Gender and the HIV Epidemic

MEN AND THE HIV EPIDEMIC
by Kim Rivers and Peter Aggleton
Thomas Coram Research Unit
Institute of Education, University of London, 1999

I. INTRODUCTION

As the epidemics of HIV and AIDS have developed over time, international organisations, national authorities and non-governmental organisations (NGOs) have recognised that social inequalities and power relations have an important impact on HIV transmission. Factors such as poverty, migration and urbanisation have a key role to play in facilitating HIV infection (Sweat and Denison, 1995). Other variables known to influence the vulnerability of individuals and groups include social background, age, race, gender and sexuality. Not infrequently, these different variables interact with one another so as to render some groups systematically more vulnerable and other groups more protected (Piot and Aggleton, 1998).

Importantly, and for the purposes of this review, there has been increasing awareness that prevailing relationships within and between the sexes, or gender relations as they are more usually called, affect not only the development of the epidemic (Carovano, 1992), but the manner in which individuals, groups and communities respond (see, for example, Aggleton and Warwick, 1998). As used here, the term gender refers to the social shaping of femininities and masculinities, and challenges the idea that relations within and between the sexes are ordained by biology or nature (Ankrah and Attika, 1997). Unequal gender relations can be seen in many ways but are particularly visible in the special vulnerability of women to HIV and AIDS in developing countries, and in men's risk taking behaviours. Economic and social vulnerability, as well as stereotypical gender roles, influence women's and men's vulnerability to HIV infection, while fuelling the overall course of the epidemic. As Meursing and Sibindi (1995: 66) have recently written 'the AIDS epidemic thrives on rigid sex-role definitions.'

Recent reviews also suggest that women in many parts of the developing world are less likely to control how, when and where sex takes place, thereby increasing the likelihood of unwanted pregnancy, STDs and HIV (see, for example, International Center for Research on Women, 1996). Women's vulnerability to HIV infection is enhanced for several reasons including their economic dependence on men, lack of access to education, poverty, sexual exploitation, coercion and rape, as well as by the fact that women are more likely than men to sell sex in order to survive (Aggleton and Rivers, 1999). Surrounding and to some extent legitimating these inequalities are ideologies of masculinity and femininity which make it seem 'natural' that men should have the upper hand when it comes to economic decision making, opportunities for advancement, expressing their sexual desires and satisfying their sexual needs.

While traditional gender roles render women less able to control the nature and timing of sexual activity, men are more able to determine how, when and with whom sex takes place. Despite this, dominant ideologies of masculinity (which emphasise male sexual pleasure, value the display of sexual prowess and encourage men to have multiple sexual partners) place men and their partners at heightened risk of HIV and AIDS. While women may be prepared to take measures to protect themselves from HIV infection, and while men may have some investment in protecting themselves, their partners and families, women's desire for safer sex not infrequently runs 'into a wall of un-cooperation from men' (Meursing and Sibindi, 1995). In this paper we will examine what it is about gender relations and dynamics, and dominant versions of masculinity in particular, that enhances risk and hinders men from protecting themselves and their partners from HIV infection. Men's relationships with women and with other men will be examined, and the importance of involving men more fully in programmes for improved sexual health and greater gender equality will be stressed.

Before doing this, however, it is important to stress that the dangers of working from a stereotypical description of 'men' and their desires, motivations and interests. There is enormous variability between individuals, not only between societies but within them. While some men display little interest in protecting themselves and their partners against disease, perhaps believing themselves to be 'invincible', others behave with the utmost responsibility and consideration for others. Moreover, while perhaps the majority of men prefer to have sex with women, a not insubstantial number of men have sex with members of both sexes or with other men alone. Whether the individuals concerned understand this behaviour to be 'heterosexual', 'bisexual' or 'homosexual' varies considerably, since in perhaps the majority of countries these terms only enjoy currency in the scientific, medical and epidemiological literatures, and rarely form part of the local vernaculars within which sex is talked about and understood. Analysing the position of 'men' in relation to the HIV epidemic is therefore a complex and difficult task, and one which cannot adequately be accomplished within the confines of a review such as this. We are aware, therefore, that we will probably raise as many questions as offer answers, yet hope that our analysis of men and masculinities in relation to the epidemic offers some useful leads for future programme development.

II. GENDER AND THE HIV EPIDEMIC

Gender and Development

Policies and programmes to promote greater equality between men and women are considered to be crucial to HIV prevention (see, for example, Rao Gupta, 1995; d'Cruz-Grote, 1996). Despite increasing recognition of the importance of more equal gender relations, many programmes continue to work solely with women in an attempt to help empower them in sexual relationships. As Wood and Jewkes (1997) point out, however, this focus is often based on an erroneous set of assumptions about women's ability to control and sustain their sexual health. Only rarely do women have direct control over the contexts, occasions and forms within which sex takes place, and there is a substantial literature to indicate how difficult it is for women to persuade men to use condoms and/or reduce the number of partners in circumstances where the latter are unwilling to do so (see, for example, World Health Organisation, 1994). In the field of international development, and while several programmes have recently altered their terminology from 'women in development' to 'gender and development', perhaps the majority of initiatives to challenge and transform prevailing gender relations still focus on women alone. Relatively few start from a recognition of the needs of both women and men (White, 1997).

This over-emphasis on reaching women who are particularly vulnerable to HIV infection has led to a neglect of two key factors: men's participation in programmes and programming and broader social circumstances (Mbizvo and Bassett, 1996). For example, while numerous HIV prevention programmes and interventions have focused on women sex workers, considerably less attention has been given to their male clients. Even today, men are rarely written about in the literature on development and, where accounts do exist, men usually appear as background figures and are rarely centre stage within the analysis. By way of contrast, in much of the literature on gender and development, women are written about as hard working and caring with a strong orientation towards community, while men are constructed as individualists who put their own desires first. The overtones here of 'colonial stereotypes about 'lazy natives' are uncomfortable, to say the least' (White, 1997: 16). Indeed, men in developing countries have almost uniformly been characterised as inconsiderate, unreliable, predisposed to coercion, rape and violence, as well as being relatively unable to control or change their behaviour. As such, they offer a counterpart for images of women as disempowered and with little control over their social and sexual lives. A more complex situation does in fact pertain (Sweetman, 1997).

While some commentators have called for increased male participation in work towards greater gender equality and improved sexual and reproductive health, concern has also been expressed about shifting the focus, and resources, from women to men. Berer (1996: 7), for example, has suggested written that '...just as women's specific problems are finally getting some attention on the world stage ... it seems that focusing only on women is no longer acceptable.'. For Berer and other writers, the key may lie in involving men in ways which are more supportive of both women and women's concerns: 'If empowering women is to remain the end point ... policies for change that involve men must also be grounded in a woman-centered and gender-sensitive perspective, not just taking men's perspectives or needs into account.' (ibid: 9).

Gender Inequalities and Masculinity

Gender differences, and the inequalities associated with them, can be explained in a variety of ways. However, while it is widely accepted that gender roles are not 'natural' but are culturally produced (Hearn, 1987), there is no consensus as to what causes them to emerge in the first place, or what leads them to change over time. Still less have the links between gender roles and broader sexual inequalities been fully explained. This poses major problems for any effort to explain the 'position' of men in relation to the sex and sexual matters, or the ways in which masculinities 'as sets of ideologies governing thoughts, actions and behaviours ' are constituted and reproduced over time. Yet some understanding of these phenomena is important if we are to develop programmes to engender greater equality within and between the sexes, to reduce HIV related risks, and to promote sexual and reproductive health more generally.

Connell has recently argued that research has failed to produce a 'coherent science of masculinity' (Connell, 1995: 67). In his view, masculinity is not a static and unchanging social norm, rather '[it]...is simultaneously a place in gender relations, the practices through which men and women engage ... and the effects of these practices' (Connell, 1995: 71). Multiple masculinities influenced by class and race as well as gender clearly exist, and it is important to examine not only gender relations between men and women, but also gender relations between men in making sense of gender inequalities and their effects.

Notions of 'hegemonic masculinity' help explain why certain versions of masculinity become the most successful and powerful in particular environments. Men who do not meet the 'standards' set by hegemonic masculinities, which in themselves can and do change over time, are viewed as unsuccessful and powerless, since within a society one or more forms of masculinity is likely to be 'culturally exalted'. Although not all men conform to the dominant versions of masculinity that circulate at any one moment in time, those who do not often find themselves discriminated against.

Despite this, all men probably share in what Connell (1995: 82) has called the patriarchal dividend through which men gain honour, prestige, the right to command, and material advantage over women.

Challenging dominant ideologies of masculinity, and their consequences for women and men's lives, is not easy. Like hegemonic ideologies of all kinds, dominant beliefs about what 'real' men are like (and by extension what women and children are like) seek to incorporate all alternative images, accounts and explanations within their sphere of influence. Thus, hegemonic masculinities legitimize not only unequal roles and relationships between women and men, but also between men. They encourage us to see men who do not live up to the ideals of hegemonic masculinity as effeminate, weak, subservient or immature. And they seek to deny men an active role in changing prevailing gender relations and inequalities for the better (Cornwall, 1997).

Masculinities and Sexual Health

Prevailing gender relations have a serious impact on men's sexual health and the sexual health of partners and families, in addition to shaping the broader oppression of women. Estimates suggest that between 60-80 per cent of women currently infected with HIV in sub-Saharan Africa have had only one sexual partner (Adler et al, 1996). Research in many parts of the world suggests that men have a greater lifetime number of sexual partners and that there are clear double standards regarding the behaviour of men and women (de Bruyn et al, 1995; International Center for Research on Women, 1996). For example, while in many cultures women are expected to preserve their virginity until marriage, young men are encouraged to gain sexual experience (International Center For Research on Women, 1996). Indeed, having had many sexual relationships may make a man popular and important in the eyes of his peers (Abdool Karim and Morar, 1995). Male sexuality is often thought of by both men and women as unrestrained and unrestrainable, and in some parts of the world having an STD is considered a badge of honour which confirms manhood (de Bruyn et al, 1995). So, while lack of knowledge and sexual inexperience remain highly valued for young women, men may be stigmatised if they cannot demonstrate having had a wide sexual experience.

Research also suggests that sexual decision-making is usually controlled by men. In many cultures, coercive sex and sexual violence are not unusual (see, for example, de Bruyn et al, 1995; Wood and Jewkes, 1997). According to both boys and girls recently interviewed in Recife in Brazil, for example, girls and women are often coerced into sex and some young women may obey their boyfriends' wishes because they believe that girls are 'meant' to be compliant and subservient (Vasconceles, Garcia and Mendonca, 1997). While there may be differences in prevailing definitions of masculinity, greater freedom, power and control characterise male sexuality across a wide spectrum of different cultures. Furthermore, where women are most economically dependent on men, their ability to make decisions about sex may be most constrained. This reinforces the importance of economic development for enhanced levels of gender equality (Rao Gupta, Weiss and Mane, 1996).

In order to avoid the problems which come from failing to conform to dominant gender stereotypes, women risk the damage associated with conformity (Overall, 1993). Men on the other hand may find that by conforming to stereotypical versions of masculinity, they place themselves and their partners at heightened risk. These contradictions need to be exposed so as to identify the dividend that accrues to both women and men when existing gender roles are transformed or cease to be obeyed. By working to show how many men do not meet idealised forms of masculinity, discussion about how some men are marginalised can begin to take place. As Cornwall (1997: 12) has recently put it, 'If gender is to be everybody's issue, then we need to find constructive ways of working with men as well as with women to build confidence to do things differently.' The intimacy, complexity and entrenched character of prevailing gender relations and ideologies mean, however, that work of this kind will need to be sustained over time (White, 1997). While women may be the initiators of this kind of dialogue, their task will be 'impossible unless a dynamic is generated amongst men to question their personal practice' (ibid: 15-16). A first step in analysing men and masculinities, therefore, may lie in examining men's 'private stories', and how these accounts and experiences support or contradict the ideologies promulgated by more hegemonic masculinities (White, 1997).

Long and Ankrah (1996) have recently argued that sexual responsibility among men is central to the health of both men and women (ibid: 392). In their eyes, funding priority should be given to programmes and activities which aim to reach both men and women, rather than women alone. Community mobilisation and other techniques may be used to help increase awareness among men of how HIV/AIDS can affect the lives of their daughters, wives, mothers, kin and friends. For Long and Ankrah, women's empowerment cannot be achieved by women alone, but requires the support of men for its successful realisation (Long and Ankrah, 1996: 395).

Gender and Other Inequalities

Cornwall (1997: 9) has recently written that in much development work, gender analysis is used to guide planners by 'delineat[ing] distinctions between men-in-general and women-in-general'. Little is usually said about the intersection of gender with 'other differences such as age, status and wealth' (ibid: 9). In reality, gender relations and ideologies interact with other social inequalities, including those based on class, sexuality, age, religion and race.

White (1997) has recently described how some men in Bangladesh are exploited by other men because of their ethnicity, and a clear interaction between gender, ethnicity and class as determinants of sexual risk taking has also been shown among mine workers in South Africa (Campbell, 1997). Here, as in other countries, lack of employment opportunities close to home encourages men to migrate. Working in highly dangerous conditions, and removed from the usual sources of familial and social support, life in cramped conditions is both stressful and lonely. Drinking and paying for sex too readily become normative, heightening the HIV-related risks faced by the men and their partners.

In contrast, women's interests are often understood as relatively little influenced by social class, and 'gender sensitive' development programmes which aim to make women less poor are often conducted in isolation from work of other kinds. For some writers. 'Gender [has become] the justice issue, women the minority (... [and]) social development (...) at least in some agencies (...) very largely commandeered by 'gender specialists'' (White, 1997: 21). A broadening and deepening of our understanding of power and inequalities seems called for if we are to better understand the sometimes complex vulnerabilities linked to class, gender and ethnicity which structure women and men's lives. While men clearly benefit from gender inequality (i.e. through their greater access to schooling, economic advantage and power), we might profitably focus on masculinity and its effects by examining the institutions, cultures and practices that sustain both gender inequality and other forms of domination, such as those attributable to class, religion and race (White, 1997). As Cornwall (1997: 11) has put it, it is important to remember that 'not all men (...) have power; and not all of those who have power are men'.

Developing a more sophisticated understanding of gender inequalities and their determinants requires an examination of sexual divisions and ideologies beyond those that operate to structure men's relationships with women. The importance of men's relationships with one another has already been mentioned in relation to the way in which men who do not conform to dominant ideologies come to be seen as unmanly and effeminate. These social perceptions not infrequently link to the homophobia and heterosexism that can be witnessed in almost every society. They also fuel the existence of homosexual relationships and roles modelled strongly on heterosexual lines, for example, the activo/passivo relationships characteristic of men who have sex with men across much of central and southern America and north Africa, and the emergence of strongly gendered 'types' of male sex work that emerge in these same contexts (Aggleton, 1996, 1998).

Sex between men remains highly stigmatised in many societies, and men who have sex with other men (and who are open about this) not infrequently experience marginalisation, stigmatisation and severe social sanctions (McKenna, 1996). In perhaps the majority of countries, homosexual masculinities lie at the bottom of the gender hierarchy among men, and overt expressions of 'gayness', for example, are often equated with femininity (Connell, 1995). While it is less useful to talk about specifically gay identities outside the West and its spheres of socio-sexual influence, men who have sex with men and who do not subscribe to dominant versions of masculinity are clearly discriminated against in the majority of societies worldwide.

Interestingly, in some cultural contexts it is not sex between men per se which generates disapproval, but rather the behaviour of those men who show attributes which are traditionally associated with women. It is important, therefore, to examine sexual identities from a culturally sensitive local standpoint rather than through Western frameworks and understandings. Khan (1997) for example, has recently written about sex between men in India and Bangladesh, both countries in which social identity is much influenced by familial relations. Here, men who have sex with other men may not be penalised so long as their activities remain hidden. In this kind of context, hegemonic masculinity seems threatened less by sexual preference and habit than by the refusal to enter into contractual and reproductive relations with women. Similar findings have been reported from research conducted in Islamic societies including Pakistan (Murray and Roscoe, 1997)

Generational issues are also important determinants of sexual inequalities and discrimination. Young people often have less access to information and services than older people, have less economic power and are at heightened risk of sexual exploitation (Aggleton and Rivers, 1999). Recent research in Tanzania (Seel, 1996), in Zimbabwe (Runganga and Aggleton, 1998) and many other countries suggests that young men may attempt to redress inter-generational inequalities through sexual activity with multiple partners, which is seen by them as symbolising adulthood and enhanced status.

Overall, analyses of gender, sexuality and inequality need to take account of the manner in which factors such as age, class, ethnicity and culture interact to determine the form that gender and sexual divisions take. It should be clear from what has been said so far that the most successful programmes and interventions are likely to be those which move beyond a narrow focus on women's concerns and needs (while recognising these as important) to look at the ways in which contemporary masculinities and constructed and reproduced in particular societies at a given moment in time. By understanding more about the relationship between hegemonic masculinities and more subordinate forms, we may be better placed to challenge the former and their divisive effects (both for women and for men), so facilitating the transformation of social relations within and between the sexes.

III. WORKING WITH MEN

A number of researchers and practitioners have recognised the importance of involving men in work designed to prevent HIV infection, as well as to address the broader inequalities which pose a threat to sexual health (Hadden, 1997; Wood and Jewkes, 1997). One of the most important 'gaps' in work for improved sexual health, however, is the absence of clear information about men's attitudes toward sex and sexuality. We need to know much more about men's perspectives and interests if we are to engage them productively in work for the prevention of HIV infection and improved sexual health.

For example, many women report that men refuse condom use, and may even become violent when safer sex is requested. Women in Thailand, for example, report that condoms might be seen as appropriate for 'casual sex', but not within the context of a longer term relationship (Cash and Anasuchatkul, 1993). Other women have reported that suggesting a partner use a condom may be tantamount to accusing him of infidelity (Heise and Elias, 1995; Ankrah and Attika, 1997). Interestingly though, we know very little about men's own perceptions on the same issues and concerns.

Orubuloye et al (1997) have argued that there has been a consistent failure to enquire into men's belief systems in relation to sex and sexuality. Where researchers have enquired into men's beliefs, findings have sometimes confounded commonly held views about male attitudes with the opinions of respondents themselves. For example, recent research conducted among South African men, suggests that the timing of requests for condom use is important in mediating likely responses. Against an overall background of reticence towards condom use, men reported that if they were asked to use condoms prior to sexual arousal, they were more likely to use them. However, they also acknowledged that if asked to use a condom when they were highly sexually aroused, they might become coercive and violent (Hadden, 1997).

Similarly, research has provided new insight into the meanings of anal sex when it takes place between men and women. In much of the development literature, heterosexual anal sex is commonly assumed to be a method of preserving virginity and preventing pregnancy. However, recent studies suggest that for some Brazilian men at least, anal sex may also be symbolic of increased power and control over women. For men interviewed, anal sex was seen as a 'conquest' to be equated with 'taking' a woman's virginity for a second time (Goldstein, 1994). Learning more about what sex means to men in different contexts is therefore an important prerequisite for the design of more effective programmes and interventions (Hadden, 1997).

Because women have less control over sexual communication, a substantial number of programmes have concentrated on work to empower girls and women. But, failures in helping women to change sexual behaviour and bringing about more equal gender roles demonstrate that boys and men too must be involved (Mbizvo and Bassett, 1996; Barnett, 1997). As Rao Gupta, Weiss and Mane (1996) have suggested, it is essential '... that interventions to strengthen women's sexual negotiation skills be conducted concurrently with educational programs designed for boys and men. Such programs must go beyond teaching condom skills by promoting men's participation as equal partners in safer sex planning,' (ibid: 345).

Reaching men in the manner advocated remains something of a challenge, however, because it remains unclear what messages will appeal to men and what are the key factors motivating safer sexual practices (Robinson, 1991). While only a small number of programmes have been designed to involve men, even fewer have attempted to systematically evaluate and report on the impact and effects of the work undertaken. Our review of the available evidence is therefore limited, and the programmes, projects and activities examined often describe work undertaken with relatively small groups of men. We will begin by reviewing work designed to increase condom use among men. Subsequently, we will look at programmes and projects that have tried to work with men considered to be at high risk of HIV infection, including truckers, migrant workers, clients of sex workers and STD patients. Next, some workplace based programmes will be described. Finally, some specific initiatives and activities addressing issues of relevance to men who have sex with men will be discussed.

Condom Use

Much of the HIV prevention work so far undertaken with men has been designed to increase condom use. Consistent condom use, one of the few effective strategies available to prevent HIV transmission, seems however to be problematic for men, and in consequence for women (Hulton and Falkingham, 1996). In Senegal, as in a number of other countries, it has been reported that men may suspect that a woman is a sex worker or has other lovers if she requests condom use (Niang, Benga and Camara, 1997). Some men in this same context reported believing that condoms could make men impotent (ibid). A programme aimed at both men and women was designed to increase safer sex and condom use in Senegal using traditional women's associations. The programme proved relatively successful with women, especially in terms of increased levels of knowledge, but the impact on men was much less pronounced. This was not perhaps surprising given that women were the main channel of communication in the programme. The authors conclude that more research is necessary in order to understand how to effectively reach men (Niang, Benga and Camara, 1997).

Hulton and Falkingham (1996) have collated survey data collected in the early 1990s in ten countries including Pakistan, Egypt, Niger, Ghana and Kenya. Data from over 69,000 women and 18,130 men was available. Reported lifetime use of condoms by men was significantly higher than that of women. Hulton and Falkingham (1996) suggest that large differences in ever-use of condoms may be because of past use by males with sexual partners before marriage and in extra-marital relationships. In Zimbabwe, for example, of those men having sex in the prior four weeks with a spouse, 12 per cent reported having used a condom, while for those men who had sex with a non-spousal partner the figure was 60 per cent (ibid).

Other research findings support the finding that condoms are not consistently popular with men, especially with their wives (Meursing and Sibindi, 1995). Amamoo (1996), for example, writes that men may interpret requests for condom use as betrayal or attempts to deprive them of their rights in sexual decision-making within the relationship. Women in a diverse range of countries have reported being unable to act upon what they know about HIV and AIDS for fear of implying through condom use that a partner is not loved or trusted. Such requests disturb the intimacy which is central to many relationships and can result in violence, abandonment or rape (Ankrah and Attika, 1997).

Wilton's work (1997) offers some interesting insights into the reasons why condom use may be so unpopular among men. She suggests that masculinity itself is threatened by condom use. There are several reasons for this: first, if condom use is requested by a woman this allows women to define the terms of sexual engagement; second, condom use may involve men having to deprioritise their own sexual pleasure; third, for men to demonstrate a degree of control over sexual behaviour may be feminising since male sexuality is most usually understood as uncontrollable; and finally, risk-taking in itself is considered to be typically masculine. Wilton (1997) points out that non-penetrative sex is rarely an option in heterosexual relationships since vaginal sex tends to be understood as adult sex, and other forms of sexual pleasure may be seen as a kind of backsliding into adolescence. Her work is important since it stresses the importance of working with men as well as women to de-construct stereotypical gender roles if HIV transmission is to be reduced.

Because of male resistance to condom use and the difficulties which women may have in negotiating the use of condoms, some authors have suggested that female controlled protection is central to HIV prevention (see, for example, Heise and Elias, 1995). The female condom, although more expensive and less widely available, provides women with an extended choice of protection, and recent research suggests that male resistance to the female condom may be less than to the male condom (Aggleton, Rivers and Scott, 1998).

Hawkins (1996) has observed that current programmes to meet women's immediate sexual and reproductive health needs, including those designed to promote condom use, may inadvertently reinforce and preserve inequalities in gender and sexuality. Marketing strategies which attempt to encourage condoms to be used often use stereotypical and 'macho' images which may further entrench gender stereotypes and inequalities. Gupta (1995) recommends that efforts be made to support the marketing of new and more egalitarian images of masculinity and femininity. Messages which promote images of predatory males and passive females may have brought about some short-term increases in condom sales to men, but have done so only at the expense of reinforcing damaging gender stereotypes.

Men at Special Risk

Programmes aimed at groups of men considered to be at special risk of HIV infection have taken place in some developing countries. These groups include truck drivers, who are highly mobile and may spend long periods of time away from home, migrant workers who are separated from their families and communities, the clients of sex workers, and STD patients.

Truck drivers in a range of countries work under conditions which directly promote risk behaviour through mobility, the time they spent away from families, and the use of sex workers (Robinson, 1991; Madrigal, 1991). Evaluations of the effectiveness of HIV prevention programmes with truckers in Africa and Asia offer important insights into what can be achieved through this kind of work. Raman (1992), for example, has recently described work recently undertaken by the AIDS Research Foundation of India (ARFI) with sex-workers' clients, including truck-drivers, in Madras. As part of this programme, condoms were distributed at transit-stops and educational cassettes played. Peer opinion leaders were also recruited to tell port and dock workers stories about men who practise safer sex, and posters were put on display in barbers and wine shops. Short street plays were performed and free STD services provided. Informal monitoring of the project's activities suggested that sales of condoms increased (Raman, 1992).

Elsewhere in India the Bhoruka AIDS Prevention (BAC) Project has concentrated its work on the trucking routes between Calcutta and Kathmandu which have been identified as important sites of high-risk sexual behaviour (Amin, 1996). Among other initiatives, the BAC Project has established a range of services including STD testing, condom distribution and counselling at Raxaul, a major intersection for trucks travelling between India and Nepal. Data collected at regular intervals during the first year of the programme showed that the number of men seeking counselling and HIV testing there increased from 136 to 2,431, and the number of condoms distributed upon request rose from 630 to 26,290 (Amin, 1996).

A linked programme of interventions collectively called Avancemos ('Let's Move Ahead') organised by NGOs in the Dominican Republic has disseminated messages to the regular partners of sex workers, their clients and to other men involved in the sex industry (AIDS Control and Prevention Project, 1997). A comic book was developed and regular workshops held to encourage the proprietors of brothels and other commercial sex establishments to support prevention efforts. These sessions approached the epidemic from the perspective of the owners and managers, and their desire to attract more customers. Impressed with the quality of services, a number of enterprises have recently began to pay small fees to support the work of Avancemos because they want the activities to continue. Project workers have concluded that working with a wide range of men involved in the commercial sex industry is essential for effective prevention efforts (AIDS Control and Prevention Project, 1997).

A number of authors have recognised that economic and social migration influences and facilitates the spread of HIV. Campbell (1997) has noted that high levels of HIV infection are characteristic of a range of unstable and economic disadvantaged social settings in Southern Africa and has looked at the ways in which dangerous and risky work may influence the attitudes of men towards sex. Forty two migrant miners were interviewed in Johannesburg. Although all the interviewees had been exposed to HIV-related information and had good levels of knowledge about AIDS, knowledge did not translate into safer sexual behaviour. Living and working conditions in the mines are highly dangerous and stressful, and drinking and sex appeared to be two of the few diversions easily available to the men. What is more, facing risks at work daily may mean that men are less inclined to worry about the long-term risks of HIV infection. For example, interviewees commented that 'the risk of HIV/AIDS appeared minimal compared to the risks of death underground, and suggested that this was the reason why many mine workers did not bother with condoms' (ibid: 277). Interviewees were relatively fatalistic about the chance of accidents at work and felt powerless to change their circumstances. Campbell (1997: 277) writes that ' ... this sense of powerlessness is an important feature of the contextual backdrop [and] is an important determinant of health-related behaviour'. Importantly, masculinity emerged as a leading narrative in interviewees accounts of their work, sexuality and health. The miners took pride in working in dangerous conditions and responsibility for providing for their distant families. Understandings of masculinity were also reinforced by male peers with whom much time was spent socialising outside the immediate work context. According to interviewees, men were defined by their bravery, fearlessness and desire for sex. Somewhat paradoxically, therefore, 'the very sense of masculinity that assists men in their day-to-day survival also serves to heighten their exposure to the risks of HIV infection' (ibid: 278). Campbell argues that her research supports the claim that an important way of reducing levels of HIV infection could be to alter the social and material conditions which facilitate and reinforce risky sexual practices.

In northeastern Thailand, interviews and focus groups have recently been undertaken with 936 men, including migrant workers involved in the harvesting of sugar-cane (Maticka-Tyndale et al, 1997). The focus of this research was on men's relationship with sex workers. High levels of knowledge about HIV infection were reported, and of those who had paid for sexual services in the past year, 76 per cent reported condom use. However, the researchers also found that the context in which sex was sold had an important bearing on whether or not condoms were used (Maticka-Tyndale et al, 1997). In part, this may be because of the insistence on condom use by bar managers, but the researchers found that men also perceived sex with women who were not working in environments where sex was traditionally sold as less risky. When men paid for sex at festivals, in markets or on the sugar plantation itself, the sex was more hurried and condoms were less often used. Moreover, some of the women selling sex in these latter circumstances were not defined as sex workers by the men, but simply as available or 'loose women' (Maticka-Tyndale et al, 1997). Similarly, the closer to home sex took place, the less risky it was perceived to be. The authors conclude that future AIDS campaigns must take account of the variety of contexts in which sex may be bought and sold, and should avoid addressing only stereotypical scenarios (for example, bars) in health promotion.

Several studies to promote safer sex with male STD patients have taken place. Hadden (1997), for example, has recently reported on findings from an experimental study aimed at both men and women STD patients in KwaZulu Natal South Africa. In the experimental group, information about HIV was supplemented by four 90-minute sessions of a skills-building group intervention designed to help men and women protect themselves from HIV infection. The control group received only information about HIV/AIDS. Single-sex sessions were held initially. Men were shown how to use a male condom, but also showed interest in the female condom, Unlike the women, men were found to be more uncomfortable using anatomically correct words to discuss genitalia and sex. Men reported in subsequent sessions that women had the right to refuse unprotected sex, but pointed out that waiting until the point of arousal before saying 'no' was likely to elicit an angry response. It was the timing of requests for condom use rather than refusal which most angered men. In common with women, men agreed that both partners should be tested for HIV if a baby was planned.

A combined session was also held. This session engendered much excitement and attendance was generally higher than for other sessions. Three role-plays were included: about sexual communication, condom negotiation and violent reactions by men towards women. While women expressed their pain and anger and described how they felt when physically abused, men recalled their experience of hostility and violence towards partners. Role play involving gender role reversal was undertaken and the men took this seriously. Subsequent to the intervention a small, but statistically significant increase in condom use was reported by members of the experimental group. The researchers concluded that more work is needed with men to explore and challenge social norms that support multiple sexual partnerships. Further research is also needed to explore different ways of engaging men in discussions about sex and sexual rights and responsibilities (Hadden, 1997).

Workplace Programmes

Some programmes have attempted to reach men through workplace activities. The Organisation of Tanzanian Trades Unions (OTTU) began its work initially with women, but in 1992 the programme was expanded to reach men as well (Hadden, 1997). During 1993, 83 peer educators conducted more than 300 educational sessions in 27 workplaces, and afterwards 75% of workers participating in these sessions reported using condoms with 'casual' or non-regular partners. The support of managers was found to be important, and informally some reported that they have noticed behavioural change among workers away for business who previously might have sought out sex workers, but now do not (Hadden, 1997).

Cash et al (1997) have built on earlier intervention research conducted with women factory workers in Northern Thailand to develop a new programme which includes male factory workers. Formative research conducted through focus group interviews established that although men commonly fear HIV infection or getting a girl pregnant, they are reluctant to take on responsibility for prevention. A variety of educational materials was designed including a comic book story about a male factory worker who is HIV positive. Peer leaders were also trained, but both young men and women expressed fears about talking about sex, STDs and HIV. The success of both single and mixed-sex sessions was found to depend on the skills of the peer leader. Among the participants were twelve couples who reported major improvements in communication about HIV and sex.

The Zimbabwe AIDS Prevention Project (ZAPP-UZ) has been following a cohort of 2,500 male factory workers living in Harare Research to determine their sexual attitudes and practices (Ray et al, 1996). The researchers found that most men preferred 'dry' sex which requires their partners to use herbs and other preparations to ensure that the vagina remains largely unlubricated during intercourse. This practice is particularly risky in terms of HIV infection, since it may predispose women to breaks in the epithelial barrier of the genital tract (Ray et al, 1996). Men reported that they generally obtained information about sex from elders and peers as they grow up, and from peers in adulthood. Communication between sexual partners was infrequent and of poor quality, especially between spouses. A number of men were subsequently recruited for a peer education programme. These men were keen to learn more about sexual practices and whether or not their was a 'scientific' basis for their beliefs about sex. Although not fully evaluated, the programme designers report that the peer educators who were trained developed new attitudes towards sexual practices. Most importantly, the authors note the importance of opportunities for men to have frank and non-judgemental discussions with an emphasis on increased male responsibility for improved reproductive and sexual health (Ray et al, 1996)

Men who have Sex with Men

Although its existence may be officially denied, sex between men occurs in every society. It is usually stigmatised and discriminated against, and the acts concerned are probably only occasionally understood as homosexual, bisexual or 'gay' (Giffin, 1998). One of the earliest insights from social research on HIV and AIDS was the understanding that sexual behaviour often fails to conform to subjective sexual identity, although the implications of this mismatch between behaviours and identities is still relatively under-explored (Aggleton, Khan and Parker, 1999). The situation is made more complicated by the existence of erotic desires and the situational specificity of much sex between men.

As Parker (1991) has argued, erotic desires are of special importance when it comes to understanding non-normative sexual behaviours and practices in some cultures, especially when these involve some kind of transgression. Context is important in making seem reasonable and acceptable patterns of behaviour that might in other circumstances be unthinkable and impossible to enact. The sexual segregation and social hierarchy characteristic of penal establishments, military environments and some religious settings, for example, may actually facilitate sex between men (Aggleton, Khan and Parker, 1999). While often not acknowledged and rarely discussed, the sex which occurs in such settings can be important in determining social prestige, gender identity within and beyond that setting, and sexual health status both positively from the point of view of sexual fulfilment and negatively from the point of view of HIV-related risks (see, for example, Schifter, 1997).

It has been widely documented how across much of Mexico, Central and South America, notions of 'activity' and 'passivity' in sex remain central to the gender constructions and identities of men who have sex with other men (see, for example, Carrier, 1995; Moya and Garcia, 1996; Schifter and Madrigal, 1992; Parker 1991; Cáceres, 1996), although there is evidence that such 'traditional' patterns of homo- and bisexuality have recently been overlain by the advent of international gay culture (Roberts, 1995). Similar role defined patterns of behaviour have been identified in Morocco and some other countries in North Africa (see, for example, Boushaba, Imane, Himmich and Tawil, 1998). In these contexts, a masculine identity remains largely unthreatened so long as the penetrative role in take in anal and oral sex, or so long as the appearance of this being the case can be sustained.

Across Asia, homosexual behaviour has been widely reported in both Islamic (Schmitt and Sofer, 1992; Murray and Roscoe, 1997) and non-Islamic societies. Even in contexts where male homosexuality has long been denied, there may be well developed homosexual networks and subcultures such as those recently documented between male sex workers and their clients in Pakistan (Mujtaba, 1997; B. Khan, 1997). Despite the existence of these networks and behaviours, in perhaps the majority of Asian countries marriage remains compulsory for men, and masculinity derives from age, economic productivity, familial relationships, getting married and having children (Khan, 1997). As a result, the social invisibility of homosexuality and bisexuality is reinforced.

In Africa too, research now suggests the existence of homosexual behaviour and relations in countries as diverse as the Sudan (Ahmed and Kheir, 1992), Kenya (Standing and Kisseka, 1989; Shepherd, 1987), Botswana (Botswana Ministry of Health, 1987) and South Africa (Gevisser and Cameron, 1995).

We are clearly dealing with universal patterns of behaviour but it must be recognised that the meanings given to sex vary widely between societies and even across sub-groups within a society. Given the clandestine nature of many of the acts concerned, and their illegality in many countries, it is perhaps not surprising that the existence of such behaviours continues to be denied. The challenge for efforts to promote the sexual and reproductive health of men who have sex with other men lies in acknowledging the existence of homosexual relations between men, the inequalities they sometimes reproduce, and the difficulties created by stigmatisation and discrimination for efforts to reach such men with HIV prevention messages as well as other kinds of work.

Partly because of its invisibility, little is known about the extent to which sex between men facilitates HIV transmission in developing countries (McKenna, 1996). A recent analysis of responses from over two hundred organisations surveyed, however, suggests that sex between men has an important role to play in HIV transmission in many contexts, with consequences for infections which may subsequently be transmitted heterosexually, or from mother to child (McKenna, 1996).

There have been relatively few well documented interventions to promote the sexual and reproductive health of men who have sex with men in Central and Southern America, Africa and Asia, but a recent review highlights some of the work which has already taken place (Aggleton, Khan and Parker, 1999). Successful projects include community based outreach work with male sex workers in Casablanca and Marrakesh (Himmich, 1992; Boushaba, Imane, Himmich and Tawil, 1998); community work with networks of men who have sex with men in Mumbai, Chennai and Cochin in India (Aggleton, Khan and Parker,1998); work with both male sex workers (Tan, 1998) and other homosexually active men in the Philippines (Nierras et al, 1992; Fleras, 1993; Tan, 1995); educational, outreach and condom promotion activities among men who have sex with men in Vietnam (Nguyen Friendship, 1997); work in the saunas and bath-houses of Mexico City (McKenna, 1996); the provision of telephone help lines and holistic workshops for men who have sex with men in Costa Rica (Madrigal and Schifter, 1992); HIV/AIDS education workshops for gay and homosexually active men in Lima (Cáceres et al., 1989); and a range of community based HIV prevention activities with gay and other homosexually active men in Rio de Janeiro (Parker and Terto Jr., 1997).

The challenge for much of this work now lies in scaling up what has so far taken place, and extending the remit of existing projects (where feasible) so as to engage with the structural factors which promote discrimination, stigmatization and repression towards men who are not exclusively heterosexual, and so as to forge links between these projects and activities and other work to promote greater gender and sexual equality. The barriers to the success of such work should not, however, be underestimated. If it has been difficult to undertake work which challenges the 'patriarchal dividend' inherent in men's existing relationships with women, it may be doubly difficult to do so in circumstances where programming and prevention efforts may be seen as supporting homosexuality and forms of behaviour which have been denied, discriminated against and stigmatised.

Gender and Care for People Living with HIV/AIDS

Stigmatisation and blame have characterised the HIV/AIDS epidemic since the start (Lawless et al, 1996). The manner in which people are blamed has consequences for the provision and receipt of care. Recent research clearly demonstrates how men are much less likely to be blamed for HIV infection than women (de Bruyn et al, 1995; Aggleton and Warwick, 1998), and are more likely to be afforded care by their partners, families and communities. Lawless et al., (1996) have suggested that women living with HIV have attracted guilt and blame partly because they are perceived to have 'failed' in their roles as nurturers and carers. It is widely believed in many societies that only certain 'kinds' of women (most usually sex workers and women who have many partners) become infected. Research also suggests that women are more likely to internalise the blame attached to them (Lawless et al, 1996).

In addition to the increased stigmatisation of women who have become infected with HIV, the burden of care for people with HIV/AIDS also falls on women. Aggleton and Warwick (1998) have recently analysed findings from a series of UNAIDS supported studies of household and community responses to HIV/AIDS in the Dominican Republic, Mexico, India, Tanzania and Thailand. In common with a number of other studies, the research highlights how women are central to the provision of care for people with HIV/AIDS in all countries. Even among gay community respondents interviewed in Mexico who received additional support from social networks of friends and lovers, men with HIV often returned home to receive care from their mothers and other female relatives when very ill.

In all sites, attitudes and responses towards people with HIV/AIDS, including the provision of care, were strongly influenced by gender and gender norms (Aggleton and Warwick, 1998). In the Dominican Republic and Mexico, however, levels and quality of care was also influenced by perceptions of innocence and guilt. But these responses too showed a gender imbalance. Men, even when considered more 'blameworthy', were nonetheless comforted and taken care of. When women needed HIV-related care, however, they generally did not expect or receive the same level of care and support as men. Women who were sick often returned to their parents for care since they were unlikely to receive this from their husbands.

Even in cases where men did offer some support and care, accounts from these recent multi-site studies suggest that gender norms influence the nature and amount of care that men offer. In the Kyela district of Tanzania, for example, there were indications that 'male heads of households would wish to do more when their partners fall ill but were curtailed by cultural definitions of maleness and the roles defined which determine masculinity' (Aggleton and Warwick, 1998: 34). There was evidence in each of the five sites in which the multi-site study was conducted to suggest the existence of clear double standards governing the care given to men and women. Whereas men with HIV disease were little questioned about how they became infected and were generally cared for (by women), women with HIV-related conditions were frequently castigated and blamed and received lower levels of support. Women also had to balance responsibility for provision of care with the need to support the family financially. In spite of such problems, however, and each of the study sites in this multi-site investigation, they continued to provide care as mothers, wives, neighbours and volunteers (Aggleton and Warwick, 1998).

Differences in attitudes towards women and men with HIV, and patterns in the provision of care for people with AIDS, are related to dominant versions of masculinity and femininity. As discussed earlier, in a wide variety of cultural contexts expectations of female and male sexuality differ. A clear dual standard exists with regard to the sexual behaviour of women and men in most cultures, so that while men are often encouraged to have large numbers of sexual partners, women are expected to remain faithful to one sexual partner. In addition, male sexuality is widely perceived as unrestrained and unrestrainable. Women who become infected with STDs or HIV are often viewed as blameworthy. Blame is less likely to be ascribed to men however, who are assumed to have little control over their sexual urges. In addition, women traditionally provide care for family members who are sick, while a care-giving role is not consistent with dominant or hegemonic versions of masculinity.

IV. LESSONS LEARNED

As stated earlier, most gender sensitive programmes aiming to reduce levels of HIV-related risk behaviour have until recently focused their work on women. Programmes and interventions involving with men are still few and far between and, where they do exist, formal evaluation has yet to take place. More research, and importantly the systematic evaluation of the impact and outcomes of HIV-related work with men, needs to take place.

While keeping in mind the limitations of the published literature in this field, it is possible to identify some issues which may be helpful in developing future programmes of HIV-related work with men. These include recognising that:

• masculinities are socially constructed and exert pressure on men to behave in particular ways. However, dominant or hegemonic masculinities are not constant and do change over time. The development of alternative versions of masculinity can, therefore, be promoted.
• Gender inequalities intersect with other social inequalities such as those organised around class, age, race, religion and sexuality. Programme design needs to be sensitive to these patterns of interaction if gender and sexual inequalities are to be properly addressed.
• Given the intersection between gender and other inequalities, the elimination of poverty for men and women through programmes of social development and other means, is crucial to the prevention of HIV transmission.
• Diversity among men has implications for efforts to meet their sexual and reproductive health needs. Stereotypical images of men (for example, as similar, as inherently 'heterosexual', or as causative of all gender inequalities) are unlikely to be helpful in programme design and do not afford men the opportunity to maximise their own sexual health and that of their partners.
• In needs assessment and in programme design it is important to allow men to express their needs, while keeping in mind that work will also need to be undertaken done to ensure that all work is sensitive to the gender inequalities which serve to silence and disadvantage women.
• Men need carefully structured opportunities to consider how dominant ideologies of masculinity, and the role relationships they reinforce, may disadvantage them as well as their lovers, partners, families and children
• The concern which many men express about the health and welfare of their children may provide a useful way of gaining attention in relation to HIV-related work.
• Condom promotion as part of broader efforts to promote sexual and reproductive health needs to be gender sensitive so as to ensure that short-term increases in sales and use do not inadvertently reinforce gender stereotypes and inequalities.
• Increasing the acceptability and use of condoms among men is crucial, since condoms provide one of the few commonly available and inexpensive means of prevention for HIV and other STIs.
• In circumstances where the male condom may be unpopular, recent research suggests that the female condom can offer a useful means of alternative protection against HIV and other STIs.
• While work with truck drivers, migrants, the clients of sex workers, and men who have sex with men is very important, in cultural environments where many men routinely have multiple partners, work with men who do not fall into any of the above especially vulnerable groups is also crucial.
• Although single sex group work is important, there is evidence to suggest that in some contexts working with men alongside women may be helpful for both men and women.
• In developing countries, as elsewhere, it is important to design, implement and scale up programmes to promote sexual and reproductive health among men who have sex with men. Such programmes need to recognise the range of contexts within which such behaviour takes place, the cultural meanings attached to sex between men, and the variations in sexual identity that exist among men who have sex with other men.
• Since women are more likely to be blamed for HIV infection, work to counter the stigmatisation and discrimination associated with such blame needs to take place with men as well as with women.
• It is crucial for HIV-related health promotion to encourage men to take a more active role in the care of people with HIV-related illnesses.
• Poor working conditions and risks of work-related injury and mortality may facilitate sexual risk taking and HIV transmission. These issues need to be addressed through programmes to promote improved working conditions as part of a broader commitment towards social development.
• Research suggests that ideologies of masculinity, and the practices associated with these, are constructed and reinforced within predominantly male groups. Working with men in groups to promote more equitable gender roles may therefore be helpful.
• There is an on-going need to evaluate the impact and outcomes of programmes to promote sexual and reproductive health among men and their partners, and to disseminate findings from such work.

Future Work

Given what has been said, it is clearly important to involve men more fully than hitherto in work linked to the prevention of HIV infection. However, responsibility for HIV infection is not just a matter for the individual. Broader social policies and actions are needed to inhibit the growth of the epidemic. Unequal gender relations, as well as other inequalities, facilitate HIV transmission and the growth of the epidemic. In the long term, greater social and gender equality must be the aim of those seeking to enhance sexual and reproductive health among both women and men in developing as well as developed countries. However, given the entrenched nature of existing gender roles, beliefs and expectations, it is unlikely that enormous advances can be made in the short term. In the face of the global pandemic of HIV and AIDS, it is important to think realistically about what is attainable, and on what timescale. While it may be possible, for example, to promote increased condom use among men, given dominant versions of masculinity, it may be less realistic to encourage all men to remain faithful to a primary partner.

Reference was made earlier to the 'patriarchal dividend' which all men share. Given this dividend, it seems unlikely that men will be prepared to relinquish the power and privilege which patriarchy affords them, in the short term at least. Although greater equality between men and women must be the ultimate goal, this may take a long time to achieve. In the interim, however, it is important that risks to the sexual health of people in developing countries are reduced. An incremental approach, which seeks to reduce the immediate risks of HIV infection within a gender sensitive framework may therefore be most helpful. In the first instance, ensuring greater male participation in programmes to promote sexual and reproductive health is crucial.

Where possible, it is important to tackle gender inequalities and the socio-economic and other inequalities with which they intersect, at a structural level as well. Policy-makers need to be encouraged to develop structural and environmental interventions to help women and men make changes in their behaviour which might help them to protect their sexual health. These interventions might include changes in law to protect women against male violence and to de-criminalise sex between men. Both of these actions would render more visible the circumstances in which HIV-related risks may be particularly acute, and could lead to the development of more effective programmes for prevention. The provision of education for girls, and increased opportunities for participation in the labour market, will help to reduce both widespread poverty and the economic dependence on men which renders women vulnerable to sexual exploitation. Labour laws which enforce improved working conditions and reduce injury and death in the workplace for men may also help men change their orientation towards certain forms of risky sexual behaviour.

Much existing information about men's behaviour and beliefs comes not from men themselves, but from women. We still know little about what men think, and what they might respond successfully to, in terms of HIV prevention. Although in the case of domestic violence, sexual coercion and rape it may be difficult to generate accurate accounts from men themselves, it is important to engage men in discussion to gain an enhanced understanding of their perceptions, attitudes and practices. Research in the following areas seems most pertinent:

• Accurate and up to date information is needed on men's beliefs and practices in relation to gender, sex, sexuality and sexual health. This is especially true in those contexts where the risk of HIV infection is high.
• Systematic enquiry into sex between men is important. Since Western typologies are rarely relevant in developing countries, it is important to develop an understanding of the meanings attached to male to male sex in local terms.
• Since risk-taking appears to be an important part of dominant ideologies of masculinity in a number of societies, it is important to develop a better understanding of risk-taking behaviour among men, especially among those who work in dangerous and/or isolated environments.
• Since condoms still provide the most useful means of preventing HIV transmission, formative research is needed to identify non-stereotypical images and messages which might appeal to men and encourage increased condom use.

V. CONCLUSIONS

This paper has suggested that involving men more fully in HIV prevention work is essential if rates of HIV transmission are to be reduced. This is likely to require a considerable scaling up of existing efforts and, in the absence of new resources, some re-orientation of existing gender sensitive programmes and interventions, many of which currently work with women alone. While such a move may not be universally popular, it seems necessary if we are to ensure that men take on greater responsibility for their own sexual and reproductive health, and that of their partners and families. Too often in the past it has been assumed that by working with women we will be able to redress the profound social inequalities of gender and sexuality that exist in the world today. While some progress has been made in this respect, too often such work has simply increased the burden of responsibilities already shouldered by women in the developing world. In relation to HIV/AIDS it may also have inadvertently reinforced the idea that women are the prime 'vectors of HIV' (de Bruyn et al, 1995).

If, in future years, men are not properly involved in work to challenge the complex inequalities of gender and sexuality which facilitate and reinforce the transmission of HIV, women are likely to have to take on responsibility for changing men's ideologies and practices as well as their own. This seems profoundly unfair and, in the face of patriarchy and the structures which reinforce it, is unlikely to yield the desired results. Work is needed to transform existing agendas of prevention, health promotion and development so as to make them more sensitive to gender and sexuality as principles structuring the lives of both women and men, and influencing HIV-related vulnerabilities in ways which could not easily be imagined only a decade or so ago.

REFERENCES

Abdool, Karim Q. and Morar, N. (1995) 'Determinants of a Woman's Ability to Adopt HIV Protective Behaviour in Natal/Kwazulu, South Africa: A Community Based Approach.' Women and AIDS Program Research Report Series. Washington DC: International Center for Research on Women.

Adler, M., Forster, S., Richens, J., and Slavin, H. (1996) 'Sexual Health and Care: Sexually Transmitted Infection: Guidelines for Prevention and Treatment', Health and Population Occasional Paper. London: Overseas Development Administration.

Aggleton, P. (Ed.) (1996) Bisexualities and AIDS - International perspectives. London: Taylor and Francis.

Aggleton, P. (Ed.) (1998) Men Who Sell Sex - International perspectives on male prostitution and AIDS. London: UCL Press.

Aggleton, P., Khan, S. & Parker, R. (1999) 'Interventions for Men who have Sex with Men in Developing Countries' in L. Gibney, R. DiClemente and S. Vermund (ed.) Preventing HIV Infection in Developing Countries, New York: Plenum Publications.

Aggleton, P. and Rivers, K. (1999) 'Behavioural Interventions for Adolescents' in L. GIBNEY, R. DiClemente and S. Vermund (ed.) Preventing HIV Infection in Developing Countries, New York: Plenum Publications.

Aggleton, P., Rivers, K. and Scott, S. (1998) Multi-Site Studies of Gender Relations, Sexual Negotiation, and the Female Condom in Developing Countries, Geneva: UNAIDS.

Aggleton, P. and Warwick, I. (1998) A Comparative Analysis of Findings from Multi-Site Studies of Household and Community Responses to HIV and AIDS in Developing Countries. Geneva: UNAIDS.

Ahmed, S. & Kheir, A. (1992) 'Sudanese Sexual Behaviour, Socio-cultural Norms and the Transmission of HIV', in T. Dyson (ed.) Sexual Behaviour and Networking: Anthroplogical and sociocultural studies on the transmission of HIV, Liège, Belgium: Editions Derouaux-Ordina.

Ankrah, M. and Attika, S. (1997) Adopting the Female Condom in Kenya and Brazil: perspectives of women and men. Arlington, VA: Family Health International/AIDSCAP.

AIDSCAP (1997) Meeting the Challenge of the HIV/AIDS Epidemic in the Dominican Republic, Arlington, VA: Family Health International/AIDSCAP.

Amamoo, N.A. (1996) Working with Men for Change: Gender and HIV Prevention in the African Community in the UK, London: Akina Mama wa Afrika.

Amin, M. (1996) 'The India-Nepal Partnership', AIDCaptions, 3, 2, 10-13.

Ankrah, E.M. and Attika, S.A. (1997) Dialogue: Expanding the Response to HIV/AIDS, A Resource Guide, Arlington: VA: Family Health International/AIDSCAP.

BManushi, 507. ang, R and Bang A. (1992) 'Why Women Hide The: Rural Women's Viewpoints on Reproductive Tract Infections'.

Barnett, B. (1997) 'Gender Norms Affect Adolescents', Network, 17, 3, 10-13.

Bassett, M. and Sherman, J. (1994) 'Female Sexual Behaviour and the Risk of HIV Infection: An Ethnographic Study in Harare, Zimbabwe.' Women and AIDS Program Research Report Series. Washington DC: International Center for Research on Women.

Berer, M. (1996) 'Men', Reproductive Health Matters, 7, 7-10.

Botswana Ministry of Health (1987) AIDS Update 1. Gabarone, Ministry of Health.

Boushaba, A., Imane, L., Himmich, H. & Tawil, O. (1998) 'Male Sex Work in Morocco' in P. Aggleton (ed.) Men Who Sell Sex - International perspectives on male prostitution and AIDS. London: UCL Press.

Cáceres, C. (1996) 'Male Bisexuality in Peru and the Prevention of AIDS' in P. Aggleton (ed.) Bisexualities and AIDS - International Perspectives. London: Taylor and Francis.

Campbell, C. (1997) 'Migrancy, Masculine Identities and AIDS: The Psychosocial Context of HIV-Transmission on the South African Gold Mines', Social Science and Medicine, 45, 2, 273-281.

Carrier, J. (1995) De los otros: Intimacy and homosexuality among Mexican men. New York: Columbia University Press.

Carovano, K. (1992) 'More Mothers and Whores: Redefining the AIDS Prevention Needs of Women', International Journal of Health Service, 21, 131-142.

Cash, K. and Anasuchatkul, B. (1993) 'Experimental Education Interventions for AIDS Prevention Among Northern Thai Single Migratory Female Factory Workers'. Women and AIDS Research Program Report-in-Brief. Washington, DC: International Center for Research on Women.

Cash, K. et al. (1997) 'AIDS Prevention Through Peer Education for Northern Thai Single Migratory Factory Workers', Women and AIDS Research Program Report-in-Brief. Washington, DC: International Center for Research on Women.

Connell, R.W. (1987) Gender and Power, Oxford: Polity Press.

Connell, R.W. (1995) Masculinities, Oxford: Polity Press.

Cornwall, A. (1997) 'Men, Masculinities and Gender in Development' in C. Sweetman (ed.) Men and Masculinity, Oxford: Oxfam.

de Bruyn, M. et al. (1995) Facing the Challenge of HIV/AIDS/STDs: A Gender-based Response, Amsterdam: Royal Tropical Institute, Southern Africa AIDS Information Dissemination Service and WHO Global Programme on AIDS.

de Moya, E.A. and Garcia, R. (1996). 'AIDS and Bisexuality in the Dominican Republic' in P. Aggleton (ed), Bisexualities and AIDS: International Perspectives, London: Taylor & Francis.

Family Health International (1998) AIDS Control and Prevention Project Final Report, Volume 1, Arlington, VA: Family Health International/AIDSCAP.

Fleras, J. (1993) Safer Sex Promotion Inside a Gay Bath-house. Poster presentation at the IXth International Conference on AIDS, Berlin.

George, A. (1997) Sexual Behaviour and Sexual Negotiation Among Poor Women and Men in Mumbai, Baroda: SAHAJ Society for Health Alternatives.

George, A. and Jaswal, S. (1994) 'Understanding Sexuality: Ethnographic Study of Poor Women in Bombay'. Women and AIDS Program Research Report Series. Washington DC: International Center for Research on Women.

Gevisser, M. & Cameron, E. (1995) Defiant Desire: Gay and lesbian lives in South Africa. New York: Routledge.

Giffin, K. (1998) 'Beyond Empowerment: Heterosexualities and the Prevention of AIDS', Social Science and Medicine, 46, 2, 151-156.

Goldstein, D. (1994) 'The Culture, Class and Gender Politics of a Modern Disease: Women and AIDS in Brazil.' Women and AIDS Program Research Report Series. Washington DC: International Center for Research on Women.

Hadden, B. (1997) 'An HIV/AIDS Prevention Intervention with Female and Male STD patients in a Peri-Urban Settlement in KwaZulu Natal, South Africa'. Women and AIDS Research Program Report-in-Brief. Washington D.C: International Research Center for Women.

Himmich, H. (1992) Interview. Global AIDS News, 2.

Hawkins, K. (1996) Resources in Social Development Practice Vol. 1: Participatory Design and Monitoring of Reproductive Health Projects, Swansea: Centre for Development Studies University of Wales Swansea.

Hearn, J. (1987) The Gender of Oppression, Brighton: Wheatsheaf.

Heise L.L. and Elias, C. (1995) 'Transforming AIDS Prevention to meet Women's Needs: A Focus on Developing Countries'. Social Science and Medicine, 40, 7, 931-943.

Henry, K. (1995) 'Workplace Prevention Programs Promote Behaviour Change in Tanzania', AIDSCaptions, 11, 1, 18-21.

Hulton, L. and Falkingham, J. (1996) 'Male Contraceptive Knowledge and Practice: What do we know'' Reproductive Health Matters, 7, 90-99.

International Center for Research on Women (1996) Vulnerability and Opportunity: Adolescents and HIV/AIDS in the Developing World, Washington DC: International Center for Research on Women.

Khan, B. (1997) 'Not-So-Gay Life in Pakistan in the 1980s and 1990s' in S. Murray and W. Roscoe (eds.) Islamic Homosexualities. New York: New York University Press.

Khan, S. (1997) Sex, Secrecy and Shamefulness, London: The Naz Foundation.

Khan, S. (1998) 'There are no Heterosexuals in India...There are Married Men and Men who will get Married', Ki Pukaar, 20, 3-5.

Lawless, S., Kippax, S., and Crawford, J. (1996)' Dirty, Diseased and Undeserving: The Positioning of HIV Positive Women', Social Science and Medicine, 43, 9, 1371-1377.

Long, L.D. and Ankrah, E.M. (1996) Women's Experiences with HIV/AIDS: An International Perspective, New York: Columbia University Press.

Madrigal, J. (1998) Al Vaivén de un cabezal - un estudio sobre los traileros en América Central y su relación con el Sida. San José, Costa Rica: Editorial ILPES.

Madrigal, J. & Schifter, J. (1992) Hombres que aman hombres. San José, Costa Rica: Ediciones Ilep-Sida.

Maticka-Tyndale, E. et al. (1997) 'Contexts and Patterns of Men's Commercial Sexual Partnerships in Northeastern Thailand: Implications for AIDS Prevention', Social Science and Medicine, 44, 2, 199-213.

Mbizvo, M.T. and Bassett M.T (1996) 'Reproductive Health and AIDS Prevention in sub-Saharan Africa: the Case for Increased Male Participation'. Health Policy and Planning, 11, 1, 84-92.

McKenna, N (1996) On the Margins: Men who Have sex with Men and HIV in the Developing World, London: Panos Institute.

Meursing, K. and Sibindi, F. (1995) 'Condoms, Family Planning and Living with HIV in Zimbabwe', Reproductive Health Matters, 5, 56-67.

Mujittaba, H. (1997) 'The Other Side of Midnight: Pakistani male prostitutes' in S. Murray and W. Roscoe (eds.) Islamic Homosexualities. New York: New York University Press.

Murray, S & Roscoe, W. (eds.) Islamic Homosexualities. New York: New York University Press.

Nguyen Friendship (1997) http://www2.bestcom/~utopia/tipsviet.htm#ngu

Niang, C., Benga, H., Camara, A. D. et al (1997) 'An Evaluation of HIV Prevention Interventions Utilizing Traditional Women's associations in Senegal'. Women and AIDS Research Program Report-in-Brief. Washington D.C: International Research Center for Women.

Nierras, T., Austero, B., Santos, J. & de Real, A, (1992) HIV/AIDS the the Filipino Gay Community. Poster presentation at the VIIIth International Conference on AIDS, Amsterdam.

Orubuloye, I.O., Caldwell, J.C., and Caldwell, P. (1997) 'Perceived Male Sexual Needs and Male Sexual Behaviour in Southwest Nigeria', Social Science and Medicine 44, 8, 1195-1207.

Overall, C. (1993) 'Heterosexuality and Feminist Theory' in D. Shogan, A Reader in Feminist Ethics, Toronto: Canadian Scholars Press.

Parker, R. (1991) Bodies, Pleasures and Passions: Sexual culture in contemporary Brazil, Boston, MA.: Beacon Press.

Parker, R. & Terto Jr. V. (eds.) (1997) Entre Homens: AIDS e homossexualidade no Brasil. Rio de Janeiro: ABIA.

Piot, P. & Aggleton, P. (1998) 'AIDS Inequality and Prevention', IAS Newsletter, 10, 8-10.

Rao Gupta, G. (1995) Integrating a Gender Perspective in UNAIDS Policies and Programmes: A Proposed Strategy, Geneva: UNAIDS.

Rao Gupta, G., Weiss, E., and Mane, P. 'Talking About Sex: A Prerequisite for AIDS Prevention' in Long, L.D. and Ankrah, E.M. (1996) Women's Experiences with HIV/AIDS: An International Perspective, New York: Columbia University Press.

Ray, S. Gumbo, N., and Mbizvo, M. (1996) 'Local Voices: What some Harare men say About Preparation for Sex', Reproductive Health Matters, 7, 34-44.

Roberts, M. (1995) 'Emergence of Gay Identity and Gay Social Movements in Developing Countries: The AIDS crisis as a catalyst', Alternatives, 20, 243-264.

Robinson, E. (1991) 'Reaching Men', Network, 12, 1, 9-16.

Runganga, A. & Aggleton, P. (1998) 'Migration, the Family and the Transformation of a Sexual Culture', Sexualities, 1, 1, 63-81.

Seel, P. (1996) 'AIDS as a Paradox of Manhood and Development in Kilimanjaro, Tanzania', Social Science and Medicine, 43, 8, 1169-1178.

Schifter, J. (1997) Amor de Machos, San José, Costa Rica: ILPES.

Schmidt, A. & Sofer, J. (1992) Sexuality andf Eroticism among Males in Moslem Societies. New York: Harrington Press.

Shepherd, G. (1987) 'Rank, Gender and Homosexuality: Mombasa as a key to understanding sexual options', in P. Caplan (ed.) The Cultural Construction of Sexuality. London: Tavistock.

Standing, H. & Kisseka, M.N. (1989) Sexual Behaviour in Sub-Saharan Africa: A review and annotated bibliography. London: Overseas Development Administration.

Sweat, M.D. and Denison, J. (1995) 'HIV Incidence in Developing Countries with Structural and Environmental Interventions', AIDS, 9, 1299-1306.

Sweetman, C. (ed.) (1997) Men and Masculinity, Oxford: Oxfam.

Tan, M. (1995) 'From Bakla to Gay: Shifting identities and sexual behaviours in the Philippines', in R. Parker and J. Gagnon (eds.) Conceiving Sexuality: Approaches to sex research in a postmodern world. New York: Routledge.

Tan, M. (1998) 'Walking the tightrope: Sexual risk and male sex work in the Philippines' in P. Aggleton (ed.) Men Who Sell Sex - International perspectives on male prostitution and AIDS. London: UCL Press.

UNAIDS. (1997) New UNAIDS World AIDS Day Report Warns that HIV Epidemic is Far Worse than Previously Thought, Press Release, 26 November.

Vasconcelos, A, Garcia, V, Mendonca, M.C. (1997) 'Sexuality and AIDS Prevention among Adolescents in Recife, Brazil'. Women and AIDS Program Research Report Series. Washington DC: International Center for Research on Women.

White, S.C. (1997) 'Men, Masculinities and the Politics of Development' in C. Sweetman (ed.) Men and Masculinity, Oxford: Oxfam.

Wilton, T. (1997) EnGendering AIDS: Deconstructing Sex, Text and Epidemic, London: Sage Publications.

Wood, K. and Jewkes, R. (1997) 'Violence, Rape, and Sexual Coercion: Everyday Love in a South African Township' in C. Sweetman (ed.) Men and Masculinity, Oxford: Oxfam.

World Health Organisation (1994) Women and AIDS: Agenda for action. Geneva: Author.

Acknowledgements

We would like to thank Peter Gordon, Shivananda Khan, Purnima Mane, Neil McKenna and Richard Parker for their advice and assistance (both direct and indirect) while producing this paper. The views expressed are those of the authors alone.

About the Authors

Kim Rivers is a senior research officer within the Thomas Coram Research Unit at the Institute of Education, University of London. She has extensive international experience working in the fields of sexual and reproductive health.

Peter Aggleton is Director of the Thomas Coram Research Unit. He is editor of the Social Aspects of AIDS series of books published by UCL Press/Taylor and Francis, and of the journal 'Culture, Health and Sexuality. He has worked internationally in HIV-related promotion for nearly fifteen years. All correspondence to Thomas Coram Research Unit, 27-28 Woburn Square, London, WC1H 0AA, UK.

Recommended citation: Rivers, K., Aggleton, P. (1999). Men and the HIV Epidemic, Gender and the HIV Epidemic. New York: UNDP HIV and Development Programme.