Testing Differences: The Implementation of Western HIV Testing Norms in Sub-Saharan Africa

Abstract

This paper examines the implementation of Western HIV testing norms– counselling, consent and confidentiality (“3Cs”) – in Malawi, a high prevalence, rural African setting. It considers the differential perspectives of three categories of stakeholders: proponents, implementers and intended beneficiaries. The proponents are members of the “Counseling and Testing Establishment.” For them, the 3Cs are human rights imports that are worth defending formally, but not always worth prioritising in practice. The implementers are HIV Counsellors. For them, knowledge of the 3Cs as Western biomedical jargon distinguishes them from villagers, but places them in situations where the ethics of testing conflict with their moral concerns for those whom they were trained to help; thus they adapt them in practice. And the intended beneficiaries, the rural Malawians whose rights are meant to be protected by the 3Cs perceive the norms as protecting themselves as individuals, but as harming rather than benefitting their communities. The case study of Malawi illustrates a tension between Western, individual rights-oriented public health norms and local concerns for the health and wellbeing of the imagined communities that they are meant to benefit.

Two health workers at a health facility in rural Malawi are in the midst of a conversation when they are interrupted. The woman who interrupts them asks for urgent assistance locating an HIV Counsellor. It just so happens that it was a counsellor she had approached. The woman tells the counsellor that she wants to be tested for HIV, immediately. The counsellor, prepared to assist her, first inquires about the nature of the urgency. The woman tells the counsellor that she is the caretaker for her husband, who had been admitted to the hospital two weeks ago with diarrhoea and a persistent fever; in Malawi, much of the care of a patient in the hospital is delegated to a family member. The husband had had his blood drawn twice since arriving, but was unwilling to share the results of the tests with his wife. Furious, the wife inquired of the clinician, who explained to her that the patient has a right to privacy and that he could not disclose the husband's test results to her. The woman, growing more frustrated, told the clinician that she does not tolerate nonsense. She immediately collected her things and approached the door, threatening to leave her husband behind in their care. The husband began to cry, pleading with his wife not to leave. The wife, unmoved by his tears, insisted on being told the truth. As she was about to leave the hospital ward, the nurse on duty called her into a separate room to discuss the matter further with the clinician. The woman told them that she wants to know exactly what her husband is suffering from and the results of his blood tests. Seeing how determined the wife was to leave her husband behind at the hospital, the clinician told her that her husband is HIV positive and suffering from HIV-related illnesses.ⁱ

The above encounter presents several themes of this paper: a woman's worries about how best to care for her sick husband and about having been unknowingly exposed to HIV; a husband's fear of abandonment by his primary caretaker in a time of great need; the situations in which health workers are often placed, ones which force them to make difficult decisions; and underlying it all, the clash between a public health norm for HIV testing that privileges individual rights, and local concerns for the health and wellbeing of the communities experiencing the AIDS epidemic first-hand. This paper aims to show why a (global) model of HIV testing – one rooted in Western notions of individualism and human rights—resonates very differently in a setting far removed from its context of origin.

The Development of an HIV Testing Regime

Particularly important for the development of the global approach to HIV testing was the early history of the AIDS epidemic in Western democracies. In the West, AIDS disproportionately affected highly stigmatised groups –gay men and injecting drug users – and concerns abounded about the prevention measures that might drive the epidemic underground (DeCock, Mbori-Ngacha, and Marum 2002). Consequently, unlike the traditional public health strategies for controlling the spread of other infectious diseases (e.g., compulsory testing, contact tracing, involuntary quarantine), the response to HIV was unique. As Bayer and Edington (2009) explain:

Out of the clash between the traditionalist perspective — one that favored routine testing conducted under conventional assumptions of presumed consent — and the viewpoints of those most at risk for the new disease would emerge a set of testing policies that stressed the importance of voluntarism, which underscored the centrality of counseling and embraced confidentiality as a prerequisite for effective programs (CDC 1986). It was in this context that those who began to speak in the name of ethics and human rights sought to provide guidance, using language that would have a lasting impact on the international discourse on testing.(302) [Italics added by author].

The Western approach to addressing the AIDS epidemic thus went global(DeCock, Mbori-Ngacha, and Marum 2002). Global HIV testing policy became anchored in a human rights framework, one that encourages testing by guarding against HIV-related stigma and discrimination. The 3Cs of HIV testing— (1) that testing includes counseling, (2) that testing be conducted with informed consent, and (3) that the test result be confidential (UNAIDS/WHO 2004)— were disseminated by international AIDS and human rights organisations and adopted by many national governments, including countries throughout sub-Saharan Africa (SSA).

The African AIDS epidemic, however, is not the same epidemic that plagued the West. In the West, a high-income, low HIV prevalence setting, the epidemic affected highly stigmatised subpopulations; in SSA, a low-income, high HIV prevalence setting, the epidemic affects the general population, with transmission occurring in the contexts of marriage and childbearing.ⁱⁱ Moreover, in SSA, cultural norms are deeply rooted in the community, rather than in the individual, with strong social obligations to others. Given these fundamental differences between the West and SSA in terms of the epidemiology of the disease, health infrastructure and cultural norms and practices, it is not surprising that the implementation of global (Western) HIV testing norms in SSA might not have the imagined effects.

Avast body of literature— spanning geography, level of analysis and empirical object of inquiry— has shown that for a policy or programme to achieve its stated goals, the various actors integral to its implementation must be invested in them equally. In the absence of shared objectives emerge latent, unexpected, or even counterproductive, consequences(Portes 2000; see Bledsoe, Banja, and Hill 1998; Swidler 2006; van der Geest, Speckmann, and Streefland 1990, for examples). Studies on HIV testing consider the direct effects of testing, such as the uptake of testing services and behavioral change as a consequence of knowing one's HIV status (see Obermeyer and Osborn 2007 for a review). These studies, however, tell us little about how the differing rationalities of various key players in the testing enterprise affect the implementation of testing, a process that is as important to its outcome as the policy itself(Roberts 2001). Such an inquiry provides an important case study of the efforts of the international AIDS community to increase the proportion of people who know their HIV status, as well as an example of the global promotion of human rights having unintended effects.

In Malawi, the SSA case study presented here, the Western 3Cs are not embraced uniformly among testing's proponents, implementers and their intended beneficiaries. These various stakeholders respond to the 3Cs according to their social position in the testing enterprise and, at times, appropriate them for purposes other than those envisioned by the global AIDS community.

Malawi: An African Case Study

Malawi is in many ways an ideal site for this inquiry: national prevalence is high, estimated at 12% (National AIDS Commission [Malawi] 2007)ⁱⁱⁱ; it is extremely poor, with the majority of its population surviving on less than $2 a day (The World Bank 2008); and Western-oriented discourses about human rights are relatively new (Englund 2006), particularly in the rural areas where over 80% of the population resides.

Testing services first became available in Malawi in the mid-1990s, though they were not widely accessible. In 2004/2005, the Malawi Ministry of Health (MOH) received donor support to expand free HIV testing to all district hospitals, and many rural hospitals and clinics. During this time, anti-retroviral treatment (ART) also became available in district hospitals for those diagnosed with AIDS and who met the eligibility criteria.

Malawi's National AIDS Commission (NAC) adopted the “3Cs” principles of the World Health Organization (WHO) and The Joint United Nations Programme on HIV/AIDS (UNAIDS) in Geneva. In 2004, and with assistance from the U.S. Centers for Disease Control (CDC), the MOH produced a set of HIV/AIDS Counseling and Testing Guidelines, which note:

Providers of CT [Counseling and Testing] services will be required to subscribe to an ethical code of conduct consistent with the principles of human rights as outlined in the Malawi Constitution and the Universal Declaration of Human Rights and Professional Ethics…(Ministry of Health [Malawi] 2004, 9)

Every effort should be made to deliver CT services in such a way as to reduce stigma, to guarantee and to protect clients' privacy and confidentiality.(Ministry of Health [Malawi] 2004, 9-10)

Clients must be given the opportunity to understand that they are being tasted [tested] for HIV so that they can provide informed consent. Client consent cannot be implied or presumed.(Ministry of Health [Malawi] 2004, 11)

These guidelines were incorporated into the national HIV Counsellor training manuals, developed in 2004 and revised in 2007.

Data and Methods

Ispent approximately eight cumulative months in Malawi between the years 2006-2009. Like this study's empirical object of inquiry (3Cs), my fieldwork moved across sites and social categories, from government ministries and NGOs in the capital city of Lilongwe where testing policies are formed, to rural clinics where they are delivered. Throughout the course of fieldwork, I observed the training of HIV counsellors; chatted informally and conducted interviews with national elites and counsellors; and, in collaboration with local research assistants, interviewed village headman, rural Malawians tested for HIV^iv and conducted focus group discussions with rural Malawians. All interviews lasted between 30 minutes to just over an hour and were not tape-recorded (except for those with the HIV-tested sample). Rather, detailed notes were taken and transcribed within a day of the interview itself. Interviews with elites and counsellors took place in English; interviews with villagers took place in chiChewa. Focus groups lasted between 40-60 minutes and took place in chiChewa. All protocols focused on the 3Cs.

I also integrate into the analysis a set of ethnographic field journals, kept by a Malawian field assistant (pseudonym Bashil Kunthani), which capture local experiences with HIV testing. The journals are part of a larger project in Malawi that began over a decade ago, and that aims to learn what people say about AIDS when they are talking with each other in natural settings. ^v The texts are written in English and bracket explanatory comments or expressions that are untranslatable in local languages. I analysed ten journals, each of which averages 30 pages in length, covering the same period of years as my fieldwork. For legibility, I insert missing or clarifying words in brackets and make minor edits to grammar.

Initial themes were developed throughout the course of fieldwork through preliminary readings of the data and extensive memo-writing (Emerson, Fretz, and Shaw 1995). After the completion of fieldwork, the data were organised and coded for content, documenting, in particular, responses to the 3Cs across the three categories of social actors who inform the HIV testing encounter (proponents, implementers and beneficiaries). Subsequent, more focused, coding was based on themes developed in the field, notably each stakeholder's differential investment in, and concerns with, the Western testing model. The various data sources were analysed in conjunction as unique, but equal, forms of data, each contributing different but equally valuable insights to the topic. The data are referenced as follows: interviews by pseudonym, category of social actor, and age of informant (e.g., James, Counsellor, age 25); focus groups by pseudonym, sex composition of the focus group, and age of informant(e.g., Charity, Women fgd, age 28); and journals in a name, year, month, day, format (e.g., Kunthani 080611). All names of individuals and places have been anonymised.

Findings

Proponents: The Counseling and Testing Establishment

The Counseling and Testing Establishment (“CTE”) is a cross-section of representatives from Western donor organisations, Malawian government entities, and non-governmental organisations. These individuals apply the “rules” from Geneva to Malawi in the form of national policies, guidelines and training manuals. Though heterogeneous in terms of the entities they represent, the stakeholders that comprise this group are homogenous in significant ways: they all are university educated, although there are a few who have a nursing degree that requires three years of schooling for a diploma in midwifery/nursing; they live in urban areas— particularly Lilongwe —and are thus largely removed from village life; their cosmopolitan identities are nascent or secure; and, most importantly, they have all contributed to the development of Malawi's testing programmes and policies by virtue of their professional background, training and/or organisational affiliation.

The majority of CTE members with whom I spoke were formerly employed in the public sector as Nurses, Midwives or Clinical Officers, before moving to work as a bureaucrat in government or for an NGO, positions they described in interviews as ‘promising,’ ‘challenging,’ or offering better pay in the AIDS industry (see Morfit 2011). Despite the relative economic security of this group as compared to counsellors and villagers, their livelihoods nonetheless depend on a dynamic, complex epidemic of which the promotion of HIV testing is but one (if key) part.

In interviews, CTE members invoked the global discourse to defend the integrity of the HIV testing process: ‘We have a code of ethics from the WHO’ (Baxter, CTE, age 45); ‘Those [3Cs] are [our] protocols, those [3Cs] are human rights issues’ (Odela, CTE, age 50); or in like fashion, ‘Even in the process of changing [making testing part of routine practice], we still have to observe consent and confidentiality’ (Esther, CTE, age 55). But despite their rhetorical commitment to the 3Cs, CTE members nonetheless expressed ambivalence about what strict adherence to them means for their more pressing priority: getting people tested and—if HIV positive—on ART. As one CTE member reasoned: ‘I try to balance both in my view [human rights and public health], but at the end of the day you have to be pragmatic—we want the [HIV] positives to access medication [ART] to avoid death’ (Baxter, CTE, age 45).

Notably, however, CTE respondents never expressed outright opposition to the 3Cs. Rather, they framed any difficulties with their implementation as something that can be remedied by the strategies that their professions have devised, such as new policies, training materials and guidelines. On the topic of consent, for example, they mentioned an urgent need for updated testing guidelines. The current guidelines, CTE members explained, were written “pre-ART” and before the official start of routine testing in Malawi, that is, before an HIV test was offered to patients by a provider as part of clinical care. Thus, the CTE is concerned that providers are working without proper instruction on how to seek informed consent for a test:

‘The [2004 VCT] guidelines say testing is voluntary, but now we are saying it should be offered as routine…They [health workers] don't understand [how to seek consent]. They are doing routine [testing] without [updated testing] guidelines.’ (Sampson, CTE, age 50)

‘We are planning now, once you go to a facility, you should be offered [an HIV test], but providers are not yet sensitised [trained]…We want to have consultative meetings with the DHOs [District Health Officers] and then train trainers…It is the WHO guidelines we will follow.’ (Annie, CTE, age 35)

CTE members also described their problems with the implementation of counseling, which has long been promoted as a way for individuals to learn how to change their behavior to avoid HIV infection (if negative) and to receive information on how to ‘live positively’ with the disease and avoid transmitting it to others (if positive). Two key difficulties with counseling, in the CTE's assessment, is that there are not enough trained counsellors to handle client demand for testing services, and counsellors do not adequately take clients' individual needs into account:

‘Counseling and advice giving are not the same thing… you are supposed to guide [the person]… [The client] has to make a contract with himself, not the counsellor told me [to do] x and y… for example, if you say, Emily, your problem is this, what do you think you should do, maybe I'll say I should use a condom, be faithful, and then when I come home, my conscious is reminding me things I'm supposed to do.’ (Emily, CTE, age 30)

Thus, they reason that more money is needed to (re)train counsellors.

The fact that these problems have yet to be addressed perhaps illustrates the point best: the 3Cs are secondary concerns for the CTE. Malawi's HIV testing reports are replete with sections that detail quantifiable successes—numbers of individuals tested for HIV, numbers of new testing sites, and numbers of individuals on ART (see Ministry of Health [Malawi] 2006; National AIDS Commission [Malawi] 2007). No section of these reports measures, monitorsor evaluates adherence to human rights principles. At the same time, however, commitment to the 3Cs is a defining characteristic of the CTE's role in the AIDS industry, so they must remain devoted to them. And this they can do: because the CTE are removed from everyday village life, they do not have to engage with the on-the-ground realities that the 3Cs imply. Counsellors and villagers, on the other hand, do.

Implementers: HIV Counsellors

Malawi's HIV counsellors include health workers such as Nurses/Midwives and Health Surveillance Assistants ^vi, as well as full-time counsellors and volunteers. Counsellors must be fluent in English, be at least 18 years old, and have a Malawi School Certificate of Education, the equivalent of a secondary school education (Ministry of Health [Malawi] 2007). The counsellors are individually selected, usually by a sponsoring donor organisation or by the health facilities in which they were employed in a different capacity, to participate in a training course sponsored by the MOH. Counsellors occupy an elite status relative to the villagers, but have considerably less status than the CTE, and their jobs and incomes are less secure. If the demand for counsellors diminishes, they would have to seek training for something else, unless they already have another job in the health sector. Thus, it is particularly important for counsellors to distinguish themselves from villagers, which they are able to do through the certificate their training affords them and in their acquired knowledge of Western biomedical jargon. ^vii

In training, counsellors are reminded by their trainers of their mandate as ‘torch bearers.’ In a training course I observed, one message was almost a daily one: ‘If we have enlightened our clients,’ the trainer would say, ‘we have done our jobs.’ Perhaps not surprisingly, counsellors talked about how they have changed their own behavior on account of their training; on the importance of clients knowing the difference between HIV and AIDS; and of the virtue of abstinence, fidelity and condom use (“ABCs”) and ‘knowing your status.’ In short, counsellors learn to see themselves as missionaries of the HIV prevention gospel. As exemplified in the remarks of one counsellor, ‘They [clients] know a lot [about HIV] because we teach them’ (Ruthie, Counsellor, age 45).

But despite the import of the 3Csas it concerns their social status and their sense of themselves as doing good (see Swidler and Watkins 2009), in practice, the 3Cs often place counsellors in difficult situations. The official protocols for counselling, and for upholding the principles of consent and confidentiality, cannot cover all the contingencies that counsellors daily confront (see Carter 2008). Counsellors thus adapt testing guidelines as they see fit. Notably, their adaptations occur when the ethics of HIV testing conflict with the desire of the counsellor to behave in a way that is consistent with his/her sense of morality, of what they perceive to be right or wrong (Angotti 2010; also see Whyte, Whyte, and Kyaddondo 2010).

Counsellors are troubled by testing guidelines because, unlike in the West where global testing norms originated, Malawi's counsellors are deeply embedded in the communities in which they live and work: they live side by side with clients and share with them a respect for religion and family. Thus, distinguishing between their role as counsellors and the competing roles they occupy in their extra-professional lives — be it a concerned friend, a village elder, a fellow church member—is, at times, quite difficult for them. To resolve these quandaries, counsellors told of the various strategies they used, such as intervening in the family lives of their clients, praying with those individuals recently diagnosed, and delivering food to those who are ill. As illustrated in the words of one counsellor, who mentioned extending counseling outside the clinic by sharing bible stories with a friend she recently tested and found HIV positive: ‘She is my friend. I can't just leave her’ (Simone, Counsellor, age 28).

Abiding strictly by the 3Cs also contradicts what counsellors take seriously as their mission: to help people mitigate the risks and worries of HIV/AIDS. They are thus loathe to enforce consent strictly when, for example, a client is quite sick and the counsellor thinks they should be tested for their own good or for the good of others. One counsellor explained that she developed a strategy of booking appointments for those she thinks should be tested, but do not want to be, so that they have time to think it over. She explained that in training, ‘They told us not to force [people to get tested]’ but reasoned, ‘It's my job to make a client have a life like I am [like I have]’ (Gertrude, Counselor, age 40).

Counsellors are also loathe to enforce strict confidentiality, namely when they feel someone they know will be infected as a result, or when they feel that those who will shoulder the responsibility of caring for them should be informed of their (positive) HIV status. In these cases, counsellors perceive the guidelines as impeding, rather than facilitating HIV prevention. One older counsellor, for example, described a case he handled involving his neighbor's son, who tested HIV positive. The counsellor persuaded the parents to ask their son about his HIV status. Afterwards, the son was angry and said to the counsellor, ‘You told my parents [that I'm HIV positive]!’ The counsellor responded to him, ‘Did I? Go and ask your parents’(Chaka, Counsellor, age 45).

Counsellors' knowledge of people's HIV status as a result of their work also puts them in positions of pressure and power in their communities, for example to provide this information to friends and family members. One counsellor, who tested returning soldiers for HIV, explained that the women in her village frequently ask her the test results of their husbands. They say to her, ‘We are going to die soon [if you don't tell us].’ She explained that they are trying to see ‘If I am telling everyone [about their husbands' test results]’ (Donna, Counsellor, age35).

Counsellors are also acutely aware that their actions are monitored by members of their communities. One way they respond is by practicing the ‘good’ behavior they preach in counselling so that they will be seen as having a good character. One counsellor remarked to me as we passed a notorious bar in his village: ‘They [people in my village] can't see me going [there]. That place is full of bitches [sex workers]’ (Aaron, Counsellor, age 25). Their behavior is also monitored for upholding confidentiality. Another counsellor, who tested a mother HIV positive and whose daughter was present in the counselling room, reflects:

I was left pondering on what would happen if the daughter would tell the results to other people not sanctioned by her mother. Obviously the mother will assume that it was me who has broken the news to other people.(Kunthani 090102)

This situation reveals a real quandary for counsellors: they are expected to maintain an ethical commitment to confidentiality and, at the same time, encourage clients to disclose their test results to someone they trust as well as honor their wishes to have someone present with them for testing. But because counsellors live in the same setting as their clients, they fear being blamed by others in their communities for lack of reticence when one's test results are discussed outside the confines of the testing room, when the counsellors themselves have disclosed nothing.

Intended Beneficiaries: Villagers

Most rural Malawians have less than a secondary school education, no participation in the formal labour market, and no salary at all. Villagers' livelihoods are tied largely to the vagaries of the weather that affect subsistence agriculture, uncertain fertiliser subsidies provided by the government, and strong patron-client relationships. Thus, unlike the CTE and counsellors, villagers have little to gain from defending the 3Cs. And while the 3Cs may matter for their individual lives, they also see them as a threat in their communities.

In Malawi, as in other African societies, families have powerful claims over their members (Izugbara and Undie 2008). Moreover, illness is not a private affliction, but rather a social one requiring the assistance of others (Whyte, Whyte, and Kyaddondo 2010), usually a woman caretaker (Chimwaza and Watkins 2004). Those going for testing are typically escorted to clinics by family members or friends, and often at their urging or encouragement. In the journal entry below, a future mother-in-law takes her daughter's fiancé to the clinic to be tested, insisting that if he were not, she would cancel the wedding. Given the grave fear in Malawi that being HIV positive will hurt one's marriage prospects, it is reasonable to assume that one may not otherwise agree to test, nor for that matter, admit to being HIV positive. The woman said to the counsellor:

‘This man should not leave without being tested, I was afraid if he had come here alone he would cheat [not tell the truth] on the results.’ I then told the woman that HIV testing is voluntary, a person has a right to refuse to be tested…The woman then said, ‘I fear for my daughter's life.’ (Kunthani 080410)

For rural Malawians, the 3Cs, particularly consent and confidentiality, are viewed as damaging to, rather than protective of, public health. This is not to suggest that people want to be forced to be tested; to the contrary, some take matters into their own hands if they think they will be, such as avoiding clinics where the testing requirement is perceived as compulsory (Angotti, Dionne, and Gaydosh 2011; see Kumwenda 2006 for a media report). Nonetheless, they see the necessity for insisting on testing and disclosure. Pregnant women, for example, spoke favourably of having been tested, which they perceive as critical for the health of the mother and her unborn child. As one woman in a focus group stated, and others echoed in agreement: ‘If people are not forced [to be tested] then there will be no one who will remain alive. We will all die’ (Memory, Women fgd, age 35). Although pregnant women are not usually offered the opportunity to opt-out of testing when they go for antenatal care, they are unlikely to do so anyway, as it would signal to others either that she suspects she is already HIV positive, or that she disregards the wellbeing of her baby.

When rural Malawians speak generally about the epidemic, they see keeping information hidden (confidential) as contributing to the spread of AIDS in their communities. This is not to suggest, however, that they want counsellors disclosing their own test results to others; many, in fact, mentioned avoiding those counsellors and health facilities with a reputation for lack of reticence. In a focus group, for example, a group of men discussed a notorious counsellor in their village, one whose behaviour they actively monitor:

Robert, Men fgd, age 35: For example you go for a [HIV] test and you meet [as your counsellor] Mr. B. …[Laughs]…By the end of the day, the news will be all over[village name]! [All laugh].

Sydney, Men, fgd, age 40:What is his [the counsellor's] behavior like? Does he drink beer? When he is drunk how [does] he speak?

Robert: Yes, how much does he speak?

Sydney: Does he speak about things happening at his work place [clinic] or things happening at a beer place only?

Though rural Malawians do not want counsellors disclosing their test results without their permission, they are nonetheless troubled by the consequences of strict confidentiality. In their communities, knowledge of the HIV status of a potential partner is vitally important. The prevention methods promoted by the global AIDS community—the “ABCs”—are considered unattractive at best, impossible to do at worst (see Schatz 2005);thus, partner reduction and gathering as much local knowledge about the sexual history of a potential partner are considered much preferable strategies of HIV prevention (Reniers 2008; Watkins 2004). A Village Headman, for example, explained that members of his village ask him to disclose the cause of death at funerals, a venue in which one can make inferences about the HIV status of those the departed left behind: ‘People here want the chiefs to say that the person has died of AIDS if the deceased suffered from any AIDS-related sickness’ (Yalani, Headman, age 45).

Popular concern about being unknowingly exposed to HIV is all the more complicated by the advent of ART. Rural Malawians know that being HIV positive is no longer a death sentence because of ART. But ART nonetheless poses a concern: it makes it difficult to rely on physical symptoms to know who is positive and who is not (Santow, Bracher, and Watkins 2008), information critical for those living in a setting where one's village and its surrounding areas are the marriage and partnership market. In the journal entry below, a group of men and women discuss ART during a group pre-test counseling session. The session begins with the counsellor asking the group how one can prevent infection from HIV, a question that ignites a populist critique:

It all started when a certain man said… ‘You people should know that AIDS will not end because of the drugs you give to those who have the virus. You make them become healthy as a result they start experiencing sexual desires^viii…About these condoms we just say it in this room [that we use them] but when we are with our partners some do not use them…Government should do something otherwise AIDS will not end.’ Then another woman said…‘The government is not to blame…you see men who have regained their strength proposing women when they know they are infected [with HIV]. Women too who have regained their strength seduce men.’ (Kunthani 080401)

As this journal entry also illustrates, villagers respond variously to the counseling messages they receive. The majority of those interviewed about their testing experience seemed satisfied enough just to be seen by a health worker. Their infrequent complaints were about long waiting times and rude health personnel, not about the content of HIV counseling messages. For others, however, HIV counselling advice is improbable and undesirable, particularly when it robs them of what they perceive to be a ‘good life,’ such as having ‘plain’ [without a condom] sex(Watkins 2004). As exemplified in the journal entry below, a young woman is engaged to marry and fearful that disclosure of her test results to her fiancé would threaten their marriage and her chance of having children(also see Mindry et al. 2011), as it had her previous relationships:

‘You people [counsellors] talk to us a verdict saying since you have been found with the HIV virus this is your way of life, condoms always and no child for the rest of your life’…She continued saying that was what [she] was told the time they found that [she] was HIV positive and [she] have [has] stayed without a hope of getting married and having a baby.(Kunthani 080905)

The young woman's conversation with the counsellor illustrates the tension between the public health messages of HIV counselling and understandable desires of young Malawian women(also see Smith and Mbakwem 2010).

Discussion

In Malawi, individual rights as a public health norm for HIV testing, as embodied in the 3Cs, are not embraced uniformly among testing's promoters, implementers and intended beneficiaries. The Counseling and Testing Establishment (CTE) that translates the Geneva model for use in rural Malawi, where the vast majority of the people live, can see the value of the 3Cs because they live a very different life than villagers. They do not interact with the intended beneficiaries of their efforts; the closest they come to the ambiguities of day-to-day practice is their work on the guidelines to train counsellors who deal with clients. It is thus not surprising that the CTE hew closest to the global human rights discourse in matters of testing. While it is clear that their greater priority is increasing the scale and uptake of testing (and treatment), they can invoke the rhetoric of the 3Csbecause they do not have to deal with the realities of everyday practice. If the implementation of the 3Cs is failing, they assume the situation will be repaired if the guidelines and manuals are amended or if counsellors had more training.

HIV Counsellors also know the rhetoric, but unlike the CTE, they must apply it in daily practice. They do interact with clients, and what is more, they live in the same setting. It is indeed important to counsellors to maintain their identity as professionals in command of the Western biomedical rules. But because they are up-close and personal with clients, they can see the problems with the 3Cs in a way that the CTE just cannot. And because they live side by side with villagers, they are more attuned to the mismatch between individual rights implied in the 3Cs package and the norms of the communities in which they live. Consequently, counsellors adapt testing guidelines: when a wife, for example, threatens not to be a guardian for her sick husband, they might acquiesce and break confidentiality because they know what it means if the wife were to leave and not change her husband's sheets, feed him, and escort him to the hospital for medication. While counsellors, as intermediaries, feel that as professionals they must take the individual rights perspective into account, as members of a community they must also take their friends, relatives and neighbours into account. Thus, at times they are forced to take sides.

For rural Malawians, the intended beneficiaries of testing programs, the 3Cs have different valences. Unlike the CTE and the counsellors, villagers do not have a professional'stake in defending human rights principles: they simply do not want to be tested against their will, and do not want counsellors disclosing their HIV status to others unless they ask them to. For a young woman thinking about getting tested and concerned about being HIV positive, for example, the norm of confidentiality permits agency: she can disclose to whom she wishes, rather than the counsellor telling people whom she might not want to know — like a potential partner who might not want to marry her or a parent who might scold her for behaving badly.^ix

But in their communities, these same human rights protections also come with tremendous risks. The protections implied in the 3Cs make it difficult to answer an important question: who is positive and who is not. This information, rural Malawians believe, is crucial for self-preservation. A wife must know if her husband is HIV positive, either so that she can decide to leave him, to prepare for caring for him, to protect herself from infection, and/or to anticipate her own diagnosis. Similarly, someone looking for a partner understandably wants to know if he or she has the virus or not, since it is assumed that “lying together is dying together” (Santow, Bracher, and Watkins 2008, 15). And AIDS is not only personal: it affects the family. Not only will there be caretaking when the person becomes ill, but the death of a breadwinner will affect the family finances and someone in the family will be responsible for raising the orphans.

The perspectives presented here lend themselves more to considerations than solutions. First and foremost, this study raises important counterfactual questions: how would HIV testing had developed had it been born out of African experience with the epidemic as opposed to the Western one? What would have been the impact on the epidemic of instituting provider-initiated, routine testing right from the beginning, or of testing without confidentiality?

At this point in time, however, it is unlikely that throwing out the 3Cs — even were this possible in the face of objections from powerful donors — would resolve the tensions presented here. Rather, what this study has shown is a critical need for those exporting a package of recommendations to take into consideration the characteristics of social life of implementers and beneficiaries, how they themselves perceive the dangers that threaten them, and what their strategies are. A case in point is the widespread demand for more training. This paper suggests that more training is unlikely to resolve the counsellors' dilemmas. Rather, an understanding of counsellors' embeddedness in the communities in which they live and work, and an appreciation of the various moral and ethical conflicts they experience in their jobs, could lead a search for more innovative solutions (Angotti 2010; also see Kaler and Watkins 2001).

Moreover, better policy might result from considering the perspectives of those who are the intended beneficiaries — that a health package based on consent and confidentiality for a communicable disease poses a threat to their efforts to avoid infection. Perhaps this would have led earlier to an emphasis on voluntary male circumcision for HIV prevention, rather than, or at least in addition to, the ABCs. Or perhaps it could lead to the development of a more flexible policy that supported counsellors in their efforts to adapt their recommendations to the individual circumstances of those they counsel.