The Constitution Cannot Hold You: Legal Equality and the Lived Reality of Lesbian and Bisexual Women in South Africa

A Note Before We Begin

This piece begins with a document. Not a body, not a statistic, not a headline. A document. Because the argument this piece makes is that the document is the problem; not in what it says, but in what it has been allowed to excuse. South Africa's Constitution is one of the most cited legal instruments in the world when the subject is LGBTQI+ rights. It is also the text behind which the state has been able to avoid the harder work of making those rights real for the women most at risk of losing them.

Before the research for this piece began, social media was searched for South African public discourse specifically about corrective rape and violence targeting lesbian and bisexual women. What was found was a general conversation about rape in South Africa; loud, frequent, and angry. What was absent was any sustained public thread naming corrective rape as a distinct category of violence, naming the women who survive it, or demanding accountability from the state for a pattern that has been documented for more than two decades. The violence is happening. The public conversation is not. On a platform whose methodology treats the absence of social media discourse as data, that silence is the first finding of this piece.

I. The Document

Section 9(3) of the Constitution of the Republic of South Africa, 1996, reads as follows: "The state may not unfairly discriminate directly or indirectly against anyone on one or more grounds, including race, gender, sex, pregnancy, marital status, ethnic or social origin, colour, sexual orientation, age, disability, religion, conscience, belief, culture, language and birth."

Sexual orientation. Explicitly named. In the foundational law of the country. In 1996.

This was not accidental. The inclusion of sexual orientation in the Constitution's equality clause was the result of sustained activism by gay and lesbian rights organisations during the negotiation process leading to the end of apartheid. It was a political achievement of considerable significance; the first national constitution in the world to explicitly prohibit discrimination on the basis of sexual orientation (International Commission of Jurists, 2019; Gaitho, 2022, William and Mary Journal of Women and the Law; Luvuno et al., 2019, African Journal of Primary Health Care and Family Medicine). It preceded the United Kingdom's Human Rights Act by two years. It preceded Canada's marriage equality legislation by nine. It preceded the United States Supreme Court's decision in Obergefell v. Hodges by nineteen years.

In 2006, South Africa became the first country in Africa and the fifth in the world to legalise same-sex marriage, through the Civil Union Act 17 of 2006 (Parliament of South Africa, 2006). In 2023, after years of delay and sustained civil society pressure, the Prevention and Combating of Hate Crimes and Hate Speech Act (Act 16 of 2023) was passed by the National Assembly in March 2023, passed by the National Council of Provinces in November 2023, and signed into law by President Cyril Ramaphosa on 9 May 2024 (The Presidency, 2024; Parliament of South Africa, 2023). The Act explicitly recognises sexual orientation, gender identity, and sex characteristics as protected categories in the definition of a hate crime (Parliament of South Africa, 2023; Centre for Human Rights, University of Pretoria, 2023).

On paper, South Africa is a model. On paper, a lesbian or bisexual woman in South Africa has more formal legal protection than in any other country on the African continent; and more than in many countries in the Global North at the time those protections were written.

This piece is about what paper cannot do.

II. The Gap Between the Clause and the Clinic

Begin at the most intimate point of contact between the state and any woman: the public healthcare system.

A 2017 study published in BMC International Health and Human Rights and conducted across South Africa's public sector found that lesbian women were being turned away from voluntary HIV counselling and testing because nurses believed lesbian women were not at risk for HIV infection (Müller, 2017). One participant from Cape Town described being refused testing explicitly on those grounds. All lesbian interviewees in the study confirmed encountering this misconception. A focus group participant summarised what nurses communicated to them: "To them it's only the straight people that can get HIV, because straight people they are usually the ones that are sleeping with men" (Müller, 2017, BMC International Health and Human Rights).

This is not a small administrative error. It is a life-threatening one. Lesbian and bisexual women in Southern Africa are at elevated HIV risk; not because of the common misrepresentation that women who sleep with women cannot transmit HIV, but because of the specific vulnerability created by forced sex. A cross-sectional study of 591 lesbian and bisexual women across Botswana, Namibia, South Africa and Zimbabwe found that 31.1% of all participants had experienced forced sex at least once in their lives; and that forced sexual experiences were the most significant independent risk factor for self-reported HIV infection in the sample (Sandfort et al., 2013, PLOS ONE; Baumann et al., 2015, BMC Women's Health). Corrective rape is not only a human rights crisis; it is a public health crisis with a direct, documented, quantified relationship to HIV transmission in this population.

The South African National Strategic Plan on HIV, STIs and TB 2023–2028 recognises the need for affirming LGBTQI+ healthcare as part of the country's HIV response strategy (National Department of Health, 2023). Yet current evidence suggests a persistent lack of LGBTQI+ health content in undergraduate health science curricula across South Africa (PMC, 2024; Luvuno et al., 2019). A 2024 protocol study published in PMC on nursing curricula in KwaZulu-Natal confirmed that "literature reveals a general lack of health worker training regarding the health needs of LGBTQI+ persons" (PMC, 2024). A scoping review of evidence on healthcare access for LGBTQI+ people in South Africa found that healthcare workers applied their own political, moral, or religious beliefs to determine what care was offered; that the Department of Health had not issued treatment guidelines or clinical algorithms for a range of LGBTQI+-specific health concerns; and that there was no nationally agreed standard of care for this population (Luvuno et al., 2019; PMC, 2024; Müller, 2017).

A nurse who turns away a lesbian woman seeking an HIV test is not necessarily acting outside her training. In a healthcare system that has produced no guidelines and delivered no instruction, she may be acting entirely within it.

III. The Constitution Meets the Police Station

The 2023 US State Department Human Rights Report on South Africa, published in February 2024, documented the following: "Black lesbians and trans persons, particularly in rural areas, were vulnerable to assault and to 'corrective' rape. LGBTQI+ individuals were particularly vulnerable to violent crime due to anti-LGBTQI+ attitudes within traditional and rural communities, in townships, and among police. Anti-LGBTQI+ attitudes of junior members of SAPS affected how they handled complaints by LGBTQI+ individuals" (US Department of State, 2024).

Read that sentence carefully: anti-LGBTQI+ attitudes of junior members of SAPS affected how they handled complaints by LGBTQI+ individuals. This is the point at which the Constitution's guarantees encounter the practical reality of the people tasked with enforcing them. A woman who has been raped in her township by men who believe they are correcting her sexuality, and who then presents at her local police station, may be met by an officer whose own attitudes about that sexuality align with those of the men who raped her (US Department of State, 2024; Outright International, 2025; GGU Law Digital Commons, 2024).

The consequences of that dynamic are measurable. The Outright International South Africa country report (2025) states that "survivors of hate crimes frequently hesitate to report incidents due to distrust of the police and prior negative experiences, and prosecution rates remain low." One NGO in Durban, documented in the 2023 Human Rights Report, claimed that most hate crime victims did not report their cases to police due to fears of secondary victimisation (US Department of State, 2024). In November 2023, a 15-year-old girl fell victim to a corrective rape hate crime in South Africa. Although a suspect was identified; a man in his 20s who reportedly told witnesses he wanted "to stop her from being lesbian"; no arrest was made (US Department of State, 2024).

A survey of rape victims in South Africa who did not report to police found that 33.3% cited fear of reprisals, 9.6% believed police would not be able to solve the crime, and 9.2% cited embarrassment as their reasons for non-disclosure (Wikipedia, citing national survey data, 2024). For lesbian and bisexual women, each of those barriers is compounded by the additional fear that disclosing a same-sex relationship to a hostile officer will produce further harm rather than protection.

IV. Corrective Rape: What the Term Obscures and What It Reveals

The term "corrective rape" was coined in South Africa in the early 2000s, when human rights organisations began documenting a surge in targeted sexual assaults against lesbian women in townships (ISS Africa, 2013; Psychology Today, 2013). It has since entered global human rights discourse; the United Nations mentioned it for the first time in its 2011 report on discrimination and violence against LGBTQI+ people (UN Human Rights Council, 2011).

UNAIDS (2015) subsequently recommended retiring the term in favour of "homophobic rape," arguing that "corrective" implies something is being fixed (UNAIDS, 2015 Terminology Guidelines). The critique is valid. The term does not describe a corrective procedure. It describes a punitive one: the use of sexual violence to enforce heteronormativity, to punish non-conformity, and to re-establish patriarchal control over women who have refused its terms (Ndlovu, 2025, Culture, Health and Sexuality; Gaitho, 2022, William and Mary Journal of Women and the Law; Ngongoma and Maweni, 2023, Critical Criminology).

A 2023 study published in Critical Criminology by Ngongoma and Maweni, focusing on Pietermaritzburg in KwaZulu-Natal, estimated that at least 500 lesbians are victims of corrective rape in South Africa annually (Ngongoma and Maweni, 2023). The authors noted explicitly that this figure is likely an undercount: societal homophobia keeps many women from reporting (Ngongoma and Maweni, 2023). The organisation Luleki Sizwe has estimated that more than ten lesbians are raped or gang-raped in Cape Town alone every week (Psychology Today, 2013, citing Luleki Sizwe data; Harvard Political Review, 2014). A peer-reviewed article published online in March 2025 in Culture, Health and Sexuality analysed corrective rape in South African townships, tracing its roots to heteronormative gender hierarchies, patriarchal cultural norms, and cultural practices such as ukuthwala: bride abduction; which normalise sexual violence against women as an exercise of male entitlement (Ndlovu, 2025). The article further argues that the strategic weaponisation of colonial histories deflects from the factors currently sustaining these harmful practices (Ndlovu, 2025, Culture, Health and Sexuality).

A study published in Critical Criminology found that corrective rape is a learnt social behaviour; absorbed through community observation, modelling, and the normalisation of homophobia as cultural self-defence (Ngongoma and Maweni, 2023, Critical Criminology; Louw, 2014, as cited in Ngongoma and Maweni, 2023). This social learning dimension is important analytically: it means that corrective rape is not the behaviour of isolated individuals acting out of aberrant impulses. It is the behaviour of communities that have constructed lesbian identity as a threat to their social order and sexual violence as the legitimate response to that threat.

The Centre for Human Rights at the University of Pretoria has confirmed that because corrective rape frequently occurs without protection, it makes lesbian and non-gender conforming persons "highly susceptible to becoming infected with HIV" (Centre for Human Rights, University of Pretoria, 2023; Sandfort et al., 2013, PLOS ONE).

V. The Criminal Justice Response: What the Numbers Mean

In September 2023, the South African government reported that the National Prosecuting Authority had secured 8 convictions during the 2022–2023 financial year for hate crimes targeting individuals for their sexual orientation (US Department of State, 2024). A government official attributed this to a "fully functioning Rapid Response Team" established under the National Intervention Strategy (US Department of State, 2024).

Eight convictions. In a country where an estimated 500 corrective rapes occur annually; a figure that, by the researchers' own admission, likely underestimates the actual number (Ngongoma and Maweni, 2023; Psychology Today, 2013; Luleki Sizwe, as cited in Harvard Political Review, 2014).

The most prominent case in the history of the country's prosecution of corrective rape remains that of Eudy Simelane, a former Banyana Banyana player and LGBTQI+ activist who was gang-raped and murdered in KwaThema township in 2008. She was stabbed 25 times. Her killers were convicted; it was described as the only prosecution of a corrective rape case to have resulted in prison sentences at the time (ISS Africa, 2013; Leiden Law Blog, 2024). The judge, however, stated in sentencing that the victim's sexual orientation was not considered a motive for the commission of the crime, and reportedly avoided using the word "lesbian" entirely during proceedings (ISS Africa, 2013; Leiden Law Blog, 2024; Gaitho, 2022). In a case that became a defining reference point for the global corrective rape conversation, the South African courts could not bring themselves to formally recognise what had happened to Eudy Simelane for what it was.

The case of Zoliswa Nkonyana, a 19-year-old lesbian murdered in Khayelitsha in 2006, was postponed 32 times before perpetrators were finally sentenced in 2012 (Psychology Today, 2013; Harvard Political Review, 2014). Five of her attackers were found innocent; those who were ultimately convicted received 18 years (Harvard Political Review, 2014). ActionAid documented that over a decade, 31 lesbians were killed in homophobic attacks in South Africa, of which only one case was successfully prosecuted (ActionAid, 2009, as cited in ISS Africa, 2013). Regardless of numerous corrective rape and murder cases, to date only three have gone to trial successfully (Leiden Law Blog, 2024; Gaitho, 2022; ISS Africa, 2013).

In August 2024; five months after the Hate Crimes Act was signed into law; Nombulelo Bhixa (28) and Minenhle Ngcobo (22), a young lesbian couple, were shot in Pietermaritzburg allegedly by Ngcobo's ex-boyfriend who could not accept her relationship with a woman (MambaOnline, 2024). The Uthingo Network described the case as a "tragic reminder of deep-rooted homophobia in South Africa, particularly in rural areas where such crimes often go unreported" (MambaOnline, 2024). A law passed in May 2024. A lesbian couple murdered in August 2024. Legislation does not, on its own, change the conditions that produce violence.

VI. What the Hate Crimes Act Does and Does Not Do

The Prevention and Combating of Hate Crimes and Hate Speech Act (Act 16 of 2023) is a genuinely significant piece of legislation. It defines a hate crime as an offence committed where the perpetrator is motivated by prejudice or intolerance toward a victim because of specified characteristics, explicitly including sexual orientation, gender identity, and sex characteristics (Parliament of South Africa, 2023). It requires the South African Police Service to issue national instructions for investigating hate crimes and requires the National Prosecuting Authority to issue directives governing how such cases are prosecuted (Act 16 of 2023, Section 9; Parliament of South Africa, 2023).

The Centre for Human Rights at the University of Pretoria described its passage as "a significant step towards the protection of all South Africans against hate crimes and hate speech" and noted that "these provisions, when implemented properly, will be a significant contribution to the fight against hate crimes" (Centre for Human Rights, University of Pretoria, 2023). The Equality Act of 2000 had already banned hate crimes based on sexual orientation; but as Ngongoma and Maweni (2023) and Avery (2017, as cited in Ngongoma and Maweni, 2023) confirm, "such cases are rarely prosecuted."

The phrase "when implemented properly" is carrying enormous weight.

The Outright International South Africa report (2025) states that implementation of the Act "remains uneven, with limited law enforcement capacity, gaps in data collection, and ongoing barriers to reporting and accessing justice for survivors." A 2025 article published in Frontiers in Sociology found that courts in South Africa currently do not discern between crime and hate crime in sentencing; and that in those instances where homophobia is acknowledged in a judgement, only the perpetrator's explicitly stated bias is cited (Van Der Schyff, 2025, Frontiers in Sociology). The article notes further that a person does not have to be demonstrably homophobic to commit a hate crime; the motive may lie in the perpetrator's awareness of a victim's vulnerability rather than in an articulated prejudice (Van Der Schyff, 2025). The South African legal system's failure to differentiate between acts of violence committed out of prejudice and other violent acts has been confirmed across multiple studies and legal analyses (Leiden Law Blog, 2024; SciELO, 2015; GGU Law Digital Commons, 2024).

The Hate Crimes Bill was introduced in Parliament in 2016. It was signed into law in 2024. Eight years of parliamentary process produced a law that has been on the books for just over a year at the time of writing. The institutions tasked with its implementation have had neither the time nor, in many cases, the cultural orientation to absorb what the law demands of them.

VII. Intimate Partner Violence: The Violence That Has No Shelter

The violence that lesbian and bisexual women experience is not only committed by strangers in townships. It also occurs within relationships; and the systems designed to address intimate partner violence are almost entirely built around a heterosexual model.

In November 2024, advocacy group Access Chapter 2 (AC2) raised concerns about the underreporting of intimate partner violence in same-sex relationships, citing stigma, minority stress, and fear of secondary victimisation as the primary barriers to disclosure (MambaOnline, 2024; OUT LGBT Well-being, 2024). AC2 called for tailored IPV awareness programmes and improved legal protections (MambaOnline, 2024).

The Domestic Violence Act 116 of 1998 applies to same-sex relationships in law. But there are no shelters in South Africa dedicated to LGBTQI+ youth or adults (US Department of State, 2023 Human Rights Report). Transgender and intersex persons are routinely refused entry into general population shelters because they do not conform to binary categories. A lesbian woman in an abusive relationship exists in a gap: unable to access LGBTQI+-specific shelters because such shelters do not exist; often unable to access general shelters without disclosing a same-sex relationship to staff who may be hostile or unprepared. The state has acknowledged this gap in its own documentation and has not resolved it (US Department of State, 2024; Outright International, 2025).

VIII. The Mental Health Burden of Surviving in Plain Sight

In South Africa, where the constitution offers protection from SOGIE-related discrimination, 57% of LGBTQI+ people surveyed showed signs of depression and 66% showed signs of anxiety (European Journal of Public Health, 2018). A 2025 study published in BMC Global and Public Health specifically examining lesbian, bisexual and queer-identifying women at a South African higher education institution found that participants reported clinically diagnosed depression, anxiety, previous suicide attempts, and substance use as coping mechanisms; with participants highlighting the strain of navigating sexual identity in unaccepting environments, the impact of homophobia on campus, and the absence of anyone to talk to, including family (Brooke-Sumner et al., 2025, BMC Global and Public Health). One participant stated: "You are suffering. You have depression. You have anxiety… it's a lot for one person to handle and you don't have anyone to talk to, especially sometimes you find that you can't even talk to your family because it's even worse on that side" (Brooke-Sumner et al., 2025).

Research confirms that bisexual women face an even higher mental health burden than lesbian women. Multiple studies have found higher rates of depression, anxiety, and suicidality among bisexual women compared to both lesbian women and heterosexual women; with bisexual-specific experiences of invisibility, erasure, and lack of bisexual-affirmative support cited as explanatory factors (Springer Nature, 2023; Salway et al., 2019, as cited in Springer Nature, 2023). A meta-analysis of 44 articles found that bias-based victimisation, bullying, and negative family treatment were significantly associated with suicidal ideation and suicide attempts among lesbian, gay, and bisexual individuals (de Lange et al., as cited in PMC, 2024).

A study on final-year medical students' preparedness to care for LGBTQI+ patients with mental illness, published in the South African Journal of Psychiatry in 2023, found significant gaps in training and confirmed that "there is limited research on the prevalence of mental illness in the South African LGBT population" (Badat et al., 2023, South African Journal of Psychiatry; Moagi et al., 2021, Health SA Gesondheid). A 2024 protocol published in PMC on nursing curricula in KwaZulu-Natal confirmed an ongoing "paucity of LGBTQI+ and key populations' health content in the undergraduate health science curricula in South Africa" (PMC, 2024). The result is a mental health system that neither has the data to understand the scope of the problem nor the trained practitioners to address it.

Civil society has reported increased rates of queerphobic bullying in schools contributing to mental health challenges and barriers to education; in several instances, LGBTQI+ youth have taken their own lives (US Department of State, 2024). Zukiswa Gaca, a 20-year-old lesbian woman, laid herself down on train tracks in South Africa in 2009 after being raped for the second time (Psychology Today, 2013). The system failed her twice: once when it could not protect her from rape, and again when it provided no pathway through which she could survive what had been done to her.

IX. Cape Town 2028 and the Limits of Visibility

In 2024, Cape Town was announced as the host city for WorldPride 2028; the first African city to receive this designation (MambaOnline, 2024). The announcement was met with celebration from LGBTQI+ communities and with hostility from homophobic religious leaders who vowed to mobilise opposition (MambaOnline, 2024).

WorldPride 2028 in Cape Town and a corrective rape in KwaThema in 2024 are not contradictions in the eyes of the South African state. They coexist, as they always have, because the state has consistently been more invested in the symbolic achievement than in the material infrastructure that would make the symbol mean something to the women most at risk.

This is not an argument against WorldPride. It is an argument about what visibility does and does not accomplish. The Constitution of 1996 was the same kind of symbolic achievement: the highest-possible legal affirmation that lesbian and bisexual women in South Africa had the same rights as everyone else. What followed was not safety. What followed was the coexistence of a progressive constitutional text and a hostile social reality; maintained by a state that pointed to the document and called the work done.

Social media followed the same logic. When this piece was being researched, South African public discourse on corrective rape was virtually absent from the platforms where the country processes its most urgent concerns. The general conversation about rape is there; furious, voluminous, and consistent. The specific conversation about what is happening to lesbian and bisexual women is not. The invisibility that defines their experience in the healthcare system and the criminal justice system extends into the digital public square. Even the forum that South Africans use to demand accountability from their state has not yet named this particular accountability gap. This platform treats that silence as data. It is the digital equivalent of a court record with no case number: proof that the system saw nothing, even when the harm was in plain sight.

X. What the State Owes These Women

South Africa now has Section 9 of the Constitution; the Civil Union Act; the Sexual Offences Act 32 of 2007; the Domestic Violence Act 116 of 1998; and the Prevention and Combating of Hate Crimes and Hate Speech Act 16 of 2023. It has more relevant legislation than almost any country in the Global South (Gaitho, 2022; International Commission of Jurists, 2019; Parliament of South Africa, 2023).

What it does not have is a healthcare system that knows lesbian women can acquire HIV through forced sex. A police force whose junior members respond to corrective rape with the same urgency as any other rape. A shelter system that includes women fleeing female abusers. A prosecutorial culture that does not require explicit proof of homophobia before it will name a crime for what it was. A mental health system with trained providers who understand minority stress, bisexual invisibility, and the specific trauma of sexual violence designed to annihilate identity. A national strategic plan on HIV that does not erase women who have sex with women. A social media conversation that says their names.

These are administrative requirements that flow directly from rights the Constitution already guarantees. A state that has guaranteed equality on the basis of sexual orientation has an obligation to build the infrastructure through which that equality becomes real (Gaitho, 2022; Centre for Human Rights, University of Pretoria, 2023; Van Der Schyff, 2025, Frontiers in Sociology). South Africa has been pointing to the document for thirty years.

The document has not held these women. The document cannot hold them.

The state was always supposed to do that. And so, eventually, was the public.