Skip to main content Accessibility help


  • Access

  • Strong NGOs and Weak States
  • Pursuing Gender Justice in the Democratic Republic of Congo and South Africa
  • Online publication date: May 2018
  • pp 33-65
  • Milli Lake, London School of Economics and Political Science


      • Send chapter to Kindle

        To send this chapter to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle.

        Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Available formats

        Send chapter to Dropbox

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Dropbox.

        Available formats

        Send chapter to Google Drive

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Google Drive.

        Available formats
Export citation

2 Researching Violence, Law, and Human Rights in South Africa’s Western Cape and DR Congo’s Eastern Provinces


One evening, in a bar on the shores of Lake Kivu, a colleague introduced me to a public relations (PR) officer from an international human rights organization. Over a glass of wine, the PR officer told me about her grueling day, which had involved photographing and interviewing a twelve-year-old rape victim for her organization’s new campaign on wartime sexual violence. The day had been frustrating, she explained, not because the story was tragic – although it was. The day had been frustrating because it was her last day filming, and the girl’s story was not quite tragic enough. She told me she had wanted someone younger, and more helpless. Someone who would really move the American public to donate money.

This anecdote was not altogether surprising; I had heard similarly flippant remarks from others, particularly those accustomed to flitting between one war zone and the next. In both South Africa and the DR Congo, researchers, journalists, and campaign staff arrive in search of stories from vulnerable individuals – especially from victims of sexual assault. Many arrive seeking stories of the worst forms of brutality and harm.

In the United States and much of Europe, stringent regulations protect vulnerable populations in their interactions with researchers and journals. While university ethics review boards attempt to moderate the risks posed to academic research subjects, their procedures are often ill suited to environments of ongoing war or violence, where states command little authority or relevance, or where literacy levels are low. This means it is often the responsibility of individual researchers to weigh the value of their research against the risks posed to interviewees, as well as the many complex power imbalances inherent within interviewer–interviewee interactions (Bouka Reference Bouka2015; Cronin-Furman and Lake Reference Cronin-Furman and Lake2018; Mitchell Reference Mitchell2013; Wood 2006).

Nevertheless, the perspectives of the intended beneficiaries of legal aid and gender justice are integral to understanding how law works for those intended to gain from it. And far too many policy interventions in sub-Saharan Africa have failed to take seriously the reflections and responses of those individuals and communities they most affect. I did not want to replicate the tendency to erase and obscure the voices of target recipients.

Given these concerns, I made a number of trade-offs and calculations throughout the course of my research. Where possible, I took the decision to rely heavily on careful firsthand research conducted by other scholars. In South Africa, for example, a number of excellent studies have already featured in-depth interviews with victims of gender-based violence. I felt it irresponsible to replicate similar interviews for the sake of academic propriety. For relevant sections of this book, therefore, I draw heavily on work carried out with victims of violence by Rachel Jewkes, Dee Smythe, Lisa Vetten, and their coauthors, supplementing their research with my own formal interviews and informal conversations with counselors, health workers, family members, community support workers, lawyers, church groups, and first responders. In a very few instances, I also conducted my own interviews with victims of violence. In eastern DR Congo, there was less existing research to draw from. At the time I commenced my research, access to law and justice had been relatively understudied. I thus took the decision to interview a carefully selected sample of individuals affected by violence and involved, in some capacity, in the pursuit of legal remedy.

This chapter discusses many of the challenges I encountered researching sensitive human rights issues in volatile or high-crime environments, as well as how these challenges affected and informed my research. It describes decisions I took in the field and prior to leaving, as well as how the particularities of each site shaped the research I undertook. It begins, however, with an overview of the key characteristics of sexual and gender-based violence in each site and a discussion of why I selected these sites for my research. After discussing the similarities and differences between the scale and nature of violence in each site, I summarize my own research approach.

The Scale and Nature of Violence in DR Congo and South Africa

Feminist scholars of international relations have repeatedly commented on the ways in which conventional understandings of both “stateness” and “security” obscure the specific concerns of women, children, minorities, or otherwise marginalized groups (Baer Reference Baer1999; Butler Reference Butler1990; MacKinnon Reference MacKinnon2006b; Tickner Reference Tickner1997; Tickner and Sjoberg Reference Tickner and Sjoberg2011; Tripp, Ferree, and Ewig Reference Tripp, Ferree and Ewig2013; Towns Reference Towns2010). These populations may remain vulnerable in very particular ways even when a state is deemed strong or secure by conventional measures. Although South Africa is one of the world’s most rapidly industrializing democracies (World Economic Forum 2012), state-centric measures of security and development typically tell us little about the insecurities faced by the country’s most vulnerable inhabitants, nor about the fate of women and men at risk of specifically gendered harms. While DR Congo and South Africa differ significantly on a number of metrics, there are nevertheless a number of informative parallels between the nature and prevalence of gendered violence in each site. These parallels are important for understanding responses to violence in each site, as well as for their role in exposing the deeply gendered assumptions embedded within mainstream security discourse and notions of statehood.

In addition to numerous similarities, citizens of eastern DR Congo face a number of distinct security concerns, stemming from the country’s twenty-year war, which do not affect citizens of South Africa. Some men, women, and children have experienced sexual torture related to the war, or other explicitly gendered conflict violence (Cohen Reference Cohen2016; Meger Reference Meger2010). Yet, in reality, only some of the gendered violence that affects Congolese civilians in the east is a direct product of the war. Much of the violence perpetrated in both public and private spheres resembles the day-to-day violence that afflicts women and men across the country, and elsewhere around the world. To attribute the scale and nature of gendered violence in eastern DR Congo exclusively to the war is theoretically and empirically misinformed. Instead, teasing out the particular forms of violence that are indirectly related to or exacerbated by conflict in eastern DR Congo, from the high rates of intimate partner violence and civilian abuse that are pervasive in South Africa and elsewhere, illuminates the analogous experiences of women and men in both research sites. Studying institutional responses to sexual violence necessitates carefully contextualizing the distinct forms of harm and vulnerability that are present in each case.

DR Congo

The forms of gendered harm felt by women and men in DR Congo and South Africa exhibit a number of similarities, as well as some clear differences. One of the most obvious differences is the presence of ongoing conflict in the DR Congo’s eastern provinces. Many scholars attribute this violence to the mass displacement that followed Rwanda’s genocide and civil war in the early 1990s. In the aftermath of the Rwandan genocide, Rwanda’s unity government feared reprisals from Hutu refugees who had fled Rwanda after the victory of the Rwandan Patriotic Front (RPF) and were living in refugee camps along the Rwanda–Zaire border. Unconvinced that the Congolese government was taking the necessary measures to curb the threat of another genocide, in 1996 a rebel group led by Laurent Désiré Kabila and backed by Rwandan and Ugandan forces led a march on Kinshasa to overthrow President Mobutu Sese Seko. Not long after this group, the Alliance des Forces Démocratiques pour la Libération du Congo-Zaïre (AFDL), took power, Kabila dismissed his Rwandan chief of staff, James Kabarebe, and ordered all Rwandan and Ugandan troops to leave the country. This move angered Kabila’s former Rwandan and Ugandan allies, who assumed they could count on Kabila’s support. In response, a movement known as the Rassemblement Congolais pour la Démocratie (RCD) emerged in the city of Goma to protect the interests of Congolese Tutsis in the eastern provinces. The Rwandan-backed RCD, supported by the Ugandan-backed Mouvement de Libération du Congo (MLC) and a coalition of local armed groups, launched an assault on government forces, instigating a five-year war that involved nine African countries. Kabila was assassinated in Kinshasa 2001, and succeeded in office by his son, Joseph.

While the war officially ended in 2003, and elections won by Kabila in 2006 and 2011, fighting has continued to destabilize the eastern provinces. The RCD officially became a political party in 2003; however, former RCD affiliates have periodically defected from the national army to create new insurgencies – most notably in the form of the Tutsi-dominated Congrès National pour la Défense du Peuple (CNDP) and, later, the M23. During their height, both the CNDP and the M23 controlled sizeable territories in North Kivu. In addition to the periodic resurrection of RCD elements, dozens of other fragmented and localized armed groups, including various Mai Mai self-defense militias, and the Forces Démocratiques de Libération du Rwanda (FDLR, a group largely comprised of so-called Hutu génocidaires who fled Rwanda in 1994) continue to control sizeable territories in the Kivus. These groups engage in continued negotiations and armed struggles over land, mining sites, administrative positions, access to infrastructures, and local authority.

Both during the war and in the years since its apparent conclusion, violence against civilians has been pervasive. The FDLR, in particular, numbering approximately 8,000 troops after the war officially ended, has been notorious for brutal acts of torture and sexual assault inflicted on local populations. The Congolese government, in partnership with MONUSCO, attempted to incentivize FDLR combatants to repatriate to Rwanda. Given the stigmatization in Rwanda at the time of anyone associated with the Interhamwe, many preferred to remain in DR Congo, prompting the Congolese government to launch an offensive, Umoja Wetu, against the FDLR in 2008. The FDLR ramped up its hostilities to coincide with the offensive, abducting civilian hostages to use as human shields and executing them when the FDLR’s headquarters in Kibua were attacked (US Department of State 2009: 161). Such atrocities were not uncommon; in 2009 alone, the FDLR was accused of burning the village of Mianga to the ground and murdering forty-one villagers, including the village chief. The FDLR has routinely been implicated in the abduction of sex slaves, the rape, mutilation, and torture of men, women, and children, and the burning of thousands of houses across the Kivus.

While violence perpetrated by the FDLR has received the most notoriety and attention, other armed groups have been implicated in similarly grave atrocities. Myriad armed groups have exacted violence against civilians, engaged in pillage and predation, or abducted men, women and children for sexual exploitation, labor or to deploy in combat (Baaz, Stearns, and Verweijen Reference Baaz, Stearns and Verweijen2013). Mass rapes and opportunistic assaults have been perpetrated with alarming frequency. Multiple integrations, defections, and reintegrations into the Congolese national army have meant that FARDC forces, as well as Congolese police officers and even armed civilians, have been responsible for committing acts of violence against unarmed civilian populations over the years since the war’s formal conclusion. A culture of impunity exacerbated by weak statehood has meant that the postwar period in the Kivus has been characterized by protracted insecurity and cyclical outbreaks of low-intensity conflict. A variety of different armed actors have resorted to increasingly violent behaviors in order to secure positions of influence, or to call attention to their political struggles (Autesserre Reference Autesserre2012; Human Rights Watch 2012).

In addition to contributing to violence perpetrated by armed groups, decades of exploitative relationships between those in positions of power and some of DR Congo’s most impoverished and vulnerable men have contributed to what many have termed a crisis of masculinity. Scholars and activists have interpreted the violence employed by soldiers, police officers, and ordinary civilians as a means of regaining a sense of power and control (Baaz et al. Reference Baaz, Maria Stern and Afrikainstitutet2010; Baaz and Stern Reference Baaz and Stern2011; Sonke Gender Justice Network 2012). This logic, combined with a long-standing societal tolerance for nonconsensual sex, has compounded problems of sexual and gender-based violence perpetrated in the context of the conflict, as well as in the home. While the vast majority of media attention to gender-based violence in eastern DR Congo has focused on the war, familial violence, violence in churches and schools, and other forms of nonconsensual sex are equally widespread.

Reliable statistics on the prevalence of rape and sexual violence are notoriously difficult to obtain, since the vast majority of incidents goes unreported. Still, a number of researchers have made efforts to arrive at reliable statistics that take into account reporting challenges and poor infrastructure. For example, a Demographic and Health Survey carried out among households in eleven Congolese provinces in 2007 found that almost ten percent of women reported that their first sexual encounter was against their will (Demographic and Health Survey 2007), with sixteen percent reporting sex against their will at some point in their lives. A separate survey – a randomized household survey conducted in sixty-seven villages in three eastern provinces in 2010 – placed this figure much higher, finding that 39.7 percent of women interviewed (and 23.6 percent of men) had experienced sexual violence at some point in their lives (Johnson et al. Reference Johnson, Scott, Rughita, Kisielewski, Asher, Ricardo and Lawry2010). The 2014 Demographic and Health Survey found that twenty-seven percent of women between the ages of fifteen and forty-nine reported that they had experienced sexual violence at some point in their lives (Demographic and Health Survey 2014).

In addition to random and opportunistic rapes perpetrated by soldiers and civilians, sexual violence has also been systematically employed as a weapon of war by a number of armed groups operating in the eastern provinces, including the government forces. Although rape has been used as an instrument of conflict for centuries to terrorize populations, torture or punish civilians, reward soldiers, and, in extreme cases, physically and psychologically prevent women from procreating, many have commented that its use as a targeted military strategy in DR Congo’s wars has been unmatched in recent conflicts (Meger Reference Meger2010, Reference Meger2011).

Surveys in DR Congo have reported varying responses to questions about wartime rape. For example, in Kirsten Johnson and her colleagues’ Reference Johnson, Scott, Rughita, Kisielewski, Asher, Ricardo and Lawry2010 household survey conducted in the eastern provinces, of the 39.9 percent of respondents who reported they had experienced sexual or gender-based violence at some point in their lives, 74.3 percent of women (and 64.5 percent of men) reported that the rape or sexual assault they experienced was conflict-related, while only thirty-one percent attributed violence to an intimate partner (Johnson et al. Reference Johnson, Scott, Rughita, Kisielewski, Asher, Ricardo and Lawry2010: 557).Footnote 1 Susan Bartels and her colleagues (Reference Bartels, Scott, Leaning, Mukwege, Lipton and VanRooyen2013) support the claim that the majority of sexual assaults is perpetrated by armed combatants. However, the 2007 Demographic and Health Survey revealed the frequency of civilian rather than military rape. In their countrywide analyses, Peterman, Palermo, and Bredenkamp (Reference Peterman, Palermo and Bredenkamp2011) found that approximately twice the number of women who admitted a history of rape reported that they had experienced sexual violence at the hands of a partner (221 per every 1,000 women of reproductive age, compared to only 121 reporting any history of rape). Further, rates of rape and intimate partner violence in North and South Kivu were comparable to many other parts of the country, including those where conflict is no longer endemic. This suggests that the ongoing conflict in the east has not necessarily played a formative role in shaping the scale or nature of violence there. While North and South Kivu experienced some of the highest rates of sexual violence, reporting in Equateur, Kasai, Bandudu, and Maniema was also high, and even higher when accounting for other forms of intimate partner violence (Peterman, Palermo, and Bredenkamp Reference Peterman, Palermo and Bredenkamp2011: 1064).Footnote 2

International organizations and government agencies such as the Ministry of Gender, MONUSCO, the United Nations Population Fund (UNFPA) have also devoted considerable time and resources to monitoring and documenting a variety of health- and population-related statistics, including incidents of sexual and gender-based violence. Under United Nations Security Council Resolution (UNSCR) 1960 on Women, Peace and Security, the UN Security Council mandated the improvement of data collection and analysis of incidents, trends, and patterns of rape and other forms of sexual violence.Footnote 3 UN Action supported the monitoring, analysis and reporting arrangements (MARA) in four priority countries, including DR Congo, and attempted to systematize MONUSCO’s approach to the collection of sexual violence statistics (MONUSCO 2014; MONUSCO and UN-OHCHR 2014: 19). MARA has built on efforts already underway by other UN agencies and government ministries, as well as local organizations and health clinics. For example, in its analysis of data collected from health care centers in seven provinces in 2011 and 2012, UNFPA found that fifty-seven percent of all reported rapes or sexual assaults were perpetrated by strangers, with only forty-three percent perpetrated by assailants known to the victim (UNFPA 2013). The vast majority of all attacks (eighty-two percent) was committed by perpetrators wearing civilian clothes. Moreover, the UNFPA study showed that only twenty-three percent of cases registered affected displaced persons; the majority of victims – even in the high-conflict provinces of North and South Kivu – were not displaced by the conflict, but remained within their domestic environments. The average age of victims in all provinces and all studies was under thirty. The Ministry of Gender (MinGenre) has engaged in similar efforts at data collection in collaboration with the UNFPA’s Data and Mapping Project, collating sexual violence reporting statistics from health centers across the country (MinGenre 2013).

It is important to take into account the reporting challenges associated with these various figures. While many victims of violence face incentives to hide the incidents from their families and communities, they are more likely to come forward to report a sexual assault if they become pregnant, need medical attention, contract HIV/AIDS, or have been abandoned by family members. Since rapes by soldiers are more likely to be violent, such cases might motivate increased reporting, particularly in medical facilities. Where rape and sexual assault are committed within the context of a partnership, and particularly where they do not result in visible physical harm, we likely know far less about them.

In spite of these barriers, qualitative and quantitative data suggest that a considerable proportion of all sexual and gender-based violence occurs not on the battlefield, but in the context of interpersonal relationships – between couples, neighbors, family members, students and teachers, or in other familiar settings. It is incorrect to assume that wartime rape is the most prevalent form of violence.Footnote 4 However, it is also necessary to recognize that civil conflict can compound and heighten problems of non-conflict violence, leading to its escalation in all spheres of life (Clark et al. Reference Clark, Everson-Rose, Suglia, Btoush, Alonso and Haj-Yahia2010; Cockburn Reference Cockburn, Giles and Hyndman2004; Horn et al. Reference Horn, Puffer, Roesch and Lehmann2014; Wolfe Reference Wolfe2014).

South Africa

Like DR Congo, the vast majority of gender-based violence in South Africa takes place in the context of intimate partner relationships (Dunkle et al. Reference Dunkle, Jewkes, Brown, Yoshihama, Gray, McIntyre and Harlow2004). Yet South Africa has also received considerable media attention, both domestically and internationally, to its high rates of opportunistic, brutal, and public sexual assaults. These assaults are more common in high-crime urban settings, although they have been common in rural areas as well (Human Rights Watch 1995, 1997, 2001, 2011).

South African health researchers have engaged in various efforts over the years to determine the prevalence of rape. Early efforts revealed that four percent of women nationwide had been forced to have sex against their will, and seven percent of women who had ever had sex had been either forced or persuaded (Department of Health 1998). Drawing from specialized sexual violence data collected in the Eastern Cape, Mpumalanga, and Northern Province, Jewkes and Abrahams (Reference Jewkes and Abrahams2002: 1232) found that, for all incidents of rape reported across the three provinces, forty-three percent were committed by strangers, twenty-one percent by acquaintances, nine percent in school, nine percent by relatives, eight percent by partners, and eleven percent by others (Jewkes and Abrahams Reference Jewkes and Abrahams2002). Other studies found that rapes are far more likely over the weekend, particularly between the hours of 6:00 PM and 10:00 PM, when people had been drinking and were most commonly perpetrated in open spaces such as fields (thirty-one percent), in the rapist’s home (twenty-nine percent), or in the victim’s home (fourteen percent) (Jewkes and Abrahams Reference Jewkes and Abrahams2002; Swart et al. Reference Swart, Gilchrist, Butchart, Seedat and Martin2000). Early studies suggested high rates of gang rape and stranger rape as compared to intimate partner violence. This presumably derives from the fact that incidents of stranger rape, and assaults resulting in severe physical injury or requiring urgent medical attention, are reported with greater frequency than sexual assaults perpetrated in the home or leaving no outwardly visible trace.

More recent studies that often adopt more nuanced or careful approaches to data collection paint a slightly different picture. In 2014, the Western Cape Gender-Based Violence Indicators Study (GBVI) revealed that thirty ninepercent of all women surveyed had experienced some form of gender-based violence in their lifetime, and the same proportion of men in the household survey admitted to perpetrating violence. Forty-four percent of the women who had experienced some form of gender violence had experienced violence at the hands of an intimate partner: Forty percent of this violence was emotional, twenty-five percent was physical but nonsexual, thirteen percent was economic, and thirteen percent was sexual (Gender Links 2015). A further seven percent of all women surveyed reported that they had experienced rape at the hands of a non-partner, and six percent reported that they had experienced other forms of non-partner sexual harassment.Footnote 5

The most recent Demographic and Health Survey in South Africa (South Africa Department of Health 2017) included anumber of questions on sexual and gender-based violence. The survey is one of the few nationally representative studies. It revealed that twenty-one percent of ever-partnered women had experienced some form of physical violence at the hands of a partner at some point in their lifetime, whereas six percent of ever-partnered women had experienced sexual violence at the hands of a partner at some point in their lifetime. The survey did not pose questions on non-partnered violence. Interestingly, the Western Cape reported among the lowest rates of sexual violence of all the South African provinces, with Northwest Province reporting the highest (South Africa Department of Health 2017).

This research raises a number of questions about the prevalence and reporting of sexual and gender-based violence in South Africa. The South African Police Service (SAPS) itself has estimated that only 2.8 percent of all rapes that occur in the country are ever reported to police or to health care professionals (South African Police Service 2014). More recent studies, including the Western Cape GBVI Study, place this figure much lower, with the Western Cape GBVI Study suggesting that only 1.6 percent of all women who had ever experienced rape reported the incident to the police, and 2.6 percent of all those experiencing other forms of nonsexual physical violence (Gender Links 2015: 54). In 2014, South African Police Service (SAPS) National Crime Statistics reported 43,219 incidents of rape that were reported to the police across the country from 2013 to 2014 (South African Police Service 2014). These figures show a decrease in the number of reported rapes of about six percent from previous years, and a decrease in the number of reported sexual assaults by five percent from previous years. However, like DR Congo, the vast majority of rapes goes unreported, and those that take place in the context of intimate partnerships or by assailants known to the victims are among the least likely to be reported to police. Survey respondents may also be unwilling to admit to intimate partner violence, even in the context of anonymous surveys. It remains perpetually unclear, therefore, whether low or varied rates across different surveys indicate varying rates of violence; variation in reporting; or variation in the survey instruments employed and how questions are posed to respondents.

To correct or corroborate reporting bias and fill in gaps in existing research, Jewkes and colleagues (Reference Jewkes, Sikweyiya, Morrell and Dunkle2011) have attempted to assess the prevalence of rape through surveys with male rather than female respondents. Through a survey of 1,686 men, Jewkes and colleagues found that thirty-three percent of respondents admitted to raping or attempting to rape a woman, whether an intimate partner, stranger, or acquaintance. Almost half of the thirty-three percent said they had either participated in a gang rape or assisted in a rape, and more than half reported having been violent on multiple occasions. The survey revealed that men admitting to rape did not significantly differ from others according to age or educational attainment, but were more likely to be cohabiting with an intimate partner, less likely to be married, and more likely to have multiple sexual partners. Moreover, there were some differences in the gender attitudes held by those admitting to rape, who were generally more adversarial in their views about women. Reported motivations usually related to anger or a sense of sexual entitlement.

Given underreporting, qualitative work has sought to fill gaps in scholarly knowledge. In this context, researchers have highlighted the significance of nonviolent coercion in marital or seemingly consensual relationships. Coercion may include implicit or explicit threats of physical violence, locking the door until a woman agrees to sex, or threatening economic insecurity if a woman does not do as her partner wishes. Coercion may also be persuasive, with boyfriends or partners using phrases such as “sex strengthens our love,” or “in relationships people must make sacrifices” (Wood, Lambert, and Jewkes Reference Wood, Lambert and Jewkes2007: 288). Understanding the ways coercion can function within relationships nuances how we understand responses to questions about forced sex, consent and non-consent. Jewkes and colleagues (Reference Jewkes, Penn-Kekana, Levin and Schrieber1999, Reference Jewkes, Sikweyiya, Morrell and Dunkle2011) have observed that coercion should be interpreted according to a culture of male sexual entitlement, which has been reinforced over time through various social and cultural institutions. A study in the Eastern Cape (Jewkes et al. Reference Jewkes, Penn-Kekana, Levin and Schrieber1999) revealed that seventy-nine percent of female survey respondents believed that if a man paid lobola (bridewealth) for his wife, she was obligated to satisfy his sexual needs whenever he demanded. A further sixty percent reported that they did not believe that a married woman could refuse sex with her husband. Additionally, the belief that if a girl accepts a boyfriend’s proposal to “love” or to be in a relationship, then she is obliged to have sex whenever her boyfriend wants, is prevalent among young South African women. Often sexual intercourse is expected in return for a combination of gifts, money, or being taken out (Wood and Jewkes Reference Wood and Jewkes1997, Reference Wood and Jewkes1998). A woman or girl’s refusal to have sex undermines this informal contract and challenges dominant norms about male sexual entitlement. These beliefs remain prevalent among youth across South Africa, in both rural and urban areas, and parallel attitudes are reported in similar studies in eastern DR Congo (see, e.g., Sonke Gender Justice Network 2012).

Sexual Entitlement and Transactional Approaches to Sex

Many scholars have written about the transactional approaches to sexual interaction that characterize male–female relationships in societies around the world. In both South Africa and DR Congo, women entering into romantic partnerships can face deeply gendered expectations vis-à-vis their relationship roles. Men may be expected to bear the costs of gifts, clothes, meals, and other items for their female partners. In return, women are often expected to satisfy their partners’ romantic needs and desires, while deprioritizing their own (Biddlecom Reference Biddlecom2007). Many (but not all) of my female interviewees and acquaintances portrayed their romantic relationships in this way when the subject arose. Despite social taboos concerning sex before marriage, women often entered into relationships with an understanding that they were obliged to fulfill their partners’ sexual needs, whether or not they wanted to do so. In order to fulfill their roles in the partnership, many unmarried, partnered (and often younger) female interviewees in both research contexts upheld that they must consent to sex whenever their partners wished because they were the recipients of clothes, school fees, or other gifts. Among married women, many communicated an understanding that consent to intercourse was given upon entering into the marriage, and not obtained for each sexual encounter.Footnote 6 In their study of transactional sex in a South African township, Wood, Lambert, and Jewkes (Reference Wood, Lambert and Jewkes2007: 293) note: “Sexual consent and sharing beer went hand in hand. A woman who drank and who did not subsequently fulfill her sexual obligation was effectively thought of as having committed theft, and this was humiliating to the man who funded her ‘enjoyment.’”

These dynamics are further complicated by a widely held social expectation in both research sites for women to demonstrate that they are not promiscuous. Marriage customs and rituals around the world capture ideas about promiscuity differently, intersecting with consent in potentially challenging ways. One Congolese marriage ritual, for instance, involves a performance of capture: the woman is expected to flee from her partner on her wedding night and feign a show of resistance while the groom engages in her pursuit.Footnote 7 In South Africa, the practice of ukuthwala (bride capture) displays similar elements, wherein the bride is captured by her prospective groom and his party and taken to be married, sometimes in secret. Versions of this ritual, practiced around the world, are intended to demonstrate that the bride is pure and chaste, and does not exhibit unwanted characteristics of sexual desire, depravity, or promiscuity. In the words of one interviewee, such practices inevitably lead to the “performance of rape being normalized in the context of marriage.”Footnote 8

If a woman consents to sex too readily, she might be perceived as “easy” or “promiscuous,” which is undesirable to many men. As a result, women and girls in DR Congo and South Africa – as in many other places around the world – face social pressure to resist sexual advances, at least at first, even in the context of genuinely consensual interactions. In her study of sexual consent in the Transkei, Marston (Reference Marston2005: 79) observes similar patterns of behavior. She writes: “if women are expected to resist sex, then being coerced is the socially ‘correct’ way to consent to sex.”

This has led to a widespread perception among many men that saying “no” to sexual intercourse really means “yes.” This perception extends far beyond DR Congo and South Africa, to communities across Europe and the United States (Muehlenhard and Hollabaugh Reference Muehlenhard and Hollabaugh1988). It leads many men (and some women) to claim difficulty in understanding the nuances of sexual consent, particularly in legal terms. To compound these dynamics, many men (and women) in both DR Congo and South Africa share the opinion that “not taking no for an answer” demonstrates a man’s love for a woman. If the suitor were to give up, it might signal that he was not interested or committed to his female partner. Wood and colleagues (Reference Wood, Lambert and Jewkes2007: 292) quote a young man who articulated this sentiment: “I’m talking about gentle force, holding her hands and lying on top of her and kissing her, and then I do my thing. I can call it forced sex [not rape] because I’m trying to be romantic. You are not rushing to penetrate, you are rushing to put her in the mood.”

These attitudes are crucial for understanding dynamics of violence and consent in South Africa, and are captured in the fact that the Xhosa language uses different words for rape committed by strangers and by partners (Wood et al. Reference Wood, Lambert and Jewkes2007).Footnote 9 This linguistic distinction hints at the sociocultural approaches underpinning consent and coercion within and outside relationship contexts. For instance, ukulala ngekani (to sleep with by force) or ukunyanzela (to force) are each used to describe unwanted intercourse within sexual partnerships, whereas ukudlwengula refers to forced sex by a non-partner or stranger (Wood et al. Reference Wood, Lambert and Jewkes2007: 277). Many men (and some legal practitioners and even judges) do not believe that a woman has been forced or coerced into sex if physical violence has not been used. Thus, accusations of rape that revolve around implicit coercion are often rejected both by society at large and by those in law enforcement infrastructures such as police, prosecutors, and judges.

Despite the fact that intimate partner violence likely constitutes the most pervasive form of sexual assault in both sites, it is the practice of gang rape, or “jack-rolling,” that has received unprecedented media attention. In the townships of Khayelitsha and Gugulethu, reports of women abducted at gunpoint or knifepoint while they were waiting for buses, walking home from school, using public restrooms, and other such activities were not uncommon. Across the country, the phenomenon of “corrective rape” has been used to target individuals with ambiguous sexualities, those who challenge conventional gender norms, and women who are thought to dress too provocatively (Human Rights Watch 2011; Lock Swarr Reference Lock Swarr2012). One interviewee captured a common sentiment:

Women here in South Africa want to be Western. They wear Western clothes, like short, short skirts. When a man sees that, he has to do something about it. He has to teach her that that is not an African way to dress. By wearing those clothes she tells him that she wants to have sex.Footnote 10

Criminologists have argued that motivations toward gang rape differ from motivations toward single-perpetrator rape, since gang rape can involve male bonding through watching and taking turns (Cohen Reference Cohen2013; Cohen et al. Reference Cohen, Green and Wood2013; Holmstrom and Burgess Reference Holmstrom and Burgess1980). In South Africa, this has taken on the additional dimension of “teaching” women lessons about how to behave and conform to gender norms (HRW 2011).

In addition to teaching women lessons, rape in South Africa cannot be delinked from a broader culture of rising levels of violent crime. One nurse interviewed by Wood and Jewkes in an early study (Reference Wood and Jewkes1998: 18) claimed:

Ngangelizwe [neighborhood] is very bad, it’s better now but it was dirty, full of running water, full of plastic, the gutters are open, and it is congested. You have seen the housing, full of over-crowding, dagga smoking people there don’t have the money to pay the rent sometimes. They want a TV, so they go and steal and buy dagga and smoke in gangs, they go about in the streets and if they see a girl they catch her and rape her.

Many interviewees referenced a culture of entitlement – “taking what’s yours,” “taking what you want,” or taking what people think society “owes” them to compensate for past injustices – as the primary reason that rape is so widespread. In Jewkes and colleagues’ Reference Jewkes, Sikweyiya, Morrell and Dunkle2011 study, men who had engaged in acts of rape were more inclined to present themselves as victims, expressing that they had not attained what they felt they deserved or were owed from life. The rhetoric of entitlement has also been invoked by soldiers in eastern DR Congo. When they have been in the bush for extended periods of time, many soldiers report that when they see a woman, they are entitled to take her by force, as compensation for all the work they have done fighting for their country or their armed group (Baaz et al. Reference Baaz, Maria Stern and Afrikainstitutet2010; Baaz and Stern Reference Baaz and Stern2008).

Research also shows that men who have been exposed to trauma in their early years, or who suffer from low self-esteem, seek validation and positive reinforcement from peers who engage in physically aggressive behavior. The physical expression of power through (often sexualized) violence represents what Jewkes and colleagues refer to as “exaggerated masculinity” (Reference Jewkes, Sikweyiya, Morrell and Dunkle2011), which allows young males to exert dominance and control in ways that they cannot do in other areas of their lives. I return to the ways in which patriarchal gender hierarchies have been reinforced in the household and in local and national politics in Chapter 4.

Historical Legacies of Violence

Decades of state-sponsored or state-sanctioned violence in South Africa and eastern DR Congo have resulted in a climate in which physical violence has been normalized as a response to dispute or discord (Bourgois Reference Bourgois2001; Godoy Reference Godoy2006 for discussions of this phenomenon in other contexts). Commanding respect, obedience, or authority through the physical expression of force or violence was an everyday feature of colonial and apartheid regimes. The subjugation of black men under apartheid (and similarly in the Belgian Congo) created what Wood (Reference Wood2005) and others have referred to as a crisis of masculinity (Walker Reference Walker2005). In South Africa and elsewhere, scholars have used this trope to refer to a sociopolitical environment wherein the systematic disenfranchisement of generations of working-class men, profoundly disadvantaged by the migrant labor economy, meant that the family domain became the primary sphere to assert power and dominance and to reclaim a waning sense of masculinity. The use of force against those less powerful, in the context of sociopolitical interaction or in the home, is thus unsurprising. Neither is it surprising that sexually coercive or violent behavior has remained inextricably bound up with postcolonial and postapartheid constructions of masculinity.Footnote 11

Given disparities in survey instruments and reporting mechanisms, as well as stigma surrounding sexual violence reporting, it is difficult to conjure a full picture of the scale and nature of sexual and gender-based crimes. What we can glean from existing research; however, is that, despite the two countries’ differences, many women and girls in eastern DR Congo and South Africa, and some men and boys, face similar threats of gendered violence, and equivalent barriers to reporting. The scale and nature of violence and insecurity felt by those I was working with, as well as the parallels between the types of gendered violence experienced across the two research sites, profoundly shaped the direction of my research. Importantly, the nature of gendered harm faced by women (and some men) across DR Congo and South Africa, and the ways in which threats to gender security have structured reporting, investigation and prosecution, made a comparison of the evolution of the legal and institutional responses to violence in the two sites particularly illuminating. The fact that legal responses in eastern DR Congo have often transcended deeply entrenched sociocultural gender norms concerning male entitlement and transactional sex, which have continued to constrain legal responses in South Africa, provided the basis of my puzzle. The ways in which similarities and differences between the dynamics of violence in each site structured my research approach is the subject of the next section.

Conducting Research on Law and Violence: The Kivus and the Western Cape

Both South Africa and DR Congo have been sites of intense and targeted advocacy over the past decade, in which international attention has been directed toward sexual and gender-based crimes. The fact that advocacy efforts in each site have been similar, but have engendered such divergent institutional responses, makes their side-by-side analysis informative.

Since I sought to understand how efforts to promote legal accountability for violent sexual crimes have been experienced differently across the two research sites, it was necessary to immerse myself, to the extent possible, in the processes and interactions through which law and violence have been produced over time. It was beyond the scope of this project to fully explore subnational variation. Instead, I focus on DR Congo’s eastern provinces and South Africa’s Western Cape over the past two decades as sites of some of the most targeted domestic and international gender advocacy. Relatedly, North and South Kivu and the Western Cape each house important justice sector institutions and large numbers of domestic and international NGOs.

The research that informed this book was carried out over multiple trips to both countries between 2008 and 2016. Between May 2012 and October 2013, I conducted approximately thirteen months of field research, eleven of which were spent in the eastern provinces of DR Congo and two in South Africa. I undertook subsequent research trips to DR Congo in 2013, 2014, and 2016, and I conducted supplementary archival research using South African court documents in 2016 and 2017. This research built on prior visits to both countries in my earlier academic and professional work, as well as many follow-up skype, email, and phone conversations in both countries.

Although I have undertaken a number of research trips to South Africa, I spent the majority of my in-country research time in the eastern Congolese provinces of North and South Kivu. The decision to focus my research efforts in eastern DR Congo was informed by a variety of factors. Most importantly, for all the development aid that has been devoted to legal capacity building in eastern DR Congo, at the time of my research there was no systematic record of cases heard by the Congolese legal system and no easy access to judgements or decisions.Footnote 12 Many judgments were never written up, and many of those that were had been handwritten and were only available in hard copy. Collecting reliable data therefore meant traveling to each court in person to examine court records and to obtain documents, as well as to monitor and observe proceedings firsthand. To supplement my legal analysis, I spent extended periods of time observing the activities of legal aid clinics, police stations, courts, and NGOs.

In the Western Cape, on the other hand, a large body of existing research was, at the project’s inception, already devoted to the treatment of sexual and gender-based violence by the country’s justice system. I was able to draw from research carried out by the Women’s Legal Centre, as well as by Lillian Artz, Rachel Jewkes, Lisa Vetten, and Dee Smythe, among others (e.g., Artz Reference Artz2003; Artz and Smythe Reference Artz and Smythe2005, Reference Artz and Smythe2007, Reference 273Artz and Smythe2008; Jewkes et al. Reference Jewkes, Christofides, Vetten, Jina, Sigsworth and Loots2009). To complement existing scholarship, I undertook original in-country fieldwork in the Western Cape in 2013 to analyze the legal treatment of sexual and gender-based violence in two courts: Khayelitsha Magistrates Court and the High Court of Western Cape. This research, discussed more fully in what follows, built on in-depth archival analysis of South African court documents, police records, and case files, as well as prior visits and ongoing communications with various South African scholars, legal practitioners, and human rights activists.

The arguments presented here are thus drawn from 193 formal in-depth interviews conducted over the past eight years, as well as countless informational interviews, conversations, and participant observations in both countries. Informally, my knowledge of the subject matter has been informed by approximately 150 supplementary interviews I have undertaken on related subject matters in association with other projects. For this book, I collected and analyzed approximately 400 judgments and case files from Congolese and South African courts.

The Ethics of Interviewing Amidst Violence

An issue that warrants careful attention here is my interaction and engagement with victims of violence. In order to develop a holistic and accurate portrayal of the repercussions of legal accountability and human rights advocacy, it was necessary to incorporate the perspectives of those most affected by violence into my analysis. In addition to specifically gendered harms, however, the legacies of war and apartheid meant that many of my interviewees and interlocutors had also experienced other forms of state and non-state violence firsthand.

One of the central questions my research sought to answer was how legal outreach, accountability efforts and access to justice programming have been implemented in DR Congo and South Africa respectively. Answering this question required understanding how victims of violence have responded to legal interventions and to prospects and opportunities (or lack thereof) for formal justice. This includes how they assessed the options available to them in the aftermath of gender-based assault and why they chose – or rejected – courts or formal legal processes as avenues of redress. I was also concerned with understanding how victims of violence interacted with formal law enforcement mechanisms in both countries, and how the work of domestic and international human rights advocates shaped these interactions.

In both sites, although more so in DR Congo, I encountered individuals in urgent need of medical attention or protection. I met women and men who had been gang raped by soldiers and had nowhere safe to seek shelter. I met young children who were pregnant, HIV positive, and rejected by their families. I frequently met women who had been robbed of all their worldly possessions and separated from their families, who asked me for advice and guidance on what they should do next.

Many scholars and researchers have battled with the ethical dilemmas associated with interviewing vulnerable populations (Clark Reference Clark2007; Cronin-Furman and Lake Reference Cronin-Furman and Lake2018; Fujii Reference Fujii2012; Wood 2006). The question of when and how to incorporate the perspectives of vulnerable or violence-affected individuals and communities is especially troubling for academic researchers, given that the benefits of academic research are rarely – if ever – visible to interviewees. Where benefits do exist, academic studies do not typically result in the much-needed food distribution, financial aid, or medical care that follow research missions undertaken by aid agencies. It was difficult to alert potential interviewees – particularly in rural parts of eastern DR Congo where education levels were low – to the difference between the kind of scholarly research I was undertaking, and the outreach and development assistance provided by the humanitarian and NGO communities they were accustomed to seeing. For some, the mere presence of a white woman in their village generated hopes of development assistance. And although vulnerable populations have frequently been photographed for human rights campaign materials (Graham Reference Graham2014) or questioned by aid organizations about their needs and experiences (in the documentation of human rights abuses or the design of aid distribution programs), few have witnessed tangible improvements to their lives as a result of these interactions.

These dynamics presented a sequence of ethical dilemmas. I was not confident that my informed consent process could fully persuade prospective interviewees facing extraordinary hardship that speaking to me would not result in material benefit. My ethics review board at the University of Washington prevented me from compensating my interviewees financially for their time. I worried that some victims of violence might consent to speak to me, not because they understood something about my research and were genuinely willing to assist me, but because they believed, despite my assurances to the contrary, that assistance of some form might result (Cronin-Furman and Lake Reference Cronin-Furman and Lake2018).

I weighed these questions carefully. I did not wish to write the perspectives of the intended beneficiaries of legal aid and gender advocacy programs out of my scholarship. And indeed, any satisfactory analysis of how law and legal accountability work in practice necessitates a bottom-up approach that centers the experiences of those most marginalized by existing injustices. Yet I did not wish to create expectations, raise false hopes, exacerbate insecurities, or place any additional burdens on those facing hardship or harm.

In the Western Cape I was fortunate enough to be able to draw from outstanding research conducted by a number of South African scholars. Given the quality, sensitivity, and rigor of this work, in order to augment my understanding of victims’ experiences with the legal system I needed to conduct only a small number of my own interviews with individuals who had interacted with legal institutions after experiencing violence. In eastern DR Congo, preexisting firsthand scholarship on the experiences of victims of violence with the law was less readily available; in order to answer my questions well, I needed to rely more heavily on firsthand research. I approached interviewing victims of violence and participants in legal processes carefully. I was fortunate enough to work with a number of excellent and incredibly sensitive local research partners who did their best to identify potential challenges in advance of my interviews and to ensure that my interviewees understood as much as possible about the nature of my research before they agreed to meet with me. I made every effort not to interview minors, and I was careful not to interact with anyone who would be endangered in any way by meeting with me.Footnote 13

In addition to the danger of raising expectations in my research, I was also concerned about trauma and revictimization. While my focus was always on the legal process, and not on the incident of violence, many interviewees volunteered to recount their histories to me, which were always painful to hear and must have been incredibly painful to tell. I stopped my interviews whenever interviewees appeared visibly distressed by the subject matter, and I employed various strategies to ensure that my research did not raise false expectations or inconvenience my research participants in any way. Despite these precautions, I was still constantly forced to reflect on whether the value of my research outweighed the time and emotional investment required of those who were generous enough to share their stories with me. The perspectives of these interviewees, and further details of the decisions I made surrounding them, are discussed in Chapter 5 and in Appendix B.

DR Congo

In DR Congo, the Kivus represent a crucial case for any study of legal and advocacy responses to gender violence. The country’s eastern most provinces have been the focus of considerable international attention since the war broke out in 1996. Additionally, the civilian and military courts in the provinces of North and South Kivu have produced the vast majority of sexual and gender-based violence rulings across the country. Given that international and domestic attention and advocacy around sexual and gender-based violence has been focused on North and South Kivu, this is perhaps unsurprising. However, in order to fully understand how and why activists have been effective in directing the attention of local courts toward sexual and gender-based crimes, it was necessary to scrutinize the local courts that have been responsible for producing these decisions.

Most of my work was carried out in the provincial capitals of Goma and Bukavu where the judicial institutions of North and South Kivu are headquartered. However, my research also involved extended travel to remote rural locations to observe trials, collect decisions, and conduct interviews. I interviewed lawyers, prosecutors, military and civilian judges, police officers, representatives from domestic and international NGOs, international organizations, members of local civil society groups, members of the Congolese national army, defendants in human rights trials, members of armed groups, and victims of violence who had attempted to access formal justice mechanisms. In addition to interview data, I monitored military and civilian court cases from their points of entry into the legal system, observed a large number of criminal trials, and traced a subset of legal processes involving gender violence since 2006. I conducted interviews in French and English and, where necessary, I relied on the assistance of interpreters to translate from local languages such as Kinande, Kiswahili, Kinyarwanda, Lingala, and Moshi. I supplemented interview and observational data with the collection of court cases and analyses of legal processes.

Eleven months of research allowed me to become familiar with the aspects of “everyday” life relevant to my research. My immersion in day-to-day life in the Kivus also enabled me to build relationships and develop trust. Combining different research techniques proved critical for my understanding and interpretation of the data I collected. For example, given the deeply unpredictable nature of conducting research in eastern DR Congo, I was constantly forced to reevaluate my approach and to reassess what kind of data were needed, which methods were appropriate for answering which questions, and what would be possible given the evolving security situation. Two incidents in the early phases of my research profoundly affected my research approach and the subsequent decisions I made in the field. In stark contrast to the formal record keeping that is standard practice in South Africa, courts in eastern DR Congo typically lack the technological capacity to record activities and keep track of cases. Thus, the simple task of knowing how many gender violence cases courts in Goma had heard in one year (or in comparison to years prior) was a momentous undertaking. I spent a period of many weeks in December 2012 and January 2013 copying out pencil-written court records from dusty brown binders into an Excel spreadsheet. Simply obtaining statistics for one court in one year took approximately two weeks of work. It rapidly became clear that collating records for just one court going back to 2006 would eat up almost all my time in the field. This would not even offer me comparative statistics from courts in other locations. After taking a break for reflection, I began observing court cases at Goma’s central prison. Many of the cases were appeals from the previous year, and thus should have shown up in my meticulously collected spreadsheet of cases from 2012. Yet only three of the eleven cases I observed that week appeared in my records. I went back to the Tribunal de Grande Instance to consult the binders. Upon finding no records, I asked the clerk why the cases didn’t appear. “C’est comme ça ici [that’s just how it is here],” he said, laughing at me. “Do you see computers here? No. We have nothing. I write the cases here myself. I write everything. But sometimes it is the job of Mâitre Olivier to write, sometimes others. We are now transferring to an online system. I will enter these cases now.” It became abundantly clear that any effort to collect court statistics would mean very little in the end. Even provided unlimited resources and capacity to collect comparative statistics from select years for specific courts, this exchange shook my confidence in the data’s reliability. If the numbers I recorded bore little relevance to the cases heard in courtrooms, then comparative records would do little good. What information would I be able to glean from them? It was clear that my insights would need to come from other sources. Information pertaining to the justice system’s response to sexual and gender-based violence would instead emerge from observation and conversation. This too would take time, but my extended immersion in the Congolese legal system provided me with the necessary tools with which to make informed assessments and analyses of these data and would allow me to incorporate an understanding of how the positionality of my interviewees might affect the information they were giving me.

M23’s takeover of Goma in November 2012 further shifted my research plan. When the rebels entered the city, they liberated Goma’s central prison, Munzenze, which served as the primary prison for the Petit Nord (the southern territories of North Kivu). Almost every conviction that I had analyzed as part of my research in North Kivu to date had resulted in a sentence at Munzenze. After the prison liberation, rumors circulated that the approximately 1,500 inmates were gone. Having spent a great deal of time at Goma’s central prison, I headed to Munzenze, harboring hopes of meeting with the director to find out what had happened. I arrived to find the place eerily deserted. The barracks outside, normally occupied by hundreds of officers from the Police Nationale Congolaise (PNC) and the national army (the Forces Armées de la République Démocratique du Congo – FARDC), were empty. Old clothes were strewn across the floor, and a few onlookers watched me curiously as I picked my way through the razor wire fence erected around the prison. It became immediately clear that I would meet neither the director nor any returned prisoners. In the director’s office, prison records filled the room – not filed in boxes along the walls as they had been previously, but in shredded, ripped, and burnt paper strewn across the floor.

Figure 1 Outside Munzenze Prison. Goma, DR Congo. November 28, 2012.

© Milli Lake.

It wasn’t clear if the office had been intentionally ransacked, or if this was just the result of 1,500 individuals making a very quick getaway. Shortly after this visit, I met with the country director of a legal capacity-building organization. “Everything we’ve worked for, everything we’ve built over the past six years is destroyed,” the country director informed me, unable to hide his despair. “We’ve seen these kinds of shifts in power time and time again in eastern Congo,” he added. “You start everything over again. That’s how it works.” With new forces in charge of the city, the rules had changed again.Footnote 14

I was to be reminded of this fact time and time again over the coming months. When I next visited the Auditorat Militaire, they didn’t have chairs left to sit on. When I visited the police seven months later on my last visit to the station before I left Goma, no one had replaced the vehicles that were stolen in the M23 occupation. The police and the courts were going about their business, going through the motions, but insurgents had looted all their possessions. All case files, records, documents, modes of transport, power cords, and chairs were stolen, burnt, or destroyed. One military judge laughed as we stood together in his office. “Look,” he said, pointing at an Internet outlet in the wall. “The United Nations pays every month for our Internet, but we have no Internet cable and no computer! What good is this?”Footnote 15 Working in Goma during this specific time period bolstered my appreciation of what it meant to build legal institutions in the midst of ongoing violence and insecurity.

While the law prescribes a Tribunal de Paix to be located in each territory and a district court (TGI) in each major provincial city, in practice the picture is more murky. In North Kivu, TGIs are located in Goma, Beni, and Butembo. In South Kivu, in Bukavu (Kavumu) and Uvira.Footnote 16 At the time of my research, the lower courts in Lubero, Rutshuru, Masisi, and Walikale were functioning to some degree, and a lower court in Nyirangongo was not functioning at all. In South Kivu, lower courts in Mwenga, Kalehe, Idjwi, and Fizi were occasionally operational, but none was functional on a day-to-day basis.

In part by design and in part by necessity, I focused the majority of my research in courts where NGOs and legal aid programs were most active. Indeed, my arguments remain most applicable to those courts – and indeed the specific cases – that have been the targets of attention and advocacy. I do not expect gender-progressive practices to spill over to courts that have little contact with legal capacity-building initiatives. Since my arguments rest on the premise that NGOs exert considerable influence over legal processes in areas of state fragility and weakness in ways that prove impossible in stronger state settings, I expected their influence to be largely unfelt in locales where NGOs have only limited presence.

In order to evaluate these arguments in full, I visited courts and police stations with varying degrees of functionality, and varying degrees of support and attention from NGOs. In each location, I conducted personal interviews with relevant local personnel, observed proceedings (where courts were open or in session), and secured court documents and police records (where available). In North Kivu, the court in Rutshuru was under M23 control for almost the entire period of my initial research. When I visited courts, prosecutors, and police stations in Rutshuru territory, I liaised with M23’s bureaucracy rather than with representatives of the Congolese state. M23 did not grant me access to court or police records. Even still, my interviews and observations there were deeply illuminating.

Political insecurity and ongoing conflict made safe, reliable access to field sites and data extremely difficult. Each trip to a new village required careful weighing of the intellectual payoff versus threats to personal safety. For example, in 2012–2013, I decided against travel to Beni and Butembo. At that time, the roads between Goma and Beni were occupied by armed groups, including M23, FDLR, and Mai Mai, making safe passage uncertain. Two colleagues who made the trip by road described it as “extremely risky” upon their return. Armed groups stopped them on multiple occasions, and they elected to fly back via commercial airplane so as not to risk their luck again. While I could have traveled to Beni by road through Rwanda and Uganda, or taken domestic flights, these options were too time-consuming, costly, or unpredictable to be productive for my purposes.

On other occasions, I had to cancel scheduled visits to courts in Walikale and Shabunda, since they could not safely be reached by road. M23 was rapidly gaining territory at the time of my scheduled visits, and the only way to reach either site was by UN flight. Since NGO workers and journalists were beginning to discuss possibilities of evacuation, I was forced to think carefully about my role as an unaffiliated researcher. Not being an employee of an international organization meant that I would have to make my own choices about when and how to leave, rather than having evacuation protocols decided for me. I ultimately decided that the logistical challenges of negotiating evacuation under such circumstances (and the potential resources that might be needed) outweighed the presumed benefits, particularly since the British and American embassies declined to assist individuals traveling under such precarious political circumstances. In other instances, however, I traveled with NGO escorts, UN convoys, humanitarian flights, or private cars to occupied territory or hard-to-reach locations, in order to visit courts, observe mobile court proceedings, and conduct interviews across the provinces.

Undertaking these calculations on a daily basis attuned me to the ways that insecurity crept into even the most banal life decisions: for the legal practitioners working in the courts I was studying; for the police and NGOs to whom cases were initially reported; for the local colleagues, drivers, and interlocutors facilitating my work; and for victims of violence.

I was most successful in collecting and analyzing written judgments from Goma and Bukavu, although still not without significant challenges. Even in Goma, many prison, police, and court records were destroyed in the M23 occupation in 2012. For those records that were not destroyed, many decisions and judgments across both provinces were never recorded or written down; others were handwritten, with missing or incomplete pages, or barely legible. I collected as many judgments and case files as I was able to dating back to 2006. Of 149 written decisions, 106 were from the civilian courts and 33 were from the military courts. Complete case files included first-instance rulings, as well as cases that had gone to appeal. The rest of my analysis was observational and interview-based.

Figure 2 Filing Office, Munzenze Prison. Goma, DR Congo. November 28, 2012.

© Milli Lake.

For the same reasons that I faced challenges accessing certain courts, so too did NGOs engaged in legal capacity building. Courts in Beni, Butembo, Masisi, Mwenga, Walikale, and other hard-to-reach locations were not, therefore, hubs for the kinds of intensive and targeted NGO advocacy as those in Goma and Bukavu, and their proceedings typically reflected more conservative gender attitudes than the more easily accessible institutions. With the exception of a few fairly high-profile mobile court hearings, the more remote the court or police station, the less likely its proceedings were to exhibit the protections described in Chapter 1. Further discussion of this variation is presented in Chapters 4 and 5.

South Africa

My work in South Africa was, predictably, very different. South Africa, transitioned from a brutally oppressive apartheid regime to one of the world’s most respected and admired democracies in the early to mid-1990s.

I selected the Western Cape as my primary field site for a number of reasons. In both DR Congo and South Africa, I elected to spend an extended period of time in one place, in order to become familiar with local responses to violence, the inner workings of the justice system, and the ways in which legal accountability has been felt by those most affected by violence. The townships of Gugulethu and Khayelitsha in the Western Cape are, like the eastern provinces of DR Congo, among the more insecure areas of the country. Crime rates in general are high according to national averages, and sexual and gender-based violence is widespread (South African Police Service 2014, 2015).Footnote 17

Moreover, like Goma and Bukavu, Gugulethu and Khayelitsha are fairly densely populated. Their close proximity to the legislative capital of Cape Town, which houses some of the wealthiest neighborhoods in the country, means that large disparities in wealth are highly visible. Yet Cape Town, like the provincial capitals of Goma and Bukavu, is home to a strong and well-connected advocacy movement working on a wide array of gender issues. A large number of domestic, regional, and international human rights and legal advocacy organizations are headquartered in Cape Town, and many of them focus resources and attention on the nearby townships of Gugulethu and Khayelitsha. In addition, there are a number of grassroots human rights organizations, civil society groups, and church groups based in Khayelitsha and Gugulethu themselves working on issues of gender violence.

In many ways, the availability of formal data in South Africa made research there less challenging. Yet extraordinarily high levels of violence and the climate of fear in which many South Africans go about their daily activities rivaled anything I observed in eastern DR Congo. In DR Congo, I spent time in refugee camps, in rebel-held territories, and with victims of brutal human rights abuses. While Congolese civilians often seemed weary, but accepting in their attitudes toward the risk of violence, it was rare to encounter the sustained sense of unease that was widespread among many urban working South Africans. In Khayelitsha, I heard stories from medical practitioners of girls being gang raped, tied up in trashcans, and left for days after getting into taxi-buses.Footnote 18 I heard stories of carjackings and of friends of friends who ran out of gas on the freeway and were raped or stabbed before they had a chance to call the police.Footnote 19 I saw medical and forensics teams sweeping local neighborhoods in their reportedly routine clean-up after the alcohol-fueled violence that was expected to follow payday. Everyone had a story. I met countless people who warned of the dangers of using public toilets after dark. Despite the fact that there were few other options given that most Khayelitsha residents do not have toilets or running water in their homes, women would rarely venture the few steps to the community toilet, because this – many told me – meant they would surely be raped.Footnote 20 Despite multiple trips to South Africa over the years, I have always found it incredibly difficult to assess the extent to which stories like these reflect the lived realities faced by the townships’ ordinary residents, or the extent to which stories of streets rife with crime are legacies of deeply racialized paranoia. Apartheid’s scars are especially sorely felt in the Western Cape, which is home to the country’s largest white population. The province remains deeply segregated, and relations between whites, blacks, coloreds, and other populations (Nigerians, Zimbabweans, and Indians) are often characterized by division and mistrust.

Segregation and fear in the Western Cape has deep roots. The Native Land Act of 1913, which formed the basis for the apartheid-era “homelands” or bantustans, confined South Africa’s eighty percent nonwhite population to only thirteen percent of the country’s land. Moreover, the land on which blacks and coloreds were permitted to live was heavily regulated and, while living conditions in white South Africa grew to resemble the high standards enjoyed by citizens of Western Europe, living conditions on the homelands were characterized by poverty, poor sanitation, overcrowding, and insufficient food, clean water, and necessary public services. A series of laws was passed that exacerbated conditions in the homelands and made life for nonwhite South Africans incredibly challenging. In 1950, after the 1948 election of South Africa’s National Party, the Group Areas Act reinforced and expanded the jurisdiction of the Native Land Act, and precipitated the forced removal of blacks from urban centers to the homelands. The Population Registration Act strengthened the country’s existing pass laws, which restricted the movement of nonwhite South Africans and required passbooks to be carried by any nonwhite South African traveling outside of his/her homeland to work, or between one homeland and another. The Separate Amenities Act and the Native Labor Act of 1953 enshrined the color bar in law. These laws restricted jobs by race and prohibited the use of “white-only” transportation, post offices, schools, and restaurants by nonwhites, even for those working as domestic or urban laborers in white-only cities and neighborhoods (Meredith Reference Meredith1988). Any South African found to be in violation of these restrictions was subject to the full force of the law. Beatings, physical and verbal abuse, torture, and other assaults on nonwhites were an everyday occurrence. In addition to extraordinarily high levels of opportunistic violence against black and colored populations, state-sponsored, extralegal assassinations, as well as targeted lynchings by vigilantes, not only persisted with impunity between 1948 and 1991, but were actively supported by South Africa’s apartheid government against those perceived to be resisting apartheid rule or demonstrating organized or ad hoc opposition to legal segregation (Motlhabi Reference Motlhabi1988). Moreover, South African security forces coordinated targeted offensives against border states thought to harbor members of the political opposition (defined as terrorists by the apartheid state). Particularly notable was South Africa’s 1978 bombing of Angola, in which more than 700 South Africans thought to be affiliated with the antiapartheid opposition movement were killed (South African Truth and Reconciliation Commission 1997).

The widespread violence, brutality, torture, and subjugation inflicted upon nonwhite South Africans have, predictably, had devastating repercussions. Following the country’s transition to democracy in 1994, a great deal of fear and mistrust remained on both sides. Many whites had, since birth, been warned about the perceived dangers of black majority rule, and had been fed unrelenting propaganda about the innate criminality and violence of black South Africans. The racist portrayal of black South Africans as inherently violent has proved a sensitive issue in the postapartheid era, particularly when media attention has focused on high crime rates. Moreover, deeply racialized fear of black violence has meant that many whites remain afraid of traveling in traditionally nonwhite areas. It can be challenging to disentangle everyday reports of rising crime from racist propaganda propagated and manufactured by apartheid apologists (Moffett Reference Moffett2006).

However, the stories of violence I describe here rarely came from white South Africans, or from those living behind security systems in gated communities. Instead, they were first- or secondhand accounts from ordinary citizens and first responders living in some of the poorest neighborhoods in the country. Conducting research in health clinics, police stations, and hospitals meant that these stories were undoubtedly some of the worst. Nevertheless, working in townships in the Western Cape, home to some of the highest crime rates and starkest inequality in the world, alerted me to the many parallels, as well as the contrasts, with day-to-day security in eastern DR Congo.

Research in the Western Cape necessitated a number of methodological and epistemological decisions that never arose in DR Congo. First and foremost, unlike eastern DR Congo, in South Africa it proved extremely difficult to assess personal risk, leading to calculations that would have significant repercussions for the research I could conduct. As a white foreigner, the extent to which I was prepared to conduct research from “behind closed windows” was a constant negotiation and renegotiation. Many white South Africans do not travel to Khayelitsha or Gugulethu, and appear to believe that doing so ensures violent assault. Many others, including those involved in advocacy or nonprofit work relating to housing, sanitation, violence, and other pressing social issues, travel to their project sites in taxis and leave before dark, echoing warnings against the use of public transport, walking, or even driving through certain neighborhoods. Others still professed that traveling around the Western Cape was no more dangerous than travel in any major city, and any expressions to the contrary were simply the product of highly racialized legacies of apartheid and a historical belief that black or colored neighborhoods are rife with violence. In order to understand what law and lawlessness looked like for victims of gender violence in South Africa, it was necessary to move – physically and mentally – beyond the narratives promoted and described in urban centers. Doing so meant that I was constantly forced to weigh legitimate warnings for personal safety against a determination not to be intimidated by racist and classist paranoia.

Thus, while I conducted many informative interviews in the administrative and legislative capitals of Cape Town and Johannesburg, much of my research in South Africa occurred in the townships of Gugulethu and Khayelitsha. There, I interviewed judges, police officers, civil society activists, health workers, counselors, and victims of violence. I visited courts, hospitals, community centers, and police stations, and I traveled around in cars, on foot, and by public transport. Even for a short time, occupying the spaces and routines that my research subjects occupied, rather than arriving and leaving by car, allowed me some insight into the knowns and unknowns that shape and define the regulation of violent crime for ordinary Khayelitsha residents. Firsthand knowledge of the impact and implementation of national and local policies and institutions, as well as knowledge of the day-to-day challenges faced by local activists and transnational human rights organizations, was critical for answering the questions central to my research. In both South Africa and DR Congo, I learned far more from these accidental, unplanned, and everyday encounters and interactions than from my formal data collection efforts (Coffey Reference Coffey1999; Schatz Reference Schatz2009; Soss Reference Soss, Yanow and Schwartz-Shea2006; Wedeen Reference Wedeen2010).

My research in South Africa was not, however, characterized by the immersion and time that I spent in DR Congo. Although I spent two months conducting original research in South Africa, I relied far more heavily on archival documents and preexisting scholarship. In addition to data held by the South African Police Service (SAPS) and National Prosecuting Authority (NPA), I conducted interviews and spent time in district courts, magistrates’ courts, specialized sexual offenses courts, and the High Court in the Western Cape. I interviewed and corresponded with a number of stakeholders, including NGOs and researchers that have been monitoring the legal treatment of sexual and gender-based violence cases for more than a decade. Finally, I collected and analyzed 219 first-instance and appeals decisions from Khayelitsha Magistrates’ Court and the High Court of Western Cape between 2006 and 2014, focusing on key moments for gender reform (for instance, prior to and following the passage of the 2007 Sexual Offences Act). Due to the sheer volume of decisions available compared to eastern DR Congo, I chose to focus on two courts in particular: Khayelitsha’s District and Magistrates’ Courts and the Western Cape High Court.

I chose these courts as two courts serving areas of high crime that have attracted targeted attention for high rates of sexual and gender-based violence. Khayelitsha Magistrates’ Court serves a poor, predominantly working-class, black and colored population, whereas the High Court is situated in an extremely affluent part of downtown Cape Town. The median annual income of Khayelitsha according to 2011 census data was approximately $445 USD per working adult per year (R6,000 per capita). Employment is approximately fifty percent of the adult population, and more than fifty percent fail to achieve a secondary education (Seekings Reference Seekings2013). Khayelitsha is comprised of approximately ninety percent black South Africans, eight percent colored, and less than one percent white and Indian.

In South Africa, rape and sexual offenses cases are heard by both High Courts and Magistrates’ Courts as the courts of first instance. South Africa has nine High Court divisions – one in each of the nine provinces, with local seats in Bhisho, Mthatha and Port Elizabeth, Johannesburg, Durban, Thohoyandou, and Lephalale (South Africa Department of Justice and Constitutional Development 2016). Below the High Courts sit the Regional Magistrates’ Courts (which deal with criminal cases) and the District Magistrates’ Courts (which deal with criminal and civil cases).

TheWestern Cape houses fifty-six District Magistrates’ Courts, many with attached Regional Magistrates’ Courts that deal with more serious cases than the District Courts (South African Government 2013). District Courts can hear civil and criminal cases carrying sentences up to a maximum of three years, or a maximum fine of R100,000 (or approximately $7,000 USD, South Africa 2007. Criminal Law Sentencing Amendment Act. No 38). With a few exceptions, Regional Magistrates’ Courts can impose a maximum sentence of fifteen years, or fines up to R300,000. For crimes of murder or rape, however, Regional Magistrates’ Courts may impose life sentences; and for armed robbery or stealing a motor vehicle, they may impose sentences up to twenty years (South Africa 2007). In practice, where complexities in a particular case emerge, Magistrates’ Courts can often refer rape and sexual offense cases to the High Courts as the court of first instance. Specialized sexual offenses courts and children’s courts are also housed at select Magistrates’ Courts, and are, in theory, equipped with expertise and resources to prosecute sexual offenses. The sexual offenses courts can impose life sentences for rape and other sexual offenses. The Western Cape High Court is the first court of appeal for cases heard by all of Western Cape’s Magistrates’ Courts and sexual offenses courts. The Supreme Court of Appeal hears appeals for any cases heard by the Western Cape High Court as the court of first instance.

Of the 219 decisions I analyzed, 150 were first-instance decisions. One hundred and twenty-nine of these were heard by the Western Cape High Court as the court of first instance, and 21 were heard by the Magistrates’ Court. Sixty-nine were appeals from lower courts heard by the Western Cape High Court. Ultimately, while I conducted fieldwork in Khayelitsha, I spent less time analyzing the judgments produced by the Khayelitsha Magistrates’ Court because these lacked written explanations of the legal reasoning. I thus drew from accompanying documentation in the Magistrates’ Court case files, and supplemented my archival analysis with observation and interviews in the legal institutions based there. As Wynberg was the site of the country’s first specialized sexual offenses court, I also conducted a limited number of interviews there.

Both Khayelitsha and Gugulethu are hubs for grassroots campaigning and mobilization around gender violence, as well as housing a number of national gender justice organizations and programs. Interviewees in the Western Cape therefore included prosecutors, registrars, clerks, judges, police, investigators, defense counsel, community organizers and representatives from local NGOs, civil society groups, and legal aid organizations.


Fluid boundaries between conflict-related sexual violence, opportunistic violence, and violence within interpersonal relationships make it difficult to gage the prevalence or pervasiveness of gendered harms. Further, a lack of clarity over what constitutes sexual consent makes measuring sexual violence notoriously challenging. Nevertheless, in both North and South Kivu and in the Western Cape, various distinct forms of sexual and gender-based violence affect large numbers of women and men. Insight into the ways in which proximity to violence shaped both my own research practices and the day-to-day experiences of many of the communities I was working with offers important context for the evidence presented in this book.

Qualitative, embedded, and contextually situated research has sometimes been criticized for its perceived lack of generalizability. However, detailed and careful analysis of particular cases affords researchers new insights into the micro processes through which legal outcomes are constituted. My embedded approach allowed me to scrutinize the lived realities of apparent legal advancements as they have been experienced by those most directly affected, illuminating those innovations that have been welcomed by their intended beneficiaries, as well as those that have created new tensions. While this chapter calls attention to the many ethical challenges associated with conducting research with and among populations affected by violence, it nevertheless emphasizes the importance of centering the experiences of those whose perspectives might otherwise be marginalized. Given that many existing studies of human rights diffusion and law reform focus on fairly narrow indicators of legal implementation, or otherwise dismiss so-called failed states altogether, they risk missing the nuanced ways in which local courts and other actors and entities in weak or war-torn countries engage with internationally promoted human rights practices. Furthermore, they risk erasing the distinct ways in which women, in particular, experience violence, insecurity, law, and the state. Given that building the rule of law relies on establishing public confidence and trust in legal institutions, foregrounding the perspectives of the law’s intended “recipients” is integral to any careful analysis of how rule of law and state-building work in differing institutional settings.

Although this project was not without its limitations, its embedded approach offers scholars of international relations a window into the politics of human rights advocacy and state-building that travels far beyond the particular sites of study. Indeed, NGOs have certainly been presented with similar types of opportunity structures elsewhere. The stories presented in this book caution, once again, against binary approaches to concepts such as “war” and “peace,” “security” and “insecurity,” and “state strength” and “state weakness,” which are not generally equipped to encompass complex lived realities. The conclusion to this book touches on this project’s significance for other issue areas and in environments similarly dominated by external advocacy around law and state-building.

1 Respondents were asked whether they or their household members had been beaten, shot, seriously injured, sexually assaulted, raped, abducted, had violent amputations, or been subjected to forced labor by combatants during the past sixteen years (Johnson et al. Reference Johnson, Scott, Rughita, Kisielewski, Asher, Ricardo and Lawry2010).

2 North Kivu scored slightly higher than other provinces in respondents answering that they had experienced a history of rape, but similar to other provinces with regard to intimate partner violence.

3 UNSCR 1960 on Women, Peace and Security. Adopted by the Security Council at its 6453rd meeting on December 16, 2010, S/RES/1960.

4 See Cohen et al. (Reference Cohen, Green and Wood2013) and Cohen (Reference Cohen2016) for further discussion of this point in other conflict environments. It is also incorrect to assume, as Cohen (Reference Cohen2013) makes clear, that the rape and sexual violence perpetrated in DR Congo is characteristic of violence in all conflict environments. In fact, there is considerable variation in the types of violence employed by armed actors both within and across conflicts.

5 The definition of sexual harassment in the survey included both physical and verbal harassment and the survey data do not differentiate between physical and nonphysical harassment (Gender Links 2015: 52).

6 Interview. Francine. DR Congo. April 26, 2013. For a discussion of transactional sex across sub-Saharan Africa, see Cole and Thomas (Reference Cole and Thomas2009).

7 Field Notes; Informal Conversation. Lynette, Public Health Practitioner. DR Congo. July 7, 2012. In reference to a Bembe marriage ritual. Similar performative marriage rituals and practices are found in Rwanda, Ethiopia, and South Africa, as well as elsewhere around the world (Woodrow Wilson Center 2013; Yarbrough Reference Yarbrough2013).

8 Interview. Lynette, Public Health Practitioner. DR Congo, July 7, 2012. Note that, in spite of social norms that function to normalize rape in the context of some romantic relationships, there is widespread societal intolerance toward accusations of rape in other contexts. Smith (Reference Smith2018) discusses vigilante violence in response to rape accusations, including a notorious incident in KwaMashu, in KwaZulu-Natal.

9 The following paragraphs draw from Wood et al. (Reference Wood, Lambert and Jewkes2007).

10 Field Notes. Khosi, Human Rights Technician. South Africa, March 17, 2013.

11 Wood and Jewkes, in their interviews with young men in 1998, note that: “Multiple sexual partners, by all accounts virtually universal among boys, seemed to be an important defining feature of ‘being a man’” (Wood and Jewkes Reference Wood and Jewkes1998: 20). They also find that men typically express a strong need to control where their girlfriends are: “From the narratives it is clear that many men had expectations that their girlfriend should wait in her home for them to visit, and not finding her there would lead them to conclude that she was seeing another man” (23). Yet underlying discourses of machismo and control was a strong sense of vulnerability. “Acquiring girlfriends in the first place was obviously a preoccupying issue for the boys. In particular, male informants who came from much poorer backgrounds as well as those who were still at school expressed their vulnerability over girls preferring wealthy partners with cars, who were said to enable them to ‘boast’ and compete with other girls. Thus one of the poorer boys complained that since he came from a home that was ‘hungry’ and ‘full of damp,’ he had ‘no status’ among women; another commented that it made him ‘feel bad’ that girls were interested in rich men, adding ‘I know that one day they will realize that that’s not love because love is what you feel on the inside’” (25). Some groups have attempted to address the association between inadequacy and the need to present a heightened masculinity to the outside world. For instance, in circumcision school, elders “actively spoke against the physical abuse of women as being ‘unmanly.’ Thus here boys were provided with alternative constructions of masculinity to those prevalent in the streets, ones which were said to emphasize respect for others and non-violence. Initiates were told that beating women, using violence to force respect, fighting with knives, drinking excessively and having large numbers of partners were among the behaviors of ‘boys’ and not men” (41). However, Wood and Jewkes found that, in the longer term, most reverted to pre-circumcision ways after receiving such trainings.

12 In 2013, to my knowledge I was the first person, for example, to compile a comprehensive list of war crimes and crimes against humanity investigations and trials in the Kivu provinces. Since I carried out this work, the International Center for Transitional Justice has compiled a similar database; however, this work had not been undertaken when I was collecting my data. Avocats Sans Frontières has conducted periodic analysis of the domestic prosecutions of international crimes in DR Congo, but these efforts have focused on legal analyses rather than a comprehensive survey of cases.

13 On various occasions, I was in situations where minors requested interviews with me and I agreed. I do not generally cite from these interviews, although they have informed my broader understanding of accountability and barriers to justice. I have conducted many interviews with minors outside the context of this project, including in the context of a USAID-funded project that focused on child prostitution. This project drew predominantly from the experiences and testimonies of girls aged between eleven and eighteen.

14 Field Notes. Stephan, Country Director. DR Congo. November 28, 2012; January 12, 2013.

15 Interview and Field Notes. Safari. Military Judge. May 23, 2013.

16 In 2014, the Décret d’organisation judiciaire n°14/015 du 08 mai 2014 fixant les siéges et les ressort des Tribunaux de Grande Instance reorganized the jurisdiction of the Tribunaux de Paix and Grande Instance, creating a number of new courts.

17 Although the most recent Gender Links survey rates sexual violence in the Western Cape lower than other provinces, rates are still high overall, and SAPS data identify the Western Cape as having one of the highest crime rates in the country.

18 Interview. Anathi, Consellor. South Africa. March 15, 2013; Interview. Phumzile, Program Officer. South Africa. March 15, 2013.

19 Field Notes. South Africa. March and April 2017; Informal Conversation. Khozi. Human Rights Technician. South Africa. March 17, 2013.

20 Field Notes. South Africa. March 2017. Informal Conversation. Thembele and Nikelwa,Community Activists. South Africa. March 2, 2017.