‘Never judge a book by its cover?’: students’ understandings of lesbian, gay and bisexual appearance

This research aimed to explore (predominantly heterosexual) students’ perceptions of sexuality and appearance. A short qualitative survey, which contained questions about the ‘typical appearance’ of lesbians, gay men, bisexual and heterosexual people, was completed by 36 university students. Previous research on dress and appearance in relation to sexuality has mainly focused on lesbian, gay or ‘queer’ individuals and communities. Minimal research has considered whether heterosexual people recognise non-heterosexuality through the dress and appearance of lesbians, gay men and bisexual people, and it would seem that previous studies have not explicitly considered the notion that heterosexuality might also be recognisable through appearance. Data were analysed using thematic analysis. Findings indicated that while students were able to provide appearance norms for lesbians and gay men (which conformed to those identified in previous research), they were also reluctant to give credence to (what they perceived as) ‘stereotypes’ of sexuality and appearance. They described heterosexual men and women in ways that conformed to ‘traditional’ gender norms, but were less able to identify any appearance norms for bisexual people, reflecting the invisibility of bisexuality within Western culture.


Introduction
It has been theorised that visibility is inextricably interwoven with domination and oppression, subjectification and objectification. Visibility is linked to validation; those who are individually and socially visible are more likely to be dominant within wider culture (Brighenti, 2007). Further, to be represented can be a powerful tool because it allows the possibility of challenging misunderstanding and misrepresentation (Gamson, 1998). Visibility can allow both literal spaces (e.g. communities) and experiential spaces (e.g. being able to recognise others 'like us' or hearing about 'ourselves' in mainstream mass media) to exist. Critically, the existence of 'space' allows room for people to position themselves; hence, a subjective, rather than an objective, identity can be expressed, acknowledged and validated. The process of visibility is self-perpetual; as individuals become visible, the groups(s) to which they belong become more represented. The more a group is represented, the more individuals become visible as members of that group (Freitas, Kaiser, & Hammidi, 1996). Therefore, to be visible can be affirmative in validating personal, social and cultural identities (Gamson, 1998;Walker, 2001). In contrast, those who are invisible are not seen, and are therefore likely to be overlooked, oppressed and 'kept in their place' as othered (Brighenti, 2007). This research is underpinned by a social constructionist approach (Burr, 2003); hence, identity and visual identity are understood to be historically and culturally created and maintained (Eves, 2004). A small body of literature indicates that lesbians and gay men can choose to adhere to particular appearance norms in order to visibly embody and express their sexual identity (e.g. Hutson, 2010;Krakauer & Rose, 2002). Shared visual identities (recognising other lesbians and gay men and being recognised as a lesbian/gay man) can be an intrinsic part of forming both individual identities and collective identities and communities, which have historically been safe spaces away from a voyeuristic and homophobic gaze (Hutson, 2010;Krakauer & Rose, 2002). Appearance norms are most commonly situated within these shared spaces because to dress appropriately can signal belonging and membership within them (Rothblum, 1994). The most documented and recognised looks within popular cultural and academic sources are of masculine or butch lesbians and effeminate gay men (Clarke & Turner, 2007). However, there are a diverse range of (everchanging) gay and lesbian visual images available within contemporary Western culture (Eves, 2004;Hutson, 2010). Research and theory indicate that these (multiple) 'looks' are created primarily through clothes (brand choice and how clothing is worn), piercings and hairstyles (Clarke & Turner, 2007;Hutson, 2010), and images may also be nuanced according to the intersections of age, race/ethnicity and social class, as well as specific sexual preferences (Rothblum, 1994;Taylor, 2007). However, in stark contrast no distinct set of appearance norms for bisexual people appear to have been recognised or acknowledged, if indeed they exist at all (Hayfield, 2011;. The term 'gaydar' (a blend of 'gay' and 'radar') specifically defines the ability of gay men and lesbians to identify each other (Woolery, 2007). However, lesbians and gay men walk a tightrope of visibility when negotiating their self-presentation because when the visual image of a group becomes 'supravisible' then there is a risk that the group are perceived as a threat to social constraint, which can result in 'moral panic'. This may put them at risk of accusations of taking up 'too much space' and result in the group being policed, controlled and constrained (Brighenti, 2007, p. 330).This can be seen in the example of the butch lesbian, whose appearance is so visible that it marks her out as different (Walker, 2001).
However, only minimal research (most of which is rather dated, as discussed below) has investigated whether heterosexual people recognise sexuality through appearance. This is important because while appearance can function affirmatively within gay and lesbian space, if heterosexual people recognise lesbian and gay sexuality then it is possible that visibility can become vulnerability within the context of wider society. Lesbians and gay men who choose to make clear their sexuality through how they dress and appear potentially risk rejection and disapproval, and may be at higher risk of discrimination, hate crimes and homophobic violence (Eves, 2004;Krakauer & Rose, 2002). It is for these reasons that more subtle codes of appearance may also exist; hence, lesbian and gay 'looks' can be subtle and about not only what is worn, but also about how it is worn (Kaiser, 1998).
Since the 1970s, US psychologists and sociologists have investigated the 'attitudes' 1 that heterosexual people hold towards gay men and lesbians (e.g. Eliason, Donelan, & Randall, 1993;Kite & Deaux, 1987;Laner & Laner, 1980;Levitt & Klassen, 1976). Some researchers have included questions about appearance within these (mainly experimental) studies which have usually been conducted with student populations. Mixed findings have been produced. Some participants believed that non-heterosexuals were identifiable through appearance and mannerisms (how they walked, talked and acted), often because of a perception that gay men and lesbians would behave like the 'opposite' sex (e.g. Eliason et al., 1993;Kite & Deaux, 1987;Laner & Laner, 1979, 1980; also see Clarke, Hayfield, & Huxley, 2012). Others did not believe that non-heterosexuals were easy to identify (Laner & Laner, 1980;Nyberg & Alston, 1977). It is possible that heterosexual people may be unable to read sexuality through appearance because they are less likely to be 'in the know' (Clarke & Turner, 2007, p. 267) about specific dress codes and clothing practices. The current study aims to explore a group of (predominantly) heterosexual students' perceptions of the (stereo)typical appearance of lesbians, gay men and bisexual and heterosexual women and men, and whether they are able to recognise appearance norms in relation to sexuality, using a qualitative survey (or open-ended questionnaire) technique.

Method
This study was developed as part of a wider research project which sought to investigate the (in)visibility of bisexuality and (lack of) bisexual visual identities (Hayfield, 2011). Thirty-six psychology undergraduate students completed a short pen-and-paper qualitative survey. A survey was chosen because it allowed students to participate anonymously, and was an effective and practical way to gather data from a relatively large sample (for a small-scale qualitative study) (Toerien & Wilkinson, 2004). The survey consisted of a demographics form, followed by seven open-ended questions about sexuality and appearance. Six questions asked about the appearance of a lesbian woman, gay man, bisexual woman, bisexual man, heterosexual woman and heterosexual man. The wording of the questions was (using the example of bisexual woman) 'If someone asked you to describe what a bisexual woman typically looks like, what would you say? (E.g. in what ways could you potentially recognise a bisexual woman from their appearance?)'. A final 'catch-all' question asked 'Is there anything else you'd like to add about appearance and sexuality?' Participants were allocated a code number, which indicated their gender and sexuality (e.g. P1HW is a heterosexual woman).
Thirty-five (out of 36) participants responded to the demographic questions. The students were aged between 18 and 23 years with an average age of 20 years, and predominantly identified as White (34), middle-class (24) and able-bodied (33). Twenty-five women and 10 men completed the survey, most of whom were heterosexual (29). The remainder included three bisexual women, one bisexual man, one lesbian and one gay man. The handwritten data were typed up and collated into one file of textual responses in the form of a Microsoft Word document (the data were not edited or 'corrected' in any way), then coded and analysed using inductive thematic analysis (Braun & Clarke, 2006). Four main themes were identified: (1) effeminate gay men and butch lesbians; (2) masculine men and feminine women; (3) resisting stereotypes; and (4) invisible bisexuality. In the section 'Results and discussion', I use examples from the data to illustrate each theme in turn.

Effeminate gay men and butch lesbians
Most of the participants 2 portrayed gay men as 'feminine' or 'effeminate' (Clarke & Turner, 2007) and accordingly gay men were consistently presented as 'look[ing] like they take a lot of care off their personal appearance' (P9HM) in ways more traditionally associated with (heterosexual) women. That gay men would value and invest in their appearance was apparent through accounts of gay male sexuality as bound up with (alleged) engagement in 'beauty' practices such as hair removal, wearing make-up and using 'fake tan' products and wearing tightly fitted clothes: 'I imagine gay men to be quite feminine. Taking pride in their appearance and maintaining it (through waxing/fake tanning etc). Often wearing bright and tight clothing which straight men would be too afraid to wear' (P3HW). Similarly, one participant wrote that a gay man 'would perhaps wear fake tan and tight fitting clothes [. . .] Some gay men also wear make-up' (P7HW), while others depicted gay men as having a 'fancy haircut' (P26HM), which could be 'gelled' (P10HM) or 'dyed' (P15GM).
The students' responses clearly mirrored (UK/US) mainstream mass media representations of gay men. In particular, many of the participants' accounts reflected the portrayal of a particular version of camp and effeminate gay male sexuality in popular culture, exemplified through 'out' gay male celebrities such as Julian Clary, Dale Winton and Graham Norton (UK), and 'makeover' shows such as Queer Eye for the Straight Guy (US and UK) and How to Look Good Naked (UK). Gay men are often visible in the media as appearance (and lifestyle) 'experts' (or style conscious 'gay best friends') who save heterosexual women and men from making serious fashion faux pas (Clarke & Turner, 2007). The conflation of gay sexuality and (ef)femininity in these caricaturised portrayals of gay man has been problematised by scholars as heteronormative and oppressive (Gorman-Murray, 2006;Ramsey & Santiago, 2004). What was apparent in the participants' accounts was that men ought to be masculine and that hegemonic masculinity requires a lack of concern with appearance (see Frith & Gleeson, 2004); therefore, gay men were presented as overvaluing their appearance through their excessive (ef)femininity. Accordingly, the gay man was portrayed as the 'slightly over feminine looking male' (P22HW) who is 'very obsessed with appearance' (P7HW).
In most of the students' accounts lesbian women were clearly portrayed as the polar opposites of gay men through images of lesbians as butch and masculine, and therefore as undervaluing their appearance. Some participants used the terms 'butch' (e.g. P4BW; P9HM; P10HM; P19HW) and 'masculine' (e.g. P13HW; P20HM; P22HM; P35HW) in their responses and on the whole lesbians were portrayed as being more akin to (heterosexual) men in their appearance. Lesbian masculinity was apparent because in these students' responses lesbians are 'generally unconcerned with appearance' (P3HW) and wear 'no make-up' (P33HW). A 'typical lesbian' was presented as 'not very girly, wearing jeans and t-shirts to show masculinity, short hair' (P34HW) (see Winn & Nutt, 2001), with 'not very good dress sense' (P30HW), presumably a reflection of their lack of femininity. However, a few participants did refer to 'femme' lesbians; for example, one (bisexual) student wrote that lesbian appearance 'depends on the "type" of lesbian. Butch lesbian -masculine, short hair etc but could also look extremely feminine' (P4BW). Femme lesbians were always discussed in relation to butch lesbians, indicating that the femme requires her butch partner for her sexuality to be revealed. This echoes previous literature which highlights the invisibility of femme lesbians within mainstream heterosexual culture (e.g. Eves, 2004).
The accounts were underpinned by a notion of (non-hetero)sexuality as adorned on the body and embodied (Entwistle, 2000). Accordingly, participants' accounts made links between appearance and mannerisms. Gay men were described as 'touching the hair and walking with a different gait' (P19HM) and as 'walk[ing] with a "skip" in their step and walk[ing] quite confidently' (P7HW). This construction of gay men as 'openly camp and... quite girlish' (P11BW) stands in stark contrast to the construction of lesbians as 'butch, built like a tank, [with] huge biceps, [and] plays rugby or golf' (P10HM). The data indicated that appearance extends beyond aesthetics, and appearance and visual identities are linked with both bodies and behaviours.

Masculine men and feminine women
In many of the participant responses, heterosexual men and women were portrayed as directly opposite to each other. It seemed that the key feature of heterosexual men's clothing practices was practicality (Frith & Gleeson, 2004) and therefore they must 'not appear to be concerned with their appearance' (P3HW); their appearance is 'thrown together' (P26HM). Heterosexual men have 'short/shaven hair' (P16LW), and clothes that reflect their enjoyment of 'masculine activities'; the masculine heterosexual man 'wears jeans (and) hoodies' (P14HW). Heterosexual men are 'blokeish, go to the pub, drink lager and like football' (P9HM). However, there were no particular behaviours associated with heterosexual women. Instead, their function was to be attractive, and they were portrayed as 'more into their appearance' (P2HW) and 'well groomed' (P3HW). Heterosexual women 'make a lot of effort with the way they look' (P7HW), they are 'stylish. Feminine' (P13HW), wearing a 'dress/skirt, long hair, make-up' (P33HW) and 'lots of jewellery' (P13HW). Apparently, heterosexual women 'would appear to be aware of the gaze of men around them' (P5), and perhaps for this reason they were portrayed as 'more likely to dress to show off body parts' (P27HW). Their bodies are 'thin' (P15GM), and 'slim' (P6BW), indicating that (heterosexual) women are expected to adhere to restrictive norms in relation to body size and weight (e.g. Tischner & Malson, 2008). In contrast, heterosexual men's bodies are 'muscular' (P19HW) and 'rugged' (P4BW; P27HW). In summary, the portrayal of heterosexual appearance was of appropriately gendered (masculine) men and (feminine) women, reflecting traditional notions of dichotomous gender (e.g. Cameron, 2007).

Resisting stereotypes
This theme captures the tentativeness in the participants' accounts in relation to appearance norms for lesbians, gay men and heterosexual women and men, even though these accounts clearly articulated dominant appearance norms. The word 'stereotype' was included in the information sheet about the study given to participants. However, the participants' accounts indicated that notions of 'typical' appearance or 'stereotypes' did not reflect what they believed about how people actually dressed, or what lesbians and gay men looked like 'in reality'. For example, P27 said: 'there are stereotypes but you just can't tell in real everyday life' (P27HW). In response to the catch-all final question P9 wrote: 'P.S. I'm sorry about the stereotypes I used to answer the questions' (P9HM), while another participant cautioned that it was best to 'never judge a book by its cover' (P13HW). Thus, the participants' accounts resisted the notion of a 'typical' lesbian/gay appearance. It is possible that this resistance reflects a lack of entitlement to comment on the appearance of a group to which most of the participants did not belong and that the term 'stereotype' was used strategically to signal that their accounts did not reflect their views on the appearances of 'real' lesbians and gay men. (While they arguably produced similarly 'stereotypical' images of heterosexual men and women, they relied on the term 'stereotype' less consistently in their responses, and instead used a range of words such as 'normal' and 'average' in their descriptions of heterosexual appearance.) It is possible that the data reflect the participants' heteronormativity and lack of subcultural capital in relation to students' lack of awareness of the operation of appearance norms within non-heterosexual communities, and the positive functions these can have in terms of the creation of individual and group identities (e.g. Peel, 2001). However, it would be interesting to engage in further research which specifically asks students not only about appearance but also about the functions of appearance for both non-heterosexual and heterosexual people.
The tentativeness in the participants' accounts may also be explained by a concern to avoid being positioned as prejudiced. There was a clear 'denial of difference' in the data, for example: 'in my opinion there is no appearance difference in people with different sexualities [ . . . ] It would be easier to make a judgement on someone's sexuality from their individual personality traits' (P17HM). A number of lesbian, gay and bisexual people (LGB) psychologists have argued that psychology students often seek to present themselves as liberal and 'open-minded' when talking about lesbian and gay issues (Clarke, 2005;Ellis, 2009). Liberal approaches to sexuality emphasise 'equality' between heterosexuals and non-heterosexuals and the need for LGB people to be 'treated no differently to anyone else' (Brickell, 2001, p. 222;emphasis added). Because 'sameness' is equated with 'equality' within liberal frameworks, the reverse is also true, so that 'difference' becomes 'inequality'. Consequently, recognising 'difference' is equated with implying 'deficient' (Hicks, 2005). While the participants' accounts provided tentative but clear (and strongly gendered) images of typical gay men and lesbians, and of typical heterosexual men and women, they provided no clear image of the typical bisexual.

Invisible bisexuality
The participants' accounts of the typical bisexual appearance reflected the invisibility of bisexuality, and the failure to acknowledge bisexuality as a valid identity, in both lesbian and gay communities and the wider culture (Barker, 2007;Barker & Langdridge, 2008;Gurevich, Bower, Mathieson, & Dhayanandhan, 2007;Hayfield, 2011). The participants' responses made little or no distinction between the appearance of bisexual men and bisexual women. Their accounts questioned whether there are recognisable appearance norms for bisexuality, for example: 'is there really an actual type? Cannot describe' (P2HM), and 'no way I would be able to distinguish!' (P17HM). While these types of responses sometimes preceded attempts to answer the question more fully, they clearly underscored the lack of a distinct bisexual (visual) identity circulating in the wider culture. For example, some participants' accounts indicated that bisexual women might look 'more heterosexual than lesbian' (P6BW), while others 'expect a bisexual woman to have similar masculine element as a lesbian woman' (P3HW). For these participants, bisexual people might look 'straight' or 'gay' but they cannot look 'bisexual'. It could be argued that the binary model of sexuality which validates heterosexual and homosexual sexual identities but denies the existence of bisexual identities (Bowes-Catton, 2007;Hayfield, 2011) may be a factor in these accounts of bisexual appearance.
Alternatively, participants sometimes reported that bisexual people could 'possibly [be] dressed in more of a Gothic/alternative/emo way' (P8HW). As one bisexual participant wrote: 'many bisexual women I have met have been "Alternative"-gothic cyber etc, so perhaps this would be my typical image [ . . . ] the bisexual men I have met have been more alternative/gothy type' (P4BW). This bisexual participant's account suggests that subcultural capital ('insider knowledge') is required to articulate bisexual appearance norms. However, as these 'alternative' looks are shared with a range of other subcultural groups, any potential for a specifically bisexual visual appearance rapidly disappears (Hayfield, 2011). What can be interpreted from these responses is that bisexual people were sometimes portrayed as likely to draw on the appearance norms of other identities, but that they were not presented as having their own shared visual identities that permitted them the opportunity to be visibly recognisable as bisexual.

Conclusion
Although the data from this particular sample of (predominantly White, heterosexual and middle-class) students indicates a concern to avoid the appearance of prejudice, it also reflects some grasp of appearance norms for lesbians and gay men. However, the participants' accounts only provided rather generalised portrayals, which lacked any appreciation of the nuances, or the functions, of appearance norms in relation to non-heterosexuality. These responses tentatively suggest that the subtleties of particular lesbian and gay 'looks' remain 'insider knowledge', which may be useful in allowing gay men and lesbians to identify each other 'without being identifiable by the dominant culture' (Rothblum, 1994, p. 92). Participants struggled to discuss heterosexuality and appearance, and resorted to drawing on 'traditional' images of (heterosexual) masculine men and feminine women as a result. However, participants grappled even more with discussing bisexuality and appearance and it was clear that they recognised no distinct appearance norms for bisexual men or women. This provides some empirical evidence of the sociocultural invisibility of bisexuality (Barker, 2007;Barker & Langdridge, 2008).
It is clear that further research (with more diverse populations) would be useful in broadening our understanding of perceptions of sexuality and appearance. One of the limitations of qualitative surveys is their lack of flexibility as a data collection tool. The use of more flexible methods such as focus groups or interviews would allow researchers to probe participants and explore contradictions in their accounts. The use of diverse samples would increase our understandings of the intersections of appearance, sexuality and other aspects of identity (such as race and class). Nonetheless, despite its limitations, this research provides evidence of how appearance can serve to make both heterosexual and non-heterosexuality visible on the (clothed) body.