Paper presented at the 3rd Annual Qualitative Methods Conference: "Touch me I'm sick"
8 & 9 September 1997, University of South Africa Regional Office, Durban

'Of pansies, perverts, and macho men':

Researching homosexuality through a discursive lens

Pravani Naidoo

Department of Psychology, Vista University, Bloemfontein

This paper explores the merits of a discursive framework in researching homosexuality. It traces how the dominant discourses have been transformed historically from those of effeminacy : 'pansies', to pathology and deviance : 'perverts', to the more recent 'butch-shift'. Signifying a breakdown in and challenge to the divide between homo- and heterosexuality, the 'butch-shift' plays itself out through cultural representations of gays as 'macho men'. This paper argues that a discourse approach is particularly valuable in that it allows us to attend to the ways in which 'normal' and 'deviant' human subjectivities are constructed, embraced and resisted discursively. Its emphasis on variation/inconsistency (as well as consistency) facilitates an exploration of the multiple, shifting, often contradictory meanings that are constructed around what it means to be homosexual. A case is made for the relevance of Parker's (1992) approach to discourse analysis in researching homosexuality.

It is unlikely that we should find a more appropriate forum for exploring how homosexual subjectivities are constructed, negotiated and resisted discursively, other than under the banner 'Touch me I'm sick : Marginality and pathology in the new millennium'. Limiting my considerations to concerns around male homosexuality in particular, this paper represents a theoretical attempt to summarise my thinking around (how to go about) researching homosexuality. I raise both theoretical and methodological issues in order to make a case for the relevance of researching homosexuality through a discursive lens.

Following the work of theorists such as McIntosh (1968),Weeks (1977), Plummer (1981), Segal (1990), and Forrest(1994) amongst others, my theoretical project entails a discursive elaboration of the ways in which homosexuality is continually re-defined. A fundamental premise here is that any consideration of homosexual identity construction warrants attention to the social processes of exclusion, marginalisation, legitimation, and transgression. Hence the title of my paper, "Of pansies, perverts and macho men" alludes to my attempt to trace how the dominant discourses of homosexuality have been negotiated, transformed and contested over time.


For over a hundred years now scientific and popular belief has held that male homosexuality derives from and expresses something 'feminine' in men - the absence of appropriate levels of masculinity (Marshall,1981;Connell,1995). This can be traced to features of 19thC society. The elevation of marriage and the family, the firm binding of sexuality to marriage, and the control of male lust all served to render homosexuality a central target for moral campaigners of the time (Marshall, 1981). In this context homosexuality referred not to deviant sexual behaviour but rather to transgressions of gender identity. Gender anomalies formed the basis of the homosexual category (Marshall,1981). In popular stereotypes this was reflected in the image of the feminine man, crudely labelled the 'pansy'. According to Marshall (1981), this connection between gender inversion (or a mixed-up gender make-up) and homosexuality served a two-fold function : To control and punish homosexual behaviour on the one hand, and to define and maintain appropriate definitions of masculine and feminine behaviour/roles on the other.


Toward the second half of the 19thC, a medicalised account of the homosexual as 'pervert' replaced the older conception of the gender invert or 'pansy'. It was now increasingly 'discrepant sexual behaviour' rather than gender anomalies which formed the basis of the homosexual category (Marshall,1981). "'Perversions' such as homosexuality were stigmatised not just as a series of sexual acts, but as a state of mind" (Forrest,1994:100); thus sexual irregularity was displaced to the realm of mental illness (Foucault, 1976; De Cecco, 1981; Wilton,1995).

With the institutional locus of control now falling in the realm of medicine and psychiatry, the dominant discourse of effeminacy came to be displaced by the competing discourses of pathology and deviance. These medicalised discourses laid down prescriptions for, and attempted to regulate how and what sexual experiences (desires, practices and relations) ought ideally to be (Tiefer,1992). Monogamous heterosexual marriage was legitimised as 'natural' and normative by the all powerful medical and psychiatric institutions. Homosexuality, a deviation from this ideal, came to be labelled as pathological. Following McIntosh's (1968) argument, the discourses of deviance and pathology work to provide a clear-cut threshold between permissible and impermissible behaviour. Further, they help to segregate those labelled as 'deviant' from others, and thus contain and limit their behaviour patterns. Implicated here are the territories of marginalisation and exclusion.

Against the backdrop of the dominant discourses of deviance and pathology, it is important to note that the late 19thC is often hailed as a critical period for the ways in which homosexuality came to be constructed and re-defined (Foucault,1976; Weeks, 1977,1981). Homosexuality came to be categorised as a separate condition (referring to characteristics inherent in a particular kind of person) for the first time (McIntosh,1968). Accompanying this was a correlative emergence of the homosexual as constituting a distinct identity (Weeks,1977;Dyer,1992;Padgug,1992;Edwards,1994; Connell,1995).

With respect to categorisation and identity construction, it is notable that the categories 'homo' and 'hetero' have been theorised as entailing an organising of desire as well as a policing of it (McIntosh,1968; Dollimore,1996). Deemed the paradox of categorisation (Plummer,1981; Wilton,1995), the medicalisation of sexuality sought not only to regulate through naming. Rather it was to produce homosexuality as an entity and identity around which individuals could define themselves and others. Moreover, these individuals could, with sufficient initiative, form collectives according to identity (Edwards,1994).

The emergence of gay subcultures may be understood as a form of resistance to the negative implications of the sexual categories. The category 'homosexual' was taken on as a basis for a way of life rather than as a condition to be overcome or cured. A significant number of homosexual people began to participate in a lifestyle, a set of tastes, fashion, language and so on that meant that their lives were, in more respects than the sexual, different from most heterosexual people. From this subculture emerged the politics of the late 1960s gay liberation movement. This was to turn the stigmatisation of homosexuality on its head, affirming (instead) a positive self-identification in terms counter to the dominant heterosexist culture (Dyer,1992; Edwards,1994; Connell,1995).

"Macho men"

Since the early 1970s, the predominant style and image of gay men has undergone a dramatic shift away from that of the effeminate 'pansy' towards that of the 'macho man'; there is now a positive identification amongst many male homosexuals with masculine style and demeanour (Marshall,1981). Documented as the 'butch-shift' and the 'masculinization of the gay man' (Plummer,1981;Gough,1989;Segal,1990; Forrest,1994), this move refers to "the emergence, of a particular accentuated form of masculinity among gay men who choose to present themselves in terms of body images, sporting activities and clothing which, in earlier stereotypes, were associated only with straight men" (Cornwall & Lindisfarne,1994:5).The shift toward the image of the 'macho man' plays itself out most clearly in terms of the body : hours in the gym are required to create, not necessarily an inflated muscle man, but a male-athletic body. This new 'masculine' identity - in style, mannerisms, certain forms of social behaviour etc - has become rooted in male physiology rather than a peculiar 'feminine' psyche (Gough,1989). Commenting on the nature of the shift, Gough (1989) asserts that it is uneven, more prevalent amongst younger rather than older men, and concentrated amongst gay men whose social life is centred around the scene in larger cities. Further, he maintains that while the shift is clear, older cultural traditions continue to be reinforced and perpetuated within the subculture.

While the discussion here has been presented in the form of an historical analysis (in terms of sequential shifts), Gough's (1989) theorising on the nature of the change serves as a reminder that these shifts in how homosexuality is constructed are not absolute, fixed and discrete. Rather the discourses compete, co-exist, interact and overlap with one another at any one point in time. Clearly then, the contemporary masculinization of the gay man is not an absolute; the style and image of the gay man as effeminate for instance still prevails and co-exists with that of the 'macho man'.

It is important to recognise that the shifts in the ways in which homosexuality is constructed, negotiated and contested discursively are not internal to a gay context only. Also at play here are considerations around masculinity. Indeed what part the discursive shift is about is contestations around what kind of men these are. Are male homosexuals effeminate 'pansies'? Are they 'perverts' suffering from mental illness or a condition to be cured? Or are they 'macho men'? Linked in a significant way with contemporary gender politics, the butch-shift can thus be seen as part of the broader 1990s project of re-defining 'what it means to be a man'. The negotiation and contestation of discourses around homosexuality takes place in a context within which it is legitimate for all men, gay men included, to think about defining themselves.

The issue of self-definition is an important feature of the butch-shift. Previously meanings and understandings around what it means to be gay tended to be constructed and imposed upon homosexual people by the 'other' (institutions like medicine and psychiatry for instance enforced a particular pathologised conception of homosexuality as disease/mental illness). Recently, there has been a shift away from meanings around homosexuality being imposed by 'others' to more of an agency at the level of self-defining discourse (Dyer,1992). In other words, there is more of an agency for gay people to be involved in the process of defining themselves.

Relevant to considerations around homosexual identity construction is the question of social acceptance. Forrest (1994:97) suggests that the gay man has "changed the image he projects to the broader society, as well as his own self-perceptions, to ones which can be seen as more acceptable". He appears to have moved away from seeing himself, and being seen by others, as a 'pervert' and a 'pansy', towards seeing himself and being seen as a complete (that is 'real') man; this suggests a union of his biological sex with what he perceives and assumes to be socially accepted as 'masculine'. The butch-shift has to some extent dislodged the association between macho masculinity and heterosexual men (Cornwall & Lindisfarne (1994). Blachford (1981) suggests that we might interpret this move as an attempt to show that masculine or 'ordinary' men can be homosexual too - thereby breaking down the stereotyped image of homosexuals as 'pansies' and 'perverts'. We need to appreciate, however, that a significant portion of gay men can and do pass as 'straight'. The fine line between a 'true' masculinity (which is heterosexual) and its opposite (which is homosexual) is being increasingly transgressed.

Some methodological considerations

Exploring the ways in which homosexuality has come to be re-constructed and negotiated over time as we have just done allows us a frame within which to understand how a multiplicity of meanings around homosexuality come to be constructed. Further, my theoretical project of tracing the historical shifts in the ways in which homosexuality is re-defined allows us to appreciate that homosexuality is not a fixed, unchanging entity. It is problematic, however, that the conceptualisation of homosexuality as a reified condition has tended to dominate much research in the field. The preoccupation with concerns around etiology, researching what causes homosexuality (McIntosh,1968; Plummer,1981;Connell,1995), has been framed by a priori assumptions of deviance and pathology. Understanding deviance and pathology as a set of discursive constructions to be embraced, negotiated and resisted thus allows us to overcome such limitations in terms of method.

A discourse approach (Potter & Wetherell, 1987; Wetherell & Potter, 1988; Edwards & Potter,1992; Parker,1992; Burman & Parker,1993; Potter & Wetherell,1994) is particularly valuable in that it allows us to attend to the ways in which amongst others, 'normal' and 'deviant' homosexual subjectivities are constructed, negotiated and contested discursively. My research entails doing a discourse analysis of multiple qualitative in-depth interviews with male homosexuals in Bloemfontein in the Free-State. I use a discursive frame to explore "how homosexual participants interpret and understand their sexual desires and practices so as to situate themselves in the world". Here the text, the interview transcripts are taken as the site of meaning/creation; meaning is created actively as individuals grapple with the process of homosexual identity construction. Attending to the ways in which meaning around homosexuality gets made brings us to an important feature of a discourse approach : the focus on how discourses work to undermine and support each other. The workings of the discursive strategies of legitimation and resistance to identity are fundamental to exploring how participants interpret and understand being homosexual. To expect a coherent, fixed/unchanging interpretation or understanding of gayness is, however, inconsistent with the nature of social reality. And it is perhaps here that a discourse approach shows its usefulness most clearly. A key feature, courtesy of Potter and Wetherell (1987), is the focus on not only consistency, but variation/inconsistency within and between accounts.

At one level then a discourse approach addresses the limitation inherent in much research which has assumed that homosexuals constitute an homogenous group. Here differing accounts, interpretations and understandings of what it means to be gay are allowed for, indeed raised as significant features. At another level, attention to variation within the account of an individual participant allows us to focus on the competing and often contradictory elements that interact to constitute the identity 'homosexual'.Thus the element of variation/inconsistency accords due precedence to the shifting, multiple often contradictory meanings that are constructed around what it means to be gay. Such a focus on the tensions between discourses and the ways in which they reproduce and transform the social world is what Parker(1992) conceptualises as a study of discourse dynamics.

In looking at the ways in which meaning around homosexuality comes to be constructed, our theoretical discussion alluded to the role of institutions, ideology and power. Parker's (1992) approach to discourse analysis is useful in that it allows us to attend to these specific issues in the course of researching homosexuality. Relevant here are criteria that discourses have ideological effects, reproduce power relations, and support institutions, as proposed by Parker. What I attempt to show briefly is that these criteria can provide a frame within which we can explore the social/political implications of the discourses of deviance and pathology for instance.

* Discourses have ideological effects : According to Parker this entails showing how a discourse connects with other discourses which sanction oppression. The discourses of deviance and pathology play themselves out through the stereotype of homosexuality as 'perversion', as a condition falling within the realm of mental illness. These medicalised discourses serve to reproduce the ideology of heterosexuality as normative. They prescribe sexual experiences (desires, practices and relations) as heterosexual by nature. The polar opposite, homosexuality, constitutes a deviation from this ideal and is thus stigmatised as pathological in nature.

* Discourses reproduce power relations: According to Parker this entails looking at which categories of person gain and lose from the employment of the discourse. Applied here, the discourses of deviance and pathology work to undermine the identity 'homosexual', which is produced as invalid and illegitimate. 'Hetero' is accorded 'master status', reinforced and perpetuated as all powerful and legitimate by the discourses of deviance and pathology.

* Discourses support institutions: According to Parker this entails identifying institutions which are reinforced when this or that discourse is used. Here our attention is drawn to the way in which the discourse of deviance and pathology frames sexuality (sexual desires, practices and relations). In subverting homosexuality and homosexual (read non-reproductive) sex/desire as pathological and deviant, the discourses of deviance and pathology could work to impinge on and constrain the options available to a person in terms of the institution of the family for instance. Working to control and police desire, the discourses force us into certain forms of family, and here I mean heteronormative families, which are deemed to be 'natural' and the preferred societal ideal. This was then briefly some of the ways in which Parker's (1992) approach is relevant for researching homosexuality through a discursive lens.

Concluding comments

Parker and Burman (1993) contend that a focus on reflexivity is one of the key moral/political consequences of 'doing' discourse analysis. In concluding my paper then, I'd like to raise some reflexive considerations around my position within the research. The question I find myself asking is "how do I reflect on the construction of meanings in the context of researching homosexuality when these meanings are so contested and contextualised in nature?". It needs to be taken into account that in the course of researching homosexuality my discursive lens is always already tinted by and framed within the discourses of my 'race', gender' (hetero)sexual orientation, and my position within the discipline of psychology (amongst others). Clearly my stance is not one of the detached and all-knowing observer. Rather I engage in what Banister and others (1994) refer to as 'doing research with people', co-constructing the varied and often inconsistent meanings that are made around homosexuality. But inherent in this process are certain tensions that we need to acknowledge; namely that I am partly wanting to construct my own meanings around homosexuality, and yet partly at the mercy of these shifts in meaning. It would appear that in researching homosexuality through a discursive lens, my stance is one of limited agency. I cannot, within the territories of research, transform the ways in which homosexuality is constructed and negotiated discursively, because the ways in which meanings around homosexuality have been contested and transformed over time have been through communal agency (at a broader social level, in other words). So, even though I might attempt to research homosexuality in a participative and reflexive way, it would appear that my position still remains, in the final analysis, one of the outsider.


Banister, P., Burman, E., Parker, I., Taylor,M. and Tindall,C. (1994) Qualitative Methods in Psychology : A Research Guide, Buckingham : Open University Press

Blachford, G. (1981) 'Male dominance and the gay world' in K. Plummer (ed) The Making of the Modern Homosexual, New Jersey : Barnes and Noble

Burman, E. and Parker, I. (1993) (eds) Discourse Analytic Research : Repertoires and readings of texts in action, London : Routledge

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

Cornwall, A. and Lindisfarne, N. (eds) (1994) Dislocating Masculinity : Comparative Ethnographies, London : Routledge

De Cecco, J. P. (1981) 'Definition and meaning of sexual orientation', Journal of Homosexuality, 6(4), 51-67

Dollimore, J. (1996) 'Bisexuality, heterosexuality, and wishful theory', Textual Practice, 10(3), 523-539

Dyer, R. (1992) 'Getting over the rainbow : identity and pleasure on gay cultural politics' in R.Dyer (ed) Only Entertainment, London : Routledge

Edwards, D. and Potter, J. (1992) Discursive Psychology, London : Sage

Edwards, T. (1994) Erotics & Politics : Gay Male Sexuality, Masculinity and Feminism, London : Routledge

Epstein, S. (1992) 'Gay politics, ethnic identity : The limits of social constructionism' , in E. Stein (ed) Forms of Desire : Sexual Orientation and the Social Constructionist Controversy, Routledge :London

Forrest, D. (1994) 'We're here, we're queer, and we're not going shopping : Changing gay male identities in contemporary Britain' in Cornwall, A. and Lindisfarne, N. (eds) Dislocating Masculinity : Comparative Ethnographies, London : Routledge

Foucault, M. (1976) The History of Sexuality, Volume 1 : An Introduction, Harmondsworth : Penguin

Gough, J. (1989) ' Theories of sexual identity and the masculinization of the gay man' in Shepherd, S. and Wallis, M. (eds) Coming On Strong : Gay Politics and Culture, London : Unwin Hyman

Marshall, J. (1981) 'Pansies, perverts and macho men : Changing conceptions of male homosexuality' in K. Plummer (ed) The Making of the Modern Homosexual, Hutchinson : London

McIntosh, M. (1968) "The homosexual role" Social Problems 16, 92-193 reprinted with postscript in E. Stein (ed.) Forms of Desire : Sexual Orientation and the Social Constructionist Controversy, Routledge :London

Padgug, R. (1992) 'Sexual matters : On conceptualizing sexuality in history' in E. Stein (ed.) Forms of Desire : Sexual Orientation and the Social Constructionist Controversy, Routledge :London

Parker, I. (1992) Discourse Dynamics : Critical Analysis for Social and Individual Psychology, Routledge : London

Plummer, K. (1981) (ed) The Making of the Modern Homosexual, Hutchinson: London

Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology : Beyond Attitudes and Behaviour, Sage : London

Potter, J. and Wetherell, M. (1994) 'Analyzing discourse' in Bryman, A & Burgess, R.G. (eds) Analyzing Qualitative Data, London : Routledge.

Segal, L. (1990) Slow Motion : Changing Masculinities, Changing Men, Virago : London

Tiefer, L. (1992) 'Social constructionism and the study of human sexuality' in E. Stein (ed) Forms of Desire : Sexual Orientation and the Social Constructionist Controversy, Routledge :London

Weeks, J. (1981) 'Discourse, desire and sexual deviance : some problems in a history of homosexuality' in K. Plummer (ed) The Making of the Modern Homosexual, Hutchinson : London

Wetherell, M. and Potter, J. (1988) 'Discourse analysis and the identification of interpretative repertoires, in C. Antaki (ed) Analysing Everyday Explanation : A Case Book of Methods, London : Sage

Wilton, T. (1995) 'What is this thing called? Models of sexual identity' in

T. Wilton (ed) Lesbian Studies : Setting an Agenda, Routledge :London

Pravani Naidoo is a lecturer in the department of psychology at Vista University, Bloemfontein.

Paper presented at the 3rd Annual Qualitative Methods Conference: "Touch me I'm sick"
8 & 9 September 1997, University of South Africa Regional Office, Durban
critical methods society - -