‘Bizarre Sapphic midlife crisis’: (Re)thinking LGBTQ representation, age and mental health

This article looks at viewers’ responses to the romance between two older women on the BBC medical drama Holby City. In the context of a continuing lack of representation of (older) women-loving women, viewers of all ages describe a transformative experience, with an emphasis on positive mental health outcomes – yet older women also orient to something implicitly problematic about this being the case. This is premised, I suggest, in the foregrounding of youth and adolescence in academic and public discussions of the role of the media in sexual self-realisation. The research demonstrates the need for qualitative case studies capturing LGBTQ portrayals, taking account of the experiences of older viewers, and of network television even in a fragmented and queer(er) digital market.


Introduction
This article presents a study of viewers' experiences of a storyline on the weekly BBC One medical drama Holby City. 1 The romance between two older female surgeons, Bernie Wolfe (played by Jemma Redgrave) and Serena Campbell (Catherine Russell), initially 2 ran across two series 3 in 2016 and 2017, attracting a slew of new viewers as well as generating fan art, fiction, videos and songs, shared on various platforms via the hashtag #Berena. 4 The article draws on written survey responses to consider the impact(s) of 'Berena', as framed by respondents. It is commonplace to see academic discussions of queer 5 media representation begin with an account of how much better things are, qualitatively and quantitatively, than has previously been the case, yet our investigations of the impact of this new visibility, particularly for LGBTQ viewers, have not kept pace with this proliferation (McInroy and Craig, 2017). Winderman and Smith (2016) call specifically for research into particular portrayals of LGB characters and their impact on LGB viewers, with emphasis on the links between television viewing and mental health. This study looks to answer that call.
In so doing, it challenges some assumptions, and therefore the article not only presents a case study of the significance for a group of viewers of a particular media portrayal of older women-loving women (wlw), but also seeks to contribute to the way that we think about the relationship between media and sexual identity. One way that neoliberalism sustains its self-definition as centrist is in the 'heterosexualisation' of gay culture (Duggan, 2002;Ghaziani, 2011): oppositional, activist culture is rhetorically rendered anachronistic by talk of compassionate conservatism, and the (finite) assimilation of LGB individuals into the mainstream via, for instance, same-sex marriage rights. Just as 'post-feminist' discourses take into account and thereby discard feminist politics (Gill, 2007;McRobbie, 2004), 'post-gay' describes an assumption that society is 'over' homosexuality, and it is therefore outdated to think of LGB identities in terms of struggle -and by turn confusion/personal difficulty. That what I find here should trouble this assumption is perhaps unsurprising; what I would emphasise instead is that this article shows that in the case of 'Berena', older women discuss experiences -learning, identification and validation of sexual identities and desires -that we most commonly associate with adolescent viewers, and the women are implicitly aware of this association, too. In various ways, they orient to a feeling (their own or on behalf of others) that this ought not to be happening. Though there is of course academic recognition of sexual fluidity later in life (Rowntree, 2015 traces these ideas through various literatures; prominent examples include Westwood, 2016 andCalasanti andSlevin, 2001), it is rarely represented on screen, and older wlw have been just as invisible in our theorisations of the impacts of media representation. 6 More specifically, there are in the study presented here references both explicit and implicit to the story's impact on respondents' mental health and wellbeing, and so I close by proposing that some rethinking of the sexual minority identity, television and mental health matrix may be due.

Context
The invisibility of older women on screen is much discussed (see Dolan, 2013) and has been for some time; McDonald's (1983) assertion that older women are 'twice unseen' remains pertinent even at a distance of more than 30 years. A similarly aged description of older lesbians as a 'triply invisible minority' (Kehoe, 1986: 139; and see Traies, 2016) also retains its accuracy -despite being able to say with certainty that there are more, and more diverse, women-loving women on screen now than ever before, Scanlon and Lewis (2016: 8) refer to a representational 'desert'. Castle's (1993: 2) discussion of the apparitional lesbian in popular culture -'elusive, vaporous, difficult to spot, even when she is there, in plain view' -is still potent, too. Reading Castle's words in the mid-1990s, one could easily have pointed to the American television series Xena: Warrior Princess (1995-2001 as an example of a lesbian love story hidden in plain sight: to multitudes of wlw fans, Xena and Gabrielle were lovers, yet this was invisible to viewers who did not want to see a lesbian pairing. 7 If we imagine that things are different now, a moment of controversy at the 2017 San Diego Comic Con suggests otherwise: recapping the latest series of Supergirl with a song, the actor Jeremy Jordan repeatedly referred to two female characters, Kara and Lena, as being 'only friends!', appearing to belittle and discourage the popular reading of their relationship as anything more, despite heavy subtext to the contrary. As Porfido (2009) points out, the 'poaching' (Jenkins, 1992) of characters by queer viewers is often borne not of playfulness, but of necessity. In turn, the skill with which wlw viewers are able to build their own storylines via subtext, and the extent to which they are prepared to do so, offers producers the opportunity to rely on this as a substitute for canon storylines (for more on lesbian viewers' fantasies and their exploitation, see Arthurs, 2004;Scanlon and Lewis, 2016).
It is not only for this reason that 'Berena', an overt wlw romance, can be considered significant. This is a BBC 'family' programme, aired on Tuesdays before the 9 pm watershed. A study of primetime BBC programming found that just 0.00059% of airtime showed homosexuality in a positive way (see Cowan and Valentine, 2006) 8 ; that lesbian and gay characters were rarely portrayed as successful people in stable jobs, or as contented middle-aged people; that such characters and their relationships lacked depth and development; and that their interactions lacked tenderness or real intimacy. Lesbians in particular, the authors noted, barely existed. Holby City's 'Berena' storyline, in this light, is something more nuanced than the BBC has typically managed in this sort of slot (I return in the discussion to why this might be especially important). Bernie and Serena are not peripheral characters, and their relationship is characterised by evident tenderness and care; the storyline sees both develop emotionally. It might be considered problematic that Serena (introduced in 2011) has hitherto been written as heterosexual, since soaps (or indeed any series chasing viewers) have shown a tendency, where they do include wlw characters, to have them appear and briefly 'turn' a previously straight character (UK viewers might think of Sonia Fowler's first flirtation with lesbianism in Eastenders; for US viewers, Abby and CJ on LA Law -see also Heffernan, 2005). However, I would argue that it is important that the storyline involves one of the longest standing and most popular characters in Holby, as well as an actress as well known and loved as Jemma Redgrave is in the UK. Describing the poverty of the representation of lesbians on screen, Castle (1993: 2-3) says, 'What we never expect is [. . .] to find her in the midst of things, as familiar and crucial as an old friend, as solid and sexy as the proverbial right-hand man, as intelligent and human and funny and real'. Bernie and Serena were, in the slow build up to their falling in love, central to Holby City. It was not uncommon for their scenes to account for between a quarter and a third of an episode, and the programme's social media channels used #Berena to post behind-the-scenes images. If further proof were needed of their centrality to the programme, it might be found in the appearance on Points of View in April 2017 of a longtime Holby City viewer who was unhappy about there being 'so many gay storylines'; besides Bernie and Serena, a gay male couple were also on screen at the time (Miller, 2017).

The present study
The 'Berena' story is by no means perfect: it does not entirely avoid cliche´(two archetypally strong women, Bernie and Serena briefly tussle for power) or the tragedy that so often accompanies wlw on screen, even if the 'dead lesbians' trope (see Millward et al., 2017) is side-stepped. Participants in this research, in common with the broader fandom online, offer criticisms of the storyline both in terms of the direction that it takes and of the specificities of key moments and dialogue. For some viewers, for instance, the show's repeated references to 'Sapphism' contribute to the ongoing erasure of bisexuality on screen. For others, though, the absence of more specific labels gives them scope to read the characters as they wish to. The point of this research is not to arbitrate on whether or not 'Berena' is an exemplary or even good representation, but to consider its impact, acknowledging that any representation of a romance between two older women is so rare. Scanlon and Lewis's (2016: 8) representational 'desert' not only offers an academic context for this research, but helps to explain my own, more personal route to the present study. Not a regular Holby City viewer, I discovered the story when I switched on the television at the moment that Bernie and Serena shared their first kiss (30 August 2016). As Cowan and Valentine (2006) suggest, starved queer audiences are keen to watch any storyline involving LGBTQ characters; after watching the perfectly awkward aftermath of that kiss the following week, I caught up on the months preceding on YouTube (NatsAshes, 2016). Before long I was searching #Berena on various social networks and befriending other viewers from the UK and beyond. When I discuss the viewers in this article I separate myself from them only in so far as I present here their views, as submitted to this research, rather than my own. I am not peering through a microscope looking at these funny creatures; I am one of them.
Among the conversations I was part of, one question kept cropping up: what on earth is going on? As relatively settled, usually coupled, usually professional women of a certain age, none of us could quite fathom the extent to which we found ourselves invested -it did not seem right, somehow. Serena's daughter's charge that this was a 'bizarre Sapphic midlife crisis' (S19 E13 'I Do, I Do, I Do') struck rather close to the bone. Isn't it embarrassing, we said, to be so emotional about a couple of soap characters at our age! Witnessing the staggeringly personal and emotional conversations that women were having about the effect that the storyline was having on their lives, a project examining this seemed increasingly worthwhile.
These conversations were often in online spaces (Twitter, Facebook, Tumblr, a chatroom called the L Chat (n.d.)) which were more or less accessible -Twitter being the most open, others requiring a login to view -and that variability posed questions of ethics. Among the arguments to be made over social media research, the question of the definition of social media platforms as public or private features prominently (see Golder et al., 2017, for the key themes in this literature); the fact that the biggest 'Berena' groups on Facebook are closed (with acceptance depending on administrators) supported my feeling that it would be inappropriate to lift and analyse disclosures on that and other platforms.
Instead, I composed a qualitative web survey in order to document and reflect on the experience of viewers. This approach has some shortcomings: first, it is unlikely to capture all viewers who had discussed 'Berena' online at some point; second, to compound the first, the granting of institutional ethical clearance was slow, meaning that the 'buzz' around the storyline had diminished somewhat when the survey was circulated; third, asking viewers to report their viewing experiences may not offer similar insights to 'in the moment' responses (Hallam and Marshment, 1995). Nonetheless, it offers benefits to the study, discussed below in terms of the survey design and dissemination.
The survey, built on institutional web-based software, introduced potential respondents to me and my research; I explained that I felt this was an important topic and wanted to give viewers a voice. I gave my email address and invited people to contact me if they wanted to. 9 The survey posed three questions: 1. Please tell me about your experience of the Berena storyline on Holby City 2. Please tell me about your experience of the Berena fandom 3. If there's anything else that you want to talk about that you haven't touched on so far, please feel free to use this box The space for responses was unlimited and the questions were deliberately as open as possible; web surveys have been found to perform better than other surveys in terms of data quality when using open-ended questions, since respondents have autonomy to act, at their convenience (Kwak and Radler, 2002;Schaefer and Dillman, 1998), and response rates tend to be better where the 'difficulty' of the questionnaire is kept low (Manfreda and Vehovar, 2002). My interest was in viewers' meanings and experiences, so it was important to allow them to determine their own focus and emphases, rather than attempting to fit their experience into researcherdesigned categories (Gibson, 2010;Singer and Couper, 2017). Respondents 10 were assured of anonymity and, to that end, were asked for no demographic information. 11 Braun, Clarke and Gray (2017) suggest that the disembodied nature of such research helps to encourage participants who might struggle to engage otherwise, either through lack of confidence or because the research tackles a sensitive topic. 12 Best and Krueger (2008) discuss creating 'anonymous response environments' so that participants not only complete and submit a survey but are also spared the pressure of presenting themselves in a favourable light -something also highlighted by Seiter (1990) in her discussion of the problems of academics interviewing audiences. Nonetheless, I discuss later some of the 'face work' (Goffman, 1959) in the responses.
Respondents were sought via several online platforms: Twitter, Tumblr and in two Facebook groups (with permission from the administrators). Though such an approach might exclude viewers without internet access or who do not engage with the fandom, as such, the research arose from fan discussions of '#Berena' on these sites. My approach is what Fricker (2008) might describe as 'judgement sampling': respondents come from a particular community/ies of internet users, though the population of interest likely extends beyond that. Pan, Woodside and Meng (2014) write about the impact on response rate of contextual cues in solicitation, and the researcher's identity; though in their experiment this hinges on the prestige/status of the researcher, I consider that here, my in-group membership may have been important in encouraging participation. Perhaps I only imagined my membership (Seiter, 1990: 69), and of course in introducing myself as an academic there was immediately some level of differentiation, but the authenticity of my fandom was not questioned. In the Facebook groups (where members, including myself, are identified by real names), there was a lot of interest in the survey as a means of giving 'Berena' fans a voice; in this way, I was co-opted as part of an 'us'. It might be suggested that such an 'insider' perspective could produce blind spots in the analysis -where, for example, I and my respondents take the same things for granted -or that my analysis might be compromised by my membership of the fan community. Although this is not a fandom study per se, researchers in that field perhaps more than others have tussled with questions pertaining to the dual identities of scholar and community member (fan) -what Jenkins (1992: 5) refers to as two modes of understanding that 'are not necessarily in conflict but are also not necessarily in perfect alignment'. Like Jenkins (1992) and many others, including, more recently, Brooker, Duffett and Hellekson (2017), I believe that more insights than pitfalls come with being 'embedded within the world of the thing you're studying' (Brooker et al., 2017: 66); it has been argued that the distant and objective academic cannot best capture the experiences of fans (Jenkins, 1992), and is in any case a fantasy (see Hills, 2007). Duffett's suggestion that this sort of scholarship is a kind of translation of experience (in Brooker et al., 2017: 64) is helpful, since it captures a sense of duty to the fan community (the key 'problem' with the insider perspective) as well as to the academic community (and its attendant expectations with regards to critical distance and analysis).
In total I received 86 responses to the questionnaire in three weeks; some ran to a few lines, some to several pages. Most were somewhere in between. My approach was a thematic analysis: reading the responses, I looked for recurring ideas, repeated words or phrases and textual indicators of intensity of feeling (see Owen, 1984). In the sections below I set out the most prevalent themes before discussing their implications. In the article I also make reference to comments made by Catherine Russell. These come from media interviews, convention appearances and a face-to-face conversation that took place while the survey was running in June 2017. Ms Russell was informed of the research aims and signed a consent form before speaking to me; our conversation was recorded and transcribed verbatim.
Seeing ourselves on screen: 'I was like wow! 2 middle aged women having a relationship on prime time telly! A true representation of me!' Unsurprisingly, given the relative paucity of wlw representation (particularly on British network television), most respondents explicitly discuss the pleasure they take in seeing themselves on screen. This is emphasised especially by women in their 30 s and above, 13 who are even less likely to see themselves. Their responses employ a vocabulary of delight, joy and rarity: 'To see a lesbian relationship between women of a certain age is really refreshing and the best thing since sliced bread'; 'To see myself (aged 56 and a professional woman) represented on TV in an honest way is joyous'. References to the story as 'refreshing' and 'a breath of fresh air' appear regularly; in their reflections on 'Berena', wlw viewers show an acute (if implicit) awareness of their typical absence from our screens. Respondents also make a point of mentioning that this is a primetime BBC show, and therefore a space in which wlw are even less likely to be seen. There is a real sense of novelty of experience in the responses, and many write of being drawn into fandom activity online for the first time, reading and writing fan fiction as well as joining Twitter and Tumblr (what Fry et al., 1989 call 'secondary viewing contexts'; see also Dhaenens and Van Bauwel, 2014, on semi-public fan participations). As a lesbian, discussing the absence of wlw on screen can feel like stating the obvious, but some heterosexual respondents describe this as being news to them, 14 and both Jemma Redgrave and Catherine Russell spoke of their initial surprise at the scale of the reaction they received. Speaking on the British daytime programme This Morning in February 2017, Russell said: 'If you don't have representation, which as a lesbian you don't, then to be on primetime television and to see yourself is occasion to hang out the bunting, light a sparkler and shout ''hurrah!'''. In fact, viewers' responses show that simply to see oneself is not enough, and that previous wlw storylines -particularly in mainstream programming -affect hopes and expectations, even once the storyline becomes canon. A number of respondents describe their initial disbelief and fear in response to 'Berena': 'I couldn't believe the subtext was more than just subtext and I couldn't believe it wasn't queerbaiting '; 15 Part of me did not want to become invested in this [. . .] it's a crushing thing to finally see oneself represented on screen, maturely etc. only to then become victim of a horrible trope. I am still a bundle of nerves over Berena . . . I'm not entirely trusting the BBC writers.
There is evident distrust of mainstream producers handling wlw relationships, and several participants refer to the BBC series Last Tango in Halifax (2012-), which also featured an older female couple until one of them was killed the day after they married (see Millward et al., 2017). As shown above, women describe their viewing experience in visceral terms -'a bundle of nerves'; elsewhere: 'I was nearly sick [. . .] it took me a few days to recover'. Some viewers discuss how the storyline after Bernie and Serena 'get together' failed to live up to the preceding narrative, and thus they struggle to decide whether Holby City is truly progressive. In seeming unsure of what to do with characters once they enter a same-sex relationship, Holby is certainly not alone -screen homosexuality has historically been more episodic than serial (McCarthy, 2003). At a fundamental level, viewers enjoy the validation offered by representation (McInroy and Craig, 2017;Scanlon and Lewis, 2016), but are not satisfied simply with that.
Self-realisation: 'My life finally made sense to me. So many moments throughout my life I now understood' Viewers write not only about the pleasures of seeing older women in love, but also of the specifics: both Bernie and Serena have been married to men and have adult children, and while Serena is written as heterosexual until developing feelings for Bernie, it is a previous affair with a woman that prompts Bernie to leave her husband. In one episode, we see Bernie confide in a colleague that she thinks about 'how wonderful life would be, if only I was brave enough' (S18 E28 'Prioritise the Heart'). Respondents pick up on the fact that Holby City showed Bernie coming to terms with her sexuality: The character development grabbed my attention as I quickly realised that a woman around my own age was actually allowed to not only be gay but to be confused about it. It's refreshing to see older women finding themselves, it serves the community well to acknowledge that not everyone is fabulous and out by the time they're 21.
The modal expression here -'actually allowed' -powerfully underscores the scarcity of such a storyline as well as its invisibility within 'the community'; it hails an existence denied. Both characters are best characterised as women-loving women, a term that captures sexual fluidity and allows viewers to read the characters per their own experiences. After Serena and Bernie kiss for the first time, Serena admits that she has 'never been more than friends with a woman before' (S18 E48 'Brave New World') but, despite being 'terrified', she soon tells a colleague that she is in love with Bernie. In conversation, Catherine Russell said the script had prompted her to rethink, 'because I had always thought that Serena, like a lot of people [. . .] maybe had sex with somebody at uni, but they were very adamant that they didn't want that. It was very much the chemistry between those two people, and if Bernie had been Bernard, she would also have fallen for Bernard'. This, too, is described in positive terms by respondents: I've been struggling to identify my sexuality for so long mainly because I believed that I should have known by my 30s that I was gay, but to see a woman in her 50s discover another side to her sexuality helped me realise it didn't matter when in life you discover yourself.
The American hospital drama Grey's Anatomy (2005-) was similarly praised for the awkwardness and epiphany in its storyline between two older women discovering same-sex attraction (Zuk, 2017). In both cases, viewers' responses echo Moran's (2008) findings: that the lack of visibility of older wlw and what she calls 'mid-life sexuality transitions' amounts to a lack of social sanction for later-in-life sexual fluidity.
Often, when women report having realised that they are attracted to women thanks to the 'Berena' storyline, they speak in terms of something that was buried or obscured but is now visible to them: 'It wasn't until these two characters, and actresses and this storyline, that I really took stock of what my age old feelings really meant'; 'Seeing two women on screen that I totally related to allowed me to understand the feelings I'd had my whole life!'. That popular culture might play a part in sexual self-realisation is not new (see, for instance, Gomillion and Giuliano, 2011;Hammack, 2005;Matthews, 2003;Plummer, 1995). Yet despite Plummer's (1995: 85) note that the point at which 'the fuller story breaks through [. . .] is largely unpredictable', there is a tendency, especially in the 21st century in the UK, to presume that realisation and/or acknowledgement of non-heterosexuality takes place in youth and adolescence. If we consider some of the proliferating media texts featuring prominent queer characters -for example, Sense8 That we tend not to see older women questioning their sexual identity is likely to be a reflection of the unreality of neoliberal sexual politics (Duggan, 2002); post-gay cultural assumptions overlook women such as one '50 something' viewer who says, 'my generation didn't agree with same-sex relations and it was a very taboo and never talked about subject'. The novelty of the 'Berena' storyline is such that some older heterosexual respondents (not quoted here) found that they, too, questioned their sexuality as never before.

Coming out: 'I'm not sure if I would ever have had the courage had it not been for the storyline'
As anticipated, the responses include a number of coming out stories, often from younger women who watch Holby City at home: 'I came out to my parents during an episode of the show, when Bernie and Serena were together on screen'; I never imagined I would be able to come out to my mum. But we watched the Berena storyline play out together [. . .] eventually I worked up enough courage and was finally comfortable with myself and with my mum to come out. We have had a much better relationship since then.
[The storyline] gave me the confidence to come out to my sister, though she wasn't entirely surprised because my love for Jemma Redgrave was blatantly obvious. Recently, my mum started watching Holby City with me. This gave me hope that if I came out to her she would react positively.
Though younger people most commonly watch mobile and on-demand services outside of family viewing contexts (Nielsen, 2017;Ofcom, 2017), here we see how such contexts might retain their value; as has been found elsewhere, a good reception for LGBTQ characters and storylines helps younger viewers to come out to friends and family (Gomillion and Giuliano, 2011;McInroy and Craig, 2017). When we spoke, Catherine Russell shared her surprise at how many young women had told her about the problems they were facing in being open about their sexuality. 'Younger women in small communities are living in a world that is very mainstream', she suggests; 'they aren't going to gay clubs, they're not there.
[Holby City] is in their sitting rooms, their parents are watching and so is their little brother and so's their big sister'.
Responses also point to other contexts in which sexual minority representationspecifically a storyline exploring sexual fluidity -on primetime shows can promote communication and improve relationships: 'I'm a lesbian, but I married a man because I buried that part of me. He's a good man and we have two children. We were able to talk about the true me for the first time'; 'I am closeted bi. [. . .] the Berena story on Holby has allowed me to be more openly affectionate to my husband as I identified with feelings of love, desire and respect'. Several respondents (typically indicating their age to be between 30 and 50) report being in relationships with men who are (or were) unaware of their same-sex attraction. They describe not just the pleasure of watching two women exploring their sexuality, but the way that this has become a topic of conversation, or has facilitated sexual contact, with their partners. Kern (2014), too, speaks to women who enjoyed the American series The L Word as an 'outlet' for their same-sex desires within opposite-sex relationships.
There are also older women, aged 60 and above, with ostensibly heterosexual life stories who orient their responses to the coming out process: I am now quite happy knowing that for all of my life I have been at the very least bi, if not completely gay, although I have never really acted upon it. I am not planning on coming out to my 96 year old mother or my daughter or grandchildren, although I would never deny it if asked. [Berena] has allowed me to interact on a level I've never found before regarding finding female stars sexy and attractive, and I find this extremely liberating.
I joined a dating site for lesbian women, feeling that, if at 50+ it was alright for Serena and Bernie, at 60+ I should give it a go as, coming out of a 35+ year heterosexual relationship, I could finally express my hidden feelings. So, although I am totally aware it is fictional, it has inspired me and given me confidence to 'come out'.
I have now met a lady of a similar age and, although it is early days, we are getting on pretty well! The contrasts in these two responses are interesting: for one, acknowledging her feelings to herself (and to others in a closed Facebook group) has been sufficiently liberating; for the other, her feelings found expression in a relationship with another woman. Westwood (2016), in her work with older LGB and 'non-labelled' people, identifies several cohorts; the respondent above who joined a dating site fits into the 'breaking out' cohort -she appears to have been aware of, but struggled with or suppressed, same-sex desires. The respondent who does not plan on coming out to her family might belong to Westwood's 'finding out' cohort -appearing to have discovered a retrospective wlw identity (Westwood, 2016: 58). For Westwood, however, such a discovery is tied to entering a same-sex sexual relationship; here, watching the 'Berena' storyline appears to have played a critical role.

Meeting others: 'I've made friends that I will always treasure'
The vast majority of women write of friends made, and of interactions with other 'Berena' fans on social media. This is by no means unique to these viewers: the prominence of social media interaction is well documented in fan studies (e.g. Bury et al., 2013;Hellekson, 2009;Hillman et al., 2014), as is the potential for online groups to connect LGBTQ individuals (see Hanmer, 2010;McKenna and Bargh, 1998;Tushnet, 2007). Yet it is worth considering, however briefly, how these particular viewers talk about these interactions and what meanings they make of them.
Participants describe the 'Berena' fandom variously as 'a nation of women', as a 'home', as being 'like a family' -essentially, a safe space of specifically (queer) female sociality (Bainbridge, 2013). They emphasise the diversity of the fans they have met or spoken to: 'a brilliant mix of ages, personalities and sexualities', and respondents identify themselves as being from the UK and other parts of Europe, north America and Australia: 'It's nice to be embraced for being who I am'; 'I can use Twitter without fear of judgement'; 'I have met my people'; 'I really feel part of a group'. In these excerpts, 'real world' experience is (implicitly) negatively evaluated against online fandom, which helps women 'to resist the hold of the local' (Hanmer, 2010: 148) and forge bonds with potentially distant others. Kern (2014: 435) also reports that The L Word viewers 'felt supported by a newfound community of viewers in ways they did not find in their daily experiences'. It is perhaps important to remind ourselves, however, of the difference between the postwatershed 'lesbian ecology' of The L Word (Sedgwick, 2006) and a BBC primetime family medical drama. Despite their geographical distance, respondents share a sense of belonging, and often seem surprised by this -especially older women who are using platforms such as Twitter and Tumblr for the first time. In a discussion about online fandom and The X Files (1993-2018), Bury et al. (2013: 301) suggest that older fans do not use Tumblr because they are 'sticking with the older technologies [. . .] because they are comfortable in the spaces they have created with those technologies'. Here, it seems, older viewers have been until now disenfranchised both offline and online by their absence from our screens -they do not have pre-existing comfortable online spaces, and have embraced new technologies as part of their 'Berena' fandom.
Many viewers had met with other 'Berena' fans, largely facilitated by Catherine Russell's theatre appearances in March and April 2017, though a considerable number indicate that they continue to meet up: 'There are a few ladies who I have met through the fandom who will now be friends for life, and we are incredibly close'; 'We met up with a few #Berena fans and I have made a few friends from the fandom, which hasn't happened since I was in my teens'; 'Friendships have been formed, relationships have been born, endless amounts of support and love have been bestowed -all because of a fictional lesbian couple'. Viewers stress the closeness of these friendships and the extent of the support they have received, and project into the future 'real and lasting friendships' (in contrast to the transience of soap storylines and, perhaps, to pre-empt any dubiousness on my part). During our conversation, Catherine Russell said: I think that #Berena will live on. [. . .] I can truly imagine in five or 10 years' time, people going, 'why is this group called Berena?' 'Why do we say, 'let's have a Berena meet up?' Wasn't it something to do with Holby City?' I can imagine that.
Mental health: 'I don't like to think where I would be now if it wasn't for these two amazing women' In addition to the themes set out above there is one that cuts across many of the responses: mental health and wellbeing. From those in their late teens who had been able to cope with the stress and anxiety of coming out, to the respondent in her mid-70 s whose grief at the sudden death of her partner was alleviated by watching two older women in love on screen; from those battling eating disorders through those experiencing the upheaval of divorce to those who have simply 'been through a shit year': 'I suffer with depression and whilst this is a deep-rooted issue, not going to be solved magically by queer media representation, Berena has honestly been a little ray of hope for me'; 'In less than a year, I have come from a place of deep unhappiness with who I am, to beginning to accept that it is alright -that I am alright'; 'I don't like to think where I would be now if it wasn't for [Berena]'; 'The Berena fandom has in no uncertain terms saved my life. [. . .] Without Berena, I know I wouldn't be here today. I would have given up the fight a long time ago'. While recognising that these viewers are likely to be among the most motivated to complete a survey of this nature, theirs are nonetheless stories worth considering. When I asked Catherine Russell what she would take away from the experience, mental health was something she picked up on, unprompted: 'I'm not surprised, if people have been made to feel unwanted or unloved, frowned upon, but the extent of mental health [problems] has shocked me'. It may be the case that there is a still a debate to be had about the prevalence of mental illness in LGBTQ populationsthough research tends overwhelmingly to support the assertion that cis heterosexuals are less likely to experience mental health problems (see Eliason, 2010), the links between sexuality and suicidality, for instance, may not be so straightforward as some scholarly and public discourse has assumed (Bryan and Maycock, 2017). However, mental health and wellbeing was evidently a concern for a number of respondents, as it was for women in the larger fan population (per Russell's comments).

Discussion
Existing studies have considered the LGBTQ identity, television and mental health matrix; some of the most recent have looked at the impact of media representation (e.g. McInroy and Craig, 2017), the connections between LGB characters and LGB viewers' sense of self (Bond and Miller, 2017), the reasons that LGB individuals watch LGB-inclusive television (Winderman and Smith, 2016), the links between queer resilience and media representation (Craig et al., 2015) and the influence of media role models (Gomillion and Giuliano, 2011). This list is not intended to be exhaustive, rather demonstrative, both of the significance of the connections between media representation and LGBTQ lives, and of a tendency to focus academic attention on youth and adolescence: only Gomillion and Guiliano (2011) and Winderman and Smith (2016) focus on broader populations, and almost half of Winderman and Smith's (2016) participants cite their occupation as 'student', from which we might infer a considerable number are aged 18-24.
The focus of these studies might be explained by the understanding that adolescence is a time of intense personal development (Matthews, 2003), and the linkage between adolescence and sexual identity and experience (Jackson, 2009): Bond and Miller (2017: 98) suggest that 'LGB teens see themselves first and foremost as LGB'. Yet I have found the same processes -learning, identification, validationhappening here, too, for viewers at different life stages. I would posit, therefore, that what has generally been approached and theorised as a youth phenomenonthe realisation and actualisation of a non-heterosexual identity through or in relation to media representation -ought not to be so tied, in our conceptualisation, to life-stage. McInroy and Craig (2017) suggest that in our academic explorations we have failed to keep pace with the increase in depictions of LGBTQ lives; Winderman and Smith (2016) call for researchers to examine specific portrayals of LGB television characters, and to explore, qualitatively, their links with mental health outcomes. The present study suggests that as depictions increase in number and depth, and therefore (one hopes) become more representative of older people and more mainstream than post-watershed public service broadcasting and ondemand services (Lambert, 1982;Needham, 2009), (sexual) self-realisation and validation, and any attendant mental health impact(s), may be more visible as extending beyond youth and adolescence. And, if we continue to describe current LGBTQ media representation as being an improvement not only qualitatively but also quantitatively on the recent past, we must also acknowledge and consider the needs and experiences of the generation(s) previously denied visual possibilities of LGBTQ self-realisation (Porfido, 2009: 167), or the kind of social sanction discussed by Moran (2008). They watch not as emerging adults but as (potentially) emerging queers. In this context at least, we are not yet post-gay. This is important.
I referred earlier to conversations that I had had with other older women who were struggling to explain the intensity of their feelings about 'Berena'; though the unfamiliar appearance on our screens of older wlw is obviously significant (see Kern, 2014 on the particular impact of 'occasional visibility'), it is not a complete explanation. Many respondents talk about how much time they have devoted to 'Berena': 'One problem is it is addictive and I can't seem to quit!'; 'In all honesty I follow the fandom a little too much to be healthy. It can be a little disruptive to my life'; 'I feel totally dumbstruck as to how this has affected me. There are more things I could say but at the worry of sounding crazy, I will leave it there'; 'Following the Berena storyline has been a very pleasant way to waste time I should have been using on something more productive; I've ended up spending far too much time watching YouTube [. . .] All the rest of my spare time is spent reading fanfic. I suppose in some ways this hasn't been great for my motivation to you know . . . do housework . . . but it is good fun! Porfido (2009) writes of the euphoria of (finally) seeing oneself on screen and its addictive quality, yet it is striking how often viewers' reflections have the air of a confession -'In all honesty' -and/or include some kind of 'face work' (Goffman, 1959): respondents seem to see in their engagement with 'Berena' behaviour that makes them susceptible to what Goffman (1963) calls stigma. Goffman suggests that the stigmatised recognise the need to hide or mitigate that stigma in interaction. In the examples above, respondents orient explicitly to their 'grown up' responsibilities -housework, jobs, 'something more productive' -as they describe their fandom; they perform knowledge of there being 'more important' things. Similarly, numerous participants describe being 'sucked in', 'gripped', 'absolutely hooked'; says one: 'I tried to resist, but the pull of two powerful and competent mature women falling in love was too much'. Their responses suggest that these women feel that they are too old or ought to 'know better' than to react to and enjoy the 'Berena' storyline in this way.
It is not only age that is relevant to this discussion. Fry, Alexander and Fry (1989: 340) apply Goffman's notion of stigma to low-status media content: 'presenting one's self as the kind of person who does or does not view soap operas is a complex interactive process in which viewers create and recreate constantly the place of television serial viewing in their lives' (see also Ang, 1985;Jenkins, 1992). Holby City is a serial drama that airs on a Tuesday evening: it is ordinary television, not something we suppose to be profound or philosophical (Fitzgerald, 2017). When respondents discuss, with embarrassment, their passion for the show, their concerns seem to stem not only from age but also from genre and the serial form: as one respondent puts it, 'There are times when I have to remind myself that this is just a TV show and it is not real'. Yet it is the very mundane constancy (or what Porfido (2009: 173) calls 'marvellous banality') of television -in this case a medical melodrama, in your living room, 52 weeks a year -that has helped to make the 'Berena' storyline important and impactful for these viewers. Thus, something else we might look to reconsider in light of this research is the impetus to diminish links between television, impact and the domestic, which have previously been so persuasive (e.g. Rymsza-Pawlowska, 2014); as scholars, we cannot be so seduced by the rhetoric of choice offered by the fragmented digital market as we suppose consumers to be. Primetime BBC programming -still often watched in the family living room -retains its potential for influence and thus for being an object of study even, perhaps especially, for non-heterosexual audiences.
This article, then, has done a number of things, beginning with the discussion of the pleasures of seeing oneself on screen, and how these pleasures impact upon the (sexual) self and its realisation and actualisation. It has shown, using responses from women across the UK and beyond 16 (and who otherwise may share little in common), how viewing translates into 'real world' interactions and relationships. Crucially, it has shown that for many of my respondents, the experience has had positive mental health outcomes 17 that trouble (a) prevailing public and academic assumptions about the relationship between sexual identity, media representation and life stage, and (b) viewers' own expectations regarding genre, self and identity. In the UK we are at a point where some structural stressors associated with queer identities are diminished (Meyer, 2003) but, as ever, trickle down is slow; we need also to take account of those who grew up in the immediate past from which neoliberal sexual politics seeks to distinguish itself. Holby City may be merely medical melodrama, and Bernie and Serena 'old farts' (Dainty, 2016), but for some viewers -triply invisible, twice stigmatised? -the 'Berena' storyline has offered a transformative experience.
3. Though Holby City runs all year round (i.e. fits the description of a serial), it is nominally divided into consecutive series of 52 episodes. 4. Fans who want two characters to become a couple (known as 'shipping') typically create a portmanteau couple name (TV Tropes, n.d.) 5. In this article I use descriptors such as 'wlw', 'lesbian', 'non-heterosexual', 'LGBTQ' and 'queer'; they are not necessarily intended to be synonymous. Sometimes term selection will be led by the preferences or scope of the source being cited (this is particularly the case when discussing existing literature). My own preference when referring to the characters of Bernie and Serena, and my participants, is wlw, since this is less exclusionary than 'lesbian' but more accurate than 'LGBTQ'. 6. There are of course notable exceptions, such as Hallam and Marshment's (1995) reception study of Oranges Are Not the Only Fruit. 7. In an interview with Entertainment Weekly in 2016, the lead actress, Lucy Lawless, said: 'You'd see people go, ''What lesbian subtext?, What are you talking about?''' (Abrams, 2016). 8. Such a survey has not been repeated more recently. As I suggest, we would be likely to find things improved -though by what margins is arguable, particularly in the case of wlw, and especially for older wlw. 9. I heard from three people who wanted to share their thoughts without anonymity. 10. In this article, I use 'respondents' and 'participants' synonymously. I also use 'viewers' to refer to survey respondents and not to the larger population of Holby City viewers. 11. In conversations with Berena fans online, I had found some women (often those who were not 'out') to be extremely cautious about sharing any kind of identifying information. Since the study was concerned not with making broader claims about Berena fans, or fandom per se, but with the stories of respondents' experiences, these were prioritised above demographic information. 12. This was not at the forefront of my design, but on reflection, given the extent to which respondents discuss mental health issues, this may have been to the benefit of both researcher and participants. 13. As previously indicated, demographic data were not sought, so can only be considered where volunteered by respondents. Indicators of age were common. 14. A small number of participants identified themselves as heterosexual women -several of them nonetheless showed an understanding of the lack of (older) wlw on screen and reported being pleased for the wlw friends they had made within the 'Berena' fandom. 15. Ng (2017) defines queerbaiting as 'situations where those officially associated with a media text court viewers interested in LGBT narratives -or become aware of such viewers -and encourage their interest in the media text without the text ever definitively confirming the nonheterosexuality of the relevant characters'. This is a more generous definition than those arrived at by scholars such as Fathallah (2015) and Nordin (2015), who charge queerbaiting text makers with denial and mockery of queer fans/readings. 16. Survey responses placed respondents across the UK and Europe, North America and Australia. Further work with the 'Berena' fandom has identified fans in south America, Asia and New Zealand. 17. That is not to say that there were not different responses -perhaps more light-hearted, briefer responses, or responses with more grievances about the storyline (and, in some cases, the fandom; not everybody found it to be so welcoming), but there is no scope to explore this further in this article.