond reasonable doubt. What becomes of central concern then, is the extent to which a reasonable rate of attrition is exacerbated by weak administrative practices and prejudicial attitudes towards rape victims. If society is to persist in demanding that women 'break the silence' and 'speak out' then the criminal justice system must ensure that it is worth their while, rather than an exercise in futility. To read the full report please visit: http://www.tlac.org.za/images/documents/tracking %20justice_web.pdf R. Jewkes, N. Christofides, L. Vetten, R. Jina, R. Sigsworth et al, Medico-Legal Findings, Legal Case Progression, and Outcomes in South African Rape Cases: Retrospective Review. PLoS Med 6(10): e1000164. doi:10.1371/journal.pmed.1000164 (2009). SA Crime Quarterly no 32 • June 2010 25 To comment on this article visit http://www.issafrica.org/sacq.php NOTES 1 R Jewkes and N Abrahams, The epidemiology of rape and sexual coercion in S outh Africa: an overview, Social Science and Medicine, 55(7), 2002, 1231-1244. 2 In this study, the word 'girl(s)' is used to refer to victims aged 0-11, 'teenager(s)' to refer to victims aged 12-17, and 'adult(s)' to refer to victims aged 18 and older. 3 Statistics South Africa, Report-03-02-44 – Prevalence of disability in South Africa, 30 March 2005, http://www.statssa.gov.za/publications/statsdownload. asp?PPN=Report-03-02-44&SCH=3359 (accessed 8 June 2010). 4 NE Groce, Disability in cross-cultural perspective: rethinking disability. The Lancet, 354(9180) (1999), 756-757; M Saxton, M Curry, L Powers, S Maley, K Eckels, and J Gross, “Bring my scooter so I c an leave you”: A study of disabled women handling abuse by personal assistance providers. Violence Against Women, 7(4) (2001), 393 – 417; MA N osek and CA Howland, Abuse and women with disabilities, 1998, http://www.vaw.umn.edu/finaldocument/vawnet/disab. htm (accessed 25 August 2008). 5 South African Law Commission, Conviction rates and other outcomes of crimes reported in eight South African police areas. Research Paper 18, Project 82, 2000. Thi s study was conducted in t he eight police areas of Western Metropole and Boland in t he Western Cape; Port Elizabeth and Cradock in t he Eastern Cape; Durban and Midlands in KwaZulu-Natal; and Johannesburg and East Rand in Gauteng. 6 SALC, 2000. 7 This referred to either the adult victim or guardian's work address (in t he case of a child). 8 After being arrested, suspects must then appear in court to be formally charged with a cr ime and apply for bail. 9 SALC, 2000. 10 Interdepartmental Management Team, Towards Developing an Anti-Rape Strategy: Report of the Interdepartmental Management Team. Report by the National Directorate of Public Prosecutions (NDPP), the SAPS, t he Department of Health, the Department of Social Development and Monitor Group, 2002. Mpumalanga, the worst performing province, recorded a 3,1% conviction rate for cases involving children and a 4,1% rate for cases involving adults. Gauteng was the third worst performing province, recording a 7% conviction rate for cases of child rape and 4,9% for adult rape. The best performing province was the Northern Cape which recorded a 16,8% co nviction rate overall.