Grassroots Participation, Peer Education, and HIV Prevention by Sex Workers in South Africa

Abstract

Objectives. This microqualitative case study of a community-based peer education program led by sex workers at a South African mine examined the role of grassroots participation in sexual health promotion.

Methods. The study involved in-depth interviews with 30 members of the target community. The interviews were analyzed in terms of social capital, empowerment, and identity.

Results. The study yielded a detailed analysis of the way in which community dynamics have shaped the peer education program's development in a deprived, violent community where existing norms and networks are inconsistent with ideal criteria for participatory health promotion.

Conclusions. Much remains to be learned about the complexities of translating theoretically and politically vital notions of “community participation” into practice among hard-to-reach groups. The fabric of local community life is shaped by nonlocal structural conditions of poverty and sexual inequality in ways that challenge those seeking to theorize the role of social capital in community development in general and in sexual health promotion in particular.

There is currently strong interest in the role of grassroots participation in health promotion. This report presents a case study of a community-led, participatory peer education program that aims to reduce HIV transmission among commercial sex workers in a South African mining community. The program is influenced by the current “paradigm drift” within health promotion. This has involved a move away from traditional health educational approaches toward a community development approach involving the participation of local people in sexual health promotional interventions.1

Although participatory peer educational approaches have increased in both popularity and practice among sexual health promoters in both rich and poor countries, they have met with varying degrees of success.2,3 Our understanding of the processes and mechanisms underlying these approaches' successes or failures is still in its infancy. Recently, peer education has been described as “a method in search of a theory.”4 It has been argued that peer education “suffers from an inadequately specified theoretical base, which does not address the important social and cultural factors implicit in the approach,”5 and that this gap undermines our ability to learn from peer education's successes and failures.6

In beginning to address this gap in the existing literature on peer education, we argue that the success of participatory peer education depends on 3 interlinked factors.7 First, the experience of participation in project implementation serves to empower members of target groupings by placing health-related knowledge (traditionally the province of experts) in the hands of ordinary people. This contributes to an enhanced perceived selfefficacy or empowerment and increases the likelihood that people will engage in health-promoting behaviors.8 Second, the peer education approach provides a context for the widespread renegotiation of the social and sexual identities of people at high risk of HIV infection. Rather than being governed by individual choice, “sexual behavior is inextricably linked with the norms characteristic of the social groups with which we identify.”9 Third, such approaches succeed to the extent that they promote the development of “health-enabling communities”10—that is, community contexts that enable or support the empowerment and social identity processes outlined above.

The concept of social capital provides a useful starting point for conceptualizing those features of community most likely to be supportive of health promotional goals.11,12 According to Putnam,13 communities that are rich in social capital are characterized by high levels of local participation in egalitarian community networks, by norms of trust and reciprocal help and support, and by a positive local identity. It has been argued that an important determinant of the success of health promotional interventions is the extent to which they mobilize existing sources of social capital or encourage the development of new sources of social capital.14 The concept of social capital has taken strong hold in the discourse of leading international development agencies, and the task of building or enhancing local social capital is increasingly regarded as a key dimension of a wide range of development initiatives in disadvantaged settings.

In this report, social capital provides a heuristic framework for a detailed microqualitative study of a recently implemented peer education and condom distribution program run by sex workers in a small shack settlement in a gold-mining community in South Africa. On the basis of interviews with community members, conducted 6 months after the inception of the program, we investigated the impact of community-level processes and dynamics on the early stages of the program's implementation. We examined the extent to which the grassroots community-led program has served to mobilize existing sources of social capital or to create new social capital. We also considered the extent to which such processes have supported or constrained the development of increased levels of empowerment and of health-enhancing social identities in the sex worker community.

METHODS

The Interviews

Our peer education program of interest is one of a series of programs being implemented by Black female commercial sex workers in a gold-mining region near Johannesburg.7 It is part of a larger HIV prevention program that was initiated by a grassroots AIDS task force set up by local township residents and community workers who were deeply concerned about rocketing levels of HIV. This initiative resulted in a large-scale program pulling together the local task force, a group of academics and development funders, and representatives of the local trade unions, mining industry, and provincial and national government health departments.15 The 3-year program was initiated in 1998.

The aim of the program is to encourage the use of condoms, as well as early and appropriate accessing of treatment for other sexually transmitted infections (STIs). Other STIs increase vulnerability to HIV infection among those who do not have HIV and compromise the health of those who are already HIV positive. A survey conducted in the early stages of the program's inception indicated that 68% of sex workers in our area of interest were HIV positive.16

Within such a context, the behavior change that the peer education sought to promote would be too late to prevent HIV infection among two thirds of the women. This point was raised at an early stage of the project when women were informed about this survey finding. In the ensuing discussion, sex workers collectively decided that they would not make any attempts to establish which of them were HIV positive as individuals. They concluded that condom use was equally important among those who were HIV positive and those who were not. For sex workers who were HIV positive, condom use would protect their clients and regular partners from HIV and other STIs and would protect the women themselves from reinfection with HIV, which would further compromise their health. For sex workers who were HIV negative, generalized condom use in the community would increase their chances of remaining negative as well as protect them from other STIs.

We are currently engaged in a longitudinal process evaluation of this project that takes the form of annual interviews with sex workers in our community of interest over a 4-year period. The first set of interviews, conducted in August 1997, provided a baseline account of factors influencing condom use before the establishment of the project.17 This article reports on the second set of interviews, conducted in August 1998, 6 months after the initiation of the sex worker–led program.

Political and Economic Context

South Africa's history has been characterized by the economic exploitation, political exclusion, and social disempowerment of the country's Black majority. The White minority government's apartheid policies stated very explicitly that Black people were inferior to White people. Blacks (who formed 70% of the population) resided in 9 politically separate “homelands” (constituting 13% of South Africa's geographical area). Restrictions on Black people's movements beyond the homelands became formally institutionalized in a draconian system of “influx control.” The country's resources were overwhelmingly directed at developing the predominantly White urban areas, with the rural areas (where most Black people lived) becoming increasingly economically unsustainable. As rural poverty grew, it became progressively more difficult for the government to stem the movement of rural people into informal shack settlements that mushroomed around the urban areas.

In the late 20th century, the White government conceded to Black majority rule, and the African National Congress took over the government in 1994. Although Black people now have full political rights, much work remains to undo the negative economic legacies of apartheid. Levels of unemployment currently stand at 38%. Employment opportunities are exceptionally poor for unskilled women with little education, and it is from this group that our sex workers of interest originate.

The gold mines form a key pillar of the economy. The industry currently employs 300 000 Black men, and these jobs are highly valued despite difficult living and working conditions. Many rural areas in and around South Africa are totally dependent on mine worker remittances, and mining is the country's major source of foreign exchange. Black African migrant workers travel to the mines, where they live in single-sex hostels, leaving their wives and families in their rural areas of origin. The hostels, with up to 18 men in a room, far from wives and families, offer scant opportunities for intimacy. Within this all-male context, a thriving commercial sex industry has sprung up, with impoverished women finding accommodation in informal shack settlements on mine perimeters. Miners' working conditions are particularly dangerous, with a mine worker having a 1 in 40 chance of being killed in a work-related accident in a 20-year working career.18 Masculine identities serve as coping mechanisms for the risks and dangers of everyday working lives (with “real men” being regarded as brave, fearless, and willing to risk death to fulfill their role as breadwinners).19 Associated with this macho masculinity is the notion that men have insatiable urges to have sex with an unlimited number of women and are not afraid to take risks. Within such a context, many miners choose not to use condoms.

Peer Education Program

Our sex worker peer education program is located in a geographically isolated shack settlement, housing about 400 people in conditions of extreme poverty. Most residents are female sex workers, landladies or landlords (who sell liquor and provide free lodging to sex workers, who attract clients to their liquor businesses), and the amaHumusha (unemployed men who make their living on the fringes of the sex and liquor business). Living conditions in the settlement are basic, consisting of windowless tin shacks with no running water or sanitation. The settlement is located near the boundary fence of a gold mine, within walking distance of a mine hostel. Our baseline interviews,17 conducted with sex workers before the establishment of the peer education program, revealed that although their knowledge of HIV was high, no one used condoms. In conditions of poverty, and in the context of male-dominated social structure, women felt powerless to insist that reluctant male clients used condoms.19 Women were disadvantaged not only economically but also psychologically. Most came from backgrounds characterized by hunger and deprivation as well as high levels of physical and emotional abuse. They expressed low levels of confidence in their ability to take control of any aspects of their lives, including their sexual health. Furthermore, they argued that the fierce competition for clients had led to disunity among women, which reduced the likelihood that sex workers would be able to present a united front and insist on condom use. In such a competitive situation, if a woman refused to have sex without a condom, the client would simply take his business to a woman who was less insistent.17

The peer program was designed with 3 goals. The first was to increase women's sense of control over their health. The second was to provide opportunities for the collective renegotiation of social and sexual identities in ways that would increase the likelihood of condom use and early obtainment of STI treatment. The third was to encourage a community context that was supportive of the first 2 goals, both through mobilizing existing social capital in support of the program's success and through creating a new source of health-enabling social capital in the form of the peer education network. The program is coordinated by the second author (previously a nurse at the mines and now part of the nongovernmental organization that oversees the project). In collaboration with the program coordinator, women in the shack settlement have elected 10 sex workers to lead the program. These women have been trained to recognize and understand the symptoms of common STIs, to understand the relationship between STIs and HIV, and to carry out participatory sexual health promotion. They are also provided with free condoms to distribute to their peers. Peer educators receive a R200 (US $30) monthly honorarium for their role in leading the project.

The interviews that form the basis of this report were conducted by the first author with 30 members of the community (7 sex worker peer educators, 15 sex workers, and 8 male amaHumusha). These interviews, which averaged 3 hours in length, had 3 parts: a detailed life history interview, questions about the peer education program, and questions focusing on various dimensions of social capital in the community.

Interview findings are presented in 2 sections. The first section provides an account of the dynamics of the community where the peer education program is located. The second section outlines the processes involved in establishing the peer education network and the way in which it has been received in the community in the first 6 months of its existence. The final section of the report discusses the way in which community dynamics have served both to support and to undermine the program. We argue that while the trend toward mobilizing existing community resources and norms in support of health promotional goals makes theoretical and political sense, the task of translating this directive into practice is fraught with difficulties.

RESULTS

Existing Social Capital: The Context of the Program

To establish the extent to which existing community networks and relationships were present to support the program, we start by providing an account of the social relations in the community.

In his definition of social capital, Putnam13 emphasizes the importance of grassroots participation in local decision making within a context of egalitarianism and solidarity. In our community of interest, life was characterized by chaos and danger, with high levels of alcohol consumption, poverty, and gang conflict that make death and injury an almost daily occurrence. Thus, for example, in the month-long period in August 1998 during which these interviews were conducted, at least 2 women were killed by fellow sex workers in fights over clients. Despite this, the community was characterized by some degree of organization and discipline. However, this cohesion came at a price. It was imposed by an unelected “committee” of men, referred to in the community as “gangsters,” who were led by 3 “chiefs” and their assistants.

The role of the committee is complex. For example, it ensures safety through community patrols to prevent certain crimes and through policing and punishing offenders (be they community members or mine worker customers), in a context where drunkenness, violence, and conflict are common. Each Monday morning, the committee organizes community cleaning drives to clear up bottles and other litter. It organizes funeral collections to ensure that community members who die will be transported to their place of origin and given a proper burial. Informants regularly cited these burial collections as one of the main advantages of living in the community.

However, the Committee also has a more sinister face. Those who did not immediately produce whatever money the chiefs and their assistants demanded for funerals were frequently beaten up with sticks or sjamboks (leather whips). Persistent offenders were handcuffed with wire to a tree near the chief's shack, where they were variously taunted by their enemies and taken pity on by their friends, one of whom would eventually come forward to make the payment so that the prisoner would be released. In this respect, the collections were a source of great resentment. Many complained that the punishments were too severe, given that people often did not have the necessary money. Furthermore, many informants felt that they had sometimes been forced to pay for funerals of fictitious people. According to one informant (identified, as elsewhere, only by descriptor and numbered code):

The very same community that protects itself from outsiders sometimes scratches its own people's faces and I don't like that. The injuries are mostly inflicted by the governors of the community if you don't have money to pay for burials. No one challenges their authority because they are murderers. They are the types of people who just shoot you with no warning. (I3, peer educator)

Women informants agreed that it was ironic that the chiefs wielded such absolute power, when women dominated the community both numerically and in terms of their earning power.

It would be better if women were involved in running the community—the men here rely on us woman because we sleep with the miners who give us money. It's us woman who support the men here, yet we have no voice on the Committee. [Why don't you suggest to the Committee that they include women?] Yo! They would attack us. (I5, peer educator)

When asked why the men had a stranglehold on power, women generally answered, “It was this way when I arrived here.” Only 2 informants offered a more substantial comment, both in the form of historical anecdotes. The first was offered by a young man.

In the past women were on the Committee, but it was found that they couldn't listen to things, and they used to disagree with other community members. They were indecisive. What is more, there would be relationships between a man and a woman on the Committee, and the couple would defraud Committee money and use it for themselves. (I25, man)

The second anecdote was offered by an older women. She commented, “Not a day goes by without us discussing how we are tired of paying in all this money, and how we should get together in a large group and confront the Committee about the money they force us to pay them—even for dead people we have never seen.” However, she went on, visibly embarrassed, to tell a story which she said illustrated the “unreliability” of women and their inability to act in their best interests.

Once, we all agreed to defy the Committee at a meeting. Sitting alone as women we were very confident that we should unite in challenging them, but when the actual meeting came only one woman stood up and voiced her defiance, and the rest of us kept quiet. She ended up having to pay a large fine. If we could have stuck together on that issue as women, we would all be living a good life now. Nowadays every time anyone suggests we should call the police when the Committee beats us up, someone refers to that story. Even now we are embarrassed. (I22, sex worker)

Both of these anecdotes illustrate the deeply ingrained nature of men's resistance to women's leadership and women's lack of confidence in challenging this resistance.

A second key dimension of social capital is the existence of relationships of trust among community members. In the baseline interviews with sex workers before the inception of the program, reference was made to the lack of trust between women. This is because they are completely dependent on mine workers for their income. The intense competition for clients leads to jealousy and conflict, often characterized by violence or treachery.17 Community trust based on a common identity is an important component of social capital. Paradoxically, the common identity existing among our sex worker informants was built on extreme poverty and deprivation and sometimes led to conflict rather than cohesion. When asked “What do women in this community have in common?” most women replied “poverty,” which forced them into a loathsome job. Yet it was precisely this poverty that forced women to compete, often viciously, to survive. Some were doubtful about the possibility of building community solidarity under such conditions.

If you have something the others don't have, they hate you for it. Maybe if everyone had the things they needed, it would be better—it's poverty that causes this jealousy. Sometimes we even poison or stab one another to death. (I21, sex worker)

I don't think people here trust one another. I should never go to anyone for help with my problems. How can you trust a woman who is a drunk, how can one trust a woman who has abandoned her children, and most of all how can one trust a woman who changes men—today this man sleeps at her house, tomorrow a different one. (I3, peer educator)

This last quotation, from a woman who was herself a sex worker and heavy drinker, illustrates not only women's lack of trust in others but also their own lack of self-respect.

There was also a consensus that people motivated by money could not be trusted.

There is total lack of trust here, because people are only here to make money, so they concentrate on themselves. The reality is that every single one of us only came here for the money, not one person came here for a holiday. (I27, man)

A third key dimension of social capital is the existence of relationships of reciprocal help and support. Despite low levels of trust, people referred to numerous instances of reciprocal help and support. These included people lending one another money, particularly for funeral collections; sex workers visiting each other in the hospital, taking food or washing the patient in her hospital bed; lending money to women who had recently given birth and were unable to sell sex. People often referred to other community members as their surrogate families. In particular, within-shack relationships often constituted an important source of support, with landlords and landladies often treating their resident sex workers with kindness and generosity.

However, people also often spoke of instances where fellow community members had refused to help. They referred to incidents where neighbors with cars would charge extortionate prices to take the seriously ill to hospital, or where the Committee would refuse to lend money to transport a sick person.

Unsurprisingly, within such a context, levels of positive community identity and personal morale were low. People's accounts of their lives reflected low levels of self-respect. They characterized the settlement as a community of losers:

Everyone is here because he or she has failed in life. Failure is the only thing that would drive anyone to a place like this. (I15, sex worker)

However, we found evidence, as we had with informants in our baseline study,17 that sex work, despite its unpleasant aspects, did give women some degree of independence from the more conventional feminine roles of wife, homemaker, and mother in impoverished rural areas. The job, for all its indignities, offered economic independence of a sort and independence from abusive relationships with men.

What I like about this job is that it's an easy way of making money, everything I want I can buy. I do not pay rent and I seldom run out of money like when I was at home. (I8, sex worker)

I have lived and stayed with men and benefited nothing out of those relationships except for suffering and frustration. As far as I'm concerned it's better to do this work. (I7, sex worker)

At the same time, women were realistic about the costs of this freedom.

We left our families and homes of our own accord, no one chased us away. We left because we wanted to govern and rule ourselves. And now we are compelled to do this job to make a living. Men are rough, sex is painful, they often have large penises which hurt you—some treat you in a dehumanizing way. You lose your sense of pride as a woman and as a human being. You are treated like dirt, sometimes that hurts. This is the price we pay for seeking our independence and freedom. (I7, sex worker)

In this section, we have sought to highlight how the peer education program was established within a community characterized by conflictual and exploitative relationships and low morale. Our findings pointed to low social capital in our community of interest. The following section provides an overview of the establishment and reception of the program, to highlight some of the complexities involved in mobilizing local community networks for health promotion in a community of women whose lives are characterized by multiple layers of disadvantage.

Building New Social Capital: The Peer Education Program

Aims and activities.

The aims of the program are to promote condom use and to encourage people to seek early treatment for STIs, through establishment of a peer education network that will provide people with basic information about sexual health risks, distribute condoms, and create a supportive context for healthy behavior. As already stated, before the program, condoms were not used, and people said that there had been a sense of fatalism and powerlessness in relation to STIs. Women had generally “hoped” that STIs would go away or relied on over-the-counter medications (such as aloe vinegar), self-made preparations with herbs from the veld, or the use of douches, drinks, or enemas concocted from Jeyes fluid (an outdoor disinfectant containing tar acids, more commonly used for cleaning drains). Medical attention was generally sought only when a woman was unable to walk, by which time she would have to be taken to the hospital in an ambulance. The peer educators encourage people to seek medical advice at the early stages of an STI—and virtually every informant reported anecdotally that since the program had started, visits by the ambulance to the community had decreased substantially.

Another aim of the program has been to increase sex workers' sense of personal vulnerability to HIV, as a way of promoting condom use. Although 68% of women were already HIV positive at the time of these interviews, the full-blown AIDS epidemic was still in its early stages, with people in a state of numb fear.

I haven't seen a person with AIDS, but I have heard rumors that people here have died of it. See—I have lost weight, perhaps I am infected too. I don't usually use condoms, only with my new clients, but not with my regulars. I do think a lot about the diseases. When I see myself getting thinner I convince myself that it is because I drink too much, or because I am homesick—but there are times when I wonder if I have AIDS or not. We often discuss these things amongst ourselves, but we usually convince ourselves that our weight loss is due to homesickness. (I7, sex worker)

People living with AIDS chose not to disclose the nature of their illnesses to other community members.

I think many people around here are HIV positive but are scared to tell others. If it was me, I would not tell anyone around here, as they would talk badly about me and stop my customers from coming to me. If the customers knew I had AIDS, they would run away. (I21, sex worker)

The program coordinator is currently working toward creating a climate of tolerance and support that might encourage HIV-positive people to disclose their status.

Initiating the program involved a lengthy and sensitive process of consultation with members of the committee, who serve as community “gatekeepers.” In gaining access to the community, the coordinator had to adopt the norm of approaching the committee leaders or “chiefs” with exaggerated gestures of humility and respect, reassuring them that she and the program would not breach their authority. Once she had gained access to the community, she initiated discussions with sex workers about possible foci of the project over a number of weekly meetings. Her main goal was to establish a sense of community ownership of the program. Once the peer educator team had been selected, they were taught about STIs and HIV/AIDS and trained in the use of participatory educational methods (e.g., songs, role plays, dramas). In addition to this, it was necessary for the coordinator to put much energy into the peer educators' personal and collective development, in a context where they had little experience of participation in a structured program or working cooperatively toward shared goals.

A key dimension of this work has involved promoting discipline and self-respect. Much effort has been put into working with sex workers to generate their own code of conduct around issues such as punctuality, personal hygiene, dressing appropriately for meetings (e.g., no nightdresses), and drinking as little as possible before and during meetings. Another important team-building dimension has been the promotion of conflict resolution skills among peer educators, who told us that the team sometimes has been plagued with conflict, jealousy, and gossip.

Community response to the program.

The program has a high profile in the community and has generated much attention and controversy. Responses have ranged from unalloyed approval to a more complex mix of skepticism and jealousy of the peer educators. Virtually every informant praised the personal qualities of the program coordinator.

We needed this program. It has made a big impact on us. When Zodwa came, she showed us all these diseases, and explained them patiently in the style of language that is best understood by us lay people. In the past, the nurses at the clinic outpatients didn't explain things thoroughly, and while some treated us with respect, others were very rude, passing insulting remarks about our profession and looking down on us. Zodwa explains everything clearly, and we feel free to ask if there's something we don't understand or anything we want clarified about HIV or STIs. She is open and kind, she is the best thing we have ever had in this community. (I8, sex worker)

The tremendous impact the program has had on the confidence of peer educators was repeatedly evident.

I know something good is going to come to me out of this project—I have made it my life. (I1, peer educator)

Until this project came here, I always felt that I was dirty, that I was nothing. I can't put my finger on what has happened to me since, but I just know that my life has changed. I feel that through this project I will get my self-respect back. (I2, peer educator)

Again and again, peer educators spoke of the personal development they had experienced through their involvement.

I myself am shy, but when the time comes to address a meeting to the whole community, I do so freely and without fear that I will not be able to express myself. (I3, peer educator)

This sense of pride and achievement spread to some community members beyond the peer educator team.

When Zodwa comes in her car, wearing nice clothes, and brings us pens and books, it does give us all a sense of pride. (I17, sex worker)

Some thought the peer educators were working hard and doing a good job. However, others were jealous of the status they received, and even more of the monthly R200 (US $30) payment they received.

People think that the peer educators are doing real work, and this is where the jealousy comes in. The difference is that the rest of us get money from using our bodies—the peer educators get theirs through using their brains. (I6, sex worker)

This jealousy caused some women to refuse to attend peer education meetings out of spite.

When the whistle is blown calling us to the meetings, I say to myself, “It's the peer educators who should attend, because they are getting paid, and I am not.” If we go to meetings it will seem as if we are part of the project, this will contribute to its success—yet at the end of the month they will brag and show us their pay envelopes, while the rest of us have nothing to show for our involvement. (I7, sex worker)

More jealous community members dismissively insisted that the peer educators were only motivated by money, that they did as little work as possible, and that if the coordinator were to withdraw, the program would fall flat.

Some informants whispered that peer educators cheated on the progress sheets (recording the number of meetings and the number of condoms distributed). Others whispered that we should not believe anyone who alleged that condom use was widespread as a result of the program's efforts.

People pretend that they are using condoms. Even the peer educators themselves, they don't use condoms. It's all a pretense to outsiders. They pretend to Sister Zodwa that they are using condoms—yet one of them was recently taken to hospital with an infection. How can you have an infection in your vagina if you use condoms? (I19, sex worker)

Several men argued that because the program was run by women, it had no chance of succeeding.

You see, it's like this, where there is no man, there is no unity. These ladies, as soon as you leave, they go back and continue with what they will always be doing. They lose interest and gamble and drink, and the project is doomed to failure. Things go wrong because of the fact that the Committee was not put in charge of this program—you have to understand that in this place women only respect men, they do not respect one another. (I25, junior Committee member)

Women informants varied in their views on the extent to which women were capable of being effective leaders. Two quotes illustrate the disparity.

In my generation there is a new dispensation altogether. We are allowed to tell men what to do, what we like and what we don't like. (I8, sex worker)

We live in a man's world. It's the men who have to agree on things first, and then we follow. This is the way God created us to be. (I22, sex worker)

The extent to which the project's activities will support the former view and undermine the latter remains to be seen.

When they were asked about community reactions to the program, people's responses reflected the degree of controversy and attention the program had generated. Some distinguished between the responses of the uneducated and those who had been to school, between younger and older members of the community, and between drinkers and nondrinkers. Others distinguished responses according to personality traits: the “stubborn” vs the “open-minded,” the “proud” vs the “humble,” the “rude” vs the “respectful” members of the community.

Older women were reported to be particularly negative, attending meetings only to heckle. Others were said to have responded to talk about AIDS with fatalism.

When we talk about AIDS, the older women tell us, “We have been having sex without condoms for a long time and we never died, what you are telling us is nonsense.” (I21, sex worker)

“Don't waste our time with your AIDS story,” the older women say. “We are just surviving here from day to day, waiting for the day they have to make our funeral collection.” (I6, sex worker)

Landladies and men often refused to lower their dignity by taking an interest in the program, which they associated with sex workers, the lowest stratum of society. “What is there that we can learn from whores?” was their frequent comment. As one peer educator commented:

I think if the peer educators were men people would listen more. You see us women, we began by selling sex, and now we think we can teach them something—we are never taken seriously. They think of us as rubbish, as tikilines—an insult for a woman who doesn't have a home and who goes around the mines sleeping with every man she meets. But it's only the men who disrespect us. Women are more tolerant. We understand one another's problems, that we do this work because of poverty. (I5, peer educator)

The difference in the responses of women and men was mentioned frequently.

It's women who take up the fight against HIV more than men. Because even when we explain about HIV in a meeting where both men and women are gathered, you'll see and feel that a woman is more moved and more touched by this than the man. (I3, peer educator)

It is clear that the program has been effective in placing the issue of sexual health and HIV on the public agenda in the community. One of our starting assumptions about the mechanisms underlying successful peer education was that collective debate and argument serve as a precondition for the development of new norms. The program has succeeded in harnessing community divisions to generate energy and debate around the issue of sexual health and condom use. We believe that the ongoing process of discussion and dispute about the program and its goals could serve as a mechanism leading to health-enhancing normative change.

Promoting Health Through Existing Community Networks

We have described the impact of community dynamics on the establishment of a participatory peer education program. In line with the philosophy of participatory health promotion, the program coordinator has striven to work through and respect existing community structures and has encouraged the peer education team to develop its own norms and approaches for ensuring condom use. However, this could not have been done without the permission of the committee, which controlled access to the physical space of the community as well as to the sex workers who came to constitute the peer education team and its target audience. Through a delicate process of consultation, the committee has been supportive of the program, giving the coordinator free and safe access to what would otherwise have been a dangerous and inaccessible workplace.

Part of gaining this support and access has involved giving the committee a stake in the project, despite the fact that sex workers lead it. Project funds have been paid to the committee leader to assist the project by transporting peer educators in his minibus when needed, for example, to enable peer educators to visit other sex worker communities in the region to assist them in setting up similar programs. Furthermore, committee members have voluntarily taken on the role of moving from shack to shack blowing their whistles to call sex workers to peer education meetings (in the same way that they call community members to their own meetings).

There are clear contradictions in having to collaborate with a violent and exploitative group of armed men in a program aiming to empower women. This situation highlights some of the complexities of following the directive by health promotion and development agencies that health workers should use existing indigenous sources of social capital as the basis for their work. While such directives make theoretic and intuitive sense, the task of implementing them in the real-world setting of our program has highlighted the complexities of working in a context where community networks are structured around a set of unequal and exploitative social relations. The program coordinator has dealt with this complexity by repeatedly reiterating her respect for the committee's authority, while at the same time working to demarcate the specific area of sexual health as “women's business.” In negotiating with the committee, she has managed to reach agreement that this is an area where men have no interest or ability to play the key leadership role that they do in other areas of women's lives. Committee members have accepted this demarcation. In other words, they have conceded that the limited area of women's sexual health is an area that they as men cannot understand and that it is thus most appropriately guided by the leadership of women.

Some sex workers were positive about the committee's role in the program, but others were skeptical about their involvement. Their accounts suggested that the committee had “colonized” the program in a way that fitted in with their autocratic and violent leadership style.

If you don't want to go to meeting, they will drag you there, and if you still refuse, they will beat you. (I17, sex worker)

Some saw the program as nothing more than a new excuse for the committee to exploit the sex workers.

The Committee forces us to attend the peer educators' meetings, saying that those who don't attend the meeting must pay a fine of 20 rand [US $3]. (I16, sex worker)

The crooks [i.e., the Committee] support the project for their own benefit and safety—so that we don't infect them with diseases. We are forced to sleep with them because they are Committee members and are in authority. Sometimes we do so willingly in the hope that when the next funeral collection comes we won't be forced to pay. (I7, sex worker)

Ironically, however, even those most opposed to the committee's involvement in the program still espoused methods of program management that drew on norms of surveillance or the threat or reality of violent punishment. A key aim of participatory peer education is to increase levels of personal empowerment among target audience members, and to improve levels of community trust by creating a context that is supportive of such empowerment processes.

However, not one person suggested that other sex workers could be trusted to use condoms out of solidarity or as a result of a public commitment to do so.

If people don't use condoms, we beat them up, and threaten to take them to the Committee. (I3, peer educator)

If we discover that a woman has not used a condom when we are selling in the veld, we demand that she hand over the 20 rand the client has paid her, and we give her a few slaps. (I17, sex worker)

If a woman is sick with an STI [which indicates that she has had unprotected sex], we encourage the Committee not to allow others to help with money for the doctor. (I4, peer educator)

In the interviews, attempts to generate discussion about the possibility of developing noncoercive methods of reinforcing condom use yielded further suggestions involving surveillance rather than trust and cooperation.

We need to have a system where women check on one another. They should follow each sex worker, and after she is finished with the client, she should be forced to produce a used condom. If she cannot do so, we should write down her name, and report her to Zodwa—and amongst ourselves we should reach a communitywide agreement that such a person will never be helped by anyone in the community any more. (I17, sex worker)

We should make women put their used condoms in a box, and take it in turns to police the box. This will also serve as a warning to each of us—when a man has an STI, his sperm will be yellow. Forcing each one of us to take turns in checking the contents of each and every condom will make us aware of how many of these men might have infected us. (I7, sex worker)

The reliance on surveillance at best and punishment at worst, rather than on trust, as a means of ensuring the program's success is consonant with indigenous community coping skills. These skills assist people in surviving in a highly deprived and violent community context, where levels of trust are low and where the threat of punishment (through public censure, fines, or physical violence) is a key means of maintaining community organization.

Do such methods have the potential for promoting a hostile climate within which persons with STIs or AIDS will continue to feel reluctant to disclose their status? Or are they a key step toward the creation of new community norms that will result in increased condom use? Our interviews were conducted 6 months into the life of a 3-year program, and it is too soon to say what the outcome of these approaches will be. However, our interviews do highlight how the theoretically important issue of locating one's health promotion program within existing local community norms is practically extremely complex. These comments echo the insights of researchers who have coined the term “anti-social capital.”20,21 It is simplistic to assume that the mobilization of local community networks and resources by development workers will automatically yield positive benefits.

DISCUSSION

In this report, we provide a microqualitative descriptive study of the early stages of a sex worker–led peer education program in a shack settlement in a South African mining region. This study was conducted 6 months into the program's 3-year life, and it would be premature to speculate about the extent to which the program will succeed in improving condom use in this community at this early stage. The aim of this report is rather to examine the way in which community dynamics have shaped the early development of the program, in light of our 3 assumptions regarding the mechanisms underlying successful participatory health promotion.

Our first assumption was that a successful program should stimulate and encourage opportunities for the collective renegotiation of social and sexual identities and norms. There is no doubt that the program has generated a great deal of attention and controversy and that existing community divisions and fault lines have provided a fertile field for the polarization of opinions and responses that has fueled much of this controversy. The extent to which this attention will result in meaningful behavior change remains to be seen. Six months into the program's life, and with cases of full-blown AIDS still rare, much of this attention has not risen out of a common interest in sexual health per se. On the contrary, it has risen out of factors such as curiosity about an unusual initiative in a hitherto isolated and neglected community, admiration for the program coordinator, and controversy about the committee's role. It has also resulted from jealousy, gossip, and conflict around the role of the peer educators, suddenly elevated from members of the most poorly respected sector of the community to a paid elite in a competitive and deprived setting.

At this early stage of the AIDS epidemic, denial of the extent of the problem is still possible. However, this will become harder to sustain as the numbers of full-blown AIDS cases and deaths increase. As this happens, we hope that the groundwork laid by this program (together with similar programs being conducted with mine workers) will provide a context for the development of a more direct interest in sexual health per se. Ideally, the foundation will be in place for the development of a coherent and united response to HIV by sex workers.

Our second assumption was that a successful peer education program should promote a sense of health-related empowerment among members of the target audience. The baseline study of sex workers in this region before the inception of the program suggested that women lacked the psychological and economic confidence to insist on condom use with reluctant clients. The program has made no significant contribution to the economic development of the community. With regard to psychological confidence, the extent to which the program succeeds in building women's confidence remains to be seen. Certainly it succeeded in transferring both supplies of condoms as well as “expert” knowledge about sexual health promotion out of the hands of sometimes unsympathetic clinic nurses and into the hands of ordinary people. Its impact on the confidence of peer educators is undeniable, and there was some evidence that the confidence of some sex workers increased through identification with peer educator colleagues. Furthermore, the fact that female peer educators became public leaders of a controversial and visible community program opened up the possibility of community debates about the potential for women to assert themselves in relation to men and even to take leadership roles. Although these debates were not always concluded in favor of women, the very fact that they were occurring could be seen as a positive sign.

The third assumption we made about the mechanisms underlying successful peer education relates to the importance of promoting community contexts that enable and support desired behavior changes, particularly increased condom use. In this report, we pointed to low levels of social capital in the community. We also examined the way in which the program has been established and received within this unpromising context, and how the coordinator and the peer education team have sought to promote the program's goals within a deprived, chaotic, and exploitative set of community relations. A key goal of the program is to create a sense of unity among sex workers in relation to condom use, through promoting their common identification with the project and its goals—wherever possible through building relations of trust and of reciprocal help and support among sex workers. Following state-of-the-art participatory principles, it was assumed that wherever possible such cooperation and unity should be established through mobilizing existing community networks and norms in pursuit of program goals. We have referred to the complexity of having to work through a male-dominated, hierarchical, and exploitative gangster committee in a program seeking to empower sex workers, the lowest-status community members. We have also looked at the way in which sex workers' recommended methods for ensuring condom use involved surveillance and punishment rather than the development of trusting cooperation.

If the program should achieve its goal of increasing condom use over its 3-year life, it will be by a route that is somewhat different from that proposed by textbook advocates of concepts of participatory development and of social capital. Such advocates argue that program goals are best achieved through cooperative participation of target audience members in networks characterized by trust and reciprocity. While this might be the case in more stable or less deprived contexts, such ideals may be more difficult to implement in contexts such as our study community. In such contexts, health promotional workers will have to work with the raw materials available to them, as has been the case in our program of interest. Communities with the highest levels of HIV infection may often be those that, like our study community, are the most disrupted or deprived. Existing norms and networks that health workers have to build on will often be characterized by violence and exploitation, or by conflict and competition, rather than by cooperation and trust.

Our evidence for the impact of poverty and gender relations on community-level attempts to limit HIV transmission highlights the need for multilevel efforts to fight HIV, at the short-, medium-, and long-term levels.22 In the short term, there is the need for biomedically driven STI interventions, which have been shown to reduce the incidence of HIV in other settings in Africa. In the medium term, there is a need for community development programs, such as our peer education program of interest, that seek to promote condom use and to create community contexts that enable and support sexual health. However, given that broader societal forces constrain the ability of the peer education process to enable or support behavior change, the impacts of short- and medium-term programs are most likely to be maximized in the context of parallel long-term efforts. Those concerned with reducing HIV need to add their support to ongoing programs to address macrosocial problems—such as migrant labor, poverty, and sexual inequality—that lead to the social conditions in which HIV flourishes and that undermine HIV prevention efforts.23,24

We hope that our case study of a peer education program in a disrupted and “hard-to-reach” community has contributed to 3 interlinked areas in the health promotion literature. The first area is that of peer education, in the context of calls for attention to be paid to the mechanisms underlying the successes or failures of such programs. The second area is that of health and social capital. We hope our case study has pointed to some of the complexities inherent in recommendations that health promotion workers should strive to locate their programs within existing indigenous community norms and networks. We also hope that our case study has highlighted the way in which the fabric of local community life is shaped by nonlocal structural conditions of poverty and sexual inequality in ways that present a strong challenge to those seeking to theorize the role of social capital in health. Third, we hope that our work will contribute to much-needed research into the community-level processes whereby grassroots participation may or may not serve as a health promotional strategy. If we are to extend our understanding of the potential of grassroots participation in health promotional activities, we will have to pay more attention to the dynamics of communities whose networks and norms fall very short of the ideal starting points for grassroots health promotional efforts. Much remains to be learned about the complexities and ambiguities involved in translating theoretically and politically important notions of participatory health promotion into practice in hard-to-reach communities.