Making prevention public: The co-production of gender and technology in HIV prevention research

Case study: Vaginal microbicides for HIV prevention

Vaginal microbicides are a class of substances under development, and are designed to reduce the sexual transmission of HIV and other sexually transmitted diseases, when applied locally to genital mucosal surfaces. Various mechanisms of action are being investigated, as well as different product formulations, including gels, films, creams, rings and suppositories (for an overview, see Cutler and Justman, 2008). In July 2010, the first ‘proof of concept’ for microbicides was established in the CAPRISA trial, which demonstrated a 39% reduction of HIV infection risk in women on Tenofovir gel compared with placebo (Abdool Karim et al., 2010). News articles reporting this successful result quoted the principal investigators as follows:

‘This new technology has the potential to alter the course of the HIV epidemic, especially in southern Africa where young women bear the brunt of this devastating disease,’ observed Dr. Quarraisha Abdool Karim. ‘This is an advance because it empowers women,’ said Dr. Salim S. Abdool Karim. (Anonymous, 2010)

Here, I take the example of one specific candidate product, PRO 2000, which entered a large phase III clinical trial, called MDP301, in 2005. I explore how it came to be developed by networks of actors from diverse social worlds working in a specific historical context, and demonstrate how scientific practice is intimately bound up with prevailing social norms and political agendas. The basis of this analysis is a set of interview data from a purposively sampled cross-section of scientists, advocates and trial participants in the UK and Zambia, groups that were central to developing and testing this experimental pharmaceutical technology. Among the ‘elite’ informants were investigators of the MDP301 trial, coordinators of the various disciplinary groups involved in the trial (statistics, social science, community liaison, social marketing, communications), the pharmaceutical industry partner, and representatives of government and international advocacy organizations. Twenty-one interviews were conducted with MDP301 staff (seven European, 14 Zambian) and eleven interviews with key stakeholders (three European, six Zambian, two American). I also conducted 15 interviews with trial participants and their partners in the MDP301 trial in Zambia, as well as five focus groups with trial participants and community members.

Interviews and focus groups took place between October 2007 and June 2009, roughly in the middle of the phase III clinical trial of PRO 2000. Interviews with the Northern researchers took place at universities and trial coordinating offices around the UK, as well as at an international investigator’s meeting in Mozambique. Interviews with Zambian researchers and community members took place in the social science offices of the MDP 301 trial, as well as in key stakeholders’ places of work. During the interviews, I asked both about the historical development of PRO 2000 and the current research, which aimed to bring PRO 2000 to licensure. Therefore, although the technology had to some extent already been produced, it was by no means fixed or stable.

My analysis of the trial is an inevitably situated account: at the time I conducted this research, I was working for the Microbicides Development Programme as a social science coordinator. Key informants who worked on the trial knew me as an insider and someone with whom they could safely and legitimately discuss the trial. By the time I started interviewing, I had been a member of the team for 3 years, and had travelled extensively with colleagues from across the trial network, building up strong rapport with them. This relationship gave me unprecedented access, but also shaped the data in important ways. For example, my choice of informants was based on privileged prior knowledge about power and decision-making dynamics within the Programme, knowledge that would not have been immediately available to an outsider. The style of interviews was often informal and ‘familiar’, mirroring common day-to-day interactions between members of a research team; senior colleagues appeared not to find my questions threatening, perhaps because I was a junior, relatively powerless actor in the trial hierarchy. Aware of the risk of taking too much for granted in our shared knowledge and understanding of the trial, I had to negotiate an awkward balance between presenting myself as an insider and questioning ‘the obvious’. At times, this broke the spell of complicity with my informants and highlighted unconventional dynamics in the researcher–researched relationship. The analysis below is a product of my experience working on the MDP301 trial, and is therefore an account grounded deeply in participation.

Pharmaceuticals and the politics of ‘empowerment’

In the field of HIV prevention, microbicides are now a given. Posited as a woman-controlled or -initiated form of HIV prevention, they are also commonly referred to as a tool for women’s empowerment. On its website, the Microbicides Development Programme provides the following definition: ‘A microbicide is an HIV-prevention method specifically for women’ (Microbicides Development Programme, 2009). Typifying the discourse on microbicides, the World Health Organization focuses on three key facets of this new technology: power, protection and women’s independence.

The availability of microbicides would greatly empower women to protect themselves and their partners. Unlike male or female condoms, microbicides are a potential preventive option that women can easily control and do not require the cooperation, consent or even knowledge of the partner. (World Health Organization, 2009)

So pervasive is this discussion of women’s empowerment from all quarters – not only advocates, but also scientists, government and the media – that the fact that this is a pharmaceutical product almost gets overlooked.

The idea that a pharmaceutical product could not only address physical and biological problems but also confront social ailments is not new; however, the social remedies provided by pharmaceutical products tend to be seen as secondary or collateral benefits, and have not usually been proposed as solutions to structural inequalities. With microbicides, the social and the biological have been thoroughly fused, and pre-clinical and clinical development have come to depend on the notion of women’s empowerment.

PRO 2000, the candidate microbicide that was taken forward into the MDP phase III trial, was initially developed by a small biotech company in the United States. When the molecule was first found to have anti-HIV activity, the initial idea was to use it as an intravenous therapy for HIV infection. At the time, AZT was the only approved antiretroviral agent, so that this application would have had a strong place in the market. However, following early clinical trials administering PRO 2000 intravenously to HIV-positive gay men, this application was deemed unsuitable, given the side effects associated with intravenous use (Weber et al., 2005). At this point, PRO 2000 was a molecule looking for a ‘technology’; uses and user groups were being constructed and contested. Following, and contributing to, the trajectory of discourses on AIDS, the drug itself was to move from the domain of the sick gay male body (HIV-infected gay men as a ‘core group’) to the unprotected female body (uninfected, vulnerable women as the potential victims of a generalised epidemic). This move was ostensibly a commercial one; Neil, ² who led pre-clinical development of the drug at the biotech company, describes the point at which women became the new user group for PRO 2000:

So the way we came to microbicides was, essentially we were trying to figure out if there was anything we could do with PRO 2000 since it wasn’t gonna work as an antiretroviral, and the idea of, you know, a female controlled prevention method had been kind of kicked around a little bit so I looked into it a little further and Lori Heise and Chris Elias had recently published a paper … it’s one of the seminal papers in the microbicide field … . And I read their paper and I said, ‘Wow, PRO 2000 would be perfect for this’, and so we re-focussed the whole programme on its development as an intra-vaginal microbicide.

The article Neil refers to provides an impassioned argument for a complete restructuring of AIDS prevention focused on women’s vulnerability to HIV infection (Heise and Elias, 1995). The authors couple women’s empowerment with HIV prevention, calling for women-controlled prevention technologies:

The AIDS epidemic … creates two imperatives: to begin in earnest to work on changing the underlying causes of women’s vulnerability and to pursue vigorously every means possible to strengthen women’s immediate ability to protect themselves in the face of the economic and cultural forces currently allied against them. This, in turn, means placing greater emphasis within existing AIDS programmes on empowering women and committing major resources to developing new prevention technologies … that women can use without their partner’s knowledge or consent. (Heise and Elias, 1995: 931)

Microbicides advocacy such as this started in the field of women’s health and contraceptive research and development. Zena Stein, a South African epidemiologist, was one of the first major voices in the field; her article, ‘HIV prevention: The need for methods women can use’, published in the American Journal of Public Health, drew widespread public attention to the issue. Stein was unambiguous in linking women’s empowerment with HIV prevention using women-controlled technologies:

The proposition of this paper is that the empowerment of women is crucial for the prevention of HIV transmission to women. It follows that prophylaxis must include procedures that rely on the woman and are under her control. A wider range of chemical and physical barriers that block transmission through the vaginal route must be developed and tested. (Stein, 1990: 460)

The link made between HIV prevention, empowerment and technology was a fortuitous one for the biotech company at this point in time, when they had a molecule with anti-HIV activity to develop and commercialise. Because of the global interest in women-controlled technologies and the government funding being made available to test them, PRO 2000 was a good commercial venture for the company at that time. Neil was frank that women’s empowerment was interesting from a commercial point of view; unsurprising, perhaps, given that microbicides were forecasted in 2002 to have a peak market size of US$5 billion (Pharmaco-Economics Working Group of the Rockefeller Foundation Microbicide Initiative, 2002):

our job really and our obligation to the share-holders is to be commercially successful and you know the way to do that is to meet people’s needs, to identify what the needs are and to meet them and to that extent, you know, the whole idea of women’s empowerment is very important, ’cause the more product is used, the better the commercialisation.

In the interview, Neil observed that when empowerment becomes an ‘unmet need’ in the medical arena, it becomes possible to meet that need and commercialise the solution. As Jasanoff has observed in relation to genetically modified (GM) food innovation for the developing world, ‘making a saleable product is to create a need … the needs that most products meet are more often made than found’ (Jasanoff, 2005: 189). In this case, by framing women’s lack of social power as a medically-unmet need in the HIV arena, it became possible to provide a pharmaceutical solution to empower women – a vaginal microbicide.

Neil was not alone in configuring women’s needs in this way. Adrian, another scientist (often identified as one of the first advocates for microbicides) was persuaded of the need to give women a product they could control without the consent of their male partners. During the 1990s, the discourse on women and HIV took on a new dimension; by 2002, when Kofi Annan stated that ‘AIDS has a woman’s face’ he appealed to a wave of sentiment that had grown through repeated discussion of the ‘feminization’ of the epidemic (Annan, 2002). Although this discourse has now become so entrenched as to be treated as a self-evident fact, ³ another leading women’s health advocate, Ellen, drew attention to the work that advocates have had to do to effect this discursive shift: ‘As the power of gay men in HIV was on the descendency, I think you started to see the rise of women and HIV and that became … a mantra and became very mainstream then. … But the microbicide thing was always ahead of that. So we didn’t actually ride that wave, we were like pushing the wave almost.’

In part through women’s health advocacy, then, the idea that AIDS was a women’s disease became common currency, and carried significant persuasive force in the microbicides field. The fusing of women’s biological vulnerability to HIV with their social vulnerability to infection seamlessly led to the medicalisation of powerlessness and the search for a medical solution to it:

If you go back to my very original paper … I said that the reason people aren’t using condoms is because the man won’t use them … and that a point about this: the woman would be able to take the initiative herself. In other words, empowerment. So that was a major thing and later on – it was mainly the men who were getting infected, I suppose from sex workers … but when it became a woman’s disease, you know, HIV wears the face of a woman and now more women are being infected than men, once that was appreciated, then of course it became not only women’s empowerment but women were very vulnerable and needed … these microbicides to be able to protect themselves. (Adrian)

In this excerpt, we see again the characteristic shift in discourse from women as a circumscribed core group (‘it was mainly the men who were getting infected, I suppose from sex workers’) to all women as a vulnerable population in need of medical intervention (‘it became a woman’s disease … women were very vulnerable’). Not only that, but the discourse of women’s vulnerability and powerlessness constructed men as the enemy in the fight against HIV; Adrian recalled, ‘We used to call them “stealth virucides” like the stealth bomber.’ ‘Stealth virucides’ refers to the idea, widely propounded in the early days of microbicide development, that women would use microbicides secretly, without the knowledge of their partners. By referencing the stealth bomber, Adrian invokes the notion of the technology as a weapon that women could use surreptitiously to combat their risk of acquiring HIV from men. Covert use became a presumed feature of microbicidal products (see for example Koo et al., 2005; Woodsong, 2004). As I will discuss below, this sort of furtive use was directly linked to the discursive production of gendered subjects by advocates and investigators working in this field.

The publics of prevention

The seed had now been sown for a prevention product to empower women in the fight against HIV and the concept of female-controlled technologies was rooted. However, in order for work on microbicides to obtain funding and to progress, it was imperative that there be a coherent subject in need of the technology. It was not enough to characterise the users simply as ‘women’; as noted above, during the early 1990s, female sex workers were seen as a core group, the reservoir of infection from which the disease was spreading to the general population (Eckholm, 1985; Kreiss et al., 1986; Padian, 1988; Rosenberg and Weiner, 1988; Van de Perre et al., 1985). A new type of woman was needed, an acceptable face of the epidemic that would garner sympathy and could be used to mobilise support for funding research on microbicides. As Ellen remarked on the subject of political lobbying, ‘we had to play up – which I always hate – the kind of vulnerable married woman thing’. One of the first scientists to develop microbicides in the UK, Harry, was also clear about the need to capitalise on the notion of the powerless female user:

Harry was not alone in arguing that the development and testing of microbicides was first and foremost a scientific question – did this technology work or not? However, the appeal to social interests and the construction of gendered subjects this entailed, became a necessary part of drug development. Discussions with senior investigators showed how an essentialising discourse of women and men made it possible to produce a subject in need of this new medical technology.

While it would be wrong to suggest that the investigators all shared a common view of women and men that they mobilised to further the research agenda, a discourse around disempowered women and irresponsible men came to the foreground during the development of the trial. This was remarked upon by the senior social scientist, who bemoaned the fact that:

there’s always such an assumption that, you know, poor old women are always the victim and, you know, these guys are dreadful, they won’t use condoms and, you know … women can’t persuade them to use condoms and it’s that whole sort of … feminist politically correct discourse that makes people think that it’s essential that women use these things secretly.

Some of the most senior investigators were core proponents of the discourse that placed women in the position of ‘victims’ and characterised men as ‘dreadful’. Their talk of women and men drew heavily on a biologically deterministic view of sexual difference, with women frequently described in terms of their reproductive capacity and men presented as promiscuous inseminators. As one senior researcher noted, ‘women are the, you know, repositories of the family and the creation of the next generation’. The exculpation of women was further established by conferring on them ‘innate’ qualities associated with caring and nurturing, sacrifice for the good of the family, communication rather than obstinacy, and bearing responsibility for family life more generally. Women’s nature could only be elaborated in this way by reference to their binary opposite in men, and thus men were typically described as irresponsible, uncaring, pleasure-seeking and driven by their sexual urges. Various examples were drawn upon which appeared to legitimate these givens, for example evidence that men wouldn’t use condoms and a study in which men said they would only accept microbicides if the products didn’t interfere with their own sexual pleasure.

While some investigators drew on their own previous experience and social scientific findings to portray the need for a product women could use alone, others went ‘back to biological basics’ to argue the same point. Describing the conceptual appeal of microbicides, Chris, a clinical principal investigator, said:

For a heterosexual sexually transmitted disease, you don’t have to actually prevent both sexes; for example, a single sex vaccine will work in a heterosexual sexually transmitted disease. You only need to block one partner, so it’s ok if you have a female-only intervention, it’s gonna block the transmission cycle.

As we continued to discuss women-controlled prevention methods, Chris went on:

In this case, the biological and epidemiological rationale for microbicides takes precedence over moral arguments about women’s right to protection. Epidemiologically speaking, you can stop the transmission cycle by intervening in one sex or the other. Biologically speaking, it makes more sense to intervene in women than men, because women, by their very nature, are more compliant (and men, due to their testosterone levels, cannot help being non-compliant). The tone and vocabulary might vary, but the overall discourse that emerged around microbicides tended to construct women as monogamous, maternal and passive, and men as promiscuous, uncaring and irresponsible.

Having established women’s vulnerability to infection, their biological suitability for intervention and their moral need of protection, the rationale for a woman-controlled technology became ‘obvious’. As mentioned above, microbicides were initially conceived as ‘stealth’ products that could circumvent men altogether. However, it was not only women’s use of the product that was proposed to be covert; the fact that the technology might also protect men was also ‘covered up’. As I have been discussing and will elaborate below, there were reasons why it was advantageous to position women as the sole beneficiaries of the technology, and to exclude men from microbicides discourse except in negative terms.

In discussion with Adrian it became clear that the absence of men from the microbicides prevention discourse was socially rather than scientifically motivated. He admitted: ‘I would be very surprised, if Nonoxynol-9 had worked, for instance, or Savvy more recently, if that had worked and protected women. … I just wouldn’t believe it if it didn’t protect men.’ ⁴ And reverting to a scientific discourse, he described the potential for bi-directional protection in PRO 2000’s mechanism of action:

We don’t know that microbicides will protect a woman, it’s never been shown yet. But … on the same grounds it’s protecting a woman it should be possible to protect a man, because the principle on which these things work, let’s take for example PRO 2000 gel, which binds to the V3 loop and prevents the virus from attaching. I mean that’s a crude explanation, but let’s say that’s how it works. The healthy woman inserts this gel and … has sex with a man that’s infected, his semen has to mix with that gel or somehow the gel coats every vulnerable surface, and it has to attach to the V3 loop of the virus – and the PRO2000 molecule does that – and that will protect the woman. Now, how long does it have to attach, what’s the time window? Not very long. The virus enters the woman, it’s going to attach to its receptor lymphocytes fairly quickly, so it doesn’t have very long. So what we’re saying is we think that’s going to protect the woman. Think of it the other way – the man is uninfected, the woman is infected … she inserts the gel, she is shedding virus into her fluids, that is going to be around for an hour or longer, I don’t know, it depends, but it’s going to be around for much longer than in the case of the man’s semen. So the chances of that microbicide destroying or attaching to the virus that she herself is shedding is much greater than … . So in other words … if it protects the woman, it’s jolly well going to protect an uninfected man.

From this biological point of view, then, men were more likely than women, or at least equally likely, to be protected by PRO 2000, and so rather than the technology being a women’s product, it could also have been a men’s product or a product for both sexes. So why has bi-directionality been absent in discussions and descriptions of vaginal microbicides? Adrian went on to locate the silence on bi-directionality in terms of the speculative basis of male protection: ‘Now that has never been published, it’s never, as I say, it’s never really discussed unless I raise it. Because it’s speculative.’ However, as he noted previously, protection in women was also speculative given that, at this time, the product had not yet been tested. The sticking point, as he saw it, was that you can prove in trials that the product protects women, but you can’t prove that it protects their male partners. ⁵ Therefore, on the one hand, the silence around bi-directionality could be seen as an inevitable response to the limits of scientific knowledge-production. However, if bi-directionality were simply not discussed because it wasn’t possible to prove it, it’s unlikely that the technology would have been framed so completely as a woman-controlled product. In fact, I suggest that the silencing of the potential protection of men was a political and commercial move, subsequently legitimated by the scientific impossibility of proving it. Labelling protection in men as ‘speculative’ thrust it outside the domain of scientific knowledge, thereby demoting its ontological status.

The limits of scientific possibility were invoked not only to legitimate why microbicides were only being tested in women, but also to explain why PRO2000 was only being developed as a vaginal, as opposed to rectal, microbicide. Several informants pointed out the fact that it is much more challenging, scientifically, to develop a rectal microbicide because of the physiology of the rectum and the efficiency of HIV transmission via this route (for an overview of biological challenges to rectal microbicide development see McGowan, 2008). However, it soon became apparent that whatever scientific hurdles stood in its way, the social conditions and consequences of developing a rectal product made it a very problematic proposition. Neil, from the biotech company, said they had considered rectal development, but that it was ‘obviously a little bit more of a delicate thing’ and that, while their marketing department thought it could be highly successful to pitch it to gay men in industrialised countries, this was ‘definitely not a high priority’. Although he was not explicit about the reason why rectal use was not high on the agenda, the difficulties involved in marketing a product utilised by ‘deviant’ risk groups to those engaged in ‘mainstream’ sexual behaviour (heterosexual men and women) may have been a deterrent. It is worth noting that the US Food and Drug Administration (FDA) has never approved a condom or other device for anal sex and that approval of technologies for rectal use has historically been considered a ‘political minefield’ (Scarce, 1999). ⁶

Chris provided additional reasons for why the development of a rectal product was not forthcoming. First, he discussed how the available research expertise was in the domain of women’s reproductive health and not rectal physiology:

He related the popularity of gynaecological research to the social acceptance of examining the female reproductive tract, as opposed to the rectum: ‘a no-go area, the bottom really, for heterosexual male conceptions’. This is perhaps indicative of the long history of the medicalisation of women’s bodies (Corea, 1985; Ehrenreich and English, 1979; Oudshoorn, 1994) and the continuing stigma of homosexuality and its practices in some spheres of public life. In a similar vein, Epstein has discussed anal cancer as the ‘“great undiscussable” in public discourse about HPV vaccines’ and the resultant epistemic void in gay men’s health (Epstein, 2010: 63). It is notable that the baseline against which Chris measures conceptual deviation is the heterosexual male, hierarchically the most dominant group and clearly still powerful in determining the knowledge base from which health research proceeds.

At the time of the interview, Chris was in fact conducting a phase I rectal safety study of PRO 2000, but had said it wasn’t possible to do the study in heterosexual couples:

consent would be an issue … maybe there are lots of women out there who are having consensual anal sex within their heterosexual relationships and both partners maybe would be willing to agree to having consensual anal sex within their relationship for a trial. But you know it’s difficult, the ethics committee … it’s a bit like the jury, it’s got to have normal men in true, you know, men and women and the average folk, so you’ve got a bit of a job persuading average folk about consensual anal sex, it’s just – it’s a bit like legalising heroin or something, actually there’s a very powerful argument to do it, but it’s so … it’s … it’s a difficult area for ordinary people to actually see the logic in legalising heroin. It would be very difficult for an ethics committee to see … there’s this notion of rape around heterosexual anal sex, that’s what comes to people and you know, the domination of men over women is subconsciously a more powerful metaphor in terms of anal sex than vaginal sex because it’s not a … route that results in conception …

Chris is clear that the design of the study was directly related to social conceptions of appropriate heterosexual behaviour, which does not include consensual anal sex. The bounds between the social and the scientific are blurred by the institution of the ethics committee – composed of ‘the average folk’, but arbitrating on the realm of the scientific. In a good example of how the social, the scientific and the technological are co-produced, we see how gender is produced by reference to sexual difference and reproduction, how compulsory heterosexuality is instituted through denial of anal sex as a legitimate heterosexual activity, and how science follows and thereby reproduces social prescriptions on gender. Finally, relating this to the development of PRO 2000 first and foremost as a vaginal microbicide, Chris linked the social to the political:

To get into specific research to prevent sexual transmission through alternative sexual practices, if that’s how you want to describe anal sex, it is a bit … challenging for many of the Daily Mail readers, yup. And the government, I’m afraid, has to go by the Daily … that’s, that’s … they’re all so centrist now, that you know, we’re not living in a socialist, progressive … we’re not living in an era when we can expect any more socialist progressive administrations for the foreseeable future.

Similar apprehensions were raised by US advocates around lobbying for rectal microbicides at a time when to do so might jeopardise federal funding for microbicides. Ellen shared Chris’s belief that the association of anal sex with homosexuality, rather than heterosexuality, inhibited the development of rectal technologies:

People don’t like to think about that (rectal use), they don’t feel comfortable, there’s been fear that … sort of raising the issue of rectal use could – and it could, quite frankly – undermine the vaginal microbicide programme, because of the crazies that are in Congress and you know, all of a sudden you get wind that NIH is funding rectal microbicide studies and the whole funding for the microbicide programme could be cut off tomorrow. ⁷

The political discomfort with acknowledging heterosexual anal sex arguably led to the delay in securing socio-political support for rectal microbicides. ⁸ Although an act of unprotected anal sex is reportedly many times more likely to result in HIV infection than unprotected vaginal sex (Baggaley et al., 2010), and although research suggests it is not uncommon between heterosexual partners, ⁹ funding levels for rectal microbicides have historically been well below those for vaginal candidates (Feuer, 2006). Furthermore, by letting themselves be led by dominant social proscriptions around appropriate heterosexual behaviour, scientists have avoided the rectum as an area of research and thus, as Chris originally pointed out, there is a limited knowledge base in the scientific arena.

‘Making things public’: Re-configurations in Zambia

The MDP301 trial to test the safety and efficacy of PRO 2000 took place in four countries in East and Southern Africa. Approximately 10,000 women were recruited from six clinical trial sites and asked to use the experimental gel every time they had sex. This provided an opportunity to explore not only the production of gender and technology in the North, but also the processes through which ideas and technologies travel, how scientific knowledge and its products bridge different geographic and social locations, and how scientific and social intelligibility are produced in an African trial community.

The trial proceeded on the assumption that women were autonomous agents who could participate freely in directing their own sexual and reproductive lives, but who were not empowered to negotiate HIV prevention with their partners. Its protocol did not address the dynamics of gendered power relations that might impact the running of the trial, such as men’s role in their partner’s recruitment, retention or adherence to the protocol for using the study gel. As Mort and colleagues have underscored in their work on telemedicine, ‘local innovations have no place in trial methodologies: indeed, they undermine the very notion of a trial’ (Mort et al., 2004: 120). This is a point also made by May: ‘clinical trials are founded on denying interpretative flexibility in practice to those working within them because they rely on the imposition of a rigorous trial protocol on everyday practice and thus the standardisation of clinical practice’ (May, 2006: 525).

As an example of this, the process of informed consent was directed at the woman alone. In clinical trials, the subject who signs the informed consent is construed as an autonomous individual, acting rationally and independently in a context of choice. The process is largely viewed as countering medical paternalism by investing research subjects with the information (and therefore power) to freely decide about their participation. However, as Corrigan observes, ‘the dualistic opposition between liberal concepts of freedom and autonomy versus powerful autocratic medical practices fails to recognise that power is not just a phenomenon that is exercised as an external constraint, but that prevailing cultural norms, values and systems of expertise shape the field of choice’ (Corrigan, 2003: 789). In the Zambian context, the prevailing cultural norm was for men to be the gatekeepers to women’s choices.

Local discourses of gender at the Zambian trial site were deeply at odds with the autonomy paradigm embodied both in microbicides and in the clinical trial designed to test them. In the setting in which the trial took place, women are not generally regarded as autonomous individuals; customary law treats them as dependents, with property and inheritance rights contingent on marriage or family ties (Byrne, 1994). As one woman put it, ‘We are one body. We are just one person. My husband and I are one.’ The discourse promulgated by both Zambian trial staff and participants was one of marital unity and togetherness. This extended to all matters in life, and use of the microbicide was no exception. Therefore, contrary to the UK researchers’ designs, the majority of women asked their partners for permission to enrol in the study and use the gel (Montgomery et al., 2010).

Women’s and men’s accounts of using the gel suggested that the autonomy paradigm from which it had emerged was deeply at odds with the paradigm of conjugal unity within which married life was discussed. Women’s control was neither a presumed feature of the research nor a desired attribute of the gel. On the contrary, male control was presented as deeply implicated in the product’s effectiveness and the success of the trial – to the extent that one of the local trialists commented:

What I’m seeing is that in future, if microbicides are made available and they’re effective, what everyone would want to see is that a man goes into a shop … gets the gel, and then takes this gel to the woman and says, ‘Today we have to use this. From today onwards, we have to use this.’Every time before we have sex I will remind this woman to say, ‘Have you applied it?’

Men were said to be responsible for allowing their wives to enrol; for agreeing to gel use, reminding their wives to insert it or inserting it for them; for facilitating wives’ continuing participation throughout the follow-up period; and in some cases for contributing to data collection. Therefore, although the discourse on women’s empowerment, which had grown out of a rights-based approach to women’s health in the West, led to the production and stabilisation of the gel as a product for women in Europe and the US, this was not a given in Zambia. In fact, when transposed to the Zambian setting, the nature, form and capacity of PRO 2000 were a complete unknown. In this setting, the gel was not interpreted as an object of women’s control; on the contrary, it was seen as a product that men and women use together:

I don’t think the way people understand is that it’s purely under the control of women. … I don’t think they perceive it as that. Because, I don’t know, maybe it’s just because of the way our society is, there’s nothing which is totally under the woman’s control [laughs], so not even this can be totally under her control. So I don’t think they really perceive it as totally woman-controlled. (Senior staff member, Zambia)

The ones who are supposed to participate in the trial, it is us women, including men, we are supposed to unite. Because like these diseases, when they come they are not supposed to come for one person. When they come like in this house then both of us will be affected. So even on the programme we are supposed to unite the two of us. (Female trial participant)

It is my wife who is using it. So in using it, we are using it together, so it means that even me, I am on the programme. (Male partner)

The fact that women involved their partners in the trial and in the use of the microbicide gel was not to say, however, that women were the passive victims of inequitable gender relations. Whereas women have historically been portrayed as the passive and disempowered objects of surveillance in clinical research – as so-called ‘living laboratories’ – data from this study showed how women actively manipulated the techniques of power to their own advantage. Not only did the trial use women to generate data and ‘make science’, but women also used the trial to shape their relationships with their partners. For example, in seeking their husbands’ consent, women were not simply being submissive to their partners, but were actively seeking to strengthen their marital relationships. Getting their partner’s consent was constitutive of the very ideal of unity and togetherness which characterised a good marriage, but in addition, it provided women with an opportunity to address aspects of their sexual relationship that were wanting. For example, by sharing trial information and materials with their husbands, women could raise the topic of HIV and discuss risk. Rather than seeing the need for their partner’s consent as disempowering, some women recognised the opportunity inherent in it.

Not only that, but in spite of the medicalisation of women’s powerlessness as the linchpin of microbicides development in the UK, one of the most prominent discourses that the gel engendered at its point of use was that of sexual pleasure. Although women’s bodies have been posited as a site of control in terms of the HIV epidemic, and historically in terms of fertility, what women themselves talked about most animatedly was the gel as a catalyst for desire:

People of God, it stimulates someone quickly! Before I started using gel he was only able to have just one sex act, that was all, but now we have sex more than once. After a sex act he still remains on heat and wants some more. He liked it. It makes the vagina warm inside and after he has ejaculated, the woman also still insists on some more sex, ‘don’t come out, please continue!’ [Laughter] (Focus group participant, 30+ age group)

The pleasure women spoke about occurred on many levels: their own physical pleasure from inserting the applicator and using a lubricant during sex; pleasure from being sexually pleasing to their partners; pleasure from the intimacy and trust of introducing and talking about a novel sexual accessory; pleasure from the knowledge that they could strategically use gel use to negotiate other aspects of their sexual and reproductive lives, such as the use of condoms and contraception.

It was not only among women that the pleasure discourse circulated, but also among men, and among men and women together, as the following extract from a mixed male/female focus group illustrates:

In effect, gel became a sexual signifier, associated at least in part with being ‘sexually active’. The latter was a stated criterion of trial participation, but was construed locally to mean someone who is active in sex and, linked to this, enjoys frequent sex (this diverging from the biomedical meaning which refers simply to someone who has sex, regardless of its quantity or quality). Changes in open communication about sex within the trial context undoubtedly allowed this discourse to be voiced, so the technology and the research can be seen to have mutually configured each other. Not only that, but the new technosexual scripts that gel-use engendered cast women not as victims, as has so often been the case in HIV discourse, but as desiring sexual subjects in their own right.

Conclusion

In this paper, I have analysed micro-processes of scientific practice in the field of microbicides to show the politics at work in HIV prevention research. I have illustrated how HIV science, far from being a dispassionate process, is a normative undertaking shaped by pharmaceutical technologies, end-users, politicians and tax-payers. In this case, these included the PRO 2000 molecule, scientists from the worlds of biotech and academia, civil servants at the Department for International Development, imagined readers of the Daily Mail, women’s and gay men’s health advocates, Zambian trial participants and their partners, and Community Advisory Board members. In bringing together the design and use phases of this particular drug’s story, I have demonstrated the contested and distributed labour of representation that is effected through transnational clinical research, work that is usefully captured by Latour’s notion of ‘making things public’:

Each object gathers around itself a different assembly of relevant parties. Each object triggers new occasions to passionately differ and dispute. Each object may also offer new ways of achieving closure without having to agree on much else. In other words, objects – taken as so many issues – bind all of us in ways that map out a public space profoundly different from what is usually recognized under the label of ‘the political’. (Latour, 2005: 15)

In this example, the co-production of gender and technology involves a public space that exceeds the designer-user dyad implied by scripting.

The gender stereotypes that northern designers mobilised to shape this space were not only deeply problematic, but also highly productive, as they reproduced norms about heterosexual marriage, desire, and sexual practice that marginalised other ‘deviant’ groups (such as commercial sex workers and gay men) and practices (such as heterosexual anal sex) in the microbicide research agenda. The privileging and marginalising of such groups and behaviours occurred regardless of location, illustrating that the key dimensions of power concerned not only North–South relations, but also heteronormativity. A key point this analysis illustrates, therefore, is the fluidity and unpredictability of North–South research relations and the need to consider multiple axes of influence. The concept of ‘making things public’ helps us to do this by drawing a more distributed assembly of parties into the analytical frame and directing our attention to the divisions and connections, voices and representations they perform.

Microbicides have drawn around them an assembly of parties representing diverse political interests that go well beyond the conceptual limits of disease prevention. Indeed, this story merely adds to a well-established body of literature describing the representational politics of AIDS throughout its 30-year history (for example Doyal et al., 1994; Epstein, 1996; Treichler, 1999; Waldby, 1996; Wilton, 1997). Like some of this earlier work, this paper suggests that what is at stake are not simple divisions between discrete genders or geographies, but the very public politics of desire. Microbicides discourse in the US and the UK configured the end-user as a vulnerable, powerless woman, often married or in a committed relationship, who would use the gel to protect herself from vaginally-acquired HIV infection. This image found traction with politicians and the general public, who funded the research. Linked to this, and to the specific chemical properties of the PRO 2000 molecule, the designers and testers of the drug candidate did not allow for high volumes of use, ¹⁰ nor for rectal use, thus implicitly delimiting the bounds of ‘normal’ heterosexual behaviour. They neither anticipated nor promoted the open use of the product by partners together.

By contrast, the users of the product who participated in the MDP301 trial in Zambia predominantly conformed to local norms regarding marital relations, extolling the virtues of openness and sharing. Within this context, sexual partners together spoke of the new technology in terms of its lubricating properties and its stimulus to sexual desire. They did not perform the gender scripts that evolved in the global microbicides movement; rather, they disrupted this script through localised reconfigurations, demonstrating that neither gender, nor ‘gendered’ technologies have intrinsic qualities.

We might ask what this means for the ‘success’ of this particular HIV prevention technology; as de Laet and Mol (2000) observe of the Zimbabwe bush pump, success is not necessarily a binary matter. At the end of the MDP301 trial, PRO2000 gel was found not to be successful in preventing HIV (McCormack et al., 2010); it was successful, however, in increasing sexual pleasure (Montgomery et al., 2010). Arguably, it may not have been successful as a tool for women’s empowerment, but it was successful as a product that women and men wanted to use together to prevent infection. As far as HIV prevention is concerned, finding something that people can and want to use or do is key. As Moore has written of latex devices such as condoms, ‘successful configuration has much to do with associating latex with pleasure and creating particular conditions in the [sex worker–client] interaction that enable the client to feel cared for’ (Moore, 1997: 460). Here, successful configuration is not only about the technology but also about the user. In the context of this paper, it is not just a question of ‘the biopolitical production of subjects’ (Nguyen, 2005: 132), but of the mutual constitution of subjects and technology in transnational contexts – a process that bears not just on therapeutic citizenship, but ultimately on therapeutic success. The coupling of the women’s empowerment agenda with pharmaceuticals for HIV prevention, while being a success with Northern politicians and publics, does not necessarily translate well in other locations.

Latour’s notion of ‘making things public’ pushes us to question the public performance not only of science in action, but science as it enacts ‘politics by other means’ (Latour, 1988: 229). With the transnational testing of new technologies for HIV prevention, these performances create new publics with a stake in the representational claims being scripted ‘for them’. Technologies designed to intervene simultaneously in the social and the biological body – such as microbicides for women’s empowerment and HIV prevention – therefore open themselves up to contestation and reconfiguration on multiple levels. The images of AIDS as having the face of a woman, of African men as promiscuous and uncaring, and of empowerment arriving in the form of a vaginal gel – these representations are deeply normative and political, particularly when exported from one side of the world to the other. Although use of the technology in this case study was successful in spite of a set of ideas that were incongruous to the actual users, this is not always the case. Where imagined and emergent users diverge dramatically, it is plausible that a technology will fail. It is not enough, then, to ask simply about the gendering of a technology, or about the ethics of the trials through which it is tested; we must go back and ask about the transformative co-constitution of technologies, users and publics, and what this might mean for therapeutic success.