‘It’s in our veins’: caring natures and material motivations of community health workers in contexts of economic marginalisation

Khayelitsha, an economically marginal peri-urban settlement in Cape Town, is home to a number of ‘flagship’ public health interventions aimed at HIV/AIDS and TB. Alongside these high-profile, foreign donor-driven treatment and care programmes are a plethora of NGOs that provide a wide range of community-based carework. Some of these organisations are large, well funded and well connected globally, while others are run by a few unemployed women responding to care needs in their neighbourhoods. This article explores the ways that community health workers (CHWs) who work for these organisations understand and speak about their involvement in carework as volunteers, employees or managers of community-based care organisations. Many CHWs framed their work through discourses of gender, religion or culture (‘African-ness’). They also described forms of material or economic benefits of providing carework, but many were concerned that these might be seen as existing in tension with more socially accepted, altruistic motivations for care. We explore here how CHWs narrate and understand their roles and motivations as carers and members of a resource-constrained community.


Introduction
In South Africa, and throughout Sub-Saharan Africa, community health workers (CHWs) have played an important role in meeting the health needs of populations in resource-constrained settings (Haines et al., 2007;WHO, 2006WHO, , 2008. CHWs have been part of the South African health system for almost a hundred years. Today, over 65,000 CHWs perform a wide range of care-related tasks in impoverished communities, mainly linked to treatment and counselling support for HIV and TB (Department of Health, 2011).
Although it is widely believed that CHWs can perform a long-term strategic role in 'bridging the gap' between the health system and the community (Schneider, Hlophe, & van Rensburg, 2008), most recent CHW programmes in Southern Africa have emerged in an ad hoc fashion, in response to the immediate health crises of HIV and TB. These programmes have also relied on the volunteer labour of community members willing to address these urgent needs. As these programmes have matured, however, and as the government moves towards the large-scale formalisation of the CHW role within the health system (van Pletzen, Zulliger, Moshabela, & Schneider, 2013;Zachariah et al., 2009), it is increasingly clear that too little is known about who these CHWs are and the motivations that keep them engaged in this work. The South African government is, for example, currently in the process of rolling out a 're-engineering' of its primary health care services and CHWs are set to play a key role as core members of these new service's community outreach teams (National Department of Health, 2011). As CHWs' roles and scopes of work shift and potentially expand in response to new policies and programmes, it is imperative to explore CHW motivations more closely.
Despite the fact that many programmes rely on altruism as a primary source of motivation, and complement this with small monetary 'incentives', CHW motivation is much more complex and contextually determined. Even when considering the incentives used by programmes to recruit and retain CHWs, there is a range of possible incentives that can be offered, including direct or indirect, financial or non-financial, each of which can have varying effects depending on individuals and contexts (Bhattacharyya, Winch, LeBan, & Tien, 2001).
This article explores the multidimensional character of CHW motivation in an economically deprived community in Cape Town. We sought to understand how CHWs themselves speak about their involvement in carework as volunteers, employees or managers of small, community-based care organisations. Using ethnographic data, we highlight the ways CHWs framed their carework strongly in terms of gender, religious or cultural discourses of care. They also described carework's economic benefits but were concerned that these motivations would be seen as in tension with more socially acceptable rationales for altruistic forms of care. We describe, therefore, how CHWs navigate their experiences as both carers and members of a resource-constrained community in order to produce socially legitimate forms of caring labour and relationship. Ultimately, we argue that neither altruism nor financial incentives on their own are sufficient explanations of CHW motivations in resource-constrained settings like Khayelitsha.

This study
Ethnographic fieldwork was conducted in Khayelitsha, Cape Town, a 'township' created in 1983 during the closing decades of the apartheid era. It is the quintessential dormitory suburb for black workers close to, but not encroaching on, the 'white' cities (Spiegel & Mehlwana, 1997). Khayelitsha has high rates of poverty and ill health, a substantial burden of unemployment and poor service delivery, including inadequate access to housing, sanitation, education and security (De Swardt, Puoane, Chopra, & du Toit, 2005;Xali, 2002). In the formal areas, 'serviced sites' have tarred roads, mast lighting, water and sewage services, and, more recently, access to electricity. 'Informal areas' have few, if any of these, basic services, and political struggles to increase access to these services are ongoing (Nleya, 2011).
As Khayelitsha grew, so did the burdens of both infectious diseases, mainly HIV and TB, and non-communicable diseases including cancers, hypertension and diabetes. In response, a significant number of CHWs have emerged to perform a wide range of tasks to meet these health needs. The range of tasks performed by CHWs is farreaching and includes home-based care, treatment support, counselling, health education and promotion, referrals to health facilities, and nutritional and social service support. The large majority of them work with specific health conditions, most commonly HIV and TB. More than half of the CHWs earn a small stipend; about 20% work solely as volunteers and the remainder earn some form of salary.
Once ethical approval was secured from the University of Cape Town's Faculty of Health Sciences Human Research Ethics Committee (#042/2010), we spent approximately six months conducting research with a number of CHWs and organisations. During the fieldwork, AS and CC engaged with a wide range of people but conducted more formal, in-depth interviews with 23 key informants including CHWs, NGO staff and managers running CHW programmes, community members, and religious and political leaders. While in the field, AS repeatedly relayed initial interpretations of CHW experience and motivation back to people to ensure that interpretations were valid. Data were analysed using thematic analysis to identify initial patterns and themes. Narrative and critical discourse analyses were used to look closely at the words and narrative forms people used to describe their experiences and explain their motivations (Pope, Ziebland, & Mays, 2000;Ulin, Robinson, & Tolley, 2005). Our approach to data analysis was 'critical' in the sense that, we explored how relations of power shaped thought and practice, and 'interpretative' in the sense that we provided a holistic account of how motivation emerges and changes in this context. Initial access to the field site was negotiated through Monwabisi Maqogi, known to us as Monwa. Monwa is a well-known pastor and community activist who introduced us to many people to whom we might not have otherwise had access. Though Monwa accompanied both authors on some of the fieldwork, we also developed independent relationships in the field.

Naturalised discourses of care
CHW motivation is based not only on individual motivating factors, but is also shaped by the relationships between CHWs, community members and the health system. In the following sections, we explore three naturalising discourses of care commonly used to explain and justify CHW motivation: gender, African-ness and Christianity.

Gendered natures
Gender is one of the strongest naturalising discourses of care. Globally, there is a common expectation that women will naturally engage in the caring labour required for physical health and social reproduction. Gender theorists and feminist scholars have explored the relationships between paid/unpaid labour, productive/reproductive labour, gender inequality and the 'double burden of care' (Akintola, 2006;Budlender, 2003;Fisher & Tronto, 1990;Hervey & Shaw, 1998). Women are seen as the 'natural' providers of care and it is assumed that this is what shapes women's ability and their motivation to engage in CHW work (Akintola, 2005;Clarke, 2005;Daniels, Clarke, & Ringsberg, 2012;Patel, 2009).
In Khayelitsha, Thembi, a young CHW explained the gendered ways she thinks about care: The work we are doing is for the women. The men don't feel as bad as women. The women feel that pain. Men feel a little bit but men don't feel a lot. There was one man but he left and he doesn't work with the organisation anymore.
When we asked women how they became involved in offering care-related services, many narrated painful experiences of caring for loved ones as a primary motivation for becoming involved in community health work. Many women have to shoulder these care burdens individually within families and communities.
Although carework is highly feminised in Khayelitsha, as it is globally, some men also provide care. Monwa, the local pastor mentioned above, highlighted some of the complexities of gendered care as he told the story of when he first went to wash a chronically ill female patient, saying that it was 'so painful' for him and describing how she was first silent, withdrawn, even resistant but finally deeply grateful to him for offering her care. What made it possible for Monwa, as a man, to perform this intimate and gendered task of care was not a strong personal commitment to challenging gender norms but rather a Christian ethic of care. His identity and status as a pastor enabled him to cross gendered lines of care that are typically strongly defended.
The ways gender shapes the work of caring thus depends not only on broad gender norms but also on the specific question of who is providing care and why (Patel, 2009). In his mobilisation of religion to explain his caring practice, Monwa works around but does not directly confront hegemonic notions of gendered care.
There are times when the gendering of care comes into conflict with other discourses of CHW motivation. For example, despite the fact that most CHWs see gender, carework and empathy as strongly linked, we also found that many women first enter into formal carework in search of financial remuneration. When they start working in the community as compensated careworkers however, they find themselves in tension between being community members as well as service providers who have access to (limited) resources. To be able to argue that one has become a CHW because, as a woman, you are more 'naturally' inclined to care may be an important way to downplay the fact that carework allows for access to scarce resources (Patel, 2009;Swartz, 2013).
'It's in our veins': the 'African-ness' of caring Another discourse that frames and justifies CHW motivation is the notion of ubuntu, generally described as a uniquely Southern African humanistic ethic of care (De Wet, 2012;Manda, 2010;Tutu & Allen, 1994). Many CHWs explained their carework in terms of this African ethic of ubuntu. One morning, after visiting an organisation of home-based carers, we asked Monwa why these women choose to volunteer. He answered, as though it was obvious, that 'it's in our veins to care like this'. The 'our' in his statement referred to his community in Khayelitsha; however, this notion of ubuntu is also often indexed to rural village life in the neighbouring Eastern Cape province, a place still considered 'home' by many Khayelitsha residents. The discourse of ubuntu highlights the importance of close family and community relationships that value interconnectedness and responsiveness to collective needs.
This discourse of the natural capacity and responsibility of African people to care for one another has also framed the work of CHW organisations. Across the range of organisations, large and small, the discourse of ubuntu is pervasive in accounts of why individuals and organisations care for the needs of fellow community members.
Ubuntu is seen by many, however, to be under threat. Older CHWs argued that ubuntu was being 'killed off' by contemporary urban life, saying that younger careworkers are only interested in employment and personal career development. Mzi, an older CHW with whom AS spent many hours, explained, 'ubuntu is gone', adding, 'It's only the old people who have that real passion for caring for people'. But Mzi believes that there is another layer to the loss of ubuntu. She believes ubuntu is gone because people literally do not have anything to give one another anymore. The ideal of reciprocity is challenged in spaces like Khayelitsha where illness, migration and economic deprivation make living up to the demands of ubuntu difficult. In this context, it is often people's neighbours and wider networks of community members, rather than their immediate families, who often end up responding to care needs.
The practice of ubuntu thus becomes both thinner and broader under these pressures. Many CHWs said that though life in the city and in poverty makes ubuntu difficult, it is still something 'in their veins' that they need to do. They said it is home-based carers' 'responsibility' to help. This discourse was often used to explain the continued broad commitment to community care, despite the severe economic constraints faced by CHWs.
Ironically, the argument about 'African-ness' that is used to explain CHW volunteering is also used to account for the reasons HIV is so prevalent here. One CHW explained that although 'Khayelitsha' means 'new home', it is 'a terrible place' where people have 'a township mind'. He said that many of the people living in Khayelitsha have been 'taught culturally', and do not know enough about their rights and responsibilities with respect to HIV and sexual relationships. This contradictionthe idealisation of 'African-ness' and its denigrationis but one clear example of the complex terrain into which caring in Khayelitsha is embedded.
A Christian ethic of care All of the CHWs we met in Khayelitsha were women, although some men found themselves in caring roles. For Monwa, caring is very closely linked to his Christianity, the third naturalised discourse of care. A Christian ethic is certainly not limited to leaders in the church, or to Khayelitsha. Links between care, volunteerism and Christianity have been documented in South Africa and beyond (Akintola, 2010(Akintola, , 2011Jack, Kirton, Birakurataki, & Merriman, 2012;Kegler, Hall, & Kiser, 2010;Patel, 2009).
When AS first met Monwa, there were 17 people living in his small three-bedroom house. He says that to open your house to people is 'the Christian way'. Monwa's framing of his caring practices through Christianity also allows him and his church members to care in ways that challenge conventional ideas about care and HIV. Monwa uses Christian moral principles to confront local churches where the stigmatisation of people of HIV is still very prevalent (Campbell, Foulis, Maimane, & Sibiya, 2005;Levy, Miksad, & Fein, 2005). Another pastor and his wife said they set up their CHW organisation when they saw people being cruelly excluded from churches after HIV diagnosis.
There are, however, limits to the use of Christian ethics to support HIV-positive patients. One local pastor who was very inclusive of HIV-positive people in his congregation also encouraged other local pastors to remember that it is always possible that people acquired their HIV infection 'through a car accident' or blood transfusion. His encouragement to care for those with HIV and not to exclude them from the congregation was rooted therefore in ambivalence about the moral problem of HIV infection and a need to suggest alternative routes of infection. Christian discourses of care are thus being used to challenge the dominant discourses through which people with HIV are excluded, stigmatised and discriminated against, one of which, ironically, is also Christianity.

Economic motivations for carework
The naturalised discourses of care described above are strong and pervasive in CHWs' more public accounts of what drives them. There is, however, another complicated set of discourses around economic motivations for carework that often sits in tension with the more essentialising explanations above. Both CHWs and community members typically perceive economic motivations, and indeed extrinsic motivations more generally, as less legitimate than intrinsic ones. In a context of economic deprivation, however, economic motivations create a powerful pull towards carework. The sections below discuss some of the ways CHWs have learned to negotiate conflicting discourses around the motivation to care.
The local political economy of carework, past and present In the early twentieth century, a number of progressive health reforms in South Africa featured CHWs prominently (Jeeves, 2000;Marks, 1997). These initiatives were the vanguard of what would later emerge as the primary health care (PHC) movement in South Africa and elsewhere (Marks & Andersson, 1992;Tollman & Pick, 2002). The National Party, which ascended to power in 1948, reversed many of these reforms, however, and CHW programmes struggled or disappeared. In the 1970s and 1980s, growing political crisis led to a renewal of CHW programmes by politically progressive health activists (van Ginneken, Lewin, & Berridge, 2010). The funding for these programmes was both local and international and had few strings attached.
While many hoped that this enlightened PHC approach would provide the framework for post-apartheid public health, the new democratic government, elected in 1994, incorporated relatively few PHC principles into its work. Health system reform immediately after apartheid was mostly technicist and disease oriented and CHWs were left out of public health policy and strategy (Coovadia, Jewkes, Barron, Sanders, & McIntyre, 2009;van Ginneken et al., 2010).
Instead of fading away, however, the number of CHWs post apartheid ballooned, driven by HIV and the Mbeki administration's weak response to the epidemic. As Mbeki's Department of Health dragged its feet and got caught up in a long-running battle over its AIDS denialism, the massive health needs created by HIV persisted and deepened in many communities. Donors and civil society organisations quickly filled the gap and in the space of a few years, there were thousands of CBOs and NGOs performing a wide range of volunteer activities. There were also millions of rands moving through these programmes, often bypassing the national government entirely. Khayelitsha, an historically popular site for pilot public health interventions, was at the centre of many of these developments.
The response from civil society had a number of unanticipated consequences. Rapid growth in CHW programmes led to fragmentation and verticalisation of care, duplication of services, competition between providers for funding and for patients, and rapidly increasing numbers of CHWs, often poorly trained and ineffectively supervised (Lehmann, Matwa, Schneider, & Colvin, 2009;Schneider et al., 2008). The formalisation of some aspects of carework by the government also resulted in the creation of new categories of CHWs (e.g. HIV treatment supporters) as well as new models for state payment of CHWs via NGO mediators (Schneider et al., 2008).
In the process, though CHW work became somewhat more formalised, it also remained highly fragmented and poorly regulated. Most carework was delivered by NGOs with little regulation or supervision by the state health services. While this created space for creative local responses to health needs, it also resulted in a confusing terrain of 'fly-by-night' CHW training programmes, financially unsustainable CHW initiatives and highly unevenly distributed CHW services across the country.

Intrinsic and extrinsic motivations to care
This was the complex environment recent CHWs have had to negotiate. In light of post-apartheid's epidemiological crises (especially HIV and TB) and its politicaleconomic legacies (of deep social, political and economic inequalities) questions quickly came to the surface about the relationship between intrinsic and extrinsic motivations. The urgent and growing need for care in communities was palpable. At the same time, however, the dramatic in-rush of local and foreign funding fostered new forms of caring relationships and incentives to care, especially for those new to CHW work.
For many CHWs in Khayelitsha, this established a tension between the CHW as 'volunteer' vs. the CHW as 'worker'. This reduction of CHW motivation to two, presumably incompatible drivesaltruism or personal benefitis a common one in broader policy debates (Akintola, 2011;Bhattacharyya et al., 2001;Kironde & Klaasen, 2002). And indeed, the notion that intrinsic and extrinsic motives are different in kind and often in conflict, is a long-standing premise of moral philosophies around the world (England, 2005). The intrinsic is often privileged over the extrinsic and is even often considered a threat to loftier moral principles. These forms of motivation are also often strongly gendered, with women typically linked to intrinsic motivations and less in need of either social recognition or economic reward (England, 2005;Nelson, 1999).
Extrinsic and intrinsic motivations are often even seen to be in active conflict. The potential, for example, of money to corrupt or corrode genuine or altruistic caring relationships has been extensively explored. This is an old concern but has recently found new expression in the idea that incentives that address extrinsic motivations may risk 'crowding out' presumably less substantial intrinsic motivations (Frey & Jegen, 2000;Georgellis, Iossayz, & Tabvuma, 2010;Pink, 2009). Paying 'too much', these authors fear, may 'crowd out' the forms of intrinsic motivation that led CHWs to caring practice in the first place, reducing care to mere labour. There is also an assumption built into this perspective that people have differing and unsteady capacities for intrinsically motivated care (depending on personality and/or culture) while the power of extrinsic motivations remains constant, mechanistic and universal (Nelson, 1999).
These perceived tensions between intrinsic and extrinsic motivations are heightened, however, in economically marginalised settings, where health needs are great, resources are scarce and work opportunities are few. In these contexts, suffering is pervasive and the urge to respond is activated daily. At the same time, those who feel the urge to care are also themselves subject to the same set of challenges. For many, providing care is one of the few ways to secure a small income. Doing carework 'for money', however, threatens local models of intrinsic motivation bound up in the idea of ubuntu.
CHWs experience this tension acutely and daily negotiate, for themselves and others, what are seen locally as conflicting forms of motivation. CHWs, however, understand these complex motivations in ways that are not always a zero-sum game. The following sections describe some of the ways they experience and interpret the relationships between economic circumstance and extrinsic motivations in their caring labour.

Economic motivations at the individual level
In Khayelitsha, the naturalised discourses for why CHWs' care are so strong that the first time a CHW told AS that she was doing it because she needed a job, AS was quite surprised. The CHW's statement was quickly followed up with an explanation that she and the other women who work with her do it because they 'like to help people', and because it is 'better than nothing'. This tension between contrasting forms of motivation was to become a common theme. In Khayelitsha, CHWs carefully manage their selfpresentation as careworkers to others, while simultaneously attempting to make sense of their own roles as providers of care.
When they do speak about economic motivations, many CHWs focus on the survival and reproduction of their families and communities. As one explained, 'We have to take our children to school but we are not after money more than we want to help the community. But we need money for crèche [daycare]'. There is clearly ambivalence about the economic rationale for carework, but it can be moderated through reference to earning money to care for the needs of others.
The public presentation and management of the relationship between care and economic resources is often carefully scrutinised by CHWs and community members. As England (2005) explains, part of the way that low pay for carework is justified is through its perceived intrinsic fulfilment. But, in the context of extreme economic scarcity, this is often impossible. It is precisely because resources are scarce that carework is so closely scrutinised, and judgements about people's legitimacy as carers are made. Accusations of 'profiting off' of the care needs of others are not unique to contexts of poverty (Le Grand, 2003); however, these concerns are brought into very sharp relief when meagre resources for caring need to be distributed among many people.
In addition to the fact that local resources are scarce, the project-driven nature of foreign development initiatives in contexts like Khayelitsha and elsewhere in the developing world, has significantly shaped how people understand caring practices (Mosse, 2004). These foreign aid-driven initiatives have, in many places, made previously local and informal relationships and practices of care increasingly bureaucratic, professionalised and commodified.
CHWs also speak about the importance of developing a particular set of skills through CHW training and work experience. It is usually younger rather than older CHWs who see their volunteering this way. It is still difficult, though, for most women to talk about how they have been able to use carework to access training and work experience. One young carer whom AS has known for some time, only recently spoke about her work as a means of gaining other employment. She, like many other home-based carers and CHWs, hopes that with further training she can get a job as an assistant nurse at a local hospital.

Economic motivations in relationship
These sorts of tensions are also experienced in the relationships among and between individual CHWs, CHW NGOs, the state, and local and foreign donors. Most CHWs work in a highly 'projectified landscape' (Whyte et al., 2013) of development project-driven responses to local care needs. This complicates how they understand and experience their work and they can find themselves struggling to legitimate their access to resources as CHWs. One way they can do this is to explain how other CHWs or community members are less deserving of available funds. The head of one small home-based care organisation said that some of her CHWs are just there because they need a job, whereas others 'really want to help people'. She says that you can see the differences between these two groups of people.
Individual motivations are clearly and significantly shaped by relationships with other CHWs, but also shift and change throughout the lifecourse of individuals. One interesting dimension of this struggle for legitimacy and access to resources plays out in terms of a 'generational' dimension of CHW work. When we describe CHWs as 'older' or 'younger', we are referring to their age as well as their accumulated experience of growing up, living and working in South Africa. CHWs of different generations bring different political and cultural experiences and expectations to their work (Lehmann, Van Damme, Barten, & Sanders, 2009;Swartz, 2013). There exists a great deal of ambivalence, however, commonly described by younger CHWs as 'jealousy', on the part of older CHWs towards younger CHWs. Jealousy was spoken about in relation to the fact that some older CHWs have said that they, rather than the younger CHWs, are the professionals, given their long work experience. On the other hand, it is also often said that older CHWs are jealous of the more formal trainings and projects that the younger CHWs have greater access to.
Economic rewards not only complicate the relationships between individuals, but they also make the relationships between organisations difficult as well. When funding for carework is scarce, competition soon develops. Judgements are made about how funds are distributed or spent, and people are weighed up according to whether they are perceived to be worthy of receiving support. When there are no funds available, a clearer moral economy of care re-emerges, but as money becomes more present, carework becomes commodified, professionalised and more closely monitored.
The amounts of money at stake and the ways it is introduced shape this moral economy of carework. In some cases, when individuals or organisations are perceived as having too much money, accusations of corruption develop. As Mzi spoke about funding cuts to her organisation, Ichibi Single Mothers, she explained that some organisations have been getting too much funding for the wrong sorts of activities. She was angry that another organisation's soup kitchen that 'was started long after' Ichibi, was getting more funding. She said the government is now giving money to the 'wrong people' and that those working at the other organisation are taking advantage of the sick and 'are running a business'.
National and international donor-funding agendas, and recent shifts in those agendas, also affect the NGOs that employ CHWs in Khayelitsha. Mzi often spoke about how her programming had frequently had to shift in response to changes in donor interests. Donor agendas do change frequently but there have been two recent changes in South Africa that have impacted CHW work more dramatically than usual. The first is the recognition that HIV-related funding may have diverted vital resources away from other necessary health-related interventions, and a consequent drying up of 'HIV money' among many South African NGOs relatively accustomed to plentiful funding. The second is a shift towards channelling donor funding through the state (van Ginneken et al., 2010) and growing international trust in South Africa's political leadership to effectively use global public health funding.
These two changes not only put further pressure on NGOs, but also intensified competition among them. Organisations that had stronger internal systems and that could employ people to navigate the complex world of funding applications had a better chance of gaining access to resources. This fact has in turn shaped how CHWs and the NGOs that employ them understand the landscape of future opportunities in carework and how they need to position themselves to survive going forward.

Discussion
The South African context in which CHWs make sense of their motivations and understand the work they do is complex. This paper explores the multidimensional nature of CHW motivation in Khayelitsha, an economically marginalised settlement in Cape Town. Individual motivations for engaging in particular kinds of carework are carefully managed, by both individuals and CHW organisations. Three naturalising discourses of care predominate: those linked to constructions of gender, African identity and Christianity. CHWs' motivations and experiences are also shaped by the temporal context in two major ways: individual motivations do not remain static, and there are marked generational differences in the experiences of younger and older CHWs (Swartz, 2013).
We have also highlighted the dangers of understanding intrinsic and extrinsic rewards in reductionist or simplistic ways, as CHWs have complex, competing and overlapping narratives for their provision of care in Khayelitsha. Individual economic motivations are often spoken about in relation to the care of families and community members, rather than in relation to individual benefit. Individual rewards for CHWs also include developing skills in hope of gaining access to other employment. CHWs scrutinise other individuals, as well as other CHW organisations, in order to judge whether they are deserving of access to limited funds. All of this occurs in a context where funding is limited, and the recognition of their value within the health system does not always translate into careful thinking about how to incentivise and encourage strong CHW programmes.
Although investment in CHW programmes has increased both nationally and internationally, using CHWs to address complex health-related challenges is often done in ways that are overly simplistic and mechanistic (De Wet, 2012). Governments and international organisations too often fail to actively engage with the highly variable contextual factors that shape the spaces where CHWs work, and their motivations for engaging in the work. In Khayelitsha, whether a CHW runs her own small, partially funded organisation or works as a volunteer in an established larger care organisation, they are all caught in the same structural position linked to severe economic deprivation (De Swardt et al., 2005;Thomas, 2009). Irrespective of where and how these women work, CHWs are not fully financially recognised or valued in their positions. The complex and ambiguous nature of caring motivations in Khayelitsha is, therefore, inextricably linked to this structural lack of recognition. Although CHWs speak about and engage with their work in different ways, they also remain marginalised in a system that forces them to use whichever avenues they have in order to make their carework produce some tangible personal benefit.
One limitation of this study is that it explores the experiences, understandings and motivations of a particular group of CHWs living and working in Khayelitsha. Although some of the key challenges facing this group of CHWs may be possible to generalise to other CHWs in South Africa, the precise articulation of individual motivations with contextual factors is of course highly variable. It is precisely this variability, however, which demands much closer attention. In this sense, we have argued for more attention to be paid to the complexities of CHW motivation, as a lack of understanding in this respect can ultimately compromise CHW programmes.
There are calls in the current policy and health activist environment in South Africa to offer greater financial rewards for those in this carework economy. With the changing policy context associated with the National Health Insurance, two major changes are likely to affect how CHWs experience and express their motivations. Firstly, such CHW policy changes are likely to bring more coherence and less variability to CHW work and secondly, such work is likely to become more formalised, professionalised and state-driven. This policy context may steer CHWs away from being work strongly associated with various individual and collective motivations, to being a profession that may interest people from a more vocational perspective. However, moves to 'motivate' and 'incentivise' CHWs financially may have short-term benefits but will also likely have a range of other, less predictable effects, both positive and negative, on caring motivation, practice and relationship.
At the same time, though, the current system that relies largely on volunteer labour to respond to significant gaps in the state's response to public health needs is both politically unsustainable and morally problematic. Some of the health care worker unions, for example, have been pushing for the formalisation and protection of CHW carework as formal work that would enjoy all of the labour protections afforded to others in the formal economy. There are also calls from the progressive public health sector in South Africa, most notably the People's Health Movement, for the government to recognise the labour value of CHW's carework and stop exploiting vulnerable volunteer workers and pay them a minimum wage. Better understanding the context in which CHWs work will help policy-makers and programme managers better navigate these longer term dilemmas.