“But the category of exposure also has to respect identity”: MSM, classifications and disputes in AIDS policy

This essay explores the relationship between diversity and public health by address-ing tensions related to classifications and recognition in the field of HIV and AIDS policy. The objective is to reflect on how classificatory and operative categories are articulated within the scope of programmatic responses towards the social production of differences and inequalities. To do so it draws from the theoretical framework of studies on vulnerability and recognition and from a methodology that includes a critical review of the literature on the category men who have sex with men (MSM) and ethnographic material, de-rived from the authors’ research and a literature review related to social movements, and research and policies focused on lesbians, gays, bisexuals, transvestites and transsexuals (LGBT). It reviews how the MSM category was constructed in the field of HIV and AIDS prevention policies at an international level, situating political actors and tensions. It problematizes these tensions by analyzing processes of production of political subjects as well as changes in socio-state relations that involve LGBT. It emphasizes the importance of considering how differences and inequalities emerge in socio-political processes and of dedicating studies to improve policies, ensuring an effectively more respectful care.


Introduction
The expression used in the title of this article emerged from an interview with an activist conducted for a study 1 about policies for the prevention of HIV transmission and AIDS, aimed at gays and other men who have sex with men (MSM). The persistence of the concentration of the epidemic in the epidemiological category of MSM and the increased prevalence of HIV infections in this group for more than ten years led us to question if the policies and actions for prevention conducted in Brazil have been sufficient for confronting the determinants of this epidemic.
Vulnerability is used as a reference to examine the social and political determinants of the multi-causal processes and relations that lead to different conformations of the HIV and AIDS epidemic in particular contexts. It is distinguished by its understanding that healthcare practices and the associated set of knowledge and concepts are determinants of the health conditions about which efforts are made to understand and transform, and by inclusion of the perspective of the subjects affected and their contexts of intersubjectivity 2 . It is in this sense that, mobilized by the disturbances expressed in the field, we propose to reflect on the meanings of adoption of the category MSM in the realm of policies for epidemiological surveillance and prevention of HIV infections and AIDS in Brazil.
We understand that, since the beginning of the HIV epidemic in the country, a dialog has been established which becomes effective through policies, actions and services designed to offer measures for prevention and control of HIV and AIDS. This dialog takes place amid a dispute over normative horizons about what is understood to be necessary, in terms of public health, and what is understood as desirable, or at least acceptable, by those at whom the policies are aimed.
By articulating vulnerability and human rights 2 to the theory of recognition 3 , we seek to understand how processes of (non)recognition, or disrespect, in public care 1,4 , understood as a set of policies, services and actions aimed at the prevention of HIV infection and AIDS, contribute to processes that make different population segments vulnerable to HIV infection and AIDS.
Our objective is to contribute to thinking about the relationship between diversity and public health by reflecting on processes of categorization and classification, understood as forms in which language, power and practices are articulated in the realm of healthcare policies. We thus consider the field of HIV/AIDS as a privileged space for thinking of questions that involve collective health and diversity, difference and inequalities.
To do so, our methodology involved a genealogical investigation of the category MSM based on a critical review of the literature about its construction in the field of HIV and AIDS prevention policies, on an international level, locating political actors and tensions between the production of operative categories and an examination of a diversity of practices, subjectivities and contexts. We problematize these tensions by analyzing ethnographic material from studies by the authors 1,4-8 and conducting a review of the literature about processes to produce policies and changes in socio-state relations involving lesbians, gays, bisexuals, transvestites and transexuals (LGBT), as well as studies about HIV/AIDS in transvestites, trans women and bisexual men in the Brazilian context.

The category MSM: its origins and path through a disputed field
Although there is no definitive production that has been dedicated to the historization of the category MSM, various scholars have dedicated themselves to criticizing it [9][10][11][12][13] . Many contradictory hypotheses about its origin have been proposed 10,14 , until the researchers Peter Aggleton and Richard Parker 11 reported that the category was created by English community activists, upon reflecting on the reception of strategies for HIV prevention among gay men who are "relatively self-confident about their sexual identity" e "'other' more closeted men" (p. 1554).
Contrary to the way that it came to be understood, it was not an inclusive or aggregating category, but one of differentiation. Initially, it distinguished those with a more open and public sexual identity from those who are more closeted and hidden; those linked to organized gay communities from those who do not have such attachments 15 . Later, it reached a distinction that was crystalized in the field: between men who identify as gay or homosexuals and those "men who have sex with men but do not identify as gay" 10 (p. 291).
Despite its community origin, the broad dissemination of the category was made by multilateral bodies, such as the World Health Organization (WHO) Global Programme on AIDS (GPA) and, later, the Joint United Nations Program on HIV/AIDS (UNAIDS) 13 , international financing organizations 15 and international networks for research and activism 10 . This adoption by bodies with the stature of the WHO and UNAIDS caused it to be identified, by some of its critics, as a category coined in the realm of science and bureaucracy "to signify behavior without identity" 10 (p. 291).
Considering the history of other operative categories of prevention, the proposal of the category "men who have sex with men" coincided with the shift from consideration of "risk groups" to "risk behaviors", that took place in the second half of the 1980s 16,17 . Thus, an axiomatic effort was articulated to affirm that the risk of infection by HIV did not come from "who you are" but "what you do" 10 (p. 291), transferring the focus from identities to practices.
The notion of "risk group" is based on epidemiological knowledge and the establishment of analytical categories that sought to identify characteristics associated to a greater chance of exposure to HIV infection, but came to configure the concrete identities of those affected by HIV and AIDS 17 . Thus, it guided preventive actions focused on abstinence and the isolation of people who were part of these groups, resulting in intense processes of stigmatization and discrimination 17 . Homosexuals were one of the groups most affected by this exclusionary approach and were also among those who most criticized it and defended focuses that valued community support and a positive perspective of sex and homosexuality 18 .
The recognition of the impact of the processes of stigmatization and discrimination related to HIV and AIDS that strongly marked the experience with the operative concept of "risk groups", required innovations in prevention efforts. The more conventional measures involved initiatives to provide information and education and social and healthcare services (distribution of condoms, testing and counseling, treatment for other sexually transmissible infections, needle exchanges, treatments for users of injectable drugs and provision of safe blood and blood products) 16 . An innovative element was also incorporated to HIV and AIDS prevention through initiatives to confront processes of stigmatization and discrimination of people living with or affected by HIV and AIDS and those social segments at greater risk of exposure to HIV 16,19 .
It is probably through associations between the categories of "risk behaviors" and MSM, which describe practices instead of reified iden-tities, that it was commonly affirmed that the objective of the category MSM was to reduce the stigma against gay and bisexual men, transexual women and self-identified heterosexual men who engage in sex with other men 14 . This did not take place without criticism. As Boellstorff 10 questioned, it is not revealed why "MSM" would be considered "less stigmatizing" than "gay", given that it is homophobia and heterosexism that make explicit use of the category "gay" undesirable.
But how did the broad adoption and stabilization of the category MSM in the global AIDS field take place? To respond to this question, it is necessary to consider the institutional structure of this field over time. In 1986, the Special Programme on AIDS was established within the realm of the WHO -and later known as GPA, in 1987 -and made responsible for implementing a global strategy to fight the epidemic. In 1996, UNAIDS was created, which included, in addition to the WHO, five other agencies in the UN system, because it understood AIDS as a transversal theme 19 .
According to critical studies 10,11,13 , in the late 1980s and early 1990s, the GPA, whose task was to guide strategies to prevent and control AIDS and provide technical and financial assistance to countries, promoted meetings to discuss HIV prevention among gays, bisexuals and other MSM. Despite its significant influence in the global response to AIDS, with an emphasis on the human rights of individuals and groups affected, including gay men, and the organizations and responses of civil society, the GPA directed limited attention to homosexual and bisexual men in the prevention of AIDS and global policy related to it. Thus, the WHO, under GPA leadership, did not publish any guidelines for the development of surveillance or prevention programs aimed at gays, bisexuals or other MSM 13 .
A few factors contributed to this according to McKay 13 , who studied the role of international organizations in the development and dissemination of the category MSM, with special attention to the GPA and UNAIDS. One is the GPA's emphasis on a dual pattern of the global AIDS epidemic, which distinguishes sexual transmission among homosexual and bisexual men in developed countries from sexual transmission among heterosexual men and women in developing countries. This emphasis minimized the recognition of the presence of gender and sexual diversity in developing contexts, and conveniently spared the GPA from the task of defending the extension of surveillance and prevention to address new infections among homosexuals, bisexuals and other MSM and to confront intolerant national governments, maintaining their commitment to other WHO programs 13 .
A second factor is related to the decision-making structure of the GPA, which was dependent on small groups of specialists organized to provide quick responses to new information and needs. This structure was identified as problematic because it makes it difficult to build consensus, in a context in which there is a lack of information based on research and deep polemics around and a politicization of the discussions. Thus, the recommendations for prevention policies made by WHO used a vague language, which grouped all forms of sexual transmission of HIV and discarded open references to sexual behavior and marginalized groups 13 .
It was in this context that the adoption of the expression "men who have sex with men" was presented as a suitable term -epidemiologically and politically -for the United Nations, distancing it from a gay and political agenda, as reported by Gary Dowsett, who was present in meetings promoted by the GPA 13 .
As is characteristic of the field of HIV and AIDS, these dialogs included public health workers, epidemiologists, researchers and activists. According to Boellstorff 10 , the concept originated from two insights: first, that there are men who have sex with men and who do not see themselves as gay and may be hostile to the term; and second, that prevention programs aimed at gays can exclude a range of these MSM who do not identify themselves as gay.
In this dialog, the main concerns of the researchers was with the limits of terms used because of their biomedical (and psychiatric) origins like homosexual and bisexual, with which few men identified in that context, and to point to the large diversity of groups of men who have sex with men, which could be associated to situational and circumstantial factors, intersectioned by race, gender, class and age, in a wide variety of social contexts, multiplicity of subjectivities and practices 11 . Aggleton et al. 11 indicate that although social research in the 1990s and 2000s documented a significant diversity among gays, bisexuals, and other MSM and trans people, the adoption of categories such as MSM, understood as unifiers in epidemiological and programmatic discourses, wound up having the opposite effect, erasing the perceptions of diversity among these populations. It is in this sense that Young et al. 9 mention the adoption, since 1994, of the acronym "MSM" and we can indicate its progressive "substantivization" in the field: it is no longer constituted as a category of differentiation and comes to progressively operate as an aggregating and invisibilizing category of the diversity it intended to present.
In terms of recognizing the validity of efforts to shift the focus of prevention actions from socalled "risk groups" to "risk behaviors", as early as the 1990s homosexual activists involved in the field of AIDS indicated concern that this shift could imply an international "de-gaying" 20 (p. 432) of AIDS, or even, that it involved adoption of a euphemizing language about sex 10 (p. 294, citing King, 1994).

Sociopolitical changes and specificities: challenges to healthcare policies in Brazil
In Brazil, in 2007, anthropologists Sérgio Carrara and Júlio Simões 21 identified the proposition of the term MSM as part of a strategy that "supposes to contemplate the specificity of the contingent of men who have sexual relations with people of the same sex and do not recognize themselves in identity categories, such as 'homosexuals' , 'gays' , 'entendidos', etc. " 21 (p. 94). They add: One problem with the MSM category is that it dissolves the question of non-correspondence between desires, practices and identities into a formulation that recreates the universal category of "man" based on a a supposed foundational stability of biological sex, as it simultaneously allows evoking the well-known representations of male sexuality as inherently unruly and disturbing 21 (p. 94).
This critical reflection emerged amid sociopolitical processes of change that include both the strengthening and the expansion of visibility of social movements currently known as LGBT, as well as the recognition, even if partial and precarious, of demands made by these populations. These processes initiated in the mid 1990s, and took shape with the creation of the "Programa Brasil sem Homofobia" [Brazil Without Homophobia Program] and a series of public policies and norms developed between 2003 and 2011 22 .
The socio-state dialogue required political subjects who could clearly explain specificities of certain populations, implying transformations in the forms of self-identification, of both activists and the LGBT population in general. These transformations include the decentering of the category of homosexuality, but also the complexification and diversification of the political subjects of the movement, with various effects on the forms of classification of this population 22 . In scholarly works of the social sciences, this is noted as a "dual epistemological fracture" of the broader category "homosexuality": with the distinction between gays and lesbians, still in the passage to the 1980s; and the emergence of the trans identities (transvestites and transsexuals) at the turn of the century 23 .
More recent studies point to a process in which the strategic activation of identities -strategic essentialism -typical of the two first decades of socio-state dialog, came to coexist with a more intense presence: 1) of demands for more direct representation of experience of various subjects; and 2) of a diffusion of an intersectional perspective, which sought to articulate multiple differences and inequalities in their dimensions of experience and social structures. This diversification of perspectives and of activist repertoires takes place amid an intensification of criticisms of the centrality of socio-state dialog and to the advance of political actors opposed to the agenda of sexual rights and human rights, as they have been conceived and constructed in recent decades 6 .
This involves a situation of intense changes both in the field of socio-state relations and in the daily life and modes of classification of LGBT populations. During this process, categories made visible through the expanded effects of the HIV and AIDS epidemic in its first decade, such as "transvestites" and "bisexuals", gradually came to inhabit the political scene. However, as we will see below, they did so under different conditions while confronting deeply rooted although different stigmas.
The category transvestite emerged politically through the collective organization made to confront the epidemic, which advanced with great strides, surrounded by stigmas, over an already extremely marginalized and discriminated population. In the passage to the 1990s, initiatives such as shelters to support transvestites affected by AIDS combined with efforts at political organization to face the epidemic and the growth of violence against those who spread the "gay plague" that specifically victimized transvestites who worked as street prostitutes 7 .
As a population marked by strong vulnerability and visibility, transvestites began to organize politically and take part in the broader social movement since the mid 1990s. They were included in the acronym used to represent the movement in 1995 and founded their first national network of activists, the National Association of Transvestites and Transsexuals (ANTRA) in 2000, which followed the emergence of transsexuals as political subjects in the late 1990s 24 .
The living conditions of transvestites and transsexuals have made positive and noticeable advances since the 2000s, particularly among subjects located in more favored social strata or those supported by their families of origin. These changes were related to both the greater social legitimacy attained by efforts of the LGBT movement and by transvestites and transsexuals specifically, and through public policies. In the field of healthcare, the Transsexualizing Process was instituted in the federal "Sistema Único de Saúde" [Unified Healthcare System] along with policies for HIV prevention and AIDS treatment. In the field of specific rights, norms emerged that allowed the use of the "social name", which is the new name chosen by trans people, in spaces such as schools, banks and public agencies, and a later Federal Supreme Court (STF) decision that authorized a change in a person's first name without the need for a medical statement or surgery 8 .
These transformations in political organization and in the recognition of demands by state entities were related to the process of citizenship 25 , but also to criticisms 23 of the dilutive and essentializing character of the category MSM. These changes, although they were very important, should be considered, recognizing their precariousness, and the tension and constant threat of reversal in which they are immersed. It is necessary to consider the unequal scope of the effects of these policies among subjects of different regional, and generational insertions, and those of class and race, as well as the harmful impacts of the rise of political sectors that give priority to blocking the advance of policies favorable to LGBT rights 8 .
Bisexuals also emerged as a visible category in the first decade of the epidemic in Brazil, based on the homogenizing presumptions of a "dual life" and the non-assumption of socio-sexual identities related to their practices and their association with the notion of a "bridge" for HIV between homo and heterosexual populations. Socio-behavioral studies conducted in various important Brazilian cities in the 1990s indicated that approximately 12-16% self-identified bisexuals in cohorts of men who have sex with men 26 .
The category bisexual has had a quite rocky path of insertion and political organization. In the late 1990s, after the emergence, growth, and diffusion of Pride Parades, beginning in São Paulo, and of efforts to articulate with international movements, the international conformation of the acronym GLBT was adopted, which includes the letter B, for bisexuals. Internal resistance to the movement was expressed through measures such as the placement of the B at the end of the acronym. And in 2003, a national encounter of the movement deliberated to exclude the letter. This extreme measure provoked the mobilization of bisexual activists who had previously been dispersed 7,27 .
The first initiative of the regional and national organization of collectives and activists around bisexuality culminated in the creation, in 2005, of the Brazilian Collective of Bisexuals (CBB), which was discontinued in 2007, amid conflicts that involved representation in spaces of socio-state dialog and difficulties in gaining recognition from other segments of the GLBT movement. Another initiative arose in 2010, with the organization of collectives supported by the popularization of internet access and the use of social networks 27 . The separate organization of the movement, which came to call itself LGBT, provided stability to the activism of bisexuals, culminating, in 2020, in the creation of the Brazilian Bisexual Front (FBB), which articulates collectives and individuals throughout the country, seeking to give visibility to activities, demands and to work for their implementation.
Scientific knowledge about the trans and bisexual populations in Brazil is precarious, characterized by small groups of studies and a predominance of qualitative studies in specific contexts. Studies about the production of public policies are nearly exclusively restricted to trans people, and until recently produced predominantly by the activist organizations themselves 28 .
Although there are no national sociodemographic data that allow estimating the bisexual population, studies conducted at LGBT Parades in Rio de Janeiro, Recife, and São Paulo, between 2004 and 2006, found among women and men participants, 8.8-12.9% self-declared bisexuals while 35.9-63% identified themselves in categories that refer to homosexualities [29][30][31] . Among these studies, only a few separated by sex the respondents self-identified as bisexuals. These results refute stigmas, indicating similar percentages of assumption of socio-sexual identity in social life among men who are self-declared as homo-or bisexuals, as well as reports of situations of discrimination and aggressions motivated by sexuality 30,31 .
In terms of HIV/AIDS, in addition to the fact that the studies do not necessarily separate homoand bisexuals, they predominantly address sexual conduct, and when they consider identities, they do not do so in a way that correlates the dimensions of the conducts and the identities to the results. Despite this, specificities identified include a trajectory of relative risk that is not declining 32 and more frequent situations of risk 33 among men with bisexual conduct. A socio-behavioral study with serological testing conducted in São Paulo between 2011 and 2012 found 14.3% self-identified bisexuals while 38.9% had conduct involving more than one sex. The prevalence of HIV found was lower among self-identified bisexuals (9.4%) compared to 17.6% among gays/ homosexuals, although both are much higher than the prevalence in the general population 34 .
Quantitative studies of an applied nature among trans people are still scarce. Studies about HIV/AIDS specifically with transvestites and trans women are recent and indicate significantly higher prevalences 35 and sufficiently diverse sexual and preventive behaviors to justify research to continue to be carried out separately The situation presented allows affirming that the category MSM currently encompasses not only self-identified homosexuals and men who have relations with other men, who do not derive an identity from this. When the category MSM was introduced in the country, there were already, among cohorts of studies, subjects self-identified as bisexuals or transvestites 26 . Over the years, specificities were thickening, related to the trajectory of political organization, the permeability to public policy agendas, but also to levels of knowledge about the various identities aggregated in the category MSM. From the point of view of knowledge about HIV, the variation in prevalence and in sexual and preventive behaviors among the different categories of identity is now known. However, the programmatic responses have not proved to be aligned to these socio-political changes and to the knowledge that has been produced.
This dissonance became evident when, in 2007, negotiations were realized, coordinated by the National Program of STDs and AIDS, concerning the development of plans to confront the epidemic. At that time, transvestites organized in a movement decided to participate in the "National Plan for Confronting the AIDS Epidemic and STDs among Gays, other Men who have Sex with Men (MSM), and Transvestites". Transsexual women joined the "Integrated Plan to Confront the Feminization of the Epidemic of AIDS and other STDs". During this period, tensions between transvestites and transsexuals deepened, with a tendency to distance the transsexuals both from the LGBT movement and the category transsexual. At the same time that gays demanded the removal from the acronym MSM, bisexuals could not do the same, and transsexuals held intense debates around categories such as "women who experience transsexuality" 36 (p. 183). In this context, the category MSM, in addition to being substantivized, began to be used, with an identity or accusatory connotation, by men with homo or bisexual conduct.
Despite the conflict over making specificities visible, the plans to confront AIDS appeared to have been weakly incorporated by state and municipal governments -and hidden, in the programmatic realm, by the development of the National Policy for the Integral Health of Lesbians, Gays, Bisexuals, Transvestites and Transsexuals (LGBT), launched in 2011 1 . The policy attended the demands for visibility and recognition, but to do so framed the LGBT population separately from other populations, such as that of women. Moreover, it did not come to fully address the confrontation of AIDS. Tensions related to the stigma that linked gays and HIV/AIDS contributed to having the issue not be properly addressed in the Policy 1,4 . In this period, the LGBT policies and agenda came to suffer brutal attacks, with scandals around educational materials and communication aimed at specific publics 1 .

Final considerations
The analyses conducted show us that the maintenance of the adoption of the category MSM in the realm of HIV/AIDS policies violates normative expectations of recognition in public care in health by clouding and homogenizing different social segments under the assumption that they are, generically, gays and other MSM.
In terms of bisexuals, although the diversity of situations of conduct or identity to which the category can refer to is recognized, a systematic erasure is found in studies and polices of the use of the category in relation to socio-sexual identity, as well as the homogenization and dissolution of specificities. This takes place both through the use of the category MSM and the expression "gays and bisexuals", the latter, without an indication of the specificities of each of the categories, but also by studies and specific actions that refer to "bisexual people", without disaggregating them by sex. Thus, the sociosexual identities of the subjects are disrespected and the possibilities for production of scientific knowledge, policies and suitable and effective programmatic actions are obliterated.
Trans women and transvestites, systematically categorized as male cases in the epidemiological data, are disrespected in their gender identities and excluded from the production of public data about HIV and AIDS. This erasure carries grave implications for the production of information about health, that fail when guiding efforts at prevention and care, or to properly steer technical and financial resources. Limited information has been produced by isolated studies 35,37,38 , but does not assure historic series and the accompaniment of the evolution of the conditions of health of trans people, or their production throughout Brazil, indicating that strategies for epidemiological surveillance contribute to the programmatic vulnerability of these population segments. This situation also expresses how sex and genitalia continue to be central categories in the production of healthcare policies, despite the growing process of citizenship for people dissident from cisgenerity.
By steering prevention policies, the adoption of the MSM category impedes the communication of preventive messages, given that it does not make clear with whom it is establishing interlocution. The arguments that initially supported its adoption -the diversity of subjectivities, practices and contexts and the distinction between identities and practices -have been shifted to a strict reference to the sex of individuals and their sexual partners (conceived as a penis in relation with vaginas and anuses) and to the constitution and presumption of an MSM identity that universalizes, encompasses and homogenizes all and any diversity of desires, practices and identities. This shift, and the consequent attempt to universalize the category MSM, culminated in limitations of healthcare policies to deal with complexities and differences in the realm of experience, producing erasures that engender inequalities of care.
Currently, based on these disjunctions and on a political situation that is refractory to recognition of LGBT demands, the very diverse group of trans subjects, who were constituted in a context of attending to the needs of corporal changes and of adaptation of the civil register to gender identity and expression, require the formal validation of their parenting and of the register of this experience through alterations in the Declaration of Live Births, health document that originates the official birth certificates, which for the first time creates the opportunity, (although still uncertain) to register their dissident gender identities in the information systems of the Ministry of Health. This is a complex and challenging theme, particularly in the current context, which transcends our focus and would require careful and specific treatment. We only mention it to emphasize that, when considering the relationship between collective health and diversity, perhaps this does not involve producing and sustaining categories considered to be correct with a focus on a supposed care and respect for individuals.
Care and respect should certainly be sought at the level of policies, in the daily offer and provision of care in health and in the production of information. But perhaps the question is not focused on multiplying categories. It is import-ant that we are attentive to what in fact establishes difference, considering that difference is not something given or static, nor does it have an essential character, but is produced in social relations, which are relations of power, in which differences potentially operate as inequalities 39 . Therefore, the dimension of production of knowledge is essential to producing responses more suitable to health needs.
The HIV and AIDS epidemic, as well as the current Covid-19 pandemic, have been privileged places for observing relations among differences, inequalities, and health. Observing how difference is profiled in given social contexts, how it articulates with other differences, and how it produces inequalities can help guide us to produce better programmatic responses, to assure effectively more respectful care in the daily health care. collaborations G Calazans and R Facchini contributed equally to the conception and outline of the article; its drafting and critical editing; and the approval of the version to be published.