Photo by Marc Steenbeke 2018, Stock image.

Introduction

Globally, almost 37 million people currently live with HIV. Around one fifth of them (7.2 million people) reside in South Africa, which has responded with one of the world’s most comprehensive public treatment programmes, providing antiretroviral drugs (ARVs) for 4.4 million people. In a population of 57.7 million, this is equivalent to one in every thirteen adults taking ARVs daily to manage HIV, making it the country’s most common chronic illness. Meanwhile, South Africa hosts 2.4 million foreign-born immigrants, who come from African countries with some of the highest HIV prevalence rates in the world, ranging from between 12 and 28 percent (UNAIDS 2019; DESA 2016; StatSA 2016). The relationship between migration and HIV in southern Africa is important because of the mobility of this region’s labour. While much research has been conducted on obstacles to healthcare integration and antiretroviral therapy (ART) – such as barriers to patient counselling and adherence to drug regimens – migrant populations are not well integrated into healthcare services and little is known about how migrant groups behave in regards to HIV within changing clinical and cultural landscapes. HIV is a biosocial disease, meaning that its lived experience consists of clinical features overlain with psychological experiences of stigma and social angst (Pearson et al. 2009). Investigating such phenomena is pivotal as these latter experiences of HIV reflect directly upon care-seeking behaviour and outcomes of treatment.

While ‘structure’ as a concept has an intricate theoretical past, health scholarship essentially explores social and economic forces, which inflect health outcomes and create symptoms in systems that produce and reproduce the social world. Such systems are inherently linked to culture as they generate and sustain the moral and aesthetic values affixed to bodies and disease. In conjunction, structure and culture are thus ‘mutually co-implicated in producing stigma and inequalities’ (Metzl and Hansen 2014, 19), which, in turn, influence medical decisions. Medical anthropologists have argued that epidemics such as HIV are outcomes of ‘structural violence’ in that ‘neither culture nor pure individual will is at fault; rather, historically given (and often economically driven) processes and forces conspire to constrain individual agency’ (Farmer 2001, 79). This idea of structural causation has been linked to the concept of stigma, as experiences of stigma are tied inseparably to the economic and political processes that produce them (Parker and Aggleton 2003). In line with the notion of structure as a form of violence, entire populations can be described as ‘structurally vulnerable’ to forces that constrain decision making, frame choices, and limit life options, and, therefore, the health of the marginalised (Adams et al. 2011; Metzl and Hansen 2014; Quesada, Hart, and Bourgois 2011).

This article locates stigma and identifies vulnerabilities to HIV among Mozambican immigrants in South Africa not merely in interpersonal behaviour, but in the (dys)functions of markets and institutions. It does this by exploring structural avenues through which migrants are rendered particularly vulnerable to HIV: perilous journeys, deportation and xenophobic violence, exclusion from public services, destitution and labour exploitation, low levels of education and language barriers, medical pluralism (co-consulting with traditional healers), social isolation in urban African immigrant enclaves, and the stigmas of living with HIV. These structural circumstances, I argue, compound to form both visible and invisible barriers to ART and the wider integration of African migrants into South African healthcare services.

Mozambican migrants in South Africa

Employed primarily in mining and farming, the three largest migrant groups living in South Africa consist of an estimated 525,000 Zimbabweans, 372,000 Mozambicans, and 256,000 Basotho. Today, due to proximity, many Mozambicans stay in South Africa’s bordering provinces of Limpopo and Mpumalanga, while Gauteng, the region’s financial centre, accounts for more foreign-born immigrants than all other provinces combined: 9.5 percent of the populace. In central Johannesburg, African immigrants reside in communities with relatively cheap housing, where they make up around 13 percent of the residents. Of the inner city’s 4.4 million inhabitants, 0.57 million are foreigners and 130,000 are Mozambicans (DESA 2016; StatSA 2016; Wentzel and Tlabela 2006).

In Johannesburg, eastern and southern suburbs with close ties to the history of Portuguese immigration to South Africa have absorbed the majority of Lusophone migrants (Glaser 2010). Portuguese people settled in three waves from the late nineteenth century until the late 1970s, yet since the political transformations of the mid-1990s there has been considerable Portuguese emigration out of the country, with African immigrants subsequently occupying their areas. This is particularly true of Mozambicans, who originally came to work for the Portuguese that fled Mozambique following its independence in 1974 and settled in Johannesburg. Today, Portuguese and Mozambican Bantu languages are spoken on every street corner in these suburbs, which display rich features of Portuguese architecture and culture. Poorer townships such as Alexandra and Soweto have also received Mozambicans. Most come from the nearest provinces in central and southern Mozambique, where HIV prevalence rates are high and as many as 20 percent may be HIV-positive (IaM 2017). Meanwhile, in South Africa, all dimensions of living with HIV are affected by extensive stigma (Gilbert and Walker 2010; Chao et al. 2017; Steenberg 2019a). It was in this ambiance that I investigated the structural vulnerabilities of migrants.

Methods

This research was conducted as part of a larger study focused on disclosure, loneliness, stigma, and Mozambican immigrants’ perceptions of HIV (see Steenberg 2019a, 2019b). Initial fieldwork was conducted between August and November 2015 with HIV-positive Mozambican migrants in Helen Joseph Hospital’s Our Hope (isiZulu: Themba Lethu) HIV and Tuberculosis Clinic in Johannesburg. Then, between February and June 2016, fieldwork was carried out in four city hubs for African immigrants, namely La Rochelle, Jeppestown, Regent’s Park, and Rosettenville. Ethnographic research was undertaken, involving informal conversations, observations, interaction, and elaborate field notes (Browner and Sargent 2011; Emerson, Fretz, and Shaw 2011). In addition, twenty-one semi-structured life history interviews were conducted in Portuguese with Mozambican patients in Our Hope. These recordings were then transcribed by an assistant. The ethnographic life histories described the circumstances of patients, not only in regards to living with HIV, but also as African migrants living in South Africa.

NVivo 11 computer software was used to code and analyse the data thematically. The informants came from Manica (3) in central Mozambique; the two southern provinces of Gaza (3) and Inhambane (3); and Maputo province (12). They varied in terms of gender (9 men, 12 women); age (mean 37, range 20–48 years); time since initiation of ART (mean 5.6 years ago, range 0–13 years); and time since migrating to South Africa (mean 16.5 years ago, range 1–28 years). Sample representation was achieved through ‘knowledge saturation’ (Bertaux 1980). Collaborating personnel at Our Hope identified Mozambicans with pre-scheduled consultations and gave notice as they prepared for the day in a patient records archive. All names appearing in this article are pseudonyms to guarantee anonymity and followed by age, occupation, and duration of stay in South Africa in years. The Medical Research Ethics Committee of the University of the Witwatersrand provided ethical clearance (no. M150496).

The circle: Perilous crossings

Legal and illegal labour migration to South Africa has been a Mozambican livelihood strategy for so long that Mozambican social networks have become transnationally entrenched on either side of the border. These networks now serve to sustain this movement in a self-perpetuating circular movement, facilitated by kinship and friendship relations in sending areas and by established immigrants in destination areas (Muanamoha, Maharaj, and Preston-Whyte 2010).

I came here because there wasn’t work. There was absolutely nothing to do in Maputo. So, I decided to come here. (Maria, aged forty-six, spaza shop keeper, twenty-eight years in South Africa)

My husband was here working [in South Africa]. Later he came to fetch me at home and now we’re both here. I’m looking for work too now. (Victória, aged forty, take-away waitress, twenty-three years in South Africa)

A host of actors buttress this transience: corrupt border agents, smugglers and conveyors, well-established peers, South African employers, and police or deportation officers. The journey is perilous with considerable transactional costs and risks in the form of bribes, arrests, and robberies. A 2005 bilateral agreement between Mozambique and South Africa that provides an automatic thirty-day visa has eased the lives of people oscillating between the two countries considerably. However, most immigrants lack current legal papers and are nervous to even address the issue.

Please don’t tell anyone, but I stay here in South Africa on a passport that already ended [expired]. (Reniesa, aged thirty-five, domestic worker, twenty-one years in South Africa)

I go by the end of each month. When I enter [South Africa], they give me thirty days and after that I have to go back and forth again and again. (Hélena, aged thirty-nine, unemployed, nineteen years in South Africa)

While some Mozambicans have settled permanently – particularly those who arrived as refugees – the overwhelming part of the migratory movement remains circular as Mozambicans maintain emotional and interpersonal ties to homelands, families, traditional beliefs, and culture. Hence, they generally maintain two homes: one in South Africa and one in Mozambique, supporting the latter through remittances and visits as often as they can.

Yes, I always go back to Chókwè each and every year in December, but actually sometimes twice a year, sometimes once a year. (Oliveira, aged thirty-six, carpenter, seven years in South Africa)

I return to Mozambique every month. Everybody’s there! Yeah, I have brothers, there are four of us. My kids are also in Maputo. They study there. (Gonçalves, aged thirty, metalworker, nine years in South Africa)

It has been argued that in South Africa, ‘there are not specific Mozambican ethnic niches and enclaves as you see with immigrant communities elsewhere in the world’ (Vidal 2010, 60). Perhaps this oversight was caused by their relatively low visibility, but nothing could be further from the truth: once in Johannesburg, Mozambicans commonly settle in communities aided by relatives, friends, neighbours, workmates, or even church groups. The stronger the ties among migrants, the slower their integration and the deeper their embeddedness into migrant communities in one of the suburbs. [note 1] Some migrants eventually tire of life in South Africa and return to Mozambique; others roam the circle indefinitely, usually embracing a distant dream of going back home for good one day.

You’re gonna see what it’s like, you know, running from your own country. It’s not like I wanted to come this side; it was the situation that forced me to come and work this side, see? (Arnaldo, aged forty-two, water proofer, twenty-three years in South Africa)

Once in South Africa, African immigrants can expect little to no protection from local authorities, and are at times actively barred from essential public services, such as in hospitals and schools.

Mass deportations and xenophobia

Post-apartheid South Africa has yet to develop progressive immigration policies. Instead of seeking ways to accommodate and integrate new African migrants, ‘South Africans began to rail against them, to blame them from everything from crime to HIV/AIDS, to unemployment and to deport them in their hundreds of thousands’ (Crush 2008, 6). Since 1994, close to 2.5 million undocumented migrants may have been deported, particularly to Mozambique, Zimbabwe, and Lesotho (McConnell 2009, 38; Africa Check 2016). Hence, among undocumented Mozambicans, I observed an all-pervading fear of incarceration into the Lindela Repatriation Centre outside Johannesburg and subsequent deportation, which involves being driven out by lorry and abandoned on the Mozambican side of the Ressano Garcia/Komatipoort border, stranded at night with no means of transport, perhaps thousands of kilometres away from home. Needless to say, it represents a challenge for sustained healthcare provision if patients are forcibly removed from their therapeutic environments and barred from re-entering. Worse, the anxiety of deportation permeates public life, resulting in the shunning of authorities or having to manoeuvre services without legal documentation. Faustine Kyungu Nkulu Kalengayi and colleagues found that a fear of deportation among migrants ‘emerged as the most important determinant of reluctance to seek [HIV] care’ (Kalengayi et al. 2012, 39), even in developed settings such as Scandinavia. Adding to this, intimidated migrants often aim to conceal their identities as foreigners or as coming from foreign descent (Vidal 2010). This transmutes into a strange form of dual concealment: first, to socially contain the stigma of being a foreigner, and then second, to ward off the stigma of being associated with HIV, which foreigners are widely accused of spreading. For instance, one earlier survey showed that a striking 61 percent of the South African population ‘support the deportation of foreign nationals who test positive for HIV or have AIDS’ (Crush 2008, 2). In short, prejudiced nationalistic sentiments and hostility towards African foreigners is the general stance of citizens. This exoantagonism is known as xenophobia, although ironically the real phobos (fear) is hardly one of foreigners, but rather one felt by foreigners.

In that place where we work, sometimes they hear me greet other Mozambicans in [Xitsonga-]Shangaan and so other South Africans say that here works a person from the outside, here works a makwerekwere. They like to use that word, makwerekwere. It means ‘foreigner’. (Victória, aged forty, take-away waitress, twenty-three years in South Africa)

Principally, xenophobia stems from ethnic friction combined with socioeconomic hardship. Foreigners, the makwerekwere, are blamed for taking jobs, spouses, and housing; pushing down already meagre wages; invading business communities with spaza shops; spreading HIV; holding odd beliefs; committing crimes; and so on. If anything, such notions have historically been (ab)used by populist political assemblages, scapegoating a ‘them’ as opposed to an ‘us’ and blaming migrants for broader structural problems for which they have come to pay a heavy price in South Africa.

When xenophobia last happened, I had a small cottage in Springs […]. They destroyed my cottage, destroyed it and took everything, you see, they stole everything from me. (Anita, aged thirty-seven, domestic worker, twenty years in South Africa)

We had to flee that zone [suburb] where we live in Honeydew […]. So, we fled and slept in the house of this White [person]. Then we waited for two days and when things had calmed down, we went back again. (Luciano, aged forty-two, construction worker, twelve years in South Africa)

On more than one occasion, xenophobia has flared up in South Africa in waves of disturbing incidents, during which fatal atrocities were committed against African migrants by disgruntled nationals. Roughly half of my informants in the clinic had somehow been affected; the remainder were very worried because, as Rosalina described it: ‘many people die, many people get hurt’ (field notes).

Yeah, I know a lad that was killed in his house in the squatter camp and another guy also died. He was my neighbour […]. He died, they burned him. They poured petrol over him, lit him up, and burned him alive there in Boksburg. (Rosa, aged twenty, hairdresser, twelve years in South Africa)

Eish! That one guy, they beat him up, you know, then he went to hospital. Even now he’s not okay. (Pico, aged forty-eight, pastor, twenty-five years in South Africa)

Whereas xenophobia erupted most violently in 2008, 2015, and again in 2019, it continues to cause what have been called ‘little tremors’, in reference to its ever-menacing presence in everyday life (McConnell 2009, 34).

This always happens to us [migrants]. What’s it called? Xenophobia, right? Many people die, many people end up crushed, many people lose their lives, really. I’m always worried. (Reniesa, aged thirty-five, domestic worker, twenty-one years in South Africa)

Yes, many of my friends [were affected], even one that ended up dead because they knifed him there in Thembisa. Now I’m nervous because of it. (Arnaldo, aged forty-two, water proofer, twenty-three years in South Africa)

Always there, as a structural factor, xenophobia and threats of violence keep migrants from vital services to which they are entitled, including healthcare. This is because of fear and self-stigmatisation, but certainly also because of exclusion actively imposed by locals.

Exclusion and exploitation

Apart from living with everyday preoccupations about police round-ups, mass deportations, and violence, African migrants also share concerns about being exploited by South African employers, or simply finding employment, as many struggle to get permanent jobs in the formal sector without valid documentation. Most manage only short-term contracts with low salaries and no benefits like medical aid or pension (Muanamoha, Maharaj, and Preston-Whyte 2010). Although they are less likely to be unemployed than local South Africans – possibly because it is the most active family members who migrate in the first place – migrants are more likely to work in areas without trade union representation. As such, they remain unshielded from exploitation, unlawful dismissal, and deportation instigated by South African employers who report them to authorities and then fraudulently retain their salaries as they are detained.

They don’t pay well cos I ain’t got that, whatcha call it? Work register. Well, I don’t have one, do I? (Luciano, aged forty-two, construction worker, twelve years in South Africa)

Here, recently? Yuh! I’m not doing anything. My husband was working, but now he’s not working anymore. He was a taxi driver, but had a bad accident with his arm […] so now we can’t pay the rent and the little ones won’t even be able to make it to school. (Rosalina, aged thirty-nine, unemployed, seventeen years in South Africa)

Migrants hold the most precarious positions on the labour market, which puts them at risk of other forms of abuse. One Mozambican youth angrily recounted to me how a security company had coerced its employees to regularly test for HIV and then dismissed those found to be positive (field notes). Commonly, when not jobless, Mozambicans in Gauteng work in construction, service, informal trade, industry, and on farms. While the nature of my informants’ positions in metropolitan Johannesburg was varied, jobs were very much determined by gender. Common city jobs for women were as domestic workers, in hair salons and beauty parlours, and as waitresses, street hawkers, or stallholders. Men typically found work on construction sites, in transport, or gardening. Other low-skill positions were available in car washes, garages, filling stations, and as cooks, shoemakers, tailors, cleaners, car guards, and in spaza shops or shebeens (pubs). A fair share still laboured for descendants of Portuguese migrants, typically as subordinate workers. Informants either battled to make ends meet or expressed profound dissatisfaction with their work.

I got sick in my lungs from working with all those damn chemicals and all that cold water, always wet and freezing. (Eugénio, aged thirty-four, carwash employee, one year in South Africa)

I don’t like this work that I’m doing, it’s just that, I work so that I won’t have to steal. It’s so that I won’t have to go hungry, you see? That’s why. (Anita, aged thirty-seven, domestic worker, twenty years in South Africa)

Another informal occupation seemed particularly common: while visiting Rosettenville one morning, a hairdresser from Beira gave me a piece of advice. ‘Look, you’ll find more Mozzies in the evening, because that’s when they all come back – the shoplifters and the pickpockets. They work all over town’ (field notes).

Though evidence of the relationship between poverty and HIV transmission remains mixed, penury in Africa has long been regarded as a contributing factor, with some theorists describing HIV as ‘one of the diseases of poverty’ (Mufune 2015, 1). Certainly, links between socioeconomic circumstances and treatment outcomes, such as chronic food insecurity and ART, remain strong, and it has long been shown how structural forces described as ‘neighbourhood effects’ and income dynamics can have calamitous consequences for the mortality of the marginalised (McDonough 2000; Schulz et al. 2000). Taking the lowest paying, most exploited, shortest-term jobs available, or not having one at all, leading at times to crime, African migrants must be regarded as among those most at risk of HIV and most likely to experience medical complications if treatment is entered despite of the structural barriers discussed here.

The piecemeal implementation of migrants’ rights and services in South Africa means that they are likely to be excluded, not only from better jobs with benefits, but also from vital public services to which they are often denied access. Many migrants report ‘being refused access to treatment at public clinics and hospitals, as well as being refused antiretroviral treatment for HIV/AIDS, because they do not carry a South African identification card’ (McConnell 2009, 39). In this way, even if a Mozambican migrant defies xenophobia and seeks care for HIV – in spite of the stigmas attached to the disease and to the makwerekwere label – he or she may be turned away, perhaps out of malice or due to a poor understanding among government service providers about migrants’ rights to be included.

Linguistic challenges

Another form of poor understanding – and an additional burden on migrants – is caused by language barriers. There is significant obstruction between what is chiefly their second language, Portuguese, and English, such that many may never learn English.

I dunno how to speak and have a conversation in English. (Belarmina, aged thirty-two, unemployed, five years in South Africa)

However, the learning curves between their first Bantu language and related, widely spoken languages in the region (including isiNdebele, chiShona, isiZulu, and isiXhosa) are not steep at all. This comes to determine with whom migrants might have meaningful interactions, and effectively prevents them from communicating well with at least some (wealthier) strata of society. For instance, one patient (Hélena) related that, while she did not speak a word of English, she got by quite well speaking a handful of African languages (field notes).

Although less than 1 percent of South Africa’s majority population speaks English as their home language, the dominant language of business and government is that of the former coloniser, Britain (Casale and Posel 2011). In exploring the relationship between English language proficiency and earnings in South Africa, Daniela Casale and Dorrit Posel (2011, 392) found that reading and writing English very well produces large returns, whereas, in contrast, there is ‘little evidence to suggest that language skills in an individual’s African home language are independently rewarded in the South African labour market’. Most informants spoke only their first Bantu language, usually Xitsonga-Shangaan, and Portuguese, in spite of having lived in South Africa for an average of sixteen-and-a-half years. This is because Mozambican populations in migrant communities can be so dense that it is entirely possible to exist cocooned within their cultural substructure and never (need to) have any noteworthy contact with locals, especially not those who speak neither a Bantu language nor Portuguese. Thus isolated, migrants may have little to no contact with speakers of other languages in South African society. Furthermore, partly due to xenophobia, Mozambicans prefer to mingle with other marginalised African migrants rather than locals: Zimbabwean, Malawian, Nigerian, and Congolese people in particular. It is likely not for a lack of effort that migrants do not engage with English or Afrikaans speakers, but simply for a lack of communicative and social contact with speakers of those languages. Migrants are offered few opportunities to learn English or any other skills needed to move beyond bottom-rung jobs.

While language barriers thus no doubt limit Mozambicans’ options and contribute to socioeconomic destitution, lacking language proficiency becomes especially problematic as they engage with healthcare services. Rosa despondently told me of her issue with the clinic: the doctors did not speak Portuguese and she did not comprehend their counsel. Similarly, Reniesa did not speak any English, but claimed to understand when staff would speak in isiZulu (field notes). Not comprehending what healthcare workers say is problematic, and, unsurprisingly, causes medical complications and misunderstandings.

What sometimes makes someone’s illness worsen is to have a doctor schedule a date for you to come in and do treatment, but then you don’t understand and don’t come, you see? Illness will continue when you don’t understand what the doctor says. If you don’t come on the day they scheduled, you’ll just get sick again. (Luciano, aged forty-two, construction worker, twelve years in South Africa)

For successful outcomes of ART and healthcare service integration, it is imperative that patients comprehend basic counselling and treatment guidelines, which was not always the case in Our Hope. This problem might be somewhat ameliorated either by staff or patients themselves, but ultimately there is no way around it other than translating printed material or affiliating multilingual interpreters to treatment programmes.

If I understand Zulu, they [staff] will find me a Zulu woman or a Zulu man so that I can understand. (Oliveira, aged thirty-six, carpenter, seven years in South Africa)

If I don’t understand, for example, when they speak too fast, I ask them to repeat and explain it to me when it’s in English. (Rosalina, aged thirty-nine, unemployed, seventeen years in South Africa)

Some informants expressed frustration at not being able to properly communicate ailments to the clinic’s staff and frequently asked me to translate during medical consultations. Even if a migrant wants to speak of medical complications, he or she might not find the words to do so – and much less understand the physician’s reply – adding to their social isolation. Clinical presentation of individuals is thus shaped by both structural and cultural variables, and poor language skills is certainly a factor that contributes to migrants’ erratic adherence to drugs and miscomprehensions about how HIV may be managed.

Medical pluralism: Western and traditional medicine

In sub-Saharan Africa, medical pluralism, defined as the use of multiple health systems (allopathic, complementary and alternative, faith-based, and traditional) is common, and traditional healers and herbalists play a pivotal role in care due to their accessibility and social acceptability. Most Mozambicans in South Africa consult with traditional healers and herbalists, referred to as n’angas or curandeiros.

In our culture, it’s like, we believe in it, you know? We want to speak to our ancestors. (Dudu, aged forty-seven, painter, eleven years in South Africa)

As Pico explained, most Mozambicans see healers as they are required to perform traditional spiritual ceremonies, although some will certainly try to hide it since it is frowned upon by church and medical staff and is sometimes considered ‘backwards’ (field notes). In Bantu ontology, there is little distinction between the ‘physical-psychological’, in the sense that disease pathogenesis is attributed to spiritual, not biological, aetiologies – for example, curses, spirits, ancestors, and so on (Bagnol 2017; Passador 2009).

So, when you get there [to the curandeiro], he’ll tell you so and so: ‘It was this lady that cursed your brother’. (Florinda, aged twenty, waitress, six years in South Africa)

Doctor, won’t you please help me get rid of this demon [HIV], please won’t you? (Felismina, aged thirty-four, domestic worker, seven years in South Africa)

You think: ‘Somebody bewitched me’. So you go to a sangoma [isiZulu: healer] and he says, ‘Somebody’s killing you’, or something like that, but only to find later that no, it’s nothing like that. (Luana, aged twenty-nine, unemployed, six years in South Africa)

In other words, traditional treatment persists in the era of ART, not as a secondary but rather as a persevering parallel pluralism, consisting of spiritual aspects of traditional Bantu ontology (Audet, Ngobeni, and Wagner 2017). However, particularly during early stages of HIV care, medical pluralism has been shown to contribute to delayed diagnosis and interrupted or abandoned therapy (Moshabela et al. 2017).

The curandeiro sometimes will give you medication that doesn’t take away HIV – it just makes it worse. Even here [in South Africa] there are people like that, we always see them, these people that are sick in this way [HIV] and what do they do? They go straight to the curandeiro. (Neves, aged forty-eight, gardener, twenty-six years in South Africa)

Studies from South Africa and Mozambique have consistently recommended that therapeutic alliances be forged between allopathic and traditional health practitioners to improve healthcare by generating synergies between health-worlds (Audet et al. 2015). Though national programmes throughout southern Africa work to ‘biomedicalise’ healers, such alliances have yet to efficiently materialise, principally due to mistrust and animosity between medical and traditional providers. Healthcare workers point to patients’ preference for traditional medicine, doubts as to the benefits of ART, and concerns about side effects as reasons for poor retention and adherence (Groh et al. 2011).

Later the curandeiros even told me that the pain in my feet [once a common side-effect of ARVs]: ‘It’s that demon you have inside who’s doing that’. (Reniesa, aged thirty-five, domestic worker, twenty-one years in South Africa)

It cannot be ignored that engagements with n’angas may be detrimental to therapeutic outcomes for unexperienced patients, and particularly for migrants from rural areas with deep-rooted traditional beliefs and little access to healthcare or education. While traditional healing may not always be entirely contradictory to HIV care – and may in fact be complementary when used for non-curative, spiritual purposes – it must be regarded as another potential obstacle to ART and exemplifies how cultural structures produce inequalities and create barriers to inclusion.

Sociocultural estrangement and stigma

Cultural and linguistic estrangement may retain migrants in deprived socioeconomic environments and limit their interaction with wealthier segments of society, but nothing isolates like the social stigma of HIV, even among peers. Recent studies from South Africa demonstrate how HIV continues to be severely stigmatised within families and local communities (Gamarel et al. 2017; Steenberg 2019a). In one telling illustration of this, businesses with known HIV-positive workers were shown to lose up to half of their customers out of fear of transmission and disdain for people living with HIV (Chao et al. 2017). Stigma also represents a significant barrier to testing and makes people reluctant to disclose their serostatus to partners (Simbayi et al. 2017; Treves-Kagan et al. 2017). Contemplating this topic, Rosa explained that people are afraid of HIV, and that when someone is infected, it is not spoken of and then becomes a secret to the extent that no one can know, ‘not even your own family’ (field notes).

I have previously described a positive HIV test as a ‘transformative othering’, in that it is ‘a schism between patienthood and personhood, in which a patient’s dread of physical demise in the future transforms into fears of social death in her present-day life-world, resulting in loneliness and secrecy’ (Steenberg 2019b, 3). A positive diagnosis, and the succeeding counselling and ART initiation, is a defining, life-changing moment where the perception of the disease transitions from ‘death sentence’ to ‘chronic but manageable’. Maria recalled how she had been terrified that HIV would kill her, but after starting on ARVs, she realised it would not and was no longer afraid (field notes).

Following this transformation, patients begin to perceive that, lacking their own experience, HIV-negative community members are ignorant, in the sense that they (still) associate HIV with suffering and death, and are therefore likely to shun and expel anyone known to be infected from their lives. Stigma, as an ever-present social angst, thus brings about an othering of ‘normals’ (non-patients). Jorge insisted that others would cut all ties and think of him as dead should they learn of his serostatus. Others, such as Rosa, think of HIV as a death sentence and do not believe it can be treated. Pico believed that disclosure would only lead others to hate him and gossip about him, while Hélena clarified that, even in this day and age of treatment, she had family members who refused to sit next to ‘a sick person’ (field notes).

Given HIV’s high prevalence (one in every five), one might expect patients to know others with the same diagnosis, especially in their families. Yet, because of stigma and concealment, this is not so and ‘disclosure among migrants is rare and confined exclusively to a tight, inner social circle of trusted confidants’ (Steenberg 2019a, 9). No one in Gonçalves’s family was aware of his illness. Arnaldo said that he had not told anyone in his family, because he could not bring himself to do so. A polygamous patient, Luciano, described how only his first wife was aware, but not the second one, because he was ‘still afraid to tell her’. Jorge explained that the mother of his children knew, but no one else in his family, as that would surely result in harassment (field notes).

Instead, most informants reported disclosing to one or two close personal friends, who might also be family, a brother or a sister, but not necessarily. Disclosure is thus linked not exclusively to familial bonds, but rather to interpersonal levels of intimacy and trust. The fact that many migrants’ close acquaintances and potential confidants still reside in a now distant Mozambique significantly limits their deeply personal choice of confidants and contributes to solitude and silence in the community. Tearful, Rosalina told me how she had cried in solitude and that a feeling of profound loneliness had come over her following her diagnosis. Hélena specifically characterised HIV as ‘a lonely secret’ as she recollected the terror of testing positive, and then not wanting to tell anyone, fearing the consequences (field notes).

Intensifying such fears of enacted stigma and gossip, Mozambican migrants have relatively small social networks. In their communities, secrecy and silence result in non-disclosure and evasion of exposure in order to ward off the stigmas of HIV (Davis and Manderson 2014; Moyer and Hardon 2014). Concealment thus becomes an act of containment and a strategy for social survival, which, in turn, sustains the psychological processes that replicate stigma in society (Monteiro et al. 2018). This social mechanism is detrimental not only to clinical outcomes of care, but also to HIV’s social normalisation alongside its medicalisation. Take as an example Luciano, who had informed his wife in South Africa of his HIV infection, but not his wife in Mozambique, or anyone else there, because that would make it ‘a lot harder to find more wives’ (field notes).

Conclusion

I have explored how a larger structural context of socioeconomic forces, determinants, and inequalities comes to shape health and illness even before Mozambican migrants encounter a clinician (if they do so at all). I have also examined how subsequent biosocial experiences of therapy and care are further moulded by cultural and ontological circumstances of living with HIV. In conjunction, these structures thwart immigrants’ integration into South African healthcare services and complicate access to HIV testing, counselling, treatment-seeking, and care. Concomitantly, conditions particular to immigrants frustrate the social implications of disclosure in isolated communities.

Firstly, in journeying from sending to destination areas, Mozambican migrants take part in an ever-flowing, self-sustained circle of labour migration, which is precarious in itself, involving risks of robberies, abuse, and payoffs. Further afield, once settled, migrants have few resources, and consequently find themselves in enclaves or destitute townships among locals who strongly dislike them. Besides the risk of being removed from care entirely (deportation), the ubiquitous menace of xenophobia, combined with institutional discrimination against the makwerekwere, represents a significant challenge for healthcare provision. Alongside this threat of violence, those without legal documentation live in ever-present fear of incarceration, which constricts interaction with social services and healthcare facilities. Secondly, employed within the most exploitative and least protected sectors of the economy, migrants are not only bereft of benefits such as trade unions, pensions, or health insurance, but are also frequently taken advantage of by employers. This abuse extends even to South African public service providers, such as clinics and schools, which often exclude migrants, which further limits access and integration. Thirdly, language barriers and scarce opportunities to engage with English contribute to the embeddedness of migrants in niche communities and belated societal integration, which, in turn, increases vulnerability to HIV and other illnesses. Most importantly here, apart from keeping migrants tied down in small-time, informal economies, lacking language proficiencies hampers communication in the clinic, where advice and counselling offered by healthcare staff is misunderstood leading to medical complications. Moreover, for that same reason, migrant patients are uneasy about inquiring into their condition. Fourth, adding much to confusion and miscomprehension, the clinic may not be the only health-world of migrants. In parallel to consulting with clinical practitioners, Mozambicans customarily co-consult with traditional herbalists whose spiritual aetiology of disease is quite incompatible with medical care, which may contribute to delayed diagnosis and interrupted or abandoned ART. Finally, and perhaps most problematically, HIV-positive Mozambican migrants are further isolated due to extensive stigma in small, closely knit communities. Stigma is already hugely problematic for local South Africans, but even more so for migrants. This is because those able to ameliorate the loneliness of the condition are often far away and because social networks abroad are relatively small, which makes disclosure risky and undesirable. Whereas there may be some social benefits to not being exposed to family networks in the South African clinic far from home, potential exposure there is to social networks tied to spaces of employment and livelihood, which makes exposure and disclosure even more risky due to financial concerns. This illustrates that stigma is tied to location, because location is tied to social networks. Non-disclosure and concealment as an act of containment thus become a preferred strategy for social survival.

Compounded, these structural vulnerabilities are detrimental to HIV care and may increase rates of morbidity and mortality and contribute to viral mutation and drug resistance. National South African social and health policies can only truly be effective when they begin to address migration, as well as areas such as entry into labour markets, social protection services, access to healthcare, medical pluralism, xenophobia, and the stigmas of living with HIV.

Acknowledgements

I am immensely grateful to Professor Lenore Manderson and my colleagues at Wits’ School of Public Health. I thank Dr. Sello Mashamaite and Right to Care for opening the doors to Themba Lethu Clinic and Ms. Erma Cossa for painstaking research assistance.

About the author

Bent Steenberg is a medical anthropologist affiliated to the University of the Witwatersrand, who studies African immigration and social and political aspects of HIV care. He has undertaken extensive fieldwork in HIV clinics and hospitals in Mozambique and South Africa.

References

Adams, Vincanne, Sharon R. Kaufman, Taslim van Hattum, and Sandra Moody. 2011. ‘Aging Disaster: Mortality, Vulnerability, and Long-Term Recovery among Katrina Survivors’. Medical Anthropology 30 (3): 247–270. https://doi.org/10.1080/01459740.2011.560777

Africa Check. 2016. Factsheet: Detention and Deportation of Undocumented Migrants in South Africa at the Lindela Repatriation Centre. https://africacheck.org/factsheets/lindela-repatriation-centre-migrants/.

Audet, Carolyn M., Erin Hamilton, Leighann Hughart, and Jose Salato. 2015. ‘Engagement of Traditional Healers and Birth Attendants as a Controversial Proposal to Extend the HIV Health Workforce’. Current HIV/AIDS Reports 12 (2): 238–245. https://doi.org/10.1007/s11904-015-0258-8.

Audet, Carolyn M., Sizzy Ngobeni, and Ryan G. Wagner. 2017. ‘Traditional Healer Treatment of HIV Persists in the Era of ART: A Mixed Methods Study from Rural South Africa’. BMC Complementary and Alternative Medicine 17 (1): 434. https://doi.org/10.1186/s12906-017-1934-6.

Bagnol, Brigitte. 2017. ‘The Aetiology of Diseases in Central Mozambique: With a Special Focus on HIV/AIDS’. African Studies 76 (2): 205–220. https://doi.org/10.1080/00020184.2017.1322867.

Bertaux, Daniel. 1980. ‘L'approche Biographique: Sa Validité Méthodologique, ses Potentialités’. Cahiers Internationaux de Sociologie 69: 197–225.

Browner, Carole H., and Carolyn F. Sargent. 2011. Reproduction, Globalization, and the State: New Theoretical and Ethnographic Perspectives. Durham, NC: Duke University Press.

Casale, Daniela, and Dorrit Posel. 2011. ‘English Language Proficiency and Earnings in a Developing Country: The Case of South Africa’. The Journal of Socio-Economics 40 (4): 385–393. https://doi.org/10.1016/j.socec.2011.04.009.

Chao, Li-Wei, Helena Szrek, Rui Leite, Shandir Ramlagan, and Karl Peltzer. 2017. ‘Do Customers Flee From HIV? A Survey of HIV Stigma and Its Potential Economic Consequences on Small Businesses in Tshwane (Pretoria), South Africa’. AIDS and Behavior 21 (1): 217–226. https://doi.org/10.1007/s10461-016-1463-1.

Crush, Jonathan. 2008. ‘The Perfect Storm: The Realities of Xenophobia in Contemporary South Africa.’ The Southern African Migration Project. Migration Policy Series no. 50.

Davis, Mark, and Lenore Manderson. 2014. Disclosure in Health and Illness. London: Routledge.

DESA. 2016. International Migration Report 2015. New York: United Nations.

Emerson, Robert M., Rachel I. Fretz, and Linda L. Shaw. 2011. Writing Ethnographic Fieldnotes. Chicago: University of Chicago Press.

Farmer, Paul. 2001. Infections and Inequalities: The Modern Plagues. Berkeley: University of California Press.

Gamarel, Kristi E., Caroline C. Kuo, Mark E. Boyes, and Lucie D. Cluver. 2017. ‘The Dyadic Effects of HIV Stigma on the Mental Health of Children and Their Parents in South Africa’. Journal of HIV/AIDS & Social Services 16 (4): 351–366. https://doi.org/10.1080/15381501.2017.1320619.

Gilbert, Leah, and Liz Walker. 2010. ‘“My Biggest Fear was that People Would Reject Me Once They Knew My Status…”: Stigma as Experienced by Patients in an HIV/AIDS Clinic in Johannesburg, South Africa’. Health & Social Care in the Community 18 (2): 139–146. https://doi.org/10.1111/j.1365-2524.2009.00881.x.

Glaser, Clive. 2010. ‘Portuguese Immigrant History in Twentieth Century South Africa: A Preliminary Overview’. African Historical Review 42 (2): 61–83. https://doi.org/10.1080/17532523.2010.517398.

Groh, Kate, Carolyn M. Audet, Alberto Baptista, Mohsin Sidat, Alfredo Vergara, Sten H. Vermund, and Troy D. Moon. 2011. ‘Barriers to Antiretroviral Therapy Adherence in Rural Mozambique’. BMC Public Health 11 (1): 650. https://doi.org/10.1186/1471-2458-11-650.

IaM. 2017. Inquérito de Indicadores de Imunização, Malária e HIV/SIDA em Moçambique (IMASIDA 2015). Maputo: INS.

Kalengayi, Faustine Kyungu Nkulu, Anna-Karin Hurtig, Clas Ahlm, and Ingela Krantz. 2012. ‘Fear of Deportation may Limit Legal Immigrants’ Access to HIV/AIDS-Related Care: A Survey of Swedish Language School Students in Northern Sweden’. Journal of Immigrant and Minority Health 14 (1): 39–47. https://doi.org/10.1007/s10903-011-9509-y.

McConnell, Christy. 2009. ‘Migration and Xenophobia in South Africa’. Conflict Trends 2009 (1): 34–40.

McDonough, Peggy, Greg J. Duncan, David B. Williams, and James S. House. 2000. ‘The Impact of Income Dynamics on Mortality in the United States’. In The Mortality Crisis in Transitional Economies, edited by Giovanni Andrea Cornia and Renato Paniccià, 392–419. New York: Oxford University Press.

Metzl, Jonathan M., and Helena Hansen. 2014. ‘Structural Competency: Theorizing a New Medical Engagement with Stigma and Inequality’. Social Science & Medicine 103: 126–133. https://doi.org/10.1016/j.socscimed.2013.06.032.

Monteiro, Simone, Wilza Villela, Priscilla Soares, Adriana Pinho, and Livia Fraga. 2018. ‘“Protective” Silence Surrounding AIDS: Reasons and Implications of Non-Disclosure among Pregnant Women Living with HIV in Rio de Janeiro (Brazil)’. Global Public Health 13 (1): 51­–64. https://doi.org/10.1080/17441692.2016.1211163.

Moshabela, Mosa, Dominic Bukenya, Gabriel Darong, Joyce Wamoyi, Estelle McLean, Morten Skovdal, William Ddaaki, Kenneth Ondeng’e, Oliver Bonnington, and Janet Seeley. 2017. ‘Traditional Healers, Faith Healers and Medical Practitioners: The Contribution of Medical Pluralism to Bottlenecks along the Cascade of Care for HIV/AIDS in Eastern and Southern Africa’. Sexually Transmitted Infections 93 (3): 1–5. https://doi.org/10.1136/sextrans-2016-052974.

Moyer, Eileen, and Anita Hardon. 2014. ‘A Disease Unlike Any Other? Why HIV Remains Exceptional in the Age of Treatment’. Medical Anthropology 33 (4): 263–269. https://doi.org/10.1080/01459740.2014.890618.

Muanamoha, Ramos Cardoso, Brij Maharaj, and Eleanor Preston-Whyte. 2010. ‘Social Networks and Undocumented Mozambican Migration to South Africa’. Geoforum 41 (6): 885–896. https://doi.org/10.1016/j.geoforum.2010.06.001.

Mufune, Pempelani. 2015. ‘Poverty and HIV/AIDS in Africa: Specifying the Connections’. Social Theory & Health 13 (1): 1–29. https://doi.org/10.1057/sth.2014.14.

Parker, Richard, and Peter Aggleton. 2003. ‘HIV and AIDS-Related Stigma and Discrimination: A Conceptual Framework and Implications for Action’. Social Science & Medicine 57 (1): 13–24. https://doi.org/10.1016/s0277-9536(02)00304-0.

Passador, Luiz Henrique. 2009. ‘“Tradition”, Person, Gender, and STD/HIV/AIDS in Southern Mozambique’. Cadernos de Saúde Publica 25 (3): 687–693. https://doi.org/10.1590/S0102-311X2009000300024.

Pearson, C. R. , M. A. Micek, J. Pfeiffer, P. Montoya, E. Matediane, T. Jonasse, A. Cunguara, D. Rao, and S. S. Gloyd. 2009. ‘One Year After ART Initiation: Psychosocial Factors Associated with Stigma among HIV-positive Mozambicans’. AIDS and Behavior 13 (6): 1189. https://doi.org/10.1007/s10461-009-9596-0.

Quesada, James, Laurie Kain Hart, and Philippe Bourgois. 2011. ‘Structural Vulnerability and Health: Latino Migrant Laborers in the United States. .Medical Anthropology 30 (4): 339–362. https://doi.org/10.1080/01459740.2011.576725.

Schulz, Amy, David Williams, Barbara Israel, Adam Becker, Edith Parker, Sherman A. James, and James Jackson. 2000. ‘Unfair Treatment, Neighborhood Effects, and Mental Health in the Detroit Metropolitan Area’. Journal of Health and Social Behavior 41 (3): 314–332. https://doi.org/10.2307/2676323.

Simbayi, Leickness C., Nompumelelo Zungu, Meredith Evans, Vuyelwa Mehlomakulu, Takura Kupamupindi, Goitseone Mafoko, and Khangelani Zuma. 2017. ‘HIV Serostatus Disclosure to Sexual Partners among Sexually Active People Living with HIV in South Africa: Results from the 2012 National Population-Based Household Survey’. AIDS and Behavior 21 (1): 82–92.

StatSA. 2016. Community Survey 2016. Statistical Release P0301. Pretoria: Statistics South Africa.

Steenberg, Bent. 2019a. ‘HIV-Positive Mozambican Migrants in South Africa: Loneliness, Secrecy and Disclosure’. Culture, Health & Sexuality: 1–16. https://doi.org/10.1080/13691058.2019.1571230.

Steenberg, Bent. 2019b. ‘Patients and Personhood: Perceptions of HIV in Mozambican Immigrants in South Africa’. Medical Anthropology: 1–14. https://doi.org/10.1080/01459740.2019.1677646.

Treves-Kagan, Sarah, Alison M. El Ayadi, Audrey Pettifor, Catherine MacPhail, Rhian Twine, Suzanne Maman, Dean Peacock, Kathleen Kahn, and Sheri A. Lippman. 2017. ‘Gender, HIV Testing and Stigma: The Association of HIV Testing Behaviors and Community-Level and Individual-Level Stigma in Rural South Africa Differ for Men and Women’. AIDS and Behavior 21 (9): 2579–2588. https://doi.org/10.1007/s10461-016-1671-8.

UNAIDS. 2019. Factsheet South Africa. Joint United Nations Programme on HIV/AIDS 2019. http://www.unaids.org/en/regionscountries/countries/southafrica.

Vidal, Dominique. 2010. ‘Living in, Out of, and Between Two Cities: Migrants from Maputo in Johannesburg’. Urban Forum 21: 55–68. https://doi.org/10.1007/s12132-010-9080-y.

Wentzel, Marie, and Kholadi Tlabela. 2006. ‘Historical Background to South African Migration’. In Migration in South and Southern Africa: Dynamics and Determinants, edited by Pieter Kok, Derik Gelderblom, John Oucho, and Johan van Zyl, 71–96. Cape Town: HSRC Press.