Abstract
This case note explores the landmark decision of the South African Constitutional Court in Mahlangu and Another v. Minister of Labour and Others, which recognised intersectional discrimination under section 9(3) of the Constitution. It shows that the Court went beyond that in fact and recognised intersectionality not just as part of discrimination law, but also as part of general constitutional law, using it as a theory of constitutional interpretation in adjudication.
Keywords
Introduction
Mahlangu and Another v. Minister of Labour and Others 1 was a challenge to the constitutionality of section 1(xix)(v) of the Compensation for Occupational Injuries and Diseases Act 1993 (‘COIDA’). The impugned provision expressly excluded domestic workers from the definition of an ‘employee’ for accessing social security assistance in the event of injury, disablement or death in the workplace. In its decision of November 2020, the South African Constitutional Court held that the provision was constitutionally invalid because it violated the rights of domestic workers who were predominantly Black women. The rights of the class of domestic workers thus came to be decided in reference to Black women and their constitutional rights, especially their right to equality and non-discrimination under section 9, the right to dignity under section 10, and the right to access social assistance under section 27 of the South African Constitution. With this, the Constitutional Court did something quite unique: it reinforced the prohibition of intersectional discrimination (discrimination based on a combination of grounds) under the Constitution, but for the first time explicitly; and recognised intersectionality (as a theory which helps understand the nature of discrimination based on a combination of grounds) as a general theory of interpretation for the bill of rights under the Constitution. In a sea of international and comparative jurisprudence which barely recognises intersectionality, let alone deploys it successfully; the decision in Mahlangu serves as an exemplar in discrimination law and more broadly, constitutional law, in its treatment of intersectionality. This case note parses through the reasoning in Mahlangu to appreciate how the South African Constitutional Court accomplished this.
The judgment
Ms Mahlangu was a domestic worker who had drowned in the swimming pool of her employer of over 22 years. She was partially blind and unable to swim. Her daughter (‘the appellant’) was financially dependent on her. On approaching the Department of Labour after her mother’s death, the appellant discovered that she was not entitled to any compensation under COIDA. Assisted by the South African Domestic Service and Allied Workers Union, and with the Commission for Gender Equality and the Women’s Legal Centre Trust as amici, she challenged the exclusion of domestic workers from compensation under COIDA. The High Court at Pretoria upheld the challenge, and the matter was appealed to the South African Constitutional Court.
The case before the Constitutional Court was that the exclusion of domestic workers infringed their right to equality, the right to human dignity and the right to have access to social security under sections 9, 10 and 27 respectively of the South African Constitution. In particular, because domestic workers, like Ms Mahlangu, were predominantly Black women, the claim was that the exclusion constituted indirect discrimination against Black women on the basis of their social status, gender, race and class. 2 Furthermore, the amici submitted that the case be analysed ‘within an intersectional framework’ for ‘a nuanced, purposive and socio-contextual consideration’. 3 In the oral hearings before the Constitutional Court, the respondents from the government conceded that the provision be struck down. But it was still for the Constitutional Court to satisfy itself that the provision was indeed invalid and infringed the Constitution. 4 It went about establishing this thus.
Victor AJ, writing for the majority, first, considered the challenge to section 27(1)(c) which guarantees the right to have access to social security. She confirmed that COIDA was a social security legislation for the purposes of section 27, and that in turn, social security assistance included support and sustenance for those affected by the injury, disease or death of a worker. 5 To her, this was a conclusion flowing from the express text of the Constitution in section 27(1)(c) which guaranteed social assistance for the purpose of supporting persons and their dependents when they were unable to support themselves. COIDA was considered exactly the kind of legislation which was meant to provide such social assistance, insofar as its purpose was ‘to ameliorate the circumstances of those who would otherwise be condemned to living in abject poverty’. 6 The only real question that remained was whether the exclusion of domestic workers from COIDA was reasonable. Herein, Victor AJ relied on the amici’s submissions to draw on an intersectional approach to appreciate the impact of the exclusion on domestic workers in this context, and to consider ‘those who are most vulnerable or most in need…[and] take cognisance of those who fall at the intersection of compounded vulnerabilities due to intersecting oppression based on race, sex, gender, class and other grounds’. 7 To her, domestic workers as a category was defined by disadvantages associated with race, sex, gender and class – which rendered them extremely vulnerable and poor. She thus concluded that the exclusion of domestic workers was manifestly unreasonable and against the constitutional values of dignity, equality and ubuntu. 8
Secondly, the majority judgment considered the equality challenge under both section 9(1) and 9(3) of the Constitution on the right to equality as well as the right against unfair discrimination. With no basis of differentiation offered between all employees on the one hand and domestic workers on the other under COIDA, Victor AJ concluded that the impugned provision was violative of section 9(1). 9 It was then that the inquiry turned to section 9(3). Titled ‘Section 9(3) challenge and the application of intersectionality’, the inquiry was entirely constructed in intersectional terms from the very outset. Three observations were made before launching into the discrimination inquiry. First, that the discrimination here was indirect in nature, because the domestic workers were predominantly Black women. 10 Secondly, that the discrimination against them, as based on race, sex and gender, was to be considered presumptively unfair based on section 9(5) of the Constitution when the grounds of discrimination involved in the claim were those that were explicitly recognised by the Constitution. 11 Thirdly, that because the case concerned more than one ground and that the Constitution explicitly prohibited discrimination based on ‘one or more grounds’ under section 9(3), that the ‘importance of an intersectionality analysis [became] unavoidable’. 12 The discussion then outlined the intersectional approach, one which helped unravel the patterns of group disadvantage associated with multiple grounds which defined the group of Black female domestic workers. 13 This was done through a perusal of ‘the nature and context of the individual or group at issue, their history, as well as the social and legal history of society’s treatment of that group’, that is ‘the particular history of social security in South Africa, as it relates to domestic workers [and]…the historical disadvantage that Black women have faced as a group’. 14 Herein the majority judgment explored ‘the gendered implications of apartheid’s racist system’ which relegated Black women to the bottom of the social hierarchy, worse off than white men and women, and Black men. 15 But according to the majority, ‘the case of domestic workers was particularly severe’. 16 Not only did they suffer from the racist and gendered implications under apartheid, they also suffered a personal loss of social and family life, living in poor conditions of their own, and devoting their lives to the lives of their employers. This marginalisation during apartheid has continued to date and reflected, as Mhlantla J described in the concurring opinion, ‘a complex history entrenching racism, sexism and social class’. 17 And it was this understanding of their intersectional disadvantage that led the majority to conclude that the exclusion from social security assistance under COIDA constituted unfair discrimination which impinged on the Constitution’s goal of achieving substantive equality and transformation for everyone. 18 Leaving domestic workers (Black women) behind in this goal was unjustifiable. 19
Thirdly, the majority judgment considered the dignity challenge under section 10 of the Constitution which provides that ‘everyone has inherent dignity and the right to have their dignity respected and protected’. For the majority, there was no doubt that the exclusion conveyed that domestic work was not real work. The exclusion thus undervalued the contribution of domestic workers to the social and economic life of the country and ultimately treated the domestic workers triflingly. 20 Furthermore, according to Victor AJ, ‘domestic work has been undervalued precisely because of who performs this work: poor Black women’. 21 Thus, an intersectional analysis once again explained the harm involved, revealing how both gendered and racist assumptions defined the way in which domestic workers had suffered a loss of dignity.
Lastly, the Constitutional Court had to address the question of remedy and whether the invalidity could have immediate and retrospective effect. The government had filed a ‘high-level’ abbreviated actuarial report claiming that the economic and administrative costs of past injuries and diseases could have a negative impact on future claims. 22 The majority was unimpressed with the lack of detail in the report which made no suggestion that the state was in fact unable to meet all claims. 23 There was thus little evidence to actually substantiate the practical realities faced by the state in giving retrospective effect to the invalidity of the exclusion. Additionally, Victor AJ considered that the intersectional discrimination in this case compelled a retrospective remedy which could ‘contribute towards the amelioration of systemic disadvantage suffered by these women and contribute to breaking the cycle of poverty they suffer’. 24 With this, the majority not only declared section 1(xix)(v) of COIDA constitutionally invalid to the extent that it excluded domestic workers, but did so with immediate and retrospective effect.
In his dissenting opinion, Jafta J too considered the impugned provision to be constitutionally invalid but for reasons materially different than those of the majority. Thus, the dissenting opinion found that not only was there no violation of section 27 or section 10 because other workers such as police and soldiers too were excluded; there was also no unfair discrimination in this case under section 9(3). But there was, according to Jafta J, a clear violation of section 9(1) on the grounds of rationality, since the exclusion did not serve any rational purpose whatsoever. 25 Although he considered ‘the conclusion on rationality [as] sufficient for striking the provision down’, 26 he also considered the unfair discrimination analysis because the majority had. He did so by reframing the starting point of the case as discrimination on the ground of ‘domestic worker’ status, a ground evidently not listed in section 9(3). 27 Those excluded as domestic workers or as police or as members of the defence forces could be Black or white, and thus there was no reason to consider discrimination based on employment status or type of work to be based on race, sex or gender. So, for him, the absence of any direct invocation of these named grounds in the impugned provision under COIDA debarred an unfair discrimination challenge under section 9(3), or at least, demanded greater proof of evidence to be established than deemed sufficient by the majority.
Analysis
Two things stand out about the majority judgment in Mahlangu in the way that it is framed, reasoned, and remedied: the use of intersectionality as a general interpretive theory of the Constitution, and the analysis of intersectional discrimination in particular.
The first and the most striking aspect of the case is that intersectionality is deemed to be the main theory of constitutional interpretation applicable in the case. According to Victor AJ: Adopting intersectionality as an interpretative criterion enables courts to consider the social structures that shape the experience of marginalised people. It also reveals how individual experiences vary according to multiple combinations of privilege, power, and vulnerability as structural elements of discrimination. An intersectional approach is the kind of interpretative approach which will achieve “the progressive realisation of our transformative constitutionalism.”
28
Moreover, the Court’s invocation of intersectionality as a general interpretive theory feeds into the analysis of the remedy in this case. Under section 172 of the Constitution, the Constitutional Court has power to declare any law or conduct inconsistent with the Constitution and make any order that is just and equitable, including a retrospective order. What is just and equitable is often assessed in reference to that which may ‘effectively vindicate [the violated] rights’. 37 In Mahlangu, the Court added that ‘[t]he fact that this case concerns intersectional discrimination is a relevant factor in determining whether a retrospective order should be granted’. 38 In fact, according to Victor AJ, because a retrospective remedy had the potential to ‘contribute towards the amelioration of systemic disadvantage suffered by [the domestic workers] and contribute to breaking the cycle of poverty they suffer’, 39 it trumped any argument against limiting the retrospective effect of the declaration of invalidity.
Much of this analysis is aided by the South African Constitutional Court’s view of itself, which has always been imagined as playing a transformative role in the post-apartheid South African society for upturning the racist and gendered impact of the history of the country. 40 What is more, this intersectional view of transformation is even recognised in the first provision of the Constitution in section 1 which sets out the founding values of the country as based on ‘non-racialism and non-sexism’. Intersectionality is thus truly central to South African constitutionalism and its use as a general theory of constitutional interpretation – beyond its recognition in section 9(3) of the Constitution which prohibits unfair discrimination on ‘one or more grounds’ – appears entirely expected.
But of course, secondly, because intersectionality is a theory of discrimination and disadvantage, 41 it is best at home in section 9(3) which is where the Constitutional Court in Mahlangu used it liberally. In its most significant pronouncement on intersectional discrimination to date, the Court suggested that the simple fact the claim involved multiple grounds, made ‘the importance of an intersectionality analysis…unavoidable’. 42 This is a grand shift from previous jurisprudence in South Africa which had always recognised discrimination based on multiple grounds, but seldom as unavoidably intersectional in nature. Thus, as I have shown in my work, the South African Constitutional Court’s treatment of discrimination based on multiple grounds can be traced along an interpretive continuum of categories which includes: substantially single-axis, capacious single-axis, contextual single-axis, multiple, and additive forms of discrimination. 43 All of these categories of analysis of multi-ground discriminaiton differ from intersectional discrimination in one significant way: in that each of them fragment the discrimination in question instead of seeing it as both similar and different to discrimination based on individualgrounds and thus being a unique admixture which can only be appreciated as a whole and in reference to the complex patterns of disadvantage which define such discrimination. However, the Court in Mahlangu seems to settle this conclusively by making intersectional discrimination as the go-to category for analysing discrimination claims based on multiple grounds. 44 Indeed, the fact that it does so by explicitly recognising intersectional discrimination not only as a concept (which it seems to have done implicitly over a decade ago in Hassam v. Jacobs 45 anyway 46 ) but by adopting the locutions of ‘intersectionality’ and ‘intersectional discrimination’, seems to make this development in discrimination jurisprudence hugely significant both as a precedent in the South African context and as an exemplar for the adjudication of intersectional discrimination everywhere.
Furthermore, unlike the dissent, the majority judgment did not even begin with seeing this case as a possible case of direct discrimination against the class of domestic workers but proceeded from the standpoint where the case was framed as a case of indirect intersectional discrimination against Black women since domestic workers in South Africa were predominantly Black women. Therefore, discrimination in this case was defined as based on the grounds of race, sex and gender from the outset – grounds which are recognised in section 9(3) and thus presumed to constitute unfair discrimination under section 9(5) of the Constitution. What is particularly interesting here is that although the Court was cognisant that section 9(5) on the presumption of unfairness could only be invoked when discrimination was based on grounds already recognised, it did not require that discrimination be based exclusively on grounds named in the Constitution. Thus, the majority opinion weaves in elements of class, poverty and socio-economic status of Black women to complete its intersectional analysis alongside race, sex and gender; but does not seem to require those elements to all be recognised as grounds for section 9(5) to be invoked. This expands the possibilities of intersectional discrimination to be attracting the presumption of unfairness when it is based on at least one of the named grounds under the Constitution. In the absence of any express requirement that all grounds in a discrimination claim be explicitly named for section 9(5) to apply, the Constitutional Court seems justified in adopting a less demanding view for establishing intersectional discrimination as unfair in fact. After all, the presumption of unfairness in section 9(5) makes all the difference by shifting the burden of proof on the state or the respondent to show that discrimination is still justified or fair. The insistence of the dissenting opinion on the proof of indirect discrimination seems to read past this key manoeuvre of the majority which extended section 9(5) to the claim in Mahlangu.
What is also noteworthy about the approach to unfair discrimination in section 9(3) is that the Court seems to have navigated a rather complex category of indirect intersectional discrimination with ease. Indirect intersectional discrimination is obviously more complicated that direct intersectional discrimination based explicitly on multiple grounds. But it is also internally complex. It thus comes in at least four forms – the disproportionate impact of a neutral policy on an intersectionally disadvantaged group such as Black women (as was the case in Mahlangu); the disproportionate impact of the intersection of two or more neutral policies on an intersectionally disadvantaged group such as Black women (as was the case in Tilern de Bique v. Ministry of Defence 47 ); or the disproportionate impact of a policy based on a specific ground (like marital status) on a group defined by that and other grounds (such as unmarried female partners defined by marital status and gender, as was the case in Volks v. Robinson 48 ), or the disproportionate impact of a policy based on a specific ground(s) (such as age) on disadvantaged groups defined by other grounds (such as sex and disability such as was the case in Gosselin v. Quebec 49 ). 50 Indirect intersectional discrimination thus complicates our view of how discrimination is produced and reproduced in the society, rather than giving a false assurance that claims can be neatly divided into basic categories of direct and indirect discrimination and addressed without more. Each case of indirect intersectional discrimination has the potential to reveal something hitherto rendered ‘invisible’ or ‘erased’. 51 So, while the claim in Mahlangu was perhaps the most straightforward type of indirect intersectional discrimination, the Court’s openness to using the category of indirect intersectional discrimination to unearth discrimination which is as yet not visible or straightforward may prove to be crucial in future cases.
Conclusion
It is only customary to acknowledge that the origins of intersectional discrimination, in fact, the term ‘intersectionality’ itself, lie in Kimberlé W Crenshaw’s seminal article from 1989. 52 Efforts to recognise intersectionality theory in discrimination law 53 and human rights law 54 more broadly, have never ceased since. But in over three decades, no court, not even the South African Constitutional Court, had accomplished the task of naming and applying intersectionality theory squarely to a case of intersectional discrimination. Mahlangu is a breakthrough in discrimination law in this respect, and as I have shown above, in human rights and constitutional law more broadly. The reasoning of the South African Constitutional Court in Mahlangu thus has the potential to serve as the template for other courts and adjudicative bodies to employ intersectionality as an interpretive framework in discrimination and human rights cases. One can only hope that, with this, the next case to follow Mahlangu’s ambitious lead would not take as long to materialise.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.