Barebacking and sexual position

Abstract

Bareback sex continues to fuel the HIV epidemic among men who have sex with men but despite the fact that much academic attention has been focused on the sexual behaviour of this population few authors have considered the significance of sexual position. In order to explore this relatively under-examined factor, interviews were conducted with 13 HIV-negative and unknown status gay men who had recently engaged in bareback sex. Using Interpretative Phenomenological Analysis and through the lens of sexual position, the findings were organized across three super-ordinal themes. There were some areas in which there was little difference between the men's experiences of engaging in bareback as tops or bottoms (for instance, how participants connected with barebacking partners). In other areas, however, there were clear differences in men's experiences according to sexual position, particularly in the interpersonal dynamic between tops and bottoms during bareback sex encounters, which, it is argued, were acted out in accordance with a barebacking ‘sexual script’. There were further differences by position in how individuals overcame ‘cognitive dissonance’ by invoking strategies to make their engagement in bareback sex safer and in the meanings men ascribed to bareback sex and semen exchange. These findings provide valuable insights for those working with MSM around HIV prevention.

Keywords

Bareback sex gay men sexual behaviour sexual position

Introduction

In the UK, men who have sex with men (MSM) remain disproportionately affected by human immunodeficiency virus (HIV). For the first time since the mid-1990s the number of new diagnoses among MSM has surpassed that of heterosexuals with MSM now accounting for just under half of the 98,400 people living with HIV in the UK and for 51% of all new HIV diagnoses (Public Health England, 2013). HIV in MSM is almost exclusively acquired through condomless anal sex (Golden et al., 2004), which is known colloquially by gay men in particular as ‘barebacking’ (Carballo-Dieguez et al., 2009; Huebner et al., 2006). During a barebacking encounter there are a range of factors that can increase the likelihood of HIV transmission, including the viral load of the HIV-positive partner (Miller et al., 2010), whether there is a co-existent sexually transmitted infection (STI) (Ward and Ronn, 2010), and the sexual position an individual adopts (Caceres and van Griensven, 1994). In addition to the manifest physical differences between the insertive and receptive positions in anal sex, there are marked differences in the risk in relation to HIV for each position. The risk of transmission per sexual act for men adopting the insertive role during anal sex is 0.06, whereas the risk for men adopting the receptive role is 1.11 and this is further elevated if internal ejaculation occurs (Benn et al., 2011).

Within gay culture the use of self-ascribed labels to designate the two positions (insertive and receptive) adopted in anal sex is common and such labels are viewed by many men as an important aspect of their identity (Moskowitz et al., 2008; Wei et al., 2011; Zheng et al., 2012). The man who inserts his penis is known colloquially as the ‘top’, or ‘active’, partner and the man who receives the penis is known as a ‘bottom’, or the ‘passive’ partner, with the term ‘versatile’ being used for someone who engages in both (Moskowitz et al., 2008; Zheng et al., 2012). These terms, however, are subject to a process of constant refinement (Dowsett et al., 2008) and have developed to include a multitude of prefixes that further specify an individual's particular preferences, practices, abilities and attributes (Moskowitz et al., 2008; Wei et al., 2011), such as ‘versatile-top’ or ‘fisting-bottom’. In addition, while some men do engage in anal sex practices outside of their self-label (Jameson et al., 2010; Templeton, Jin et al., 2009; Templeton, Millet et al., 2010) there also appears to be a correlation between self-labels for anal sex and sexual position, or role, in other sexual practices. For example, men who identify as top are also more likely to top in other practices such as urination, fisting and role-play (Hart et al., 2003; Moskowitz et al., 2008; Zheng, Hart and Zheng, 2012). For the purposes of this article, we will use the terms top and bottom, as these remain close to the language used by the participants in the present study.

There are socio-cultural dimensions to sexual position that would suggest that it is more than simply a personal preference or predilection. Wei and Fisher (2011) found that white men tended to be evenly distributed across the range of sexual positions but that black men are statistically overrepresented as tops and Asian men overrepresented as bottoms. They also argued that men with a lower educational attainment were more likely to prefer bottoming (Wei and Fisher, 2011). There appears to be an association between bottoming, perceptions of femininity and power differentials, with tops perceived to have an ‘elevated status over the bottom’ (Hart et al., 2003; Wegesin and Mayer-Bahlburg, 2000: 56).

Given the risk differentials between men who adopt the top and those who adopt the bottom positions during bareback sex and the wider cultural meanings attached to sexual position by many MSM, it is perhaps surprising that despite a wealth of literature devoted to bareback sex and sexual risk-taking few authors have examined the relationship between sexual position and bareback sex. Among older men, barebacking has been found to be associated, for tops, with optimism for the future, while older men who bottom were more likely to experience ‘internalised negativity’ such as low self-esteem, self-hatred and beliefs in their own inferiority (Jacobs et al., 2010). Rates of HIV and substance use in this population were high, with 30% of new HIV diagnoses made after the age of 40 years (Jacobs et al., 2010). In relation to substance use, there appears to be an intersection between the types of drug used, partner type, sexual position and bareback sex (Rusch et al., 2004). For example, Rusch et al. (2004) identified that situation-specific methamphetamine use was associated with bareback sex as a bottom, whereas gamma-hydroxybutyric acid (GHB) and ketamine (K) were associated with bareback sex as a top with regular partners. In addition, alcohol use was significantly associated with bareback sex as a bottom and marginally associated with bareback sex as a top, while the association of ecstasy with bareback sex as a bottom was dependant on the use of methamphetamine at or around the time of sex (Rusch et al., 2004).

In contrast to this literature, some report that there are no differences between tops and bottoms in relation to the reasons given for engaging in condomless anal sex (Ostergren et al., 2011) and others assert that there appears to be little difference by sexual position in the willingness to engage in bareback sex or the frequency of bareback encounters (Hart et al., 2003). Given this under-examined and contradictory picture it is the aim of this article to examine the phenomenon of barebacking through the analytical lens of sexual position.

Methods

Following ethical approval, participants were recruited in London between November 2010 and November 2011 using a range of strategies including advertisements in the gay press, posters and flyers in gay venues, email announcements and snowballing (asking participants to identify others who may be willing to be involved). Once recruited, the interviews were conducted in an office at the University, with the exception of one, which was conducted in a participant's home at his request. To secure informed consent the participants were advised of the confidential nature of the interview and were given written information about the study. In addition, participants were asked to provide a pseudonym. The interviews, which were digitally recorded, were unstructured in order to allow the interview to track the concerns and interests of the participant. Although a topic guide was developed and present during the interview, this was primarily to assist in the event of a difficult or reluctant interviewee. At the end of the interview the participant was then given the opportunity to clarify any points that he had made and asked if he had any further points he wished to make. The recordings were then transcribed verbatim; the transcripts were checked against the digital recording for accuracy and any errors were corrected.

The interview material was analyzed using Interpretative Phenomenological Analysis. A preliminary reading of the transcript was undertaken, whilst listening to the digital recording, and initial thoughts were noted. The transcripts were then read and reread so that a depth of analysis was developed through an iterative process, moving from a descriptive analysis to a more interpretative account (Smith et al., 2009). First-order interpretations involved descriptive coding and attempted to explain the subject in context, while the second-order analysis was more critical and conceptual and attempted to consider meaning, or how the participant made sense of his experience of barebacking (Larkin et al., 2006). The aim of the analysis was to produce a transparent, plausible and sensitive thematic account (Smith, 2004). The themes were not chosen for their frequency in the data; rather, the range of themes was mapped. Once the preliminary analysis was complete and each case had been analyzed the initial coded transcripts were uploaded to NVivo9™ and cross-case analysis began. Using the functionality within NVivo9™ the generated themes were subsumed or abstracted into three super-ordinal themes, thus grounding the findings in the data. Yardley's (2000) four principles for ensuring quality in qualitative research (sensitivity to context, commitment and rigor, transparency and coherence, and impact and importance) were applied.

Participant characteristics

A total of 13 HIV-negative or unknown status men who had engaged in recent condomless anal sex were recruited in London. Their average age was 30 years old (range: 29 years to 55 years). Five of the participants stated that they were HIV-negative, following HIV testing outside of the window period and bareback sex that occurred within the parameters of negotiated safety. The other eight participants’ last HIV tests were negative; however, they were deemed of unknown status due to discordant or anonymous bareback sex. Five of the men identified as being ‘tops’ (i.e. anally insertive), five identified as ‘bottoms’ (i.e. anally receptive) and three identified as ‘versatile’ (i.e. engaging in both insertive and receptive anal sex), however the sexual role that participants disclosed did not necessarily reflect the sexual behaviour in their narratives. Participants in the sample were predominantly white. Five of the participants originated from the UK, two from Australia, and one each from The Republic of Ireland, France, Ukraine, North America and Indonesia. With the exception of one participant, who had only lived in London for two months, they had all lived in London for at least three years, the majority for 10 or more. Most of the participants were in a relationship (n = 9). Four of the men in relationships stated that the relationship was ‘closed’ (i.e. they did not have sex with anyone outside the relationship); the other men in relationships stated their relationships were ‘open’ (i.e. they had sex with other men outside the relationship). Most men in open relationships only had sex with other men when their partner was present (i.e. threesomes), with only two men having completely open relationships. One participant had had condomless sex on the day of the interview, seven others had engaged in condomless sex in the preceding week and the rest within the previous three months.

Findings and discussion

There were three super-ordinal themes identified through the analysis:

How men locate their barebacking encounters

The act of bareback sex

The meanings men ascribe to bareback sex

Across these three super-ordinal themes there were areas where there was little difference between the experiences of tops and bottoms, while in other areas differences according to sexual position could be observed. Due to the limitations of space the main focus of this article will be on areas where there were differences between top and bottom narratives.

1. How men locate their barebacking encounters

How participants located their barebacking encounters provided the context for their barebacking experiences and, for men in this study, was an important part of their narratives. Within this first super-ordinal theme there were four subthemes: affective states, connecting with barebacking partners, partner attributes and substance use. There were examples from both top and bottom narratives in each subtheme. Overall, how participants located their barebacking encounters did not appear vastly different between sexual positions, especially in relation to how individuals connected with partners. Where there were differences in men's narratives between sexual positions these tended to be subtle.

Affective states

Affective states are the experiencing of moods, emotions and feelings and can be either positive or negative. Consistent with the literature (BD Adam, Husbands et al., 2005; BD Adam, Sears et al., 2000), the experiencing of negative affective states was associated with barebacking among participants. Although men in both top and bottom narratives discussed negative affective states in relation to barebacking encounters, loneliness characterized by being single as well as being older was specific to men who adopted the bottom position, as exemplified by an excerpt from Richard:

I've never been on my own before. And really struggled to come to terms with living on my own. Uhmm — by this time, I'm coming up to a 48 uhmm — all these conflicting uhmm — things going on in your life uhmm, and, really come to yeah, really come to the conclusion. Really, I – I, I guess really what I'm getting at is really low confidence, low self-esteem. Certain that you know, you're gonna live out the rest of your days as a lonely man. (Richard, 50: bottom narrative)

As in the case of other participants, loneliness exacerbated by age led Richard to visit a local ‘pick up bar’ where he got drunk and met a younger partner with whom he engaged in bareback sex as a bottom. There appears to be an intersection between negative states, such as low mood, substance use and bareback sex, with the latter two being used instrumentally to address the former. While younger people remain most at risk of STIs (Public Health England, 2013) and may be more likely to engage in risky behaviours (Crepaz et al., 2000) there appear to be increases in both risky behaviours, including substance use and STIs among older MSM (Health Protection Agency, 2008; Osmond et al., 2007). In relation to HIV, more than half of the older adults diagnosed with HIV between 2000 and 2007 were over 50 years of age and late diagnosis within this group is common (Smith et al., 2010).

Connecting with barebacking partners

Overall, how participants located their barebacking encounters did not appear vastly different between sexual positions, especially in relation to how individuals connected with partners. Where there were differences in men's narratives between sexual positions these tended to be subtle.

Partner attributes

Having selected a partner, participants stated that specific attributes of that partner could influence their decision whether to bareback. Consistent with the literature there was a complex interplay between conceptions and performances of masculinity (Holmes et al., 2008; Ridge, 2004). Regardless of sexual position, men in this study were attracted to men who displayed masculine physical characteristics such as being ‘built’ or ‘muscular’ as well as behaviours such as being sexually dominant. There were, however, subtle ways in which top and bottom narratives differed in relation to partner attributes. Men who adopted the top position were more likely to engage in bareback sex with casual partners. While in bottom narratives, men would not only engage in bareback sex with an attractive partner, but would relinquish themselves entirely, as James-Lee explains:

when I have sex with somebody that actually is, actually is much better looking than me, I feel like he is actually much better looking than me I will do just everything he wanted me to do. You know. So there is a level of superiority you know, what I like. (James-Lee, 36: bottom narrative)

The perceived difference in partner characteristics was often bound to self-esteem and created a shift in the interpersonal dynamic. This has been described as ‘erotic capital’ (Hakim, 2010), whereby one partner is perceived to have greater ‘sex currency’ than the other. This ‘currency’ is in a constant state of flux, residing in one partner in one situation but not in another, and was more than a matter of traditional aesthetics, including also other physical as well as behavioural aspects, along with factors such as ethnicity, class, education and economics (Ho and Tsang, 2000). This discrepancy can make individuals less equipped to refuse a partner for fear or rejection, placing participants at greater risk of HIV acquisition.

Substance use

Substance use was common among participants and was an integral part of their social and sexual lives. It was accordingly a frequent occurrence in men's barebacking narratives. The use of substances was nuanced. With regard to sexual position, the only difference between top and bottom narratives concerned being rendered incapable because of the substance use:

at one stage I slightly OD'd and slightly passed out and didn't know quite what was going on and at that stage I was more, during the risk, during the drug taking I tend to be more passive, I tend to be more versatile. I'm more up for being passive let's say, or being versatile let's say. Erm so put myself on PEP¹ after the second or third time because of the situation where I'd literally passed out for a while I didn't quite know what was going on, I think there was a certain amount of activity where I wasn't fully aware. (Andrew, 32: bottom narrative)

While none of the participants classed these encounters as non-consensual, drug facilitated sexual assault (DFSA) are encounters in which individuals are subjected to sexual acts while incapacitated or unconscious due to either involuntary or voluntary substance use (Hall and Moore, 2008). In the example given, Andrew consented to take recreational drugs and attend a bareback sex party, and he even actively engaged in bareback sex at the party both as a top and as a bottom with multiple partners, yet his conscious state (i.e. becoming unconscious) brings into question the validity of his consent (Hall and Moore, 2008). Andrew was not aware and nor, I suspect, were his sexual partners that the sexual activity could be construed as opportunistic DFSA. In conjunction with the widespread use of recreational drugs among MSM, especially for sex (Colfax et al., 2004; Purcell et al., 2005), this makes raising awareness of the issue of substance use, consent and sexual activity particularly salient.

2. The act of bareback sex

This super-ordinal theme was concerned with the act of bareback sex: where sexual encounter occurred, how it was negotiated and how participants addressed the ‘cognitive dissonance’ (Festinger, 1957) that engaging in bareback sex caused. In relation to where the bareback sex occurred, there were unsurprisingly no differences between top and bottom narratives.

The negotiation of bareback sex

While there were participants for whom the bareback sex had been negotiated prior to the barebacking encounter, such as those who had used the internet to connect with partners, many barebacking encounters did not begin with an explicit understanding that bareback sex was to occur. Consistent with the literature (P Adam, Teva et al., 2008; Ridge, 2004), most communication and negotiation occurred in silence, with participants and their partners relying on non-verbal modes of communication to convey their desire to engage in bareback sex or assess their partner's willingness to engage in bareback sex. Within men's narratives there were commonalities in the way they described the unfolding encounter which was suggestive of a ‘sexual script’ (Gagnon and Simon, 1974). This barebacking sexual script enabled the negotiation of bareback sex and revealed the interpersonal dynamic between men who adopted the top and men who adopted the bottom position. The barebacking sexual script involved two stages: (i) the presentation of self and (ii) positioning, manoeuvring and testing.

(i) Presentation of self

before we had started actually, erm out of the drawer I took some condoms, some lube and some poppers and put them on the side so they were there … I wouldn't enforce the situation, let other people take control I've laid out the condoms kind of saying I'm happy to use them, there they are. (Andrew, 32: top narrative)

Goffman (1959) suggests that sexual actors use symbols and substitutes to convey information to support their performance and this laying out of condoms and lubricant could be seen in several narratives and it fulfils several functions. First, as seen in this excerpt it abdicates responsibility for initiating bareback sex to a sexual partner. Second, it provides a narrative which fits within the safer sex paradigm by suggesting that an individual intends to use them and because of their long association with safer sex, it could be argued that they could be interpreted as a prop which conveys that an individual is safe. In addition, later in his interview Andrew states that, in his own experience, when he has felt pressured into engaging in bareback sex he was less inclined to do so, therefore he places the condoms out as he doesn't want to ‘enforce the situation’ and put a potential barebacking partner off. Ironically, the laying out of condoms could be used to facilitate a barebacking encounter. This highlights the fact that what may appear to be innocuous within a sexual encounter such as the laying out of condoms, can actually play a significant role in the communication and negotiation of bareback sex.

(ii) Positioning, manoeuvring and testing

there is another process this is what we call it teasing when you do play your dick in front of his arse you know you don't actually stick it in you just rub it in you know like normally it gives a massive turn on when you do that and then this is another process of teasing you slide it in a bit if they guy doesn't refuse it, doesn't mention anything about condoms means you can fuck him until you go all the way through. (James-Lee, 36: top narrative)

The second stage in the barebacking script thus involves a process of positioning, manoeuvring and testing. This second stage would involve participants, or their partner(s), lubricating the top's penis and bottom's anus and stimulating them; this would then progress to placing the penis at the entrance to the anus. Undertaken as part of foreplay, this process also allows for the reading of a partner's willingness to proceed with bareback sex. Foreplay additionally provides a credible alternative to barebacking that fits within a safer sex paradigm, should the behaviour be challenged. Consistent with the literature (BD Adam, Husbands et al., 2008) silence was read as agreement to, or non-refusal of, bareback sex. The reliance on non-verbal means of communicating and assessing is not without its problems, particularly in relation to a shared understanding of the unfolding encounter. For example, non-refusal is not the same as agreement to bareback sex, as silence could also be used to communicate reticence by participants or, as seen in the section on substance use, the non-refusal may be the result of incapacity due to intoxication. The lack of verbal negotiation makes both partners vulnerable during a barebacking encounter.

Overcoming cognitive dissonance

Many participants did not enter sexual encounters having decided that they were going to engage in bareback sex and they talked of an internal dialogue during a barebacking encounter in which they desired bareback sex while at the same time wanted to avoid acquiring HIV. This can be described as ‘cognitive dissonance’ (Festinger, 1957): in order to engage in bareback sex the participant needed to overcome this dilemma as none of the participants sought to acquire HIV. Participants developed their own sophisticated prevention strategies that were based on considerable knowledge regarding HIV transmission and HIV prevention messages which they considered would make their engagement in bareback sex safer: ‘Um and you know there are things that I do to reduce the risk.’ (Peter, 40: top narrative). Some of the strategies were across sexual position, the most common cited being choosing partners that they considered to be low risk for HIV, a practice also known as ‘sero-sorting’ (Dubois-Arber et al., 2012). Sero-sorting was often employed at an individual level with little or no verbal communication. It must therefore have been based on assumptions made about a sexual partner during the encounter, in particular from their physical appearance:

I judge people who I have sex with, if that person, appearance in appearance he has to look really convincing, you know what I'm talking about, he looks like he's got something with him, skinny you know he doesn't look healthy, I wouldn't do it, I wouldn't even go there. I would normally do it with someone who looked perfectly healthy. (James-Lee, 36: versatile narrative)

Sero-sorting could also be employed at a population level, using technology to filter partners perceived to be ‘risky’:

… if a guy on Gaydar usually will say I like barebacking with big letters or whatever I will usually avoid him, but strangely enough that's probably unconsciously to do with risk if somebody is clearly a big barebacker they are positive and that is the assumption that I make. (Peter, 40: top narrative)

Other strategies that were found in both top and bottom narratives included reducing the duration of bareback sex, such as by using condoms for some of the penetration:

So er and in my head goes reduced risk by probably eighty per cent because total length of sexual intercourse is this long and eighty per cent with a condom well that's ok. (Barry, 55: top narrative)

Strategies also included the use of anti-retroviral therapy (ARTs), either directly in the form of Post Exposure Prophylaxis or indirectly by the knowledge of a discordant partner's undetectable viral load: ‘I know that they were HIV-positive, erm and one of them had basically said my viral load “I am undetectable”.’ (Andrew, 32: top narrative). There were, however, some strategies that were specific to the sexual position that an individual adopted.

Most of the participants considered engaging in bareback sex as a top to be less risky than having bareback sex as a bottom and as such this was commonly cited as a strategy for making bareback sex safer. Men would be willing to have bareback sex as a top, but would insist on condoms if they were a bottom:

Yeah without a doubt because as I've said the risk of a top getting HIV it is less than the bottom and if I was the bottom I dare say I would be wanting them to use a condom much more often. (Peter, 40: top narrative)

In addition, being top for bareback sex was often presented with other factors that participants considered would make bareback sex safer such as circumcision status, washing and urinating after sex, using lots of lubricant and not being rough:

I'm less likely to get it than, I wouldn't say most people but than a lot of people because I'm a top, because I'm cut, um those two things help also I do other things like when I'm finished I go and wash my dick and have a piss straight away. (Peter, 40: top narrative)

I generally um I am always the top and um it's the sex is not sort of rough or anything. (Barry, 55: top narrative)

These methods are commonly believed to reduce the likelihood of HIV acquisition (Metsch et al., 2004; Wilson et al., 2001) and some may even be biologically plausible. For example post-coital genital hygiene may reduce HIV acquisition although further research is required (Meier et al., 2006; Short, 2006). The most significant, however, is circumcision. It has been known since the 1980s that circumcised men may be less likely to acquire HIV than those who are not and the results of three randomized control trials in Africa suggest that circumcision dramatically reduced HIV acquisition by up to 60% in heterosexual men (UNAIDS, 2007; Wei et al., 2011; Wiysonge et al., 2011).

After sero-sorting, the position-specific approach to make bareback sex as a bottom safer was the prohibition on internal ejaculation, referred to in a novel coinage by one participant, James-Lee, as ‘half-bareback’. Participants, however, were aware that the prohibition on internal ejaculation was not without its problems. For example, the risks associated with pre-ejaculatory fluid or the danger that an unknown partner could ejaculate inside regardless of the request. There were two main approaches that participants took to this, one was requesting a partner to refrain from ejaculating and the other was with use of condoms for orgasm:

er one of them as with a one night stand who I knew from the pub, er and erm I took him home one night, I was drunk and erm he was hot and I was like I need to be fucked, so fuck me. And he withdrew at the point just before ejaculation … he said I don't have a condom and I said okay fine then don't cum inside me. (Paul, 38: bottom narrative)

So I knew that if I had anal sex um somebody might fuck me for a while without a condom but at the end of the day if he was going to cum he was going to wear a condom. (Mark, 51: bottom narrative)

This approach was also applied by participants when intoxicated:

… another thing is I wouldn't do it with anybody, like with every person even if I'm like high or drunk or whatever … I never allow him to cum inside me, I mean no. So that, that even when I am high it's a big no-no. (Pavel, 36: bottom narrative)

These excerpts clearly demonstrate how men in this study wanted to avoid acquiring HIV. In order therefore to overcome cognitive dissonance and engage in bareback sex participants operationalized a range of strategies (some position-specific), which were based on sophisticated levels of HIV knowledge and which they believed would reduce the likelihood of HIV transmission. The most commonly applied strategy, regardless of sexual position, was the selection of seroconcordant partners. Yet bareback sex occurs with little or no verbal communication, with both men relying on a barebacking sexual script to communicate their desire to bareback. This meant that they had to rely on nonverbal modes of ensuring concordance. A further consideration is that there are gendered constructions of risk, with risk-taking being associated with idealized notions of masculinity (Junge, 2002; Race, 2007); however, the actual risk involved in bareback sex is vastly different according to the sexual position adopted. This means that even if a man who adopts the bottom position desires bareback sex, is inclined to be risky, or desires to please their partner by engaging in bareback sex, they will be acutely aware of the greater risk of doing so. Conversely, men are more willing to engage in bareback sex as a top as they consider this to be less risky; this resulted in examples of tops being persistent in their advances or ‘trying it on’ by initiating bareback sex, which could be difficult for some of those who adopted the bottom position to resist.

3. Meanings men ascribed to bareback sex

The meanings that men ascribed to bareback sex were central to men's barebacking experiences. Regardless of their self-identified sexual position, participants gave examples of bareback sex in either sexual position and there were commonalities across top and bottom narratives with regard to the meanings they ascribed to bareback sex, for example pleasure. Yet there were also nuances according to sexual position. Many men in bottom narratives said that bareback sex felt the same as a bottom regardless of whether a condom was used or not, suggesting that for some the pleasure was psychological in nature. Conversely, many tops reported condom-related issues, and suggested that forgoing the condom had improved sexual functioning.

Meanings shared across top and bottom narratives

A commonly shared meaning related to the symbolic nature of bareback sex and semen exchange; within narratives men often drew parallels with heterosexual sex and procreation. For many participants, engaging in bareback sex was a representation of closeness, exclusivity and intimacy and was thus a behaviour which, for many, was limited to significant partners because it provides ‘greater closeness with the other person but also in the sense that this being something we definitely don't do with other people’ (Pete, 29: versatile narrative). Engaging in bareback sex within a relationship can be a potent expression of love, trust and commitment (Rhodes and Cusick, 2002; Theodore et al., 2004). In addition, internal ejaculation was considered pleasurable across top and bottom narratives:

… having somebody cum inside you is a turn on it's, you know it's intimate, it's hot it's wet, it's sticky … so yeah I mean there's always that element in mind that it's a turn on. (James, 34: versatile narrative)

For some, the pleasure of internal ejaculation was based on the eroticism that they ascribed to the act:

my biggest ever fantasy which I've only really done once or twice … is a guy getting fucked, preferably really cute guy, getting fucked by a whole pile of guys they all bareback they all cum inside him and then I have a go and when I finish someone else has a go. And er if I'm barebacking a guy it kind of feeds back into that fantasy. (Peter, 40: top narrative)

In this excerpt, insemination is central to Peter's fantasy and barebacking is a vehicle for achieving this. In his fantasy his penis is engulfed by his partner's anus, but also the other men's intermingling semen. In men's narratives there was a circularity of actual barebacking experiences and eroticism. In Peter's narrative he acknowledges that this recurring fantasy feeds into his barebacking experiences and later in his interview he describes in vivid detail a barebacking encounter with a casual partner in which his semen was ‘dripping out of his partner's arse’. While the sharing of semen was erotic and because the symbolic nature of internal ejaculation was significant for most participants, there were, however, meanings that were also specific to top and bottom narratives.

Meanings specific to top narratives

There were several themes that were specific to top narratives. For example, men in top narratives did not restrict internal ejaculation to romantic or significant partners. This perhaps reflected the different risks associated with insertive rather than receptive anal sex. Furthermore, in top narratives internal ejaculation was associated with concepts of ownership, expressions masculinity and aggression:

I suppose I feel that ok um er you know that, was going to say that I've achieved something, sort of some sort of putting a stamp on it and ownership, you know left my mark inside someone else. (Barry, 55: top narrative)

… it feels good, I mean I love it, especially if I come inside them it's kind of a slightly macho, I don't know if it's a control freak thing or if it's a macho thing but it's a turn on, a big turn on. (Peter, 40: top narrative)

… if I feel aggressive toward a guy and it really is a ‘alright you asked for it you gonna get it kind of thing’. Then if I’m feeling aggressive I cum inside him because I feel like really fucking the shit out of him um and as soon as you put on a condom that dynamic goes away. (Mark, 51: top narrative)

These excerpts conjure images of masculine penetrators exercising control and fostering ownership through internal ejaculation, which in turn was seen as erotic. The connection between partners was not just romantic but one in which the bottom is conquered, further perpetuating ‘dichotomous and fixed notions of gender differences’ (Moore, 2002:13). As suggested by Dowsett et al. (2008) this highlights the rendering of one group of MSM ‘bottoms’ as being in some way subordinate to other MSM ‘tops’ within the homosexual order of desire. The foregoing excerpts show bottoms constructed as obstacles to be overcome in pursuit of pleasure, and perhaps reflect the perception by both tops and bottoms that bottoms were the ‘gatekeepers’ to the tops’ engagement in bareback sex (Hoppe, 2011).

Meanings specific to bottom narratives

Although many men reported that sex as a bottom felt the same regardless of condom use in physiological terms, many men found bareback pleasurable because of the meanings they ascribed to it:

I enjoyed the fact that he ejaculate inside me because it's like – going back to what I said before, it's-it's just something from, it's kind of like, I-I-I think of it as kind of it's the essence of — a man really … it's kind of almost indefinable what — it's it's to me, it's him sharing — it's not just him sharing — his body with me. He's sharing, you know — yes — sharing you know — cuming inside me, you know, it's him physically giving me that, that fluid inside me. (Richard, 50: bottom narrative)

I only see him once or twice a week so again it sounds corny but I-I-I I like the feeling that knowing that he's, he's inside me as well [laugh] … the enjoyment of that-that-that feeling that they are still with you even though they've probably left the house. (Robert, 31: bottom narrative)

These excerpts show that bareback sex as a bottom was seen as intimate and when coupled with internal ejaculation could intensify their sexual experience. Internal ejaculation, however, was something that bottoms (unlike tops) limited to romantic partners, perhaps reflecting the elevated risk associated with this practice. Seminal fluid was considered the symbolic ‘essence of man’ as it was a secretion that originated from a partner's genitals and, in addition, it contained a partner's DNA and reproductive potential and was intrinsically of them. Furthermore, because of the association between semen and procreation and the fact the semen contains the partner's DNA some participants paralleled barebacking with internal ejaculation to heterosexual sex. Accordingly, during bareback sex with internal ejaculation the top is not just sharing his body but also part of himself. This added another dimension of intimacy and promoted connection. In insemination, the top leaves part of himself; the seminal fluid can then be held internally by the bottom and this physical holding can have figurative or symbolic resonances.

Conclusion

Among this group of HIV-negative and unknown status gay men sexual position is an important factor in how they make sense of their barebacking encounters and there were clearly differences in the experiences of engaging in bareback sex as a top and as a bottom. The act of bareback sex is imbued with much symbolic meaning (Holmes and Warner, 2005) and while some of these meanings were shared across top and bottom narratives, there also appeared to be two different sexual scripts, each specific to the sexual position a participant adopted. When engaging in anal sex, individuals bring into play in intrapsychic scripting a set of beliefs about how to ‘execute’ the role and about their partner's expectations (Whittier and Melendez, 2004). The present study shows how tops and bottoms engage, to some extent differently, in such scripting through barebacking sexual scripts. Yet such sexual scripts have the potential to position men in socio-sexual worlds that could hinder, or facilitate, their capacity to navigate and understand those sexual scenarios (Hoppe, 2011) which place them at risk of transmitting or acquiring sexual infections, including HIV. For example, Hoppe (2011) found that pleasure for bottoms is not only derived from the physical pleasure of anal sex but also from giving pleasure to their partner (relational pleasure) or submitting to their partner (relational power). He argues that because of these scripts, although some bottoms are able to negotiate condom use, negotiation could be problematic if pleasing a partner were central to their own pleasure (Hoppe, 2011).

When engaging in bareback sex, participants were not abandoning safer sex completely because in order to engage in bareback sex they overcame cognitive dissonance by operationalizing a range of techniques that they believed would make their bareback sex safer, some of which were specific to sexual position. This demonstrates that men are attempting to mitigate risk by utilizing complex knowledge, information and negotiation. This contrasts with the way in which individuals are often constructed in HIV-prevention discourses as unknowledgeable or incompetent (Aguinaldo and Myers, 2008). Future HIV-prevention strategies would benefit from acknowledging the significance of sexual position in how individuals make sense of bareback sex and their barebacking partners. In addition, as condoms are less of a feature in these men's sexual experiences there is a need for novel strategies that go beyond condoms (Goldhammer and Mayer, 2011).

Footnotes

Acknowledgements

We are indebted to the participants for their openness and honesty.

Funding

This research has been supported with National Institute for Health Research Clinical Doctoral Research Fellowship and an Imperial College Health Charity Trustees non-medical Research Award.

Note

Matthew Grundy-Bowers is a consultant nurse at Imperial College Healthcare NHS Trust and recently completed a PhD on a Clinical Doctoral Research Fellowship at City University London. He continues to work clinically at Imperial and holds an honorary senior lecturer post at City as well as volunteering with gay men’s health charities.

Sally Hardy is Professor of Mental Health Nursing and Practice Innovation and Head of the Department of Mental Health and Learning Disabilities at London Southbank University. She is also Honorary Professor at University College Hospital, London and Adjunct Professor at The Faculty of Medicine, Nursing and Health, Monash University, Australia.

Eamonn McKeown is Senior Lecturer at the School of Health Sciences, City University London. He obtained his BA (Hons) and PhD, both in Social Anthropology, from Queen's University, Belfast. As a Senior Research Fellow at University College London he was the lead researcher on the first large-scale study of lesbian, gay and bisexual mental health in Britain.

References

Adam

Husbands

(2005) AIDS optimism, condom fatigue, or self-esteem? Explaining unsafe sex among gay and bisexual men. The Journal of Sex Research 42(3): 238–248.

Adam

Husbands

(2008) Silence, assent and HIV risk. Culture, Health & Sexuality 10(8): 759–772.

Adam

Sears

Schellenberg

(2000) Accounting for unsafe sex: Interviews with men who have sex with men. The Journal of Sex Research 37(1): 24–36.

Adam

Teva

de Wit

(2008) Balancing risk and pleasure: Sexual self-control as a moderator of the influence of sexual desires on sexual risk taking in men who have sex with men. Sexually Transmitted Infections 84(6): 463–467.

Aguinaldo

Myers

(2008) A discursive approach to disinhibition theory: The normalisation of unsafe sex among gay men. Qualitative Health Research 18(2): 167–181.

Benn

Fisher

Kulasegaram

(2011) UK Guideline for the use of post-exposure prophylaxis for HIV following sexual exposure (produced on behalf of the BASHH PEPSE Guidelines Writing Group Clinical Effectiveness Group). International Journal of STD and AIDS 22(12): 695–708.

Caceres

van Griensven

(1994) Male homosexual transmission of HIV-1. AIDS 8(8): 1051–1061.

Carballo-Dieguez

Ventuneac

Bauermeister

(2009) Is ‘bareback’ a useful construct in primary HIV prevention? Definitions, identity and research. Culture, Health and Sexuality 11(1): 56–65.

Colfax

Vittinghoff

Husnik

(2004) Substance use and sexual risk: A participant- and episode- level analysis among a cohort of men who have sex with men. American Journal of Epidemiology 159(10): 1002–1012.

10.

Crepaz

Marks

Mansergh

(2000) Age-related risk for HIV infection in men who have sex with men: Examination of behavioral, relationship and serostatus variables. AIDS Education and Prevention 12(5): 405–415.

11.

Dowsett

Williams

(2008) Taking it like a man: Masculinity and barebacking online. Sexualities 11(1–2): 121–141.

12.

Dubois-Arber

Jeannin

(2012) Risk reduction practices in men who have sex with men in Switzerland: Serosorting, strategic positioning, and withdrawal before ejaculation. Archives of Sexual Behavior 41(5): 1263–1272.

13.

Festinger

(1957) A Theory of Cognitive Dissonance, Stanford, CA: Stanford University Press.

14.

Gagnon

Simon

(1974) Sexual Conduct: The Social Sources of Sexuality, London: Hutchinson and Co.

15.

Goffman

(1959) The Presentation of Self in Everyday Life, London: Penguin Books.

16.

Golden

Brewer

Kurth

(2004) Importance of sex partner HIV status in HIV risk assessment among men who have sex with men. Journal of Acquired Immune Deficiency Syndromes 36(2): 734–742.

17.

Goldhammer

Mayer

(2011) Focusing on sexual health promotion to enhance preventive behaviors among gay men and other men who have sex with men: Report from a state-of-the-art conference. AIDS Behavior 15: S1–S8.

18.

Hakim

(2010) Erotic capital. European Sociological Review 26(5): 499–518.

19.

Hall

Moore

(2008) Drug facilitated sexual assault: A review. Journal of Forensic and Legal Medicine 15(5): 291–297.

20.

Hart

Wolitski

Purcell

(2003) Sexual behavior among HIV-positive men who have sex with men: What's in a label? The Journal of Sex Research 40(2): 179–188.

21.

Health Protection Agency (2008) Sexually Transmitted Infections among over 45s on the increase. Available at: http://webarchive.nationalarchives.gov.uk/20140714084352/http://www.hpa.org.uk/NewsCentre/NationalPressReleases/2008PressReleases/080627sti/ (accessed 16 January 2015).

22.

Tsang

(2000) Negotiating anal intercourse in inter-racial gay relationships in Hong Kong. Sexualities 3(3): 299–323.

23.

Holmes

Warner

(2005) The anatomy of a forbidden desire: Men, penetration and semen exchange. Nurse Inquiry 12(1): 10–20.

24.

Holmes

Gasaldo

(2008) Bareback sex: A conflation of risk and masculinity. International Journal of Men's Health 7(2): 171–191.

25.

Hoppe

(2011) Circuits of power, circuits of pleasure: Sexual scripting in gay men's bottom narratives. Sexualities 14(2): 193–217.

26.

Huebner

Proescholdbell

Nemeroff

(2006) Do gay and bisexual men share researchers' definitions of barebacking? Journal of Psychology and Human Sexuality 18(1): 67–77.

27.

Jacobs

Fernandez

Ownby

(2010) Factors associated with risk for unprotected receptive and insertive anal intercourse in men aged 40 and older who have sex with men. AIDS Care 22(10): 1204–1211.

28.

Jameson

Celum

Manhart

(2010) The association between lack of circumcision and HIV, HSV-2, and other sexually transmitted infections among men who have sex with men. Sexually Transmitted Diseases 37(3): 147–152.

29.

Junge

(2002) Bareback sex, risk, and eroticism: Anthropological themes (re-) surfacing in the post-AIDS era. In: Lewin

Leep

(eds) Out in Theory: the Emergence of Lesbian and Gay Anthropology, Champaign, IL: University of Illinois Press, pp. 186–221.

30.

Larkin

Watts

Clifton

(2006) Giving voice and making sense in interpretative phenomenological analysis. Qualitative Research in Psychology 3(2): 102–120.

31.

Meier

Bukusi

(2006) Independent association of hygiene, socioeconomic status, and circumcision with reduced risk of HIV infection among Kenyan men. JAIDS Journal of Acquired Immune Deficiency Syndromes 43(1): 117–118.

32.

Metsch

McCoy

(2004) Prevention myths and HIV risk reduction by active drug users. AIDS Education and Prevention 16(2): 150–159.

33.

Miller

Rosenberg

(2010) Role of acute and early HIV infection in the sexual transmission of HIV. Current Opinion in HIV and AIDS 5(4): 277–282.

34.

Moore

(2002) Extracting men from semen: Masculinity in scientific representations of sperm. Social Text 20(4): 91–119.

35.

Moskowitz

Rieger

Roloff

(2008) Tops, bottoms and versatiles. Sexual and Relationship Therapy 23(3): 191–202.

36.

Osmond

Pollack

(2007) Changes in prevalence of HIV infection and sexual risk behavior in men who have sex with men in San Francisco: 1997–2002. American Journal of Public Health 97(9): 1677–1683.

37.

Ostergren

Rosser

Horvath

(2011) Reasons for non-use of condoms among men who have sex with men: A comparison of receptive and insertive role in sex and online and offline meeting venue. Culture, Health and Sexuality 13(2): 123–140.

38.

Public Health England (2013) HIV in the United Kingdom 2013 Report: Data till the end 2012, London: Public Health England.

39.

Purcell

Moss

Remien

(2005) Illicit substance use, sexual risk, and HIV-positive gay and bisexual men: Differences by serostatus of casual partners. AIDS 19: S37–S47.

40.

Race

(2007) Gay men and the risk of HIV prevention. In: Hannah-Moffat

O'Malley

(eds) Gendered Risks, New York: Routledge, pp. 99–126.

41.

Rhodes

Cusik

(2002) Accounting for unprotected sex: Stories of agency and acceptability. Social Science and Medicine 55(2): 211–226.

42.

Ridge

(2004) ‘It was an incredible thrill’: The social meanings and dynamics of younger gay men's experiences of barebacking in Melbourne. Sexualities 7(3): 259–279.

43.

Rusch

Lampinen

(2004) Unprotected anal intercourse associated with recreational drug use among young men who have sex with men depends on partner type and intercourse role. Sexually Transmitted Diseases 31(8): 429–498.

44.

Short

(2006) New ways of preventing HIV infection: Thinking simply, simply thinking. Philosophical Transactions of the Royal Society B: Biological Sciences 361(1469): 811–820.

45.

Smith

(2004) Reflecting on the development of interpretative phenomenological analysis and its contribution to qualitative research in psychology. Qualitative Research in Psychology 1(1): 39–54.

46.

Smith

Flowers

Larkin

(2009) Interpretative Phenomenological Analysis: Theory, Method and Research, London: SAGE.

47.

Smith

Delpech

(2010) HIV transmission and high rates of late diagnoses among adults aged 50 years and over. AIDS 24(13): 2109–2115.

48.

Templeton

Jin

Mao

(2009) Circumcision and risk of HIV infection in Australian homosexual men. AIDS 23(17): 2347–2351.

49.

Templeton

Millet

Grulich

(2010) Male circumcision to reduce the risk of HIV and sexually transmitted infections among men who have sex with men. Current Opinion in Infectious Diseases 23(1): 45–52.

50.

Theodore

Duran

(2004) Intimacy and sexual behavior among HIV-positive men-who-have-sex-with-men in primary relationships. AIDS and Behavior 8(3): 321–331.

51.

UNAIDS (2007) New Data on Male Circumcision and HIV Prevention: Policy and Programme Implications. Available at: http://data.unaids.org/pub/Report/2007/mc_recommendations_en.pdf (accessed 20 March 2015).

52.

Ward

Ronn

(2010) Contribution of sexually transmitted infections to the sexual transmission of HIV. Current Opinion in HIV and AIDS 5(4): 305–310.

53.

Wegesin

Meyer-Bahlburg

(2000) Top/bottom self-label, anal sex practices, HIV risk and gender role identity in gay men in New York City. Journal of Psychology and Human Sexuality 12(3): 43–62.

54.

Wei

Fisher

(2011) Preference for and maintenance of anal sex roles among men who have sex with men: Sociodemographic and behavior correlates. Archives of Sexual Behavior 40(4): 829–834.

55.

Wei

Raymond

McFarland

(2011) What is the potential impact of adult circumcision on the HIV epidemic among men who have sex with men in San Francisco? Sexually Transmitted Diseases 38(4): 353–355.

56.

Whittier

Melendez

(2004) Intersubjectivity in the intrapsychic sexual scripting of gay men. Culture, Health and Sexuality 6(2): 131–143.

57.

Wilson

Usküla

Feldman

(2001) A case-control study of beliefs and behaviors associated with sexually transmitted disease occurrence in Estonia. Sexually Transmitted Diseases 28(11): 624–629.

58.

Wiysonge C, Kongnyuy E, Shey M, et al. (2011) Male circumcision of homosexual acquisition of HIV in men (Review). The Cochrane Library (6). Available at: http://apps.who.int/rhl/reviews/langs/CD003362ru.pdf (accessed 20 March 2015).

59.

Yardley

(2000) Dilemmas in qualitative health research. Psychology and Health 15(2): 215–228.

60.

Zheng

Hart

Zheng

(2012) The relationship between intercourse preference positions and personality traits among gay men in China. Archives of Sexual Behaviour 41(3): 683–689.