Lay theories of decreasing homophobia in the UK among an older heterosexual cohort

Within the last 50 years, the UK has moved from prosecuting male homo-sexuality to identifying sexual orientation as a protected characteristic. There have been differing explanations for how this change has occurred. In the present study, we investigate how this political and moral change is understood by those who witnessed it, with the aim of gaining new insight into this change process. We elicited accounts from 25 heterosexual UK residents over the age of 65 years. First, we conducted individual semi-structured interviews with 21 participants. Second, we used a memory work method. Specifically, a group of four participants met four times with one of the researchers to share and discuss memories relevant to the research question. The analytic process aimed to reconstruct and make sense of the participants’ theorisations based on a thematic reading of the transcripts. Contact with LGBTQ people and increased awareness through the media were construed as central to the shift in societal homophobia. Participants remembered contact with LGBTQ people mediated by friends and family members but accounts show a complex understanding of social change that includes but also transcends direct personal experience and extant psychological theoris-ing. Overall, this was a fragmented story of changing social norms, albeit one embedded within a canonical narrative of tolerance, decency and fairness.


Introduction
Attitudes towards lesbian, gay, bisexual, trans and queer (LGBTQ) people have undergone a significant shift over the last few decades, at least within Western democracies.Specifically, in the UK, male same-sex sexual behaviour -'homosexuality' -was illegal for almost four and a half centuries , but rapid change occurred over the last 50 years: male same-sex sexual behaviour was decriminalised in 1967, and same-sex marriage was recognised in 2013.At the same time, attitudes have shifted as well: in 1983, only 17% of Britons thought there was nothing wrong with same-sex relationships, as opposed to two-thirds of the population in 2019 (BSA 2019-Curtice et al., 2019).That said, it is important to note that historical change has not been linear nor universal (Anderson & Fidler, 2018;Clements & Field, 2014;Watt & Elliot, 2019).
Public opinion towards homosexuality in the UK and U.S.A., and likely most Western societies, has undergone a stark shift from being overwhelmingly negative in the 1940s and 50s towards liberalisation into the 1960s and 70s (Clements & Field, 2014).The slow progress that was being made was however set back during the AIDS crisis in the 1980s (ibid).Anderson (2009) argues that this was the most homophobic period in recent history, with it producing an extreme 'homohysteria' -the cultural fear of being 'homosexualized' -in the UK and U.S.A.. Anderson (2009) and McCormack (2012) argue that this peak results from a unique conflagration of a moralising right-wing politics, panic around the AIDS crisis, and (in the U.S.A., in particular) evangelical religion.The 1990s onwards have been marked by arguably the most radical change in attitudes towards sexual minorities, with much greater acceptance than ever previously (Clements & Field, 2014;McCormack, 2012): a 'zeitgeist of decreasing homohysteria' (McCormack, 2012, p. 60).Watt and Elliot (2019) found that rates of homonegativity declined in every socio-demographic group.That is not to say that homophobia has disappeared entirely.A recent national social attitudes survey in the UK showed that the number of people holding negative attitudes to people who engage in same sex relationships has now reached a plateau, staying consistent at around the mid-thirty percent level for the last three years (Curtice et al., 2019).That is, to be precise, the proportion stating that sexual relations between two adults of the same sex are 'not wrong at all' has remained at around two-thirds for the last three years (ibid).Watt and Elliot (2019) found religiosity and ethnicity to be more strongly associated with homonegativity between 1990 and 2010, but with religiosity now replacing education as the characteristic most associated with homonegativity.
Regardless, there has clearly been substantial positive change but how did this change happen?The rapid change in societal attitudes towards homosexuality (and to a lesser extent other aspects of sexual and gender diversity) in Western democracies provides a unique opportunity to better understand processes of social change.Many theories have been proposed to explain these societal changes.They provide valuable individual insights about some of the processes underpinning change, even if they do not provide a singular definitive explanation.There remains uncertainty about where and when these theoretical explanations are most applicable, how they work together, if they do, and whether key factors are missing from or undervalued within extant theories.
On the one hand, there are theories that have been tested experimentally (i.e. they have led to effective and efficient techniques to change attitudes), but it is unclear whether they explain change as it happened on a societal scale.For example, and most significantly, contact with LGBTQ people has been shown time and again to reduce prejudice (e.g.Bartoș et al., 2014;Smith et al., 2009).However, there is no evidence that a large number of people made acquaintances who were LGBTQ in the 1980s and 1990s to explain the shift observed in public opinion.Similarly, educating people about sexuality and LGBTQ rights, e.g. through media exposure, can change attitudes (Bartoș et al., 2014;Bartoș & Hegarty, 2019), but we do not know if people encountered such educational experiences when public opinion shifted.Evidence suggests these theories appear to be effective at an individual level but are perhaps more limited when it comes to explaining large scale societal change.
On the other hand, we have theories that can predict what groups in a society are more likely to reject LGBTQ people, but which have less utility explaining individual attitudinal change.For example, people who espouse more traditional religious values are likely to be less accepting of LGBTQ people (Herek & McLemore, 2013).This points to the possibility that secularisation over the last few decades is what led to more accepting sexual attitudes.This theory, however, is not supported by data from large-scale surveys (Bartoș et al., 2014;Hadler, 2012).Also, after the fall of the Berlin Wall in 1989, Eastern European societies have become more religious and less homophobic at the same time (Bartoș et al., 2014).
Alternatively, it has been suggested that the decrease in homophobia is explained by older generations being replaced by younger ones ('cohort replacement').While there is clearly some veracity to this theory, the situation is somewhat complicated.For instance, evidence suggests the attitudes of older generations have changed alongside the young (see e.g.Andersen & Fetner, 2008).But in their detailed qualitative work focused on homohysteria and masculinities, Anderson and Fidler (2018) found the majority of their UK sample of older men remained strongly homophobic.Donaldson et al (2014) also found their older US participants continued to frame the notion of gay and lesbian residents in assisted living accommodation as 'other', albeit with a wide range of views expressed from general acceptance to outright hostility.
The history of change regarding attitudes towards LGBTQ folk has been told many times from the perspective of LGBTQ people, in academic and popular literatures, providing valuable insight regarding change from an insider perspective (see e.g.Weeks, 2007).There have also been significant developments that have discussed the changing nature of masculinity in the context of gay and bisexual male homophobia (see e.g.Anderson, 2009;McCormack, 2012).However, we have not heard the stories of heterosexual people who witnessed the change and who may (or may not) have changed their attitudes themselves.This gap is important because: (1) change has not happened everywhere; (2) change is being rolled back in some places; and (3) these narratives are likely important in informing a response to (1) and (2).
That is, more generally, lay theories occupy an important place in how people make sense of the world (Furnham, 1988).They have been shown to underpin individual judgements, influencing emotion and motivation, and form the basis for constructing an individual life history (Chiu et al., 1997;Ross, 1989;Tong & Chiu, 2002).Levy et al (1998) examined the role of implicit theories in stereotype formation and endorsement.Her experimental work found that 'entity theorists', those believing people's traits are fixed, rather than 'incremental theorists', those who believe people's traits are malleable, make more stereotypical trait judgements.Building on this, Yzerbyt and Rocher (2002) argue that we might thereby reduce prejudice by directly attacking beliefs regarding the inherence of group characteristics (their 'entitativity').We might also usefully reconceive such lay theories as narratives (Bruner, 1990(Bruner, , 1991;;Sarbin, 1986).In this context, entitativity and incrementalism might not be thought of as individual traits but instead narrative processes by which we make sense of different groups.Narratives are key for making links between the ordinary and exceptional, with much of our folk psychology focused on the expected or canonical.It is only when we encounter the exceptional -such as a heterosexual first encountering someone with a nonnormative sexuality -that these canonical narratives are challenged, and it is in these moments that human beings must find reasons to make sense of deviations from the norm (Bruner, 1990(Bruner, , 1991)).
Understanding the nature of such lay theories may therefore provide valuable insight into prejudice reduction strategies: what works, for whom, and through what process?Specifically, we have an opportunity to gain insight into people's real-life experiences with the factors we suspect to have played a role in shaping their attitudes, such as personal contact with LGBTQ people.They also have the potential to elicit novel hypotheses about change processes.Researchers have studied lay theories for homosexuality and gender diversity (Furnham & Taylor, 1990;Furnham & Sen, 2013), along with prejudice more generally (see e.g.Apfelbaum et al., 2017;Rattan & Dweck, 2010).However, to date there has been no attempt to explore lay theories for social change itself with regard to the decreasing homophobia we have witnessed in the West in the last fifty years or so.With this in mind, this study sets out to provide the first systematic investigation of lay theories of change for decreasing homophobia over the last 50 years or so among an older (65 years +) heterosexual sample of men and women.By garnering the stories of a generation who actually witnessed and participated in this change, we hope to gain greater insight into the process of social change itself with respect to increasing acceptance for sexual diversity in the UK, and more generally.
Given the historical focus on change, and in recognition of the argument that LGBTQ identities demand separate analyses (Worthen, 2013), we sought to analytically limit the study to homophobia.However, we found participants discussed bisexual and trans identities in addition to lesbian and gay, and therefore we treat all accounts relating to sexual and gender identities in an equivalent manner while maintaining a primary focus on homophobia.

Methodology
To systematically investigate lay theories of change for decreasing homophobia over the last 50 years among an older heterosexual sample of men and women, this study utilised two distinct but complementary data collection methods (described further below): semi-structured individual interviews (n = 21) and a memory work group (n = 4).

Participants
Participants were recruited via posters, word of mouth and community networks for elderly people in London and the South East of England.Twenty-five participants were recruited in total.Four of them agreed to participate in the memory group led by one of the researchers (J.L.).The other 15 participants were interviewed individually by another researcher (A.H.) and a further 6 were interviewed by a different researcher (A.D.).Each participant received an information sheet, signed a consent form, and a £10 Amazon voucher.The interviews and memory group discussions were transcribed by a specialised company.The study received ethical approval from the University of Surrey Ethics Committee.
All 25 participants were heterosexual individuals aged 65 or older, living in the UK at the time of data collection (mid to late 2018).As the research team was primarily based in the south of England, we generally recruited participants living in London and the South East.The three recruitment criteria (age, sexuality, and location) were emphasised on recruitment materials and the participant information sheet but demographic data were not collected or explored any further.Two white women and two white men participated in the memory group.Individual interviews were conducted with eleven women and ten men.The majority of participants were white, and about half of them were university-educated.

Individual interviews
Individual interviews were conducted either on university campus or in participants' own homes.The duration of the interviews ranged from 45 min to 1 h 50 min, with a median and mode of 1 h 10 min.The interviews were semi-structured and explored three broad topics.First, participants were asked to contrast the past and the present.Specifically, participants were asked about their earliest memory of lesbian, gay or bisexual people or issues, and their thoughts and feelings at the time.They were then asked to explore the differences and changes that have occurred in their views between then and the present day.Second, participants were asked about potential drivers of change.Specifically, they were asked about the events, people, or experiences that shaped their views.Finally, participants were asked about the context of change, namely changes in the attitudes of others, representations and thoughts about other social changes that happened over the same time period.

The memory work group
We decided to use memory work (Haug, 1987;Crawford et al., 1992;Langdridge et al., 2012) in addition to individual interviews due to its focus on collective experience (Stephenson & Papadopoulos, 2006) and for the way it allows participants to theorise their experiences as a group (Willig, 2001).Traditionally, in memory work there is no separation of researcher and participant, however, we departed from the usual process as the researcher was a different age cohort to other members of the memory group -given our focus on people aged 65+ years -and would not therefore have shared a similar cultural history.Lay participants were also not involved with overall analysis or the writing up of output from the project.
Data collection consisted of meetings in which the participants shared and discussed memories about LGBTQ issues over their lifetime.The memory work method has the advantage that it allows participants to think and remember over time, as opposed to interviews that often require an almost instant answer to the researcher's questions.Memory work is also an affect-rich methodology in which a group of people are able to reconstruct their memories together, prompting, complementing or even challenging one another.This latter aspect proved very important in this study, where participants were trying to remember events over long time periods and that did not necessarily affect their lives directly.
The memory group (MG) consisted of four one-hour meetings of the four participants and researcher (J.L.) on a university campus.In the first meeting, the researcher explained the study to the participants, facilitated planning for the group work, and supported them in agreeing ground rules for the next meetings.Participants wrote down their memories relevant to the topic and across different time periods following this initial meeting.Following group discussion, they decided to focus primarily on memories relating to the 1950-60s, 1980-90s, and 2010-present.These accounts were then read out and/or retold in the subsequent meetings, enriching the accounts and encouraging further memory production, discussion and reflection.This included discussion of: different people's 'actions' within the memory; relationships between the people involved; clichés and contradictions; things that might be missing from the account of the memory -whether there were any 'absences' in the account.These strategies were used to enhance participant recall and reduce inaccuracies associated with memory recall (see Limitations).The initial meeting was not recorded but subsequent meetings were recorded for analysis.For confidentiality, the written accounts of memories were not retained by the researcher.In the final meeting, the researcher invited the participants to generate hypotheses about the drivers of social change in the context of the memories they had produced.

Analysis
A minimal definition of a narrative requires that it is temporally structured, i.e. that it has a beginning, a middle, and an end (Sarbin, 1986).Accordingly, we structured the conversations with our participants on three narrative 'moments', specifically, the time around decriminalisation (50 years ago), the present day, and what happened in between.It is important to note that while this study is focused on homophobia rather than LGBTQ+ prejudice more broadly, given public discussion of trans and queer identities is a more recent phenomenon, participants often blurred distinctions between sexual and gender prejudice.It is for this reason that we use the acronym LGBTQ throughout to more accurately reflect the population being addressed by participants in their accounts.That said, the primary analytic focus remains change in homophobia, albeit with discussion of trans and queer identities included when it relates to processes underpinning lay theories of change in prejudice.
We conducted narrative analysis on the transcriptions of the individual interviews and memory group discussions.The analytic process largely followed the guidelines provided by Josselson (2011).We read the transcript repeatedly to identify themes within the narrative: 'These iterative readings continue until we develop a "good Gestalt" that encompasses contradictions' (Josselson, 2011, p. 228).To this end, we thematically coded the text and created memos, summarising and making interpretative comments, following Miles and Huberman (1994).We particularly focused on the initial (pre-decriminalisation) and final (present date) states and the processes that occurred in between thought to effect change.

The beginning: fifty years ago
The most prominent theme in narratives about the past is silence, echoed by essentially all participants: 'Those sorts of things were just not discussed'.(P2) Many participants could not clearly remember the first time they learnt about homosexuality, albeit some participants had memories of LGBTQ folk in families or communities but even here the topic was not discussed openly.Some were first exposed via slurs and oblique references: 'one of them', 'go the other way' (P2).Others could remember a specific cause célèbre, such as Lord Montagu's trial 1 (P1) or the Jeremy Thorpe scandal 2 (P9): 'Jeremy Thorpe was probably the first time I remember someone being gay and I was quite shocked about that.I remember being fairly shocked that a politician was gay.Not nowadays of course, but in them days it was quite unusual'.
They also remembered some 'camp' characters in the media, such as John Inman's performance in Are You Being Served 3 or Kenneth Williams's roles in the Carry On films 4 but these characters were treated as exceptional rather than representative: Well, that's right, I was going to say that despite the fact that it had been decriminalised you didn't get a sudden exodus of everyone coming out, I mean obviously there were the comedians, you know, Kenneth Williams who put on a very camp and I mean he had sort of sexual problems and there was that sort of thing on stage but we all laughed at it didn't we really because it was . . .Well, I suppose we thought it was funny not like the majority of people sort of thing.P5 Another participant remembered that the news had 'really nasty sort of vibrations', with 'men being chased and beaten up' (P11).Most notably, participants could rarely remember turning points that historians or other academic parties consider important, such as decriminalisation (1967) or Section 28 (1988) (Nunning & Sicks, 2012).These topics were not discussed in family or friendship networks, but participants were generally aware of societal change and/or significant topics such as HIV/AIDS, albeit rarely in a specific or personally relevant way.
When participants first encountered homosexuality, they found it 'strange', 'surprising', 'abnormal' (P1), 'weird' (P2), 'awkward', even 'shocking' (P3).Some participants revealed feelings of disgust, expressed facially (P2) or verbally: 'I won't say quite repulsive, but along those lines' (P3).However, most insisted that 'it wasn't malicious' (P2) and 'it's never been a thing for me' (P9).Some participants added that they were quietly questioning the negativity towards homosexuality, which was never given a rational explanation, while others were outright angered: 'my sense of justice came to the fore' (P11).Ignorance in others is sometimes viewed as the root of hostility: P9 speculated that people fail to distinguish actual harm from what intuitively feels wrong to them, and 'there are some idiots go around and just beat them up for the sake of it'.In general, perceptions of difference were framed within an underlying canonical narrative of tolerance -'British decency' -in which a 'live and let live' attitude predominates.
Language was a strong preoccupation, with most participants making efforts to remember slurs and euphemisms from their youth while also commenting on changes in (acceptable) terminology.This is perhaps not surprising given the contemporary focus on language in sexual politics.It is however notable how a number of participants understood they needed to change their language, especially in public, but this was understood as a matter of change in what was now publicly acceptable rather than any profound change in attitude (see Lyonga, 2021;Walters, 2014).P13 exemplified this very clearly: 'if I saw two chaps walking along holding hands I wouldn't sort of scowl and say "scum!", you know?'.

The middle: how and why change happened
In apparent contradiction with the premise of this study, most participants described change as slow: 'slowly, slowly just people got over their fears' (P11).The Memory Group (MG) also described the change in societal attitudes as 'the creep'.Some argued that changes in laws and education have not been in place for long enough to 'kick in' as internalised attitudes: 'nobody's come to terms with it really' (P11).The transmission of attitudes from parents to children, and within families more broadly, was thought to play an important role in effecting and also slowing down change: 'because one generation brings up another', prejudice 'takes a generation undoing' (MG).The MG agreed several times that change was self-sustaining, a virtuous cycle.This was brought up in relation to the visibility of gay communities among the wider public: 'once you see it, you see it everywhere'.
Asked why people might not change, participants often asserted the irreducible nature of individual variability.P1 was unsure why some people changed while others did not, but he cited personality, genetics, and 'they've all had their own journey'.Similarly, P3 recognises that some people may never change, but he suggests 'letting people develop'.As a specific impediment to change, religion was most often mentioned.P9 attributed some of the changes to the diminishing role of religion in society; P11 similarly appreciated that 'religion has been thrown out the back door and . . .liberalisation has been coming in the front door'.P6 cited religion as the reason for the lifelong stability of her own view on sexuality.

Contact
Contact with LGBTQ people is, unsurprisingly, discussed by many participants as a key driver of attitude change: 'I think it's only when you come across gay people, meet them, talk to them, that prejudice starts going down' (P3).Contact humanises the outgroup.For example, P2 states that her views 'changed because now I know gay people and see them as people', referring specifically to seeing gay people for more than their sexuality.However, contact may also come in the form of an intense experience.One participant in the MG remembers being taken to a gay bar for the first time: 'She 5 couldn't believe it . . .It was a revelation to her at the age of 18 . . .It felt like she'd landed on another planet'.This echoes the common feedback given by participants in diversity training that meeting gay people and/or learning about them was 'eye opening' (Bartoș & Hegarty, 2019).
Contact also provides counter-stereotypical information.P2 describes meeting a gay couple at a dinner party.They were 'extremely smart', as opposed to the 'wildly camp guys [. ..] mincing along' that she expected: 'This all sounds very naïve', she comments, 'but it's indicative of the lack of experience until that'.Of course, longer-term contact makes a more profound impact: P11 learnt a lot about diversity within the gay community from sharing a house with a lesbian couple; while another participant says that 'The friends of mine that have gay children . . .I think that has helped me become more open' (MG).
Some participants disclaimed the effect of contact on themselves: 'No, no, meeting gay people has not changed my attitude'.(P1).One participant thought contact important and pervasive, as 'most of us have got a gay friend nowadays'; but, when asked if he had gay friends himself, he answered: 'no, I don't think I have, that quite interesting, isn't it?'Another participant recognised that 'meeting people really makes an impact', but thought such direct contact impractical on a larger scale: 'I don't see trucking out a lot of gay people around the country to introduce them'.(P3), one of the arguments for the value of imagined (as opposed to direct) contact as a strategy for effecting attitudinal change (Crisp & Turner, 2009, 2010).
Examples of negative contact are rare but present (Dixon & McKeown, 2021).The coming out of someone close can be hurtful, like when a significant other leaves for a same-sex partner, or a platonic friend starts expressing their romantic or sexual interest.Equally, some found sexual advances troubling but such occurrences were only rarely reported (e.g.P13): P13: And if you were, and of course, in those -we had skirts in those days, these sort of rather full skirts, and as you came down the stairs your skirt sort of went, you know, and she'd always happen to be going up the stairs and she'd put her hand up on your leg.
Int: So, what did you make of that?P13: Oh, well I . . .I used to sort of, I would go at her. . . .And I'd say, 'I'll push you back down the stairs if you do that again', and that sort of, you know, just sort of in, as a response.

Media, visibility and changing social norms
Change started with more awareness -'my children for example, would have grown up seeing that on television and seeing it tolerated' (P8) -but that was not automatically positive.P3 argues that media exposure to LGBTQ people is important because people's sexuality is not always visible in real life, and so people may not always benefit from contact: 'Television is the most influential thing we have today.It has to be television'.As an example, he cites the first gay kiss on EastEnders, a long running UK soap opera; he does not watch the show, but the scene had such an impact he can remember seeing it on the front pages of newspapers.But other than exposure, people discussed how individuals learn specific things from the media: seeing Elton John's lasting marriage in the public eye made P11 realise same-sex relationships can be just as stable as heterosexual ones.
P1 had a particularly clear representation of the role of the media in attitude change.He speculated that new ways of thinking were 'filtering down' or transmitted through 'osmosis' from the 'educated classes' to the 'general public'.Specifically, what filtered down was an awareness of the presence of LGBTQ people in society, relatable LGBTQ people ('very, very normal and nice individuals') and counter-stereotypical examples ('very sort of macho, manly characters').It was also notable how negative perceptions emerged where figures were perceived to be stereotypical.Referring to the camp British comic Alan Carr, P13 remarked 'It irritates me. . . .I feel he's putting on a show, you know, he's making a show of his personality'.

Economic factors
Economic factors driving change feature prominently in the accounts of the MG.Participants considered that limits to economic expansion in the 1980s were an important determinant in the reduction of homophobia: LGBTQ people were one of the few untapped markets and human resources in an age that found little room for growth: 'Society changes when it has to change'.Most prominently, 'businesses now want the pink pound . . . .Do whatever you wish, as long as you buy from us'.Similarly, soap operas introduced gay characters and stories because 'they're running out of storylines'.Other examples are adoptions by same-sex couples, whereby change is largely attributed to the state's need to control the cost of raising children in care: 'economics have got an enormous amount to do with it'.This narrative provides a counterpoint to the moral progress narrative underlying the other themes: 'But they're [companies with pro-gay stances] not really doing it for social reasons, they're doing it for commercial reasons'.

Social movements
Social movements were rarely recognised by participants as being instrumental to social change.The MG briefly recounted a shift in the strategy of LGBTQ communities and organisations, from internal solidarity to outward activism: 'Communities existed, kind of closed communities . . . in the 80s and 90s . . .[they] made themselves more visible and more integrated . . .' and there was the occasional mention of a pride parade with e.g.P6 finding that a powerful instrument for change: 'the more people that turn up the more people tend to think this is real, this is reality, and people like myself and other people have to accept it because that's the way they are'.But in general LGBTQ movements and organisations are absent from the narrative of these participants, perhaps only of significance for those directly involved or benefitting from them.

The end: Today
Most participants saw the present day as different from their youth and were content with the changes in general.P3 was content that more people could have committed relationships, a view echoed by most participants.P1 was also 'pleasantly surprised' by marriage equality, because 'loving relationships are a good thing'.P9 found the acceptance of LGBTQ people was necessary for social cohesion: 'if we are going to live together as human beings . . .we have to accept people even if they are slightly different from us'.The MG, in particular, discussed at length the social necessity for all couples to be able to adopt children.Even when they are optimistic about changes in their own environment, participants recognise that change is not universal, noting the need for considerable change in some countries and communities.But there was even some optimism about change in these contexts: 'gradually, gradually everything is going to change and it'll take longer in some places than it does in others'.(P11) Other participants were more ambivalent about change.For example, P6 rejected the idea of same sex couples getting married, raising children, and displaying affection in public (while also condemning violence and discrimination against them).P9 supported marriage equality, but raised questions about adoption.P16 expressed similar concerns, like several others emphasising gender differences in parenthood: 'I don't have a problem with gay marriage or civil partnerships, I think let them get with it and do their own thing.But adopting I think a child needs a male and female in order to have a balanced upbringing'.For others, acceptance was more performative rather than personal: 'one would accept them if one's in company then one would accept them' (P5).And even among the most progressive of participants, like P14, who was supportive of same-sex marriage, adoption, gender transition etc, there were doubts about some recent aspects of contemporary sexual culture: 'My own thing is the extent of LGBT-ishness, it becomes almost absurd if you have what sex are you going to be today and all the rest of it and it's . . .oh it's one of those silly little things'.
Also, some participants raised doubts about the extent of societal change itself: 'There's a lot of hostility but it's not expressed and when you scratch the surface some places you kind of feel it's still there'.(P11) The participant repeatedly used the word 'underground' to refer to this lingering hostility.She based this on her experience at her workplace, which had anti-homophobia policies but where colleagues 'talked about poofters . . .with a smirk on their face'.In general, participants played down change in their own attitudes.They argued they were just ignorant, but not hostile when they were younger.As such, when they acknowledged their own change, they largely referred to learning.When asked whether his views changed, P1 replied 'very much so' (Similarly, P2 said 'utterly . . .completely').He described this change as learning more about the demographics and biological bases of homosexuality, but he insisted that he 'would have been just as kind to someone who was homosexual' when he was a teenager.P9 similarly said that he "never really had an issue with it anyway to be honest, but he concedes he changed 'very slightly because I suppose society moves forward and I move forward with it'.

Discussion
Our findings show the older heterosexual adults in this sample think reduced homophobia came about as a result of increased awareness, through personal contact or media exposure, and consequent personal growth.Personal change most often involved a gradual -albeit, often qualifiedacceptance of sexual diversity through it being incorporated into a valued canonical narrative of tolerance: what is often termed a 'live and let live' attitude.This involved a reduction in entitativity, such that the hitherto alien group became humanised through knowledge of LGBTQ individuals in person or through the media, supporting claims that challenging entitativity may be a valuable prejudice reduction strategy (Yzerbyt & Rocher, 2002).Some participants did not change, of course, and continued to express homophobic attitudes, as seen in previous studies with similar cohorts (e.g.Anderson & Fidler, 2018), but even here there was recognition that the public expression of such views was no longer acceptable.Not all remembered events need an explanation, and some of our participants struggled to generate explanations for the change they witnessed.People tend to seek explanations for events that run counter to canonical narratives, are recent, negative, momentous, unexpected, and selfrelevant, among others (Bruckmüller et al., 2017;Bruner, 1990;1991).To this end, a key aspect regarding change, hitherto rather ignored in the psychological literature, is the power of canonical narratives to frame wider processes of cultural change.A canonical British narrative focused on 'tolerance' -being 'polite and decent' -may mean that not all encounters with difference will demand explanation.Change may 'just happen' over time and with increased awareness as an inevitable part of a tolerant (decent) response to legitimate citizenship claims for equal rights.Where there is resistance this tends to appear when change does require new narratives, specifically where claims appear to be in tension with extant canonical narratives.
Participants' accounts echo psychological theory in many ways.The prominence of intergroup contact (Allport, 1954;Pettigrew & Tropp, 2006) as a strategy for personal prejudice reduction and social change, was present in many if not most accounts, with relatively few caveats.This theory is central in the psychological prejudice reduction literature and clearly also anchored within everyday understandings of experience.Furthermore, the possibility of having mediated contact is also present in both the psychological literature and our participants' accounts.Media, particularly television (Schiappa, Gregg, & Hewes, 2006), was perceived to be highly influential in effecting change.And ideas akin to Crisp andTurner's (2009, 2010) imagined contact hypothesis, in which it is argued that simply imagining contact with outgroup members can improve outgroup attitudes, were invoked by several participants, even though they will have had no formal exposure to these theoretical ideas.Even negative contact (see Dixon & McKeown, 2021 for a discussion) was occasionally mentioned, such as when someone presumed to be Lesbian, Gay or Bisexual attempted to sexually touch one of the participants but this reinforced rather than effected change in existing attitudes.
There remained some uncertainty about how much change is acceptable or should be permissible among these participants.The great majority disavowed holding prejudiced views, but occasionally demonstrated quite strong reservations about issues such as adoption.Tolerance does not necessarily equate to fundamentally changed attitudes or wholesale embrace of cultural change, even when it appears sincere (Lyonga, 2021;Walters, 2014).There is also a sense, even among the most liberal of participants, who are clearly very supportive of LGBTQ equality, that they find some contemporary change regarding sexual and gender diversity problematic.This tended to concern contemporary radical movements arguing against fixed identities/categories of sexuality and gender, rather than claims for equality per se.For some, this felt trivial or even undermining of the LGBTQ cause while for others this was felt to be at odds with canonical narratives concerning the protection of others perceived to be vulnerable, notably women and children.

Limitations
Our strategy of interviewing witnesses rather than participants of history has borne its fruit, but it also has some limitations.First, people may struggle to remember events of the distant past that have not affected them personally and the veracity of individual memories may of course be questioned.The accounts provided by participants cannot be considered as facts per se, given retrospective questioning will inevitably result in problems of recall (Dex, 1995;Pratt et al., 2000).This is notable with participants unable to remember key moments of historical change.And for the events and people that were remembered, there is a question whether they were remembered because of their prominence at the time when they occurred, or because they had been revisited more recently.For example, the Jeremy Thorpe scandal was the object of a recent TV series and figured in several participant narratives.That said, we sought to understand these accounts as narratives rather than memories, which provide insight into meaning making processes and are therefore particularly appropriate for theory generation in the social sciences (Mishler, 1986).
Second, the majority of discussions focused on gay men, with lesbians and bisexual people mentioned more rarely.This may of course simply be a reflection of the historical fact that gay men have been the subject of considerably more discussion than lesbians and bisexuals until relatively recently, in part because they have also been the subject of anti-gay laws, unlike lesbians and bisexual women.The memory work method seems to have mitigated some of the issues, being able to collectively reconstruct detailed accounts of topics that were present in a fragmentary form in the individual interviews, such as social movements.
Finally, the sample was not particularly diverse or reflective of any specific population.It was a relatively small convenience sample.While we did not specifically set out to explore individual or identity category distinctions and their impact on lay theories in this study, and we did not record such data, such factors may be important.Our sample was balanced for men and women, well represented across the older age range but was not diverse with respect to ethnicity, being mostly white.Ji (2005) highlights the importance of cultural diversity in lay theories of change (see also Haslam & Muller, 2017), and we must be careful about assuming our findings extend, without qualification, beyond this sample.Finally, we should note that participants were also informed about the nature of the study prior to participation and it is possible that those who came forward to participate held views that were distinct from the wider population.That said, we did not find any clear or apparent differences in accounts provided by men or women nor any perceivable difference between the very youngest (circa 65 years) and oldest (circa 85 years) participants among this sample.However, given the relatively small convenience sample employed herein, care needs to be taken regarding generalisability and future work would benefit by garnering more representative samples.

Conclusions
This study set out to explore lay theories among people who have actually witnessed the decreasing homophobia that has occurred over the last fifty years within the UK.Unlike the narrative we usually hear regarding change on this topic, focused on activism, education and key political turning points, this group of older heterosexual adults drew on a mixture of psychological theory -notably, contact theory -and sociological theorising -e.g. the influence of the media and 'the pink pound' -to make sense of their experience.For these individuals, what has often been described as rapid social change driven by activism, education and formal politics was perceived to be something more distant, albeit inevitable, grounded in a canonical narrative of 'British' tolerance, decency and fairness.We found an emphasis on contact theory, in particular, that was very strong but which was grounded in lived experience rather than academic discourse.Extension of contact theory to include imagined contact -in line with recent developments in contact theory -was also discussed among the sample, albeit mostly in the context of the power of the media, particularly television, to effect social change through recognition of LGBTQ people as individuals.The focus on changing social norms, as opposed to education, which was rarely discussed, appeared to be a key factor in changing attitudes that has been relatively ignored in the psychological literature.This was itself grounded in a canonical narrative concerning tolerance, decency and fairness that was clearly perceived to be at the heart of much of the change from discomfort -or even disgust -towards LGBTQ folk to growing tolerance or even acceptance.
Invocation and encouragement of canonical narratives and associated identities -e.g. of 'being a decent tolerant person' -appear to have worked well (in many ways) with respect to prejudice reduction and sexual diversity, even if not part of an explicit strategy.Finding ways to leverage such canonical narratives to effect lasting change may prove a valuable new line of prejudice reduction research.This canonical narrative was also relevant to understanding present limits regarding social change.While the vast majority of participants resisted any notion that they were simply prejudiced against LGBTQ folk, it was apparent that there remained some antipathy to key citizenship claims, particularly adoption and some trans rights.Some of this antipathy was related to religious beliefs, albeit here too there had been within cohort change, but it was wider than this.Resistance to change appeared mostly present in these narratives where nascent claims regarding rights challenged other canonical cultural narratives e.g.regarding the stability of core identities or need to protect women and children.There is also an open question, ripe for further study, about whether lay theories of social change are domain specific or draw upon core ideas, such as contact theory, and whether these ideas have currency beyond specific topics and cultures at particular points in history.

Data sharing
Data not available -ethics and participant consent The participants of this study did not give written consent for their data to be shared publicly, and due to the sensitive nature of the research supporting data is not available.

Notes
1.The Lord Montagu of Beaulieu was a British nobleman who was at the centre of two trials in the mid-1950s.He was eventually imprisoned for having consensual sex with another man.2. Jeremy Thorpe was a leader of the Liberal Party in the UK whose political career was ended by a scandal in the late 1970s.Specifically, Thorpe was accused (but acquitted) of conspiring to assassinate his male lover in order to hide his sexuality.3. The late actor John Inman is best remembered in the UK for playing Mr Humphries in the 1970s-1980s sitcom Are You Being Served.Mr Humphries was a friendly, humorous, and intelligent character who behaved in a somewhat effeminate manner and was sometimes implied to be gay.John Inman himself was openly gay. 4. The Carry On films are a series of popular British comedy films in the 1960s − 1980s, with unrelated plots but linked by an ensemble cast.The humour of the films was often sexual, relying on double-entendres and risqué plotlines.Kenneth Williams typically played pompous upper-class characters who were often implied to be gay.Kenneth Williams himself has described his sexuality in various ways in his extensive diaries, autobiographies, and interviews and mostly describes himself -and is described by others who knew him -as gay. 5. Memory group participants used the third person to refer to themselves in the memories they wrote at home and bought into the meetings.