Uneven and Still Insufficient

Abstract

This paper presents the findings of a quantitative study undertaken at South African Police Service stations across the Western Cape Province, measuring what we term committed intake compliance with sexual offences law. Fifty police stations were surveyed for compliance with the first layer of procedural and service requirements for receiving a victim’s report of a sexual offence. On average, stations complied with only 68% of the measured requirements. The authors analyze patterns in compliance that likely contribute to the drop-out of rape and sexual assault cases from the criminal justice system.

Keywords

comparative issues policing quantitative research rape sexual assault victim services international issues

Introduction

Widespread legal reform concerning sexual victimization has taken place worldwide over the last 40 years, and a growing body of scholarship aims to measure the impact of legislative changes on criminal justice practice (Rumney, 2001). In particular, law reform has aimed to address the chronic underreporting and underenforcement associated with sexual victimization.

This article contributes to the emerging body of global research on the impact of legislative change on citizen’s access to their rights by assessing whether South African police stations are compliant with the intake requirements laid out in South Africa’s newly reformed sexual offences legislation. The article presents findings from research that sought to understand if and how South Africa’s reformed sexual offences laws are implemented at the initial stages when the victim enters the criminal justice system. This quantitative study, undertaken at South African police stations across the Western Cape Province, measures what we term committed intake compliance with sexual offences law and policy. The study focuses on police station compliance with the procedural and service requirements when a victim enters the criminal justice system by reporting a sexual offence at a police station. The study is based on mandatory intake procedures and documentary requirements found in the Criminal Law (Sexual Offences and Related Matters) Amendment Act, No. 2 of 2007 (SOA), South African Police Service (SAPS) National Instruction 3/2008 (National Instructions), and various state-issued victim empowerment standards.

South Africa’s legal framework for sexual offences was reformed in 2007 with passage of the SOA, one of many legislative reforms since the establishment of a democratic dispensation in 1994. The new statute created new offences for sexual misconduct, addressed child prostitution and (to a limited extent) human trafficking, and enabled victims to apply for mandatory HIV testing of the alleged offender. It changed procedural and evidentiary rules related to the trial process, such as eliminating the cautionary rule (Artz & Smythe, 2005).¹ The SOA also made significant strides toward sex-neutrality in sexual offences definitions, thereby formally acknowledging the sexual victimization of men and boys, a traditionally overlooked victim population.

Many of the changes were intended to address the varied experiences of victims of sexual violence, to empower victims to report crimes, and to reduce secondary victimization within the criminal justice system.

To implement the SOA, SAPS updated its policies concerning sexual offences with the adoption of the National Instructions. The new instructions endeavor to align police victim intake and investigation procedures with the new statute and to facilitate the victim empowerment envisioned by the legislation.

In South Africa, much of the existing research examines the way reported cases make their way through the criminal justice system, in particular the “highly selective process of elimination” that occurs, referred to as “attrition” (Vetten et al., 2008). This paper situates intake compliance as an important first phase in the process of reporting, investigation, and prosecution of sexual offences.

This study defines compliance as adherence to a standard of conduct that promotes the underlying purpose of relevant laws and policies (Reiss, 1984). Compliance is defined in this manner: reaching beyond the letter of the law and examining whether the impact is promoting the purpose of the law. Elsewhere, this definition has been referred to as “committed compliance” (Braithwaite, 2003).

In terms of the SOA, this study seeks to analyze the implementation of the victim-centered nature of the Act as outlined in its preamble and as articulated in victim empowerment standards (found in three documents, the Service Charter for Victims of Crime in South Africa, the Minimum Standards on Services for Victims of Crime, and the Minimum Standards for Service Delivery in Victim Empowerment) and the National Instructions. Researchers used a checklist to record whether police stations had, for example, copies of relevant laws and policies and forms required for processing requests for compulsory HIV testing of alleged offenders. Researchers also determined whether services such as counseling and language interpretation were available at the police station. The researchers’ checklist included “soft,” or nonenforceable victim empowerment standards (named above), and “hard” enforceable requirements found in sexual offences laws and policies.

When taken together, the components of the checklist measure what we consider “committed intake compliance” in the Western Cape, a critical first step in criminal justice processes for responding to sexual offences.

Rape and Sexual Assault: Reporting and Conviction Rates

In the 2010/2011 period, SAPS received report of 68,332 cases of sexual offences nationwide, 4.4% fewer than in 2009/2010. Since the 2007 legal reform, these statistics include a wider range of offences under the single category of “sexual offences.” While the statistics are not disaggregated by type of offence, SAPS categorizes its data into prevalence rates for contact crimes (rape and sexual assault), and crimes detected as a result of police investigation (sex work-related crime and child pornography). In the current reporting period, 63,603 (93.1%) of reported sexual offences were contact crimes. The Western Cape recorded the highest rate of reported sexual offences in the 2010/2011 period, at 178 per 100,000 persons.

Prevalence research paints a different picture than the official reports, sometimes garnering international media coverage that depicts South Africa as “leading the world in rapes” (BBC News Online, 1999). A 2009 Medical Research Council study made front-page news worldwide when it reported that one in four South African men admitted to committing rape. Twenty-eight percent of those surveyed in the representative sample acknowledged that they had raped a woman or girl, and 3% stated they had raped a man or boy. Almost half of the men who admitted to rape had done so more than once. A total of 73% stated they committed their first assault before the age of 20 years, and 1 in 10 before the age of 10 years. A study of school children aged 11 to 19 years found that 9% of male respondents reported having been forced to have sex in the last year alone (Andersson & Ho-Foster, 2008). In this second study, of all respondents 18 years old at the time of the survey, 44% said they had been forced to have sex in their lifetime. Twenty-seven percent reported male and female perpetrators. Rape is also believed to be extensively underreported to officials in South Africa (Vetten, 2011). Only one in nine women who were raped and had physical force used against them reported the crime to the police—just 11% of female victims reporting cases (Jewkes & Abrahams, 2002). Reporting in other countries is somewhat higher with the following rates: 15% to 19% in the United States, 14% to 18% in England and Wales, 12% to 20% in Australia, 6% to 19% in Canada, and 12% in New Zealand (Daly & Bouhours, 2010).

The discrepancies between the official SAPS statistics and those gathered by researchers suggest underreporting of sexual offences to police. Underreporting has been attributed to a range of factors including stigma toward victims, fear of not being believed, fear of being blamed for the attack, lack of faith in the criminal justice system, close relations with the perpetrator, and the absence of physical force during the attack (Rasool et al., 2003). Access to justice systems represents another contributing factor, particularly in rural communities and those under traditional governance mechanisms where anecdotal evidence suggests that sexual victimization is often dealt with through traditional dispute resolution (mediation) rather than through the police and courts. Cases are withdrawn due to settlements, lack of evidence, and the inability of victims to identify a suspect (Vetten et al., 2008).

The SAPS reported an official conviction rate in 2009/2010 of 14.6% of sexual offences reported, one of the lowest conviction rates of serious crimes in South Africa, though not dissimilar to conviction rates in other parts of the world (Artz & Smythe, 2007a; South African Police Service, 2010). The National Prosecuting Authority (NPA; 2009/2010) report states that dedicated sexual offences courts have a higher conviction rate (67.7%), but the Department of Justice and Constitutional Development and the NPA conviction rate is based on the proportion of cases prosecuted (Waterhouse, 2008). This rate does not include cases that were reported but not prosecuted (Artz & Smythe, 2007b). As the discrepancy between SAPS and NPA conviction rates reveal, significant attrition occurs before prosecution. The first stage at which error may occur is the police reporting level.

Literature Review

Studies Examining Implementation and Impact of the Sexual Offences Law Reform

Studies examining the impact of sexual offences law reform on the dispensation of cases present mixed results, but largely find limited effects on the conviction rate, even reforms that expand the definition of rape and update evidentiary and procedural rules to bring them in line with Constitutional norms (Smythe & Waterhouse, 2008).

Many factors contribute to high levels of attrition in the criminal justice system, spanning infrastructural, procedural, and individual obstacles. Though police intake is just one component of securing a conviction, it serves as the crucial first point of contact for complainants, where the police function as gatekeepers, supporting or subverting the intent of law reform (Berger, Neuman, & Searles, 1994). In addition to police officers’ serving as gatekeepers, the negative experiences of survivors in reporting to the police can potentially disrupt or terminate their help-seeking in the criminal justice system (Burgess-Proctor, 2012). A five-country study by Daly and Bouhours including the United States, Australia, England and Wales, Canada, and Scotland found that 65% of cases drop out of the system at the “police stage,” including the time between reporting to police and referral by the police to the prosecutor’s office (Daly & Bouhours, 2010). An attrition study conducted by the Centre for the Study of Violence and Reconciliation (CSVR) in South Africa found that 45% of cases in its sample were disposed of at the police stage (Vetten et al., 2008).

Whether rape complaints move forward in the criminal justice system largely depends on law enforcement’s discretionary power to determine the “validity” of reported sexual offence, which may be informed by rape myths (Artz & Smythe, 2007a). In attrition studies from the United States, the United Kingdom, New Zealand, and South Africa, scholars point to police attitudes, shaped by “rape myths,” as a strong determinant of whether police officials processing complaints will view a case as a rape (Du Mont, Miller, & Myhr, 2003). Most commonly, rape myths hold that women often falsely accuse men of rape; that a real rape victim will immediately seek out assistance; and that rape occurs at night, outdoors, with strangers as perpetrators and must involve violent physical force (Andrias, 1992). More recently, scholars have pointed out that male victims are also overlooked due to stereotypes about female victimhood (Capers, 2011; Stemple, 2009). One study found that only 67% of surveyed police officers agreed that any man could be raped, and just 48% said they would believe a man who claimed he was raped (Page, 2008).

Patriarchal institutions such as police stations often subscribe to rape myths (Logan, Shanon, & Walker, 2006). The social norms reinforced in the police around sexuality and violence work together to normalize women’s victimization and erase men’s victimization. Studies in New Zealand and the United Kingdom have also shown that police departments are characterized by a robust subculture that insulates its members and fosters in-group cohesiveness (Skolnick & Bayley, 1993). The subculture emphasizes police solidarity and traditional, masculinist values and perspectives, including skepticism toward people who make rape allegations. Reforms that explicitly challenge patriarchal norms, including normalized male sexual aggression, may therefore be resisted (Gruber, 2009). Specialized training courses on these topics may be seen as impositions from the upper tiers of management, who may be viewed with suspicion by the rank and file (Jordan, 2005). Thus, the interconnection of rape myths with the predominant patriarchal social norms may make it especially difficult for rape law reforms to take root in police departments. A related hindrance to implementation of sexual offences reforms may be the suddenness and magnitude of the change, which influence the degree to which officers are able to successfully adapt (Robinson, 2000). Robinson posits that the inability to discard dated behavior is part of the resistance to change inherent in police culture.

In the South African context, as Smythe and Waterhouse point out, resource constraints, lack of training, and overly burdensome workloads with which South African police are faced may also make it more likely for them to draw from their “personal store of stereotypes to interpret a situation” (Smythe & Waterhouse, 2008). They also assert that resource shortages create a “perverse incentive to cut caseloads.”

Police attitude during intake may impact a victim’s commitment to seeking justice, the quality and detail in the content of the complaint, and whether a victim will access referral information to health services and the like. Research indicates that some victims voluntarily withdraw their cases because they do not want to traumatize themselves further by continuing to pursue their complaints. Smythe and Waterhouse indicate that brokering payments from the perpetrator to the victim is one of the most common reasons for unsuccessful prosecutions in South Africa, and there are allegations of police corruption and complicity in this practice.

The CSVR study on sexual offences case attrition in South Africa also found gaps in record keeping with intake documents, pointing to what Vetten et al. describe as, “a general carelessness of attention to detail . . . that may have an impact on the admissibility and reliability of the victim’s statement in court” (Vetten et al., 2008). For example, 15.5% of statements were not signed by the complainant, 8.5% were not dated, 14% did not record the scene of the crime, and 7.5% of children’s dockets had victim statements missing in their entirety. Studies in the United States that have examined the impact of rape and domestic violence law reform at the police-level have also found limited impact. A study by Campbell and Johnson found that despite decade old legal reforms, more than half of the police officers they interviewed continued to define rape according to prereform definitions, adhering to definitions of rape that require the use of force or the threat of force, as opposed to simply requiring the lack of consent (Campbell & Johnson, 1997). Promisingly, studies in the United States found that the more knowledgeable officers are about rape scenarios and victims, the less likely they will subscribe to rape myths, and that departments with more written directives on police behavior have higher arrest probabilities (Eitle, 2005). These results indicate the promise of consistent training and clear directives.

SAPS and the South African government have committed themselves on paper to addressing gender-based violence, but these commitments do not necessarily filter down to the service delivery level (Vetten, Leisegang, & Haken, 2010). For full implementation, this commitment to address violence must reach from the national and provincial levels, to the managers at the police station level, and to the officers themselves.

Civil Society Efforts to Increase Implementation of the SOA

South African civil society antirape advocates have sought to improve implementation of the SOA and increase the availability of victim-centered services. The Shukumisa Campaign, which has monitored the implementation of the SOA at police stations, courthouses, and designated health centers since 2008, is one such effort. In 2008, the Shukumisa Campaign monitored 34 police stations across the Gauteng, North West, and Western Cape provinces (Vetten et al., 2010). In 2010, the campaign monitored 59 stations across the Gauteng, Limpopo, Western Cape, and KwaZulu-Natal provinces. The Campaign sought to determine whether stations had the set of forms and documents identified within the SAPS National Instructions on hand, where and how SAPS members took statements from victims, and SAPS members’ familiarity with required regulations and procedures (Shukumisa Campaign, 2010). This study adapted the checklist from the Campaign’s monitoring forms used for police stations.

The Shukumisa Campaign, designed primarily as a means to encourage compliance by stations, rather than as a scholarly inquiry, had methodological limitations we sought to overcome—for example, the stations were not selected in a way that reduced research bias.² Still, it bears highlighting that the Campaign’s 2008 and 2010 cycles of monitoring each revealed that few stations had the required forms available, and that many of the SAPS officers interviewed were unfamiliar with the SOA and its documentary requirements to facilitate intake procedures (Vetten et al., 2010). These findings are consistent with ours.

Gender Inequality, Sexual Victimization, and HIV

The SOA reforms address, among other issues, victims’ empowerment through access to information and services for HIV testing and treatment. The revised SOA makes explicit the connection between sexual violence and HIV transmission. Gender-based violence and gender inequality—often experienced through domestic violence, sexual victimization, and socioeconomic inequality—are essential determinants of HIV risk, worldwide and within South Africa (Dunkle et al., 2004). For instance, people in abusive relationships are less likely to be able to negotiate condom use, and victims of childhood sexual abuse are more likely to practice high-risk sexual behaviors that lead to HIV exposure (Andersson, Cockroft, & Shea, 2008).

The research drawing the connections between sexual violence and HIV and AIDS is especially significant when considering the potential impact of postexposure prophylaxis (PEP) for sexual violence victims as a matter of law and policy. For example, Wulfson’s research on PEP provision for rape victims at a private health facility in Johannesburg reported that only a single seroconversion was documented of a cohort of 173 patients who started on a course of PEP and returned for follow-up testing (Mbali, 2005). This was despite the fact that a high number of the patients had been gang raped and that there was a high HIV prevalence rate in the general population.

Police Intake Duties for Victims of Sexual Offences

The SAPS’ enforceable duties to implement sexual offences laws can be found in the SOA, its accompanying regulations, and the National Instructions, which are binding directives for the police, with disciplinary steps outlined for failure to comply. There are also nonenforceable victim empowerment standards (the Service Charter for Victims of Crime in South Africa, the Minimum Standards on Services for Victims of Crime, and the Minimum Standards for Service Delivery in Victim Empowerment).

In terms of the National Instructions, each station must have copies of certain documents available at the front desk area, called the Community Service Centre (CSC), including the following: a copy of the SOA, the regulations promulgated under the SOA, the National Instructions, and a list of service providers for victims. The station commissioner must also issue station orders, which are written directives tailoring implementation of the SOA to that specific station.

The accompanying SOA regulations require stations to keep copies of forms that inform victims of their right to compulsory HIV testing of alleged offenders, the availability of PEP at state expense, confidentiality of HIV test results, and an application for a warrant of arrest for alleged offenders who do not comply with compulsory testing orders.

If the victim is unable to read and understand the document in the printed languages, the SAPS member, or another willing and able person, is required to assist the SAPS member by explaining the contents of the form to the victim in a language that they understand (South African Police Service, 2008). The SAPS members must provide a list of public health establishments in the province that are able to provide PEP to the victim. This must happen quickly due to the 72-hr time window for effective administration of PEP.

Station orders operationalize sexual offences law at the station level with consideration of the station’s available resources and the local context within which the station is situated, and provide specific steps for members to take to assist victims of sexual offences (South African Police Service, 2008).

The National Instructions clearly stipulate that SAPS members to whom a victim reports the offence at a police station must take statements away from the CSC. SAPS members must immediately determine whether the victim requires medical assistance, and if so, arrange for assistance prior to handing the case over to an investigating officer.

In the Western Cape, the investigating officers are deployed from the Family Violence, Child Protection and Sexual Offences Units (FCS Units). The FCS Units’ mandate covers family violence crimes not including murder, child protection including all offences against children, and sexual offences for all adult victims (Frank, Waterhouse, Griggs, & Rontsch, 2009). On receiving a report from a victim regarding a sexual offence, SAPS members in the Western Cape are required to call their designated FCS cluster station. An investigating officer from the FCS cluster station must then pick up the victim in an unmarked vehicle to continue the investigation.

Mandatory testing processes require great care so as to avoid false-negative test results and infringement of the alleged offender’s constitutional rights to bodily integrity and privacy. The mandatory testing provision of the SOA itself is problematic in this regard. SAPS members duties to enable testing include informing and assisting victims throughout the application process, attending the magistrate’s hearing for the application, informing parties of the magistrate’s decision, delivering the testing order to the alleged offender, and applying for a warrant if the alleged offender does not comply.

The Service Charter for Victims of Crime in South Africa (Victim’s Charter) and the Minimum Standards on Services for Victims of Crime (DoJ Minimum Standards) produced by the Department of Justice and Constitutional Development and a schedule of Minimum Standards for Service Delivery in Victim Empowerment (DSD Minimum Standards), issued by Department of Social Development together provide a set of standards for the state’s criminal justice response to victims of crime (Artz & Smythe, 2005).

The DoJ Minimum Standards urge accountability in the “delivery of an appropriate and quality service,” urge service providers to inform communities on availability of and access to resources, and to provide information on referral procedures. These standards also state that victims are entitled to request an interpreter to be interviewed in a language they understand. Victims also have the right to be treated with fairness and with respect for dignity and privacy. In particular, they have the right to be interviewed in private by the police.

The Victim’s Charter states that victims have the right to offer and receive information and to assistance, “and, where relevant, have access to available social, health and counseling services, as well as legal assistance which is responsive to [their] needs.” It states that during investigations, the police are required to take measures to minimize any inconvenience to victims by conducting interviews in a language of the victim’s choice.

Method

This study used quantitative research methods to generalize findings to the wider Western Cape province. We utilized a checklist form, adapted from a form used by the Shukumisa Campaign for monitoring police stations. Researchers administered the checklist at 50 police stations across the Western Cape over a 3-month period from March to May 2011.

Sample Selection

There are 148 police stations in the Western Cape (South African Police Service, 2011). We excluded 25 stations that had already been monitored by the Shukumisa Campaign due to the potential for response bias. This left a total of 123 total stations from which to sample. We calculated the sample size based on a 95% confidence level and a 10% margin of error, resulting in a sample size of 50 stations.

To obtain representative samples of urban and rural stations, we used stratified sampling, weighing samples according to population size (Table 3). The Western Cape SAPS organizes police stations into four divisions, each managed by an Area Commissioner. Two areas are part of the Cape Town City municipality (Eastern Metropole and Western Metropole), and two encompass the rural areas to the northwest of Cape Town (Boland), and the northeast and southern coast (Southern Cape), respectively (South African Police Service, 2002).³

We combined the two Cape Town metropolitan divisions that contain 62.2% of the provincial population, calling this unified division the “Cape Town Metropole.” The Cape Town Metropole can be classified as an urban demographic. The SAPS’ Boland area contains a mostly rural demographic consisting of 22.7% of the population, and the Southern Cape area also contains a mostly rural 15.1% of the population.

We then coded and numbered the listed police stations, and used a random number generator to make the final sample selection.

Monitoring Tool

We developed a checklist form pertaining to police intake duties as delineated in national policy and standards documents summarized in Table 1. This form was adapted from the Shukumisa Campaign’s monitoring tool, which we reorganized. We attempted to eliminate leading questions, minimize interviewers’ subjective responses, and ensure uniformity by using “yes/no” or “complete/incomplete/no copy” answer formats. To ensure anonymity, we eliminated questions that could lead to the identification of individual SAPS respondents. For the list of questions that we administered and scored for compliance, see Table 2.

Table 1.

Summary of Sources of Police Duties to Implement Sexual Offences Law.

Sexual Offences Act	National instructions	Victim empowerment standards
Enforceable	Enforceable	Nonenforceable
Provides for victims to apply for compulsory HIV testing of the alleged perpetrator. Victims can access PEP at designated public health facilities.	Maintain list of local victims services providers at the CSC. Copies of the SOA, regulations, administrative forms, and national instructions at the CSC. Requires victims’ statements to be taken away from the front desk at the station. SAPS must inform victims about availability of PEP and compulsory testing of alleged perpetrator. Requires SAPS to explain contents of forms and information in language that they understand.	Inform communities of available services for victims. Provide information on referral procedures. Provide social, health, counseling and legal services responsive to the victims needs. Victims entitled to be interviewed in private. Conduct interviews in a language of the victim’s choice.

Note. HIV = human immunodeficiency virus; CSC = community service centre; PEP = postexposure prophylaxis; SOA = Sexual Offences Act; SAPS = South African police service.

Table 2.

Questions Contained in the Monitoring Tool.

Question	Response format
1. Do you take statements away from the Community Service Centre (Front desk)?	Yes/No
2. Do you have a copy of the SAPS National Instruction 3/2008?	Complete/Incomplete/No copy
3. Do you have a copy of the Sexual Offences Act?	Complete/Incomplete/No copy
4. Do you have a copy of the compulsory HIV testing forms?	Complete/Incomplete/No copy
5. Do family violence, child protection, and sexual offences investigators come to your station to assist the victim?	Yes/No
6. Do you have information about hospitals providing PEP to rape victims?	Complete/Incomplete/No copy
7. Do you take statements in a separate room that is private?	Yes/No
8. Do you have a list of organizations providing services to rape victims?	Complete/Incomplete/No copy
9. Do you have station orders around sexual offences?	Complete/Incomplete/No copy
10. If a victim does not speak English, Xhosa, or Afrikaans, does your station work with interpreters to help take statements?	Yes/No
11. If a deaf victim reports to your station, is there a person or an organization you call to provide sign language services when taking statements?	Yes/No
12. If a mentally disabled rape victim reports to your station is there a person or an organization you call to help you take their statement?	Yes/No
13. Does your station have volunteers, members of organizations providing services for victims at your station?	Yes/No
14. Does your station have volunteers, members of organizations providing services for victims at your station of a 24/7 basis?	Yes/No
15. If someone reports that they were raped because of their sexual orientation, do you refer them to specialized counseling for LGBTI persons?	Yes/No
16. Does your separate victim support room in which victims can receive counseling and other services?	Yes/No

Note. SAPS = South African police service; HIV = human immunodeficiency virus; PEP = postexposure prophylaxis; LGBTI = lesbian, gay, bisexual, transgender, and intersex.

Table 3.

Stratified Sampling by SAPS Area Commission Divisions^a.

Location	Population no.	%	Stratified sample
Boland	906,341	22.7	11 stations
Southern cape	601,344	15.1	8 stations
Cape town metropoles	2,479,988	62.2	31 stations
Total	3,987,673	100	50 stations

Note. SAPS = South African police service.

The most recent available census data are from 2000; thus we accessed noncensus data (Elsenberg, 2005, p. 3).

Research Procedure and Approach

Five research assistants who supported the data collection underwent a 2-day training session that provided information on the laws and policies, familiarized the research assistants with the monitoring tool, and reviewed ethical boundaries of the research. Researchers undertook practice investigations at two stations already monitored by the Shukumisa Campaign, after which correction and further instruction were provided.

Researchers (including the lead researcher and research assistants) were assigned to the Cape Town Metropole, Boland, and Southern Cape stations to administer the checklist in teams of two. Stations were not informed of visits beforehand so as not to bias the investigation. Each researcher individually completed a separate identical checklist forms. After each visit, researchers reviewed the two complete forms for consistency in their recorded responses and discussed and reconciled divergent responses under the oversight of the lead investigator. More than 95% of responses were not divergent.

Researchers did not pressure the SAPS members at sampled police stations to respond, but informed respondents that under the Victims Charter and the Minimum Standards for Service Delivery, police members are allowed to “inform communities on availability of and access to resources,” and “to provide information on referral procedures,” for victims.

After the investigation was complete at each station, researchers provided copies of the SOA and/or National Instructions if no copies existed at the station, as well as copies of an NGO-produced lay persons guide to the SOA.

Ethical Approval

This research project was limited to gathering public institutional data. The UCLA internal review board therefore determined that it did not require ethical approval as it did not meet the definition of human subjects research.⁴ Thus, while researchers interacted with SAPS members, the questions were not about the SAPS members themselves, their personal opinions, or any private or identifying information. The responses pertained only to public information and were limited to the response of officers located at the front desk and charged with communicating with the public as part of their ordinary duties.

Compliance Index

We developed an index against which to score each police station for “committed intake compliance,” by which we mean compliance that comports with the intent of laws and standards governing police reception of sexual offences reports. Each item scored in the index is matched with an item on the monitoring checklist.

The index allocates up to one point per item. The items are grouped into three categories designed to signify different degrees of compliance, as detailed below. The three categories are called minimal compliance, ordinary compliance, and recommended compliance. We assigned each question from the monitoring tool with a single compliance category.

Minimal compliance (enforceable requirements)—5 points possible

All items in this category are requirements from enforceable policy documents, see Table 2. The minimal compliance category includes items such as having paperwork in place, for example copies of the SOA and compulsory HIV testing forms. These documents are contained in the National Instruction 3/2008 and thus do not require that the police make efforts beyond printing or obtaining a copy to keep on file. This category also includes the requirement to take statements away from the CSC (front desk) area and the requirement to call an investigator from the designated FCS Unit.

Ordinary compliance (enforceable requirements)—4 points possible

All items in this category are requirements from enforceable policy documents, see Table 2. Compliance with these items requires the station commissioner, victim empowerment officer, or other SAPS member to actively engage with the requirements contained in the policy documents. This includes the requirement that station commissioners create station orders that outline procedures that are tailored to the resources and context at the specific station. The station commissioner must also compile a list of victim services providers in the community, identifying the public health institutions where victims can access PEP, and provide access to language interpreters for victims.

Recommended compliance (nonenforceable requirements)—4 points possible

All items in this category are recommended services from victim empowerment standards. This category includes taking statements in a separate and private room (beyond simply moving away from the front desk, which is all that is required under the National Instructions), identifying interpreters available for deaf victims, identifying persons to help take statements from mentally disabled persons, and having volunteers or other personnel at the station to provide victim support.

Data Analysis

We analyzed the data set using univariate and bivariate statistical analysis. We calculated descriptive statistics such as mean, median, SD, range, and tallies. The univariate descriptive analysis allowed us to analyze and explain patterns of compliance for each requirement listed in the monitoring form. We used bivariate analysis to compare and contrast findings. Lastly, we conducted correlational analysis using Pearson’s correlation coefficient to examine the linear relationship between rural and urban demographics and compliance, average community income level and compliance, and sexual offences per capita and compliance.

Results

Overall the Boland and Southern Cape (rural) stations scored higher than the Cape Town Metropole (urban) stations, with compliance of an average of 72.9% of items, versus 53.1% (see Tables 4 and 5). In fact, half of the rural stations but only 32% of the urban stations had compliance with 70% or more of the measured items. The highest scoring station, Conville, in the Boland and Southern Cape, received a perfect score. The lowest scoring station, Camps Bay, received zero points and was in the Cape Town Metropole (Table 7).

Table 4.

Boland and Southern Cape SAPS Stations.

No.	Station	Family violence, Child protection and Sexual offences units (FCS)	April 2010–March 2011 Sexual offences recorded per station^a	“Minimal compliance” (Out of 5 items)	“Ordinary compliance” (Out of 4 items)	“Recommended compliance” (Out of 4 items)	Total score (Out of 13 items)	Percent of all items complied with
1	Bonnievale	—	24	5	3	3	11	84.6
2	Ceres	YES	87	2	1	3	6	46.2
3	Conville	—	161	5	4	4	13	100.0
4	De Rust	—	10	5	3	3	11	84.6
5	Dysselsdorp	—	25	5	2	3	10	76.9
6	Franschhoek	—	49	2	0	3	5	38.5
7	Groot Brakrivier	—	30	5	2	4	11	84.6
8	Groot Drakenstein	—	29	5	2	2	9	69.2
9	Heidelberg	—	22	5	3	2.5	10.5	80.8
10	Klapmuts	—	33	4.4	2.5	3	9.9	76.2
11	Klawer	—	24	2.8	2	3	7.8	60.0
12	Mossel Bay	—	9	2	0	4	6	46.2
13	Murraysburg	—	10	4	2	2	8	61.5
14	Nuwerus	—	10	—	—	—	—	—
15	Oudtshoorn	YES	102	5	4	3	12	92.3
16	Philadelphia	—	18	4.6	2	2	8.6	66.2
17	Rawsonville	—	40	5	2	2	9	69.2
18	Saldanha	YES	29	5	3	4	12	92.3
19	Vredenburg	—	63	4.8	2	4	10.8	83.1
M		—	40.8	4.3	2.2	3.0	9.5	72.9

Note. SAPS = South African police service. Vredenburg is the FCS Unit but it is housed within Saldanha’s premises.

Reported sexual offences drawn from the SAPS crime statistics reported per each station in the Western Cape Province for March 2010 to April 2011, reported on the SAPS website each year (available at http://www.saps.gov.za/statistics/reports/crimestats/2011/crime_stats.htm, last accessed July 2011).

Table 5.

Cape Town Metropole SAPS Stations.

No.	Station	FCS STATION	April 2011 to March 2011 sexual offences recorded per station	Minimal (5)	Ordinary (4)	Recommended (4)	Total Score	%
20	Athlone	—	78	4.2	4	3	11.2	86.2
21	Bellville	—	80	2	0	3	5	38.5
22	Bellville South	—	46	4.8	3	3	10.8	83.1
23	Bothasig	—	11	—	—	—	—	—
24	Camps Bay	—	13	0	0	0	0	0.0
25	Durbanville	—	51	5	2	2	9	69.2
26	Goodwood	—	33	3	2	3	8	61.5
27	Gordons Bay	—	17	4.6	2	4	10.6	81.5
28	Gugulethu	—	232	2	3	2	7	53.8
29	Harare	—	293	5	0	3	8	61.5
30	Hout Bay	—	48	2	1	2	5	38.5
31	Kensington	—	31	2	1	2	5	38.5
32	Kleinvlei	—	179	5	3	2	10	76.9
33	Kraaifontein	YES	251	5	1	2	8	61.5
34	Kuilsrivier	—	91	—	—	—	—	—
35	Lingelethu West	YES	98	5	2	2.5	9.5	73.1
36	Maitland	—	23	4	2.5	2	8.5	65.4
37	Milnerton	YES	161	4.2	4	2.5	10.7	82.3
38	Mowbray	—	10	5	2	2	9	69.2
39	Ocean View	—	57	5	2	2	9	69.2
40	Philippi	—	77	2	3	2	7	53.8
41	Philippi-East	—	124	2	2	2.5	6.5	50.0
42	Pinelands	—	19	5	1	2	8	61.5
43	Ravensmead	—	100	5	2	2	9	69.2
44	Sea Point	—	20	2	2	2.5	6.5	50.0
45	Simons Town	—	11	2	0.5	2.5	5	38.5
46	Steenberg	—	84	5	3	3	11	84.6
47	Strand	—	60	2	1	3	6	46.
48	Strandfontein	—	22	2	2	2	6	46.2
49	Table Bay Harbour	—	4	5	3	3	11	84.6
50	Tableview	—	70	5	2.5	3	10.5	80.8
M		—	77.2	3.6	1.9	2.4	8.0	53.1

Note. SAPS = South African police service; FCS = family violence, child protection and sexual offences units.

In both demographic areas, stations had the most success with the “minimal compliance” measures—They complied with a mean of 3.6/5 items in the Cape Town Metropole and a mean of 4.3/5 in the Boland and Southern Cape. These items include documentary requirements such as having forms for compulsory testing of alleged sexual offenders, and having copies of relevant laws and policies (Table 5).

Stations had the least amount of success with the “ordinary compliance” group of items, with a mean of 1.9/4 in the Cape Town Metropole and 2.2/4 in the Boland and Southern Cape (Table 4). These items include creating referral lists to local victim service providers and identifying where victims can access PEP. These items require stations to engage with the policy requirements and identify and list victim services in the surrounding community, serving as an information source for victims.

The mean score for “recommended compliance” items was higher than for ordinary compliance: 2.4/4 in the Cape Town Metropole and 3.0/4 in the Boland and Southern Cape (Tables 5 and 4). While these items are not contained in enforceable policy, they are intended to lessen the likelihood of victims’ secondary trauma by, for example, taking statements in a private room. Recommended items also aim to enable marginalized victims to access criminal justice; they include measures such as having interpreters for deaf and mentally disabled victims.

A total of five FCS Units were selected through the random sample selection process. The average compliance of the FCS Units is 74.6%, in contrast to the average non-FCS compliance rate of 70.35%. The specialized FCS Units in the Boland and Southern Cape were Ceres (46.2% compliant), Oudtshoorn (92.3% compliant), and the Vredenburg FCS (92.3% compliant), which is housed within Saldanha station (Table 4). The Cape Town Metropole FCS units that were assessed were Kraaifontein (61.5%), Lingelethu West (73.1%), and Milnerton (82.3%; Table 5).

There is no statistically relevant correlation between urban versus rural location and compliance, though it is interesting to note that of the top five highest percentages, 8 of the 11 stations were from the Boland and Southern Cape areas. Conversely, 11 of the 14 lowest scoring stations were from the Cape Town Metropole area.

To assess whether income had any relation to compliance, we gathered household income per annum from each station based on the most recent census data. We categorized the station areas as low, middle, and upper income brackets according to the guidelines provided by Statistics SA, the South African government’s statistical bureau. Because police station jurisdictions do not correspond precisely with those of the census, the income categorization is approximate.

While there is no statistically significant correlation between income and the top five highest compliance scores, there is a strong negative correlation (−0.78 correlation) with the bottom five lowest compliance scores (Table 6); high and middle-income stations were more likely to score lower (Figure 1). Notably, the highest scoring station is in a low-income area in the Boland and Southern Cape area, while the lowest scoring station is in a high-income area in the Cape Town Metropole.

Table 6.

Stations With the Top Five Highest Compliance Percentages.

Rank	Station	Demographic	Average reported vs. reported sexual offences	Compliance %	Income bracket (2001)^a
1	Conville	Boland and Southern Cape	40.8/161	100.0	Low
2	Outdshoorn	Boland and Southern Cape	40.8/102	92.3	Middle
	Saldanha	Boland and Southern Cape	40.8/29	92.3	Low
3	Athlone	Cape Town Metropole	77.2/78	86.2	Middle
4	Bonnievale	Boland and Southern Cape	40.8/24	84.6	Low
	De Rust	Boland and Southern Cape	40.8/10	84.6	Low
	Groot Brakrivier	Boland and Southern Cape	40.8/30	84.6	Middle
	Steenberg	Cape Town Metropole	77.2/84	84.6	Middle
	Table Bay Harbor	Cape Town Metropole	77.2/4	84.6	High
5	Vredenburg	Boland and Southern Cape	40.8/63	83.1	Low
	Bellville South	Cape Town Metropole	77.2/46	83.1	Middle

We used 2001 census data on household income (per annum) for each suburb and categorized incomes into low, middle and high based on the 2001 Census Handbook’s suggestions (Statistics South Africa and Human Sciences Research Council, 2007, p. 206).

Figure 1.

Income bracket versus compliance score per station.

Table 7.

Stations With the Bottom Five Lowest Compliance Percentages.

Rank	Station	Demographic	Average Reported vs. reported sexual offences	Compliance %	Income bracket (2001)
1	Camps Bay	Cape Town Metropole	77.213	0.0	High
2	Franschoek	Boland and Southern Cape	40.8/49	38.5	Middle
	Bellville	Cape Town Metropole	77.2/80	38.5	Middle
	Hout Bay	Cape Town Metropole	77.2/48	38.5	Middle
	Kensington	Cape Town Metropole	77.2/31	38.5	Middle
	Simons Town	Cape Town Metropole	77.2/11	38.5	Middle
3	Ceres	Boland and Southern Cape	40.8/87	46.2	Low
	Mossel Bay	Boland and Southern Cape	40.8/9	46.2	Middle
	Strand	Cape Town Metropole	77.2/60	46.2	Low
	Strandfontein	Cape Town Metropole	77.2/22	46.2	Middle
4	Philippi-East	Cape Town Metropole	77.2/124	50.0	Low
	Sea Point	Cape Town Metropole	77.2/20	50.0	Middle
5	Gugulethu	Cape Town Metropole	77.2/232	53.8	Low
	Philippi	Cape Town Metropole	77.2/77	53.8	Low

Just more than 40% of stations in the Boland and Southern Cape and Cape Town Metropole areas have language interpreters available for non-English, non-Afrikaans, and non-Xhosa speakers (Table 8). A greater percentage of stations in the Boland and Southern Cape (42.1%) had interpreters for deaf complainants than the Cape Town Metropole (30.4%). There is an even starker difference for interpreters who can assist with taking statements from mentally disabled complainants—68.4% of stations in the Boland and Southern Cape had services available while only 25% of stations in the Cape Town Metropole reported having services available.

Table 8.

Stations With Interpreters Available for Non-English or Afrikaans Speakers, Deaf Persons, and Mentally Disabled Persons.

Demographic	Language interpreters (%)	Interpreters for deaf complainants (%)	Interpreters for mentally disabled complainants (%)	N
Boland and Southern Cape	42.1	42.1	68.4	19
Cape Town Metropole	41.1	30.4	25.0	28
Total	41.5	35.1	42.6	47

Correlation analysis of compliance scores and sexual offences per capita showed a 0.11 correlation in the Boland and Southern Cape, and a −0.01 for the Cape Town Metropole (Figure 2). Thus, there is no statistically significant correlation between the sexual offences per capita and compliance scores.

Figure 2.

Sexual offences recorded per station versus compliance score per station.

Discussion

Our results point to a mixed picture and may help explain the high drop-out rate of cases at the police stage, as found by the Daly and Bouhours and CSVR studies. The results also confirm the urgent need for SAPS to improve its compliance with the Sexual Offences Act and its ability to meet the needs of victims of rape and sexual assault.

The results suggest that the police stations in the rural Boland and Southern Cape have higher levels of committed intake compliance than those in the urban Cape Town Metropole. One might expect the converse to be true given the urban Cape Town Metropole stations’ proximity to resources, civil society and government service providers, and better communications and other infrastructure. On the other hand, the distance from resources might motivate rural stations to be more self-reliant, particularly in more isolated locations. It is not possible for these stations to send people elsewhere for services within a reasonable distance. Rural stations may not be able to rely on specialized FCS services as much as stations in the urban areas because of the great distances between rural stations and these units.

We examined the income level of police stations’ jurisdictions to test its correlation with compliance. Research in the United States has found that spending levels for departments in high- and low-income municipalities are either equal or favorable to high-income communities, despite the generally higher crime rates in low-income communities (Inman & Rubinfeld, 1979). Likewise in South Africa, police stations receive unequal financial resources, personnel, and equipment. This occurs against a backdrop of extreme income disparity in South Africa, one of the most unequal countries in the world (World Bank, 2012).⁵ Due to the history of apartheid, these inequalities largely follow racial lines, with White incomes far exceeding those of Black people, and “colored” and Indian-descent communities occupying intermediate positions (van der Berg, 2010). South African policing, already weakened by limited resources, a centralized and bureaucratic hierarchy, and limited management capacity, also suffers from the skewed allocation of resources in poor and rural areas (Pelser, Louw, & Ntuli, 2000). With this context in mind, we hypothesized that police stations in high-income communities would have higher rates of compliance due to greater resources.

Instead, stations in high-income areas are less compliant than those in low-income areas, with a significant negative correlation between income and compliance. The lowest scoring station, Camps Bay, which scored 0% compliance, is one of the most affluent neighborhoods in the Cape Town Metropole (Table 5). This station did not have required documents available and indicated that they typically send complainants directly to the FCS unit without taking statements. They also indicated that if they do take statements, they do so at the front desk. While the complainant will be processed by specialized personnel at the FCS Unit, sending complainants to FCS on their own instead of processing and calling the investigator may cause complainants to drop out of the reporting process. Furthermore, sending complainants to FCS Units without taking a statement first may allow the station to avoid recording the incident of rape or sexual assault, perhaps enabling it to report low prevalence rates. Indeed, Camps Bay reports a low number of sexual offences recorded at 13 for 2010/2011 (see Table 5).⁶

While middle and higher income stations are less compliant, it is likely not because they generally do not receive many reported sexual offences. For example, Bellville (urban) and Franschoek (rural) are both middle income areas, but their reported sexual offences are both above their category averages. Bellville has 80 reported sexual offences and the Cape Town Metropole has an average of 77.2 reported sexual offences. Franschoek has 49 reported sexual offences and the Boland and Southern Cape has an average of 44.2 reported sexual offences.

Sending complainants directly to specialized FCS Units would especially hinder low income victims in rural locations, where there are further distances between police stations and FCS Units and public transport is less accessible. This is the case with next lowest scoring station, Franschoek, which complied with only 38.5% of items (see Table 4). The Franschoek station must refer to an FCS unit that is more than 30 km away. Franschoekis in an area that is home to a mix of wealthy wine farm owners and low-income farm laborers; the distance may disparately impact those without convenient transportation.

One untested hypothesis that may explain the negative correlation between income level and compliance is the theory that higher income stations do not feel pressed to adhere to routine compliance with detailed sexual offences laws and policies. This could be because of the lower levels of reported sexual offences in the higher income stations. With a lower demand may come the assumption that officers’ ad hoc responses to sexual offences reports are sufficient.

On average, the specialized FCS units scored higher than the non-FCS units (74.6% vs. 70.35%; 4.25% higher). Notably, studies on the implementation of domestic violence laws in the United States found that the specialized domestic violence units did not have increased compliance (Eitle, 2005). While the FCS units were more compliant on average, the station in Ceres scored 46.2%, making it one of the lowest scoring stations overall, and well below the average score for rural stations (72.9%). The FCS stations from the urban category, Kraaifontein and Lingelethu West, scored 61.5% and 73.1%, respectively—above their category average of 53.1%. These mixed results support the idea that specialized stations do not necessarily demonstrate increased compliance.

Though our findings indicate no correlation between the rate of sexual offences in the station’s jurisdiction and its rate of compliance (Figure 2), officers in higher income stations may erroneously believe that people in their jurisdiction are at lower risk, thereby lessening their efforts. Related to this theory is the hypothesis that officers in lower income areas may use extra effort, believing that their hard work is especially needed in these otherwise underserved communities.

It is also important to note that, in general, the more effort each item requires on the part of the station, the lower the level of compliance. To wit, our “ordinary compliance” category that required that officers take the initiative to identify local service providers and prepare station-specific orders had the lowest mean score. This is especially troubling given that research indicates that the provision of clear written directives, such as station orders, lead to increased implementation (Eitle, 2005).

The “recommended compliance” category, while comprising items that are not from enforceable policy, does not require as much initiative as those in the “ordinary compliance” category. For example, the availability of a private room for statements is a matter of the station’s physical layout, and almost all stations had such a room available (though officers at Camps Bay indicated that they did not utilize it). Similarly, the availability of counseling services may be influenced more by civil society initiative rather than by a station’s efforts.

Accessibility issues for marginalized populations, such as foreign language or nonofficial language speakers, and deaf or mentally disabled persons, remain a concern. A prior study found that without services in place for deaf or mentally disabled victims, SAPS members struggle to take statements, sometimes resulting in reports being discounted or missing from dockets in their entirety (Vetten et al., 2008). Compliance in this regard need not be onerous. For example, officers may call SAPS’ national radio control to identify officers or interpreters with the language skills necessary to take statements from victims in the language they require. Once again, in this respect we found that Boland and Southern Cape far outperformed Cape Town Metropole (see Table 8).

We also asked the police stations whether they had referral information for lesbian, gay, bisexual, transgender, and intersex (LGBTI)–specific counseling services. Only stations in the Cape Town Metropole reported having services available. A handful of stations in low and middle income locations had such referrals in place.

Limitations

We verified station responses as often as possible by, for example, asking to see and review the required documents, referral lists, and separate interview spaces. However, we did not observe police behavior in practice concerning compliance with required procedures, such as whether private rooms were actually used by the police station. Station scoring may have been influenced by who happened to be on duty at the time of our investigation. We did not measure all SAPS members’ individual familiarity with requirements under law and policy, nor did we measure attitudes and behavior toward victims. Future qualitative studies may shed light on these types of issues.

In addition, specific questions about each station’s jurisdictional boundaries could be used in the future to determine the specific service area of each station, as well as the corresponding demographic, language, and income data of each area. This would enable a more nuanced measure of whether services are tailored to serve the specific needs of each surrounding community. For instance, though we found that stations in the Boland and Southern Cape have more language interpretation services in place than those in the Cape Town Metropole, we did not analyze whether the available services are a match with the many possible languages spoken in that community, or whether the services are used.

Lastly, because the research focuses on a specific cultural context, the ability to generalize findings is limited. However, we hope these data from South Africa will contribute to the international comparative dialogue on the implementation of rape law reform.

Conclusion

Worldwide, the implementation of rape law reforms remains a challenge. These challenges are heightened in developing countries, such as South Africa, that have increased burdens of financial, personnel, and infrastructural resource constraints. Civil society organizations that work to end sexual violence against women and men, including those with whom the authors are affiliated, remain alarmed by the persistent high levels of sexual victimization in South Africa. NGOs have expressed skepticism toward the government’s stated commitments to addressing this problem (Vetten et al., 2010). Meanwhile, media coverage of rape and sexual assault in all its violent permutations frequently includes the mishandling of cases by SAPS (Majangaza, 2012; Moselakgomo, 2012).

An average rate of compliance of 68%, with 72% compliance in rural areas and only 53% in urban areas, presents a very mixed picture of SAPS commitment to meeting the needs of those who report rape and sexual violence. During the course of our study we encountered many individual station commissioners and SAPS members who considered rape and sexual assault to be serious priorities. However, the compliance level is still a far cry from what is needed to properly respond to these offences and ensure victims’ access to justice. An enormous amount remains to be done.

This study provides information about where progress has been made as well as where critical gaps remain. There are serious implications to noncompliance at this initial stage of the criminal justice process. Noncompliance may prevent a case from progressing or even opening, it may result in victim retraumatization, and it can prevent victims from accessing critical HIV prevention services. To better enable rape law reforms to take root in the police service, training should occur on a consistent basis, starting at police training colleges and continuing as police rise through the ranks of management. It is also critical that clear written directives for each police station, which we found to be typically lacking, must be created and implemented. Lastly, consistent and continuous evaluation of compliance is required to ensure that rape law reform takes hold. Increasing SAPS compliance with its intake obligations will bring us one step closer to addressing sexual victimization in South Africa.

Footnotes

Acknowledgements

The authors would like to acknowledge and thank Lisa Vetten and the Shukumisa Campaign members for their efforts to monitor police stations across South Africa that informed the development of this research project. Keehn would like to thank and acknowledge her fellowship mentors: Lara Stemple, Dee Smythe, Shari Dworkin and Tim Shand. Thanks to the research assistants: Jimmy Singh, Englah Samuelsson, Simphiwe Peter, Mzamo Sidelo, Jean-Marie Nukurunziza, and Curt Syster. Thanks to Desmond Lesejane for additional guidance and input.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed that they received the following support for their research and/or authorship of this article: The corresponding author conducted this research with fellowship funding from the University of California Global Health Institute Women’s Health and Empowerment Center of Expertise Post-Doctoral Fellowship Program. The fellowship program was funded by the National Institutes of Health, Fogarty International Center, grant number 1R24TW008807-01.

Notes

Author Biographies

Emily Keehn is a policy development and advocacy specialist at Sonke Gender Justice, focusing on sexual and gender based violence, criminal justice and incarceration. Keehn was awarded a UC Global Health Institute Women’s Health & Empowerment Fellowship in its inaugural year.

Lara Stemple is the director of the Health and Human Rights Law Project at UCLA School of Law. She teaches and writes in the areas of human rights, global health, gender, sexuality, HIV/AIDS, and incarceration.

Cherith Sanger is the policy development and advocacy manager at Sonke Gender Justice, a South African based human rights and social justice non-profit organisation. She is an admitted attorney specialising in the areas of women’s rights and gender law and policy.

Dean Peacock is the executive director of Sonke Gender Justice, co-chair of the MenEngage Alliance, and is a member of the United Nations Secretary General’s Network of Men Leaders formed on gender based violence prevention. Peacock is an honorary senior lecturuer at the University of Cape Town’s School of Public Health.

References

Andersson

Cockroft

Shea

(2008). Gender-based violence and HIV: Relevance for HIV prevention in hyperendemic countries of southern Africa. AIDS, 22(Suppl. 4), S73-S86.

Andersson

Ho-Foster

(2008). 13,915 Reasons For Equity In Sexual Offences Legislation: A National School-based Survey In South Africa. International Journal for Equity in Health, 7(1), 20.

Andrias

(1992). Rape myths: A persistent problem in defining and prosecuting rape. Criminal Justice, 7(2), 51-53.

Artz

Smythe

(2005). South African legislation supporting victims’ rights. In Davis

Snyman

(Eds.), Victimology in South Africa (pp. 131-151). Pretoria, South Africa: Van Schaik Publisher.

Artz

Smythe

(2007a). Case attrition in rape cases: A comparative analysis. South African Journal of Criminal Justice, 20(2), 158-181.

Artz

Smythe

(2007b). Losing ground? Making sense of attrition in rape cases. South African Crime Quarterly, 22, 13-20.

Bamford

(2007, September 29). Cop raped me in van, woman says. The Star.

BBC News Online. (1999, January 19). South Africa’s rape shock. BBC News. Retrieved from http://news.bbc.co.uk/2/hi/africa/258446.stm

van der Berg

(2010). Current poverty and income distribution in the context of South African history. Maitland, Cape Town, South Africa: University of Stellenbosch.

10.

Berger

Neuman

W. L.

Searles

(1994). The impact of rape law reform: An aggregate analysis of police reports and arrests. Criminal Justice Review, 19, 1-23.

11.

Braithwaite

V. A.

(2003). Taxing Democracy: Understanding tax avoidance and evasion. Aldershot, UK: Ashgate Publishing, Ltd.

12.

Burgess-Proctor

(2012). Backfire: Lessons learned when the criminal justice system fails help-seeking battered women. Journal of Crime and Justice, 35(1), 68-92.

13.

Campbell

Johnson

(1997). Police officers’ perceptions of rape: Is there consistency between state law and individual beliefs. Journal of Interpersonal Violence, 12(2), 255-274.

14.

Capers

(2011). Real rape, too. California Law Review, 99, 1259-1308.

15.

Daly

Bouhours

(2010). Rape and attrition in the legal process: A comparative analysis of five countries. Crime and Justice, 39, 565-650.

16.

Du Mont

Miller

Myhr

T. L

. (2003). The role of “real rape” and “real victim” stereotypes in the police reporting practices of sexually assaulted women. Violence Against Women, 466-486.

17.

Dunkle

K. L.

Jewkes

R. K

Brown

H. C.

Gray

G. E.

McIntryre

J. A.

Harlow

S. D.

(2004). Gender-based violence, relationship power, and risk of HIV infection in women attending antenatal clinics in South Africa. Lancet, 363(9419), 1415-1421.

18.

Elsenberg. (2005, August). A profile of the Western Cape Province: Demographics, poverty, inequality and unemployment. Provincial Decision-Making Enabling Project Background Paper 1,South Africa: Author.

19.

Eitle

(2005). The influence of mandatory arrest policies, police organizational characteristics, and situational variables on the probability of arrest in domestic violence cases. Crime and Delinquency, 51(4), 573-597.

20.

Frank

Waterhouse

Griggs

Rontsch

(2009). Raising the bar: A review of the restructuring of the SAPS family violence, child protection and sexual offences units. Cape Town, South Africa: RAPCAN.

21.

Gruber

(2009). Rape, feminism, and the war on crime. Washington Law Review, 84, 581-658.

22.

Inman

Rubinfeld

(1979). The judicial pursuit of local fiscal equity. Harvard Law Review, 92, 1662-1750.

23.

Jewkes

Abrahams

(2002). The epidemiology of rape and sexual coercion in South Africa: An overview. Social Science & Medicine, 55, 1231-1244.

24.

Jordan

(2005, August). Rape, reform and resistance. Paper presented at the Fourth Australasian women and policing conference: Improving policing for women in the Asia Pacific region, Australian Institute of Criminology, Darwin, Australia.

25.

Logan

Shanon

Walker

(2006). Police attitudes toward domestic violence offenders. Journal of Interpersonal Violence, 21, 1365-1374.

26.

Majangaza

. (2012, February 10). Residents dispute police claims on rape: Leaches bay neighbours state suspect “was in the shack and had not fled.” Daily Dispatch.

27.

Mbali

(2005). HIV/AIDS and gender-based violent crime in South Africa. In Davis

Snyman

(Eds.), Victimology in South Africa (pp. 263-270). Pretoria, South Africa: Van Schaik Publishers.

28.

Moselakgomo

(2012, March 14). Rape victim gets an apology at last. The Sowetan. Retrieved from http://www.sowetanlive.co.za/news/2012/03/14/rape-victim-gets-an-apology-at-last

29.

National Prosecuting Authority of South Africa. (2010). Annual Report 2009/2010. Pretoria, South Africa: Author.

30.

Page

A. D.

(2008). Judging women and defining crime: Police officers’ attitudes toward women and rape. Sociological Spectrum, 28(4), 389-411.

31.

Pelser

Louw

Ntuli

(2000). Poor safety: Crime and policing in rural areas (Issue 47 of ISS monograph series). Pretoria, South Africa: Institute for Security Studies.

32.

Rasool

Vermaak

Pharoah

Louw

Stavriu

(2003). Violence against women: A national survey. Pretoria, South Africa: Institute for Security Studies.

33.

Reiss

(1984). Consequences of compliance and deterrence models of law enforcement for the exercise of police discretion. Law and Contemporary Problems, 47(4), 83-122.

34.

Robinson

(2000). The effective of a domestic violence policy change on police officers’ schemata. Criminal Justice and Behavior, 47(4), 600-620.

35.

Rumney

(2001). The review of sex offences and rape law reform: Another false dawn? The Modern Law Review, 64, 890-910.

36.

Shukumisa Campaign. (2010). Findings of the shukumisa monitoring of police stations. Cape Town, South Africa: Author.

37.

Skolnick

Bayley

(1993). Above the law: Police and the excessive use of force. New York, NY: The Free Press.

38.

Smythe

Waterhouse

(2008). Policing sexual offences: Policies, practices and potential pitfalls. In Smythe

Artz

(Eds.), Should we consent? Rape law reform in South Africa (pp. 198-223). Cape Town, South Africa: Juta & Co Ltd.

39.

South African Police Service. (2002). Manual of the South African Police Service in accordance with Section 14 of the Promotion of Access to Information Act, 2000. Information Officer of the South African Police Service. Retrieved from http://www.saps.gov.za/docs_publs/legislation/paia/english/part1.pdf

40.

South African Police Service. (2008). National instruction 3/2008: Sexual offences. South African Police Service. Retrieved from http://www.saps.gov.za/docs_publs/legislation/sexual_offences/ni/ni0308_annexure_f.pdf

41.

South African Police Service. (2010). SAPS annual report 2009/2010. South African Police Service. Retrieved from http://www.saps.gov.za/saps_profile/strategic_framework/annual_report/ar_archive.htm

42.

South African Police Service. (2011). Profile: SAPS western cape. South African Police Service.

43.

Statistics South Africa and Human Sciences Research Council. (2007). Using the 2001 census: Approaches to analysing data. Pretoria: Statistics South Africa.

44.

Stemple

(2009). Male rape and human rights. Hastings Law Journal, 60, 605-646.

45.

Vetten

(2011). Paradox and policy: Addressing rape in post-apartheid South Africa. In Westmarland

Gangoli

(Eds.), International approaches to rape (pp. 169-190). Bristol, UK: The Policy Press.

46.

Vetten

Jewkes

Sigsworth

Christofides

Loots

Dunseith

(2008). Tracking Justice: The attrition of rape cases through the criminal justice system in Gauteng. Johannesburg, South Africa: Centre for the Study of Violence and Reconciliation.

47.

Vetten

Leisegang

Haken

(2010). The Right and the Real: A shadow report analyzing selected government departments’ implementation of the 1998 Domestic Violence Act and 2007 Sexual Offences Act. Johannesburg, South Africa: Tshwaranang Legal Advocacy Centre to End Violence Against Women.

48.

Waterhouse

(2008). The Impact of changing criminal justice responses to child victims of sexual abuse. Cape Town, South Africa: Open Society Foundation for South Africa.

49.

World Bank. (2012). South Africa economic update: Focus on inequality of opportunity. Washington, DC: Author.