Scholarly article on topic 'Intersectionality: A Challenge for Feminist HIV/AIDS Research?'

Intersectionality: A Challenge for Feminist HIV/AIDS Research? Academic research paper on "Sociology"

0
0
Share paper
Academic journal
European Journal of Women's Studies
OECD Field of science
Keywords
{""}

Academic research paper on topic "Intersectionality: A Challenge for Feminist HIV/AIDS Research?"

A Challenge for Feminist HIV/AIDS

Anna Bredström

linköping university

abstract The aim of this article is to engage critically with feminist HIV/AIDS research from an 'intersectional' perspective. Focusing in particular on the work of Tamsin Wilton (1997) and Janet Holland et al. (1998), the article examines how 'race', ethnicity and class are theorized and conceptualized in this literature. Through a scrutiny of their empirical analyses, the article points to the pitfalls of a descriptive approach to 'differences' and problematizes Wilton's and Holland et al.'s theoretical focus on gender and sexuality. The benefit of including a critical perspective on 'race' and ethnicity and other axes of domination is illustrated further using some empirical examples from the Swedish HIV/AIDS policy context. The article concludes by arguing that an intersectional perspective poses a challenge to feminist HIV/AIDS research that needs to be addressed in order to produce an effective sexual health policy.

key words cultural racism ♦ HIV/AIDS ♦ intersectionality ♦ safer sex ♦ Sweden

Feminists have shown that gender constitutes a crucial factor for understanding the HIV/AIDS phenomenon, regardless of which aspect of the pandemic we try to elucidate. As a result, feminists have questioned sexist and patriarchal influences on biomedical discourses and underlined the importance of acknowledging gender-specific needs in health policy (see, for example, Doyal et al., 1994). Feminist research has shown, moreover, that the gendered aspects of HIV/AIDS do not only apply to the medical condition and the lives of the infected, but that gender is also a decisive factor in the negotiation of safer sex. Accordingly, feminist research has contributed to the deconstruction of dominant HIV/AIDS discourses and thus challenged ideas such as that unsafe sexual practices result only from a lack of knowledge concerning routes of transmission or means of protection.

In my study of Swedish HIV/AIDS policy, feminist HIV/AIDS research constitutes an important source of inspiration. However, as I try to

European Journal of Women's Studies Copyright © 2006 SAGE Publications (London, Thousand Oaks and New Delhi), 1350-5068 Vol. 13(3): 229-243; http://ejw.sagepub.com DOI: 10.1177/1350506806065754

approach the field from an 'intersectional' (Crenshaw, 1994) perspective in which gender is not treated separately from sexuality, class or 'race'/ethnicity, I have also found parts of this feminist literature problematic. In this article, I aim to highlight and problematize some of these shortcomings, focusing on the work of Tamsin Wilton (1997) and Janet Holland et al. (1998). I start by presenting the main arguments of Wilton and Holland et al. From there, I go on to demonstrate how 'race', ethnicity and class are theorized and conceptualized in their work. Drawing from Black and postcolonial feminism, I subsequently discuss the pitfalls of their approaches. As a way of demonstrating the importance of a feminist intersectional approach to the HIV/AIDS problematic, I conclude this article by discussing some empirical examples from the Swedish HIV/AIDS policy context.

RISK-TAKING IN SEXUAL ENCOUNTERS: A GENDERED ENDEAVOUR

In EnGendering AIDS: Deconstructing Sex, Text and Epidemic (1997), Tamsin Wilton scrutinizes dominant AIDS and safer sex discourses. As a point of departure, she emphasizes that 'representational practices both reflect and construct social and psychological "reality"' and thus have 'profound consequences for the impact of the epidemic' (Wilton, 1997: 6; emphasis in original). Sharing the same epistemological starting point, sociologists Janet Holland, Sue Sharpe, Caroline Ramazanoglu and Rachel Thomson set out to make sense of young people's discussions about sexual experiences (e.g. Holland et al., 1998; Holland, 1993). Despite differences in terms of empirical material, Wilton and Holland et al. present very similar analyses. In order to understand AIDS discourses as well as unsafe sex practices, they urge us to theorize notions of risk and safety as intertwined with gender and heterosexuality. Holland et al. illuminate how risk taking is embedded in heterosexual relations. Heterosexual femininity, they argue, equals an 'unsafe sexual identity' since 'to be conventionally feminine is to appear sexually unknowing, to aspire to a relationship, to let sex "happen", to trust love, and to make men happy' (Holland et al., 1998: 6). A heterosexual woman is, accordingly, in a position where asking her partner to use a condom is either risking her femininity by being too active in the sexual encounter, or implying that he is to be considered a potential disease carrier, which, in turn, could jeopardize their relationship. If women are in a precarious situation when negotiating sexual safety, Holland et al. point out that men are equally hindered by norms of masculinity. In line with what has been identified by Wendy Hollway (1996) as the discourse of a hydraulic 'male sex drive', they argue that the hegemonic construction of heterosexual masculinity rests upon a notion

of male sexuality as an instant, not to be interrupted (biological) force. The general reluctance of heterosexual men to use condoms is thus understood as an effect of condoms 'disturbing' what is supposedly uninterruptible: 'Condoms symbolize breaking the flow and destroying his passion' (Holland et al., 1998: 37). What men then are 'defending', by resisting condom use, is their identities as 'real men'.

Wilton shows how this discourse is further reproduced in safer sex materials and other HIV/AIDS discourses. For instance, since heterosexual masculinity is more or less constructed as impossible to change, women are often the primary targets of safer sex campaigns that address heterosexuals. Women are thus handed the responsibility for both male pleasure and the general public's well-being. Wilton argues that this is, from a feminist perspective, not exclusively bad news, since it has informed women of how the HIV virus is transmitted, and pointed out means of protection. However, as women are simultaneously disem-powered by the construction of femininity as passive and disembodied, the possibilities for women to fulfil such responsibilities are rather limited.

As seen, both Wilton and Holland et al. are not content only with analysing gender. Rather, they stress that it is equally important to theorize heterosexuality. By using the term 'heteropolarity' in her work on AIDS discourses, Wilton (1997: 73) aims to 'highlight the chronic inseparability of "gender" from the erotic'. Similarly, Holland et al. focus not only on how women are the object of men's desire, but also on how this objectification is intertwined with, and made intelligible through, an understanding of heteronormativity. In a heteronormative discourse, they argue, heterosexuality is constructed as 'natural, oppositional and hierarchal' (Holland et al., 1998: 24). In fact, Holland et al. hold that heterosexuality could almost be interpreted 'as masculinity', since women's desires are not expressible within a heteronormative discourse. Likewise, Wilton claims that masculinity more or less coincides with heterosexuality:

. . . 'masculinity' means that-which-fucks-women (into submission), and sex between men is stigmatized precisely because of its escape from the discursive limits of 'authentic' masculinity. (Wilton, 1997: 32; emphasis in original)

Wilton demonstrates that there are different positions in AIDS discourses carved out for gay and straight men respectively, as well as for lesbian and heterosexual women. As AIDS for a long time was equated with gay men, gay men are highly visible and portrayed as being at risk as well as personifying risky behaviour. Wilton asserts that the ways in which heterosexual men have been directed to distance themselves from the virus stem from gay men having been much to the fore in AIDS discourse. She also argues that gay men's success in using condoms had

to do with the fact that they do not invest as much in their masculinity as heterosexual men (are forced to) do. Women, in contrast, have largely been invisible and lesbian women more or less absent from AIDS-related safer sex discourse.

'RACE', ETHNICITY AND CLASS IN THE WORK OF WILTON AND HOLLAND ET AL.

There are several reasons why I chose Wilton and Holland et al. for this discussion. First, they are prominent researchers who are frequently referred to in (feminist) scholarship on sexual health and their contributions to critical HIV/AIDS research are of great importance, not least for my own work. Thus my critique should not be read as a dismissal of their work; rather it forms part of a continuous, feminist dialogue. Second, Wilton and Holland et al. are suitable for my purpose since they explicitly address the feminist debate on how different social divisions intersect. Both Wilton and Holland et al. repeatedly emphasize that 'race', ethnicity and class - as well as age and able-bodiedness - structure and sustain inequalities between women and between men. Wilton (1997: 5) refers to the 'heterosexualisation of AIDS' as occurring through a process of racial-ization, whereby heterosexual transmission is exclusively associated with sub-Saharan Africa and she criticizes the way western epidemiology has framed AIDS in 'the so-called Third World - especially sub-Saharan Africa' - as being caused by 'pre-existing abnormalities, failures or pathologies among (black) indigenous populations'. Wilton (1997: 14) also argues that it is important to 'avoid constructing a simplistic model of patriarchy which ignores "race", class and other oppressions', since such a model 'merely reinscribes Western cultural imperialism' and 'results in a white Western feminism'. Similarly, Holland et al. (1998: 16) suggest that the depiction of women as having common experiences and political interests has been 'identified as a critical intellectual weakness' and that it is therefore necessary to address the issue of power relations between women.

These assertions are not, however, reflected in their overall work, where, as seen earlier, the main focus remains on gender and sexuality. This may appear to be an omission, given their explicit recognition of a range of different axes of domination. However, it soon turns out to be fully in accordance with how both Wilton and Holland et al. reason theoretically when they discuss these issues. Indeed, in their theoretical elaborations they give several reasons why they do not find it necessary to theorize 'race', ethnicity or class for their research purposes. Wilton, to start with, argues that it is possible to speak of gender and sexuality (the 'erotic' in her words) without addressing the issue of class or racialized

social relations, since she believes the body is separable from 'the wider social arena':

Class, age, 'race', nationality and dis/ability all impact profoundly on the social construction of AIDS, and on the ability of individuals to respond to the epidemic, to protect themselves from HIV infection or to get appropriate care if they become infected, on the policy agenda and, ultimately, on who dies and who survives. They also, of course, impact profoundly one upon another, and upon gender, the central concern of this book. Yet I believe that the social, cultural and theoretical co-dependency of gender (always already 'racialized', located in a context of nationality, age, class and dis/ability) and the erotic (always already similarly located and inflected) is more significant than the relations among age, 'race', dis/ability, class, nationality and either gender or the erotic. This is partly due to what I can only think of as the remorseless mapping of the sexual onto the always already gendered body, a mapping which, in tandem with Cartesian dualism, has structured Western thinking in the modern era. Thus a body which has a vagina and uterus is understood as for penetration (and by implication, impregnating) by a penis, while a body which has a penis is understood as for penetrating (and impregnating) a vagina/uterus. . . . The same immediate relation to the body does not exist for, say, 'race' or age, factors whose significance is located more clearly in the wider social arena, at a certain distance from matters of skin and flesh. (Wilton, 1997: 15; emphasis in original)

Accordingly, Wilton admits that 'race', for instance, is a constitutive factor ('always already racialized'). Yet she believes that there is a fundamental 'functionalism' of the body - 'penis is to vagina what plug is to socket' (Wilton, 1997: 15) - present in a gendered discourse that is separable from 'race' or class.

Albeit not referring explicitly to the body, Holland et al. come to a similar conclusion when they display their motives for not addressing the issue of race and ethnicity in their work. In keeping with Wilton, they start out by admitting that they are aware that there exist differences in terms of 'race' and ethnicity, but continue arguing that their focus is not on differences but on similarities:

Racism clearly positioned young people differently within British culture, economy and polity, making 'race/ethnicity' a significant social division in their lives, and complicating class divisions and gender relations. Racial stereotypes and racial/ethnic identities were entwined with differences of style, culture, and language, and with gendered and sexualised qualities. . . . In this book we have not focused specifically on these differences, but rather on what seemed similar across cultural, political and social structural divisions. (Holland et al., 1998: 17)

Holland et al. (1998: 18) also admit that they could identify class differences in their interviews. Yet, they claim that:

. . . being working class, or being middle class did not appear to determine how young people managed or experienced their sexual encounters. We are not saying that class differences were insignificant in the young people's lives, but we can make sense of our data in terms of distinctive masculinities and femininities across their class differences.

Moreover, they argue that a focus on masculinities - in plural - might fragment feminism:

Documenting diversity . . . does not enable us to see how socially diverse men access male power. Feminist theory needs to identify what male power is, how it works and what sustains and reproduces it. (Holland et al., 1998: 26)

Thus, in the end, Holland et al. come very close to Wilton's idea that 'race', ethnicity and class do not (really) affect gendered relations.

DIFFERENCES AS VARIATIONS

Accordingly, both Wilton and Holland et al. acknowledge the importance of taking 'differences' into account, yet they maintain a theoretical position by which it is possible to separate gender and sexuality from other social divisions. Before refuting their arguments theoretically, let me first discuss the way in which 'race'/ethnicity and class take shape in their subsequent empirical discussions and analyses.

Despite their explicit focus on gender and sexuality, both Wilton and Holland et al. occasionally bring up the issue of 'race'/ethnicity and class. Wilton mentions, for instance, that it is important to take ethnicity into account when distributing safer sex materials:

For black or minority ethnic women, simply accessing adequate and appropriate information about safer sex may be especially problematic either because such materials are not produced in their first language or appropriately distributed or because their access to any written materials is restricted by traditional patriarchal customs. (Wilton, 1997: 30-1)

In a similar vein, one of the reports from the research project carried out by Holland et al. deals specifically with the issue of racial and ethnic differences among young women. The report presents the impact of culture and religion on young girls' different experiences by statistically comparing the experiences of young women from different ethnic and racial backgrounds (Holland, 1993). In their overall work, Holland et al. also highlight class differences. When discussing how men become empowered by violence, for instance, they refer to 'varying stories of male culture of violence' and explain that 'in working-class community

cultures . . . defending oneself or protecting female relatives is a matter of preserving family honour and masculine identity' (Holland et al., 1998: 152). Moreover, in their book, both ethnic/racial and class-belongings are marked under each quote (which enables a rereading such as this).

There are, as I see it, several interrelated problems with this approach. First, it represents a distorted focus, zooming in foremost on practices that are made to stand out as 'different', and often as 'problematic'. For instance, the issues of culture and religion are only brought to bear on some women, whose experiences then are continuously 'measured' and compared in relation to what becomes an unspoken norm. As such, migrant/minority or black women are frequently represented as if they were expected to have less knowledge, be less experienced and in more need of official AIDS education due to more restricted family relations. By contrast, other ethnic or racial identifications (such as Britishness or whiteness) are not explicitly discussed, but remain invisible. 'Arranged marriage' and 'virginity before marriage' are, for instance, conceptualized as cultural or religious convictions, whereas other gendered and sexual relations are not seen as part of ethnic or cultural practice.1 Second, the concepts of culture and religion are taken for granted, as an easily summarized set of beliefs, attitudes and practices. Class differences are treated similarly, as if they are mostly about differences in style, behaviour or language. Differences are thus addressed in a descriptive rather than an analytic way. Compared to the detailed theoretical approach that both Holland et al. and Wilton apply in order to understand gender and sexuality, the concepts of 'race', ethnicity, culture, religion and class are treated in a 'pre-theoretical' fashion (see Baca Zinn et al., 1986: 297). In turn, this could also explain the contradiction discussed earlier, that differences are acknowledged as empirical realities, yet theoretically overlooked.

However unintended, such an approach has serious implications. If left unanalysed, mere descriptions could indeed serve to reproduce already existing racial/ethnic/classed social relations. The translation of ethnicity into unproblematized apprehensions of 'cultural differences' does, for instance, constitute a long-standing object of study within critically oriented theories on 'race' and ethnicity. Scholars engaged in this field assert that, although culture makes up an important realm where meanings are constructed, reproduced, deployed, challenged and subverted, notions of cultural differences have also become part of commonsense perceptions of ethnic and racial differences (Anthias and Yuval-Davis, 1992). Applied in this way, culture easily becomes a marker of deviance and is construed as static and inheritable. In this discourse, moreover, culture loses its reference to a site of contestation and instead becomes an indicator of 'otherness'. As much research has demonstrated, such an appropriation of culture always runs the risk of merely substituting for older ideas of racial biological differences, and thus risks forming

part of a new (cultural) racism (Balibar, 1991), despite the good intentions of the authors. Processes of racialization could thus as easily be based on ideas of immutable cultural differences as on earlier ideas of immutable racial biological differences. Taken together, both racisms conform to an 'ideology' that, implicitly or explicitly, depicts certain groups as different and inferior. And whereas some groups are stigmatized as 'other', hegemonic ethnic or racial identities simply pass as 'normal'. Following this line of reasoning, when describing cultural variations it is very important to take into account the part that notions of immutable cultural differences play in (neo-) racializing discourse. In a similar vein, untheorized depictions of (violent) 'class cultures' could have the effect of reproducing unequal class relations.

MUTUALLY CONSTRUCTED SYSTEMS

Aside from the matter of how differences are described, the work of Wilton and Holland et al. also points to the importance of taking 'race', ethnicity and class into account in their own right. Let me exemplify further. In Holland et al. (1998: 62) an 18-year-old woman of working-class background talks about her mother in an interview:

A: Well she got pregnant before she got married . . . she wanted to be a teacher but she ended up working in a factory.

Q: Does she want to protect you from the same thing happening?

A: Yeah, I know she does because she's always saying that.

Despite the explicit reference to the living condition of the mother, Holland et al. do not analyse the gendered discourse as a classed event. Instead they analyse the quote jointly with other mother-daughter stories of 'don't trust men' and 'sex is dirty' as an outcome of a general 'protective discourse' that supports the 'passive model of female sexuality'. Furthermore, when Holland (1993) discusses 'variations' among 'young women from different ethnic minority groups' she mentions one girl who talks about 'the pressure' to be a 'bad girl':

Well, going to that school, there were some black children, black girls in particular, who mucked about quite a lot, and, since a lot of black children - I mean there was only one or two of us that didn't join the bandwagon, and they felt that you were being 'white' if you did work, and if you - and if certain teachers liked you, you pretended to be white. And I used to get, to be really frightened to go to school at times. (Afro Caribbean woman, 19 years, quoted in Holland, 1993: 32)

In the text that precedes the quote, Holland states that peer pressure is

important when these young women perform (or resist performing) femininity characterized as a 'bad girl'. However, in the quote the young woman does not only talk about peer pressure; she also asserts that her life is affected by a racialized school situation, where the construction of white and black identities seems to permeate the relations between pupils. She talks about how her resistance to being a 'bad girl' is conceived of as betraying her racial identity ('trying to be white'). In spite of this, Holland refrains from discussing how the process of racialization affects possible identifications for young men and women.

It could of course be argued that it is only reasonable that Holland et al. only focus on gender, as that is indeed what they set out to do. However, such an argument would rest upon an understanding of 'race' and class as something that could be added, as an extra dimension, to an already established theory of gender and sexuality. I would argue that such an approach is not feasible. What these quotes illustrate is that the construction of gender is inseparable from other social divisions; 'race' and class hence form the young women's gender identities. As expressed clearly in the second quote, 'black femininity' cannot be reduced to a lesser version of white femininity since 'blackness', as in this case, is understood as an antithesis of whiteness.

In order to grasp the young women's stories, insights from Black and postcolonial feminism are helpful (e.g. hooks, 1995; Mohanty et al., 1991). These theoretical interventions have repeatedly shown that there is no such thing as a common experience of female subordination. Neither do patriarchal relations affect all women and men in the same way, and nor do they always leave all men superior to all women. In order to incorporate these insights in feminist theory, the concept of intersectionality is sometimes brought forward (Crenshaw, 1994; de los Reyes and Mulinari, 2005; Collins, 1998). Patricia Hill Collins emphasizes that intersectionality is not a question of adding one oppression on to the next, but rather, 'As opposed to examining gender, race, class and nation, as separate systems of oppression, intersectionality explores how these systems mutually construct one another' (Collins, 1998: 63; my emphasis). In her analysis of the concept of the family, Collins shows how the notion of family naturalizes not only a gendered, aged and heteronormative hierarchy, but also plays an important role in both nationalist and racist imaginaries and ideologies, and can normalize both racial hierarchies and class relations. The nation, for instance, is often communicated with reference to the family, and nationalist ideologies assign men and women different roles and responsibilities just as the traditional family ideal does.

A prerequisite for an intersectional analysis is thus a theorization of the relationship between different axes of domination. For feminist HIV/AIDS research this implies that an exclusive focus on gender and sexuality needs to be replaced by an intersectional approach. This brings

us back to the core of the theoretical arguments presented earlier. Albeit in different ways, both Wilton and Holland et al. position the rupture of heterosexual masculinity as the most urgent challenge for radical safer sex politics. In Holland et al., it is the establishment of hegemonic masculinity that empowers men and ultimately disempowers women. Wilton argues that, in spite of its strong focus on a (gay) male body, AIDS discourse works by feminizing all subjects associated with an alleged risk group (gay men, injecting street drug users, sex workers, people of African descent and people living with AIDS). Keeping a distance from the illness altogether becomes a prerequisite in order to retain masculinity. 'In this image', Wilton contends, 'woman/sex/disease/death have become almost inseparable'; and she claims that it is even inconceivable to picture 'the "easy boyfriend" causing women to become blind, mad, paralysed or dead' (1997: 62, emphasis in original).

Seen from an intersectional perspective, such claims will inevitably run into difficulties. What happens, for instance, in a situation where white women tourists become infected from the local young black men at their service? What if these white women end up transmitting the virus to their partners? Are they then held any more responsible than when male clients put women sex workers at risk? While it is true that minoritized masculinities sometimes have been feminized, it has equally often been the reverse. Throughout history, both minoritized and working-class masculinities have been assigned different 'macho' identities, or been adopted as a strategy of defence (O'Donnell and Sharpe, 2000). That is, rather than feminized, minoritized and working-class masculinities are often represented as excessively masculine. Accordingly, an intersectional approach forces us to acknowledge that masculinity is not only constructed against femininity/homosexuality, but also against other heterosexual masculinities (cf. Connell, 1995). It also impels us to enquire into which ethnicities or class positions get normalized through our exclusive focus on the intersection between gender and sexuality.

AN INTERSECTIONAL ANALYSIS AS A CHALLENGE FOR FEMINIST HIV/AIDS RESEARCH

By turning to current developments in HIV/AIDS policy, this final section highlights some ways in which feminist HIV/AIDS research would benefit from an intersectional approach. With the Campaign for World AIDS Day (1 December) 2004 launched under the heading 'Women, Girls, HIV and AIDS', the issue of gender - at least symbolically - could be said to have transcended its previous invisibility in the HIV/AIDS policy context. In fact, since gender mainstreaming became a global strategy following the Beijing Platform for Action in 1995, the concept of 'gender'

has become more commonplace, including in HIV/AIDS policy. This does not, of course, say anything about the content of such gendered strategies, or that gender mainstreaming necessarily will help achieve the UN's 'ultimate goal' of gender equality.2 On the contrary, in a context of HIV/AIDS prevention, the charged debate on abstinence vs condom use reveals that there is no consensus on how to target HIV/AIDS from a gender perspective.

Looking at Sweden in particular, however, the effects of gender mainstreaming are palpable. In general, Swedish HIV/AIDS policy proposes the need for correct information on how the virus transmits, everyone's right to have access to condoms, equal rights for sexual minorities and women's right to choose. Sweden has also explicitly dissociated itself from the current US administration's conservative programme, which the Swedish government claims to be joining forces with the Vatican and some Muslim countries as concerns sexual and reproductive rights in general and HIV/AIDS policy in particular (Utrikesdepartementet, 2005). Lately, moreover, an increasing focus on men's responsibility and on masculinity has become visible (see, for example, Herrstrom et al., 2004).

If Swedish HIV/AIDS policy at first glance corresponds fairly well with the feminist imperatives traceable to the thesis of disempowered femininity/empowered masculinity, an intersectional approach requires that we pose further questions. By enquiring into how 'race' and ethnicity are understood in this gendered discourse, a quite different picture soon emerges. For one, migrants and refugees from countries that have suffered the most from the effects of the pandemic are often stigmatized and singled out as the most potent disease carriers in Sweden (see, for example, SOU, 2004). It is not, therefore, surprising to find frequent direct and indirect references to ethnic and racialized boundaries in the discussion of gender and masculinity. This is not exclusively the case though - there is indeed an ongoing discussion about gendered power relations that does problematize sexual relations between 'Swedish' men and women. Nevertheless, whenever ethnic boundaries come to the fore 'immigrant' men are inevitably represented as 'more' patriarchal and misogynist than 'Swedish' men. For example, an editorial in a periodical on HIV/AIDS-related issues distributed by the Swedish National Institute of Public Health, starts by discussing how 'old patriarchal traditions' affect views of female and male sexuality in Sweden, prohibiting women from fully living out their sexuality. It goes on to add:

Sweden is one of few countries in the world with school based sexual education, and we have a comparably long tradition of information on sexual matters as well as a profound mutual view that everyone is entitled to their own sexuality. . . . These are traditions to defend with pride. Therefore, when more and more witnesses appear, telling us about how some girls with roots in other cultures are led by the nose, threatened and treated

with violence, and when it even results in murder because they are not living the life their families have assigned to them, then something has happened in our country. Something we need to talk about. (Renberg, 2002: 2; my translation)

In this particular case, the author refers to a public debate on honour-related violence that frequently appears as a case in discussions about migration in general and about migrant integration in particular. Lately, a similar discussion has emerged around homosexuality, in which migrant communities are described as more homophobic than the Swedish mainstream. To this, the gay community in Sweden has answered by pointing out how similarly heterosexist patterns and violence take place among Swedes (RFSL, 2005).

As seen in the quote above, there is a construction of 'immutable cultural differences' between 'Swedes' and 'immigrants',3 and in this neo-racializing discourse Swedish culture is normalized and sometimes even idealized (Bredstrom, 2005). Parallel racialized representations are also to be found in Swedish debates on AIDS in sub-Saharan Africa (cf. Patton, 1992). The then president of the Red Cross in Sweden, Anders Milton (2003), claimed that the Red Cross has failed in their fight against AIDS by not challenging patriarchal cultural patterns that, as he contends, generally exist in all sub-Saharan African countries. Milton also adds fire to other prevalent notions such as that prostitution, multiple partners among men or wicked cultural practices are the main causes for the vast spread of the virus. For an intersectional analysis, however, it is not the racialized representations per se that are of primary interest, but rather the ways in which these notions of otherness are constructed through a gendered and sexualized idiom. In this discourse, official Swedish gender equality policy comes to demarcate between 'equality minded Swedes' and 'patriarchal immigrant men' (de los Reyes et al., 2002). Similarly, Swedes are construed as sexually enlightened in relation to conservative Americans, reactionary Muslims and underdeveloped Africans.

CONCLUSION

In this article I have argued for the indispensability of an intersectional approach to the HIV/AIDS problematic. I have done so through discussing what I consider to be the chief drawbacks in feminist HIV/AIDS research that focuses only on the intersection of gender and sexuality. I would like to conclude the article by pointing out how such a narrow focus becomes not only a problem for analyses, but also limits the development of an effective sexual health policy.

First, in order to challenge hegemonic heterosexual identities, Wilton (1997: 138; emphasis in original) argues that safer sex discourses need to

offer women 'a readership position which invests them with the desiring gaze, whether the object of that gaze is male or female but especially when it is male'. Holland et al. also assert that it is essential for women to obtain sexual agency in order to be able to negotiate sexual safety in heterosexual encounters. Good intentions aside, I would argue that such a strategy could have unintended consequences if other social divisions are not accounted for. It is, for instance, necessary to make sure that it does not reinforce agency only for some women, while, in a routine fashion, depicting other women as victims of culture, religion and tradition. Likewise, it is crucial to acknowledge that the construction of masculinity as an object for female desire is not always synonymous with challenging hegemonic discourses.

Second, the ways in which heterosexual masculinity has been appointed the main problem in feminist analyses give rise to a need to address heterosexual masculinity per se. However, in a climate where certain (racialized) masculinities are constructed as excessively masculine, there is an obvious risk that only those men whose masculinity has already been stigmatized as 'macho' and misogynist are targeted. or even worse, there is a risk that 'popular discussion of masculinities as problematic and dysfunctional very quickly translates into a desire to restrict and police especially working-class young men (both white and black)' (Redman, 1996: 171). A focus only on gender and heteronormativity might serve to render some heterosexual masculinities invisible, hence unintentionally abetting their hegemonic and unproblematized status. Rather than challenging the superiority of heterosexual men, sexual health policies that lack a critical perspective on how 'race', class and ethnicity intersect with gender and sexuality might very well contribute to a reification of the very hierarchy they intend to dethrone.

I would like to thank Diana Mulinari, Erik Olsson, Marianne Winter J0rgensen, Peo Hansen, the editors and the two anonymous referees for helpful comments on this article.

1. It should be noted, however, that Wilton (1997: 5, 6) stresses that her empirical material is mainly from those countries that are most familiar to her, and she also explains that the ways in which sex and gender intersect are not necessarily universal. Holland et al. (1998: 25) similarly point out that constructions of gender and sexuality are 'historically specific and subject to change'. Neither, however, theorize how this geographical or historical location impacts upon their theories of gender and sexuality.

2. Overall principles for gender mainstreaming were established in the UN Economic and Social Council (ECOSOC) conclusions, 1997/2. See United Nations (2002).

3. Despite their heterogeneity, 'immigrants' in this discourse are, if not explicitly then implicitly, often treated as a homogeneous collective (de los Reyes et al., 2002; Bredstrom, 2005).

REFERENCES

Anthias, Floya and Nira Yuval-Davis (1992) Racialized Boundaries. London: Routledge.

Baca Zinn, Maxine, Lynn Weber Cannon, Elisabeth Higgenbotham and Bonnie Thornton Dill (1986) 'The Costs of Exclusionary Practices in Women's Studies', Signs: Journal of Women in Culture and Society 11(2): 290-303.

Balibar, Etienne (1991) 'Is There a "Neo-Racism"?', pp. 17-28 in Etienne Balibar and Immanuel Wallerstein (authors) Race, Nation, Class: Ambiguous Identities. London: Verso.

Bredstrom, Anna (2005) '"Love in Another Country": "Race", Gender and Sexuality in Sexual Education Material Targeting Migrants in Sweden', Sexualities: Studies in Culture and Society 8(4): 517-35.

Collins, Patricia Hill (1998) 'It's All in the Family: Intersections of Gender, Race and Nation', Hypatia 13(3): 62-82.

Connell, Robert W. (1995) Masculinities. Berkeley: University of California Press.

Crenshaw, Kimberle (1994) 'Mapping the Margins: Intersectionality, Identity Politics, and Violence Against Women of Color', pp. 93-120 in The Public Nature of Private Violence. New York: Routledge.

De los Reyes, Paulina and Diana Mulinari (2005) Intersektionalitet. Lund: Liber.

De los Reyes, Paulina, Irene Molina and Diana Mulinari, eds (2002) Maktens (o)lika forkladnader. Stockholm: Atlas.

Doyal, Lesley, Jennie Naidoo and Tamsin Wilton, eds (1994) AIDS: Setting a Feminist Agenda. London: Taylor and Francis.

Herrstrom, Staffan et al. (2004) 'Mans sexuella handlingar dodar', Aftonbladet 1 December.

Holland, Janet (1993) Sexuality and Ethnicity: Variations in Young Women's Sexual Knowledge and Practice. London: Tufnell.

Holland, Janet, Caroline Ramazanoglu, Sue Sharpe and Rachel Thomson (1998) The Male in the Head: Young People, Heterosexuality and Power. London: Tufnell Press. (2nd edn 2004.)

Hollway, Wendy (1996) 'Gender Difference and the Production of Subjectivity', pp. 62-73 in Stevi Jackson and Sue Scott (eds) Feminism and Sexuality. New York: Columbia University Press. (Orig. pub. 1984.)

hooks, bell (1995) Killing Rage: Ending Racism. New York: Owl Books.

Milton, Anders (2003) 'Roda korset misslyckas i kriget mot aids', Dagens Nyheter 8 October.

Mohanty, Chandra Talpade, Ann Russo and Lourdes Torres, eds (1991) Third World Women and the Politics of Feminism. Indianapolis: Indiana University Press.

O'Donnell, Mike and Sue Sharpe (2000) Uncertain Masculinities: Youth, Ethnicity, and Class in Contemporary Britain. New York: Routledge.

Patton, Cindy (1992) 'From Nation to Family: Containing "African AIDS"', pp. 218-34 in Andrew Parker (ed.) Nationalisms and Sexualities. New York: Routledge.

Redman, Peter (1996) 'Empowering Men to Disempower Themselves: Heterosexual Masculinities, HIV and the Contradictions of Anti-Oppressive

Education', pp. 168-79 in Mäirtin Mac an Ghaill (ed.) Understanding Masculinities. Buckingham: Open University Press.

Renberg, Gudrun (2002) 'Sexualitetens olika villkor', Hiv-aktuellt 2: 2.

RFSL Radgivningen i Skane (2005) Hedersrelaterat vald mot ungdomar pa grund av sexuell läggning. Malmö: Länsstyrelsen i Skane län.

SOU (2004) Samhällets insatser mot hiv/STI, SOU 2004: 13. Stockholm: Fritzes offentliga publikationer.

United Nations (2002) Gender Mainstreaming: An Overview. New York: UN.

Utrikesdepartementet (Ministry of Foreign Affairs) (2005) 'Sexuell och reproduktiv hälsa och rättigheter'; at: www.regeringen.se/sb/d/5433 (accessed 7 June 2005).

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

Anna Bredstrom is a PhD candidate in ethnic studies at Linkoping University, Sweden. Her PhD examines Swedish HIV/AIDS policy discourse, focusing on how notions of risk and safety are mediated through intersections of 'race', gender and sexuality. She has previously published in the journals Sexualities and Nora: Nordic Journal of Women's Studies. Address: Department of Ethnic Studies/ISV, Campus Norrkoping, Linkoping University, SE-601 74 Norrkoping, Sweden. [email: annbr@isv.liu.se]