Challenging Misrepresentations of Roma Through Queer Romani Visual Self-Representations

This article considers, in social semiotic terms, the visual self-representations created by lesbian, gay, bi, trans, intersex and queer (LGBTIQ, thereafter queer) Romani visual activists and artists, and some of the processes used in the course of a transdisciplinary, collaborative research project. Undertaken in 2019, its aim was to investigate, through queer intersectional research-informed interventions, the potential of this semiotic material – photographic renditions of the lived experiences of queer Roma – to challenge dominant stereotypical misrepresentations and an overall lack of visibility of queer Roma. Another goal was to further enhance the impact and accessibility of the knowledge co-produced with and by the queer Romani visual activists and artists by giving a visual form to their lived experiences of antigypsyism intersecting with homophobia, transphobia, sexism and other forms of oppression. This approach to co-producing knowledge enabled the queer Romani visual activists and artists not only to exercise control over the process of creating the visual self-representations, but also to spell out, in a visual form, the terms in which queer Roma wish to be represented.


INTRODUCTION
Historically, misrepresentations of Roma 1 have been instrumental in generating and maintaining distorted, often negative portrayals of Roma. Contrary to the reality of nuanced, plural, multi-layered experiences and intersectional identities, Roma have been trapped in a mire of myths (e.g., Sabino Salazar, 2021). The actual lives of Roma have been reduced to stark, crude representations (Tremlett, 2017: 722) and controlled by stereotypical images of 'Gypsies'. Non-Roma have often associated Roma with romanticising images of nomadism, unchained freedom, passion, voluptuousness and exoticism; or with vilifying images of criminality, deviance, backwardness, inadaptability, anti/asocial behaviour, threat, risk, conspiracy or invasion (van Baar, 2015;Clark and Campbell, 2000). Yet, '[a]ll stereotypes are inaccurate, even the positive ones' (McGarry, 2014: 761). Consequently, these damaging visual (mis)representations of Roma as fundamentally ethnically/'racially' distinct and different from the presumed non-Romani, white norm have resulted in what are often tense social relations, easily manipulated and exploited by the media, politicians and other actors and stakeholders.
In Romani Studies, too, Roma have often been essentialised as 'eternal Gypsy nomads' (Willems, 1997). Such a stigmatised, collective conception of Romani ethnic identity has led to members of this 'group' being reduced to possessing the same set of assumed, often negative characteristics and values. In the political landscape, neoliberal nation states have tended to over-visibilise and use Roma as 'bargaining chips' to generate notions of solidarity, belonging and identity among non-Roma (McGarry, 2017). To counter these processes, Tremlett (2017) believes that an accurate understanding of Roma requires a conscious effort made by non-Roma, including through research (Tremlett and McGarry, 2013), to transcend historically constructed stereotypes about Roma and to attend closely to the self-representations of Roma themselves.
In the social sciences and humanities, textual data have been traditionally privileged over visual data despite the fact that even literary studies have been forced to conclude that the world-as-a-text has been replaced by the world-as-a-picture. Such world-pictures cannot be purely visual, but by the same token, the visual disrupts and challenges any attempt to define culture in purely linguistic terms (Mirzoeff, 1999: 6).
In Romani Studies, too, scholars such as, for example, Tremlett (2017), End (2017) or Dunajeva (2018) have been examining photographic (mis)representations of Roma 2 . Yet, there is an urgent need for more critical engagement with visual content and materials (mis)representing Roma. The reason is simple: a photograph is not a neutral depiction of reality. Even seemingly neutral representations of Roma reproduce ethnicised/'racialised' meanings and visual paradigms (End, 2017). These misrepresentations -often construed along biologically or culturally essentialist lines, also by means of photography as a tool of physical anthropology and racial classification (Surdu, 2018) -have become part of dominant societies' hegemonic visual language of eclipsing the everyday lives of Roma.
Effectively, '[a]ll semiotic materials, including photographs, and their canons of use are ideological in their nature and have evolved to accomplish things that serve particular social interests' (Ledin and Machin, 2018: 39). This is especially the case in today's world dominated by visual representations disseminated on social media, where 'our bodies are now extensions of data networks' (Mirzoeff, 2015: 14). From a social semiotic perspective, the meaning of an image never lies solely in the object or person portrayed but in the context of these wider, taken for granted socio-cultural contexts, or discourses that tend to dominate in a society.
In this article, I draw on social scientific literature regarding the use of visual images when researching stereotypical representations. First, I briefly flesh out queer Roma's experiences of intersectional discrimination, which informed both the methodological and theoretical underpinning of the research. Second, I discuss the methodology, methods and processes applied during the research project. The article then moves to its empirical part, analysing and theorising the visual and textual data through the prism of Barthes' two layers of meaningdenotation and connotation -informed by queer 3 intersectionalities (Fremlova, 2017(Fremlova, , 2019(Fremlova, , 2021.

INTERSECTIONAL DISCRIMINATION EXPERIENCED BY QUEER ROMA
Lesbian, gay, bi, trans, intersex and queer (LGBTIQ, thereafter queer) Roma experience multiple discrimination at an intersection of antigypsyism -also referred to as Romaphobia -homophobia, transphobia, sexism and other forms of oppression. Additionally, research in countries such as the Czech Republic, Slovakia and Hungary has found that there is both a conceptual and a legal vacuum with respect to intersectional discrimination: vacuums that exist against a backdrop of human rights protections for members of ethnic/'racial' and sexual/gender minorities, including queer Roma, being reversed, along with a general trend of weakening equality bodies (ARA Art, 2021). Writing about a recent research study on LGBTIQ Travellers in the UK, Eminson (2021: 288) notes the following: The results presented a negative experience for LGBT+ Travellers who felt the need to hide their identity and experienced discrimination in the identified theme of stereotyping in LGBT+ spaces. This research identified a lack of visibility of LGBT+ Traveller identity, in both Traveller and British society which has led to a negative impact upon this intersectional identity, predominantly causing an inability to authentically participate in either community.
While a decade and a half ago, there was a lack of research and data on the status of queer Roma, queer Roma have been gaining visibility at an unprecedented pace. Although queer Roma have barely featured in research on Roma and/or LGBTIQ people, the ever-growing scholarship on queer Roma (Baker, 2003;Kurtić, 2013;Baker, 2015;Máté, 2015Máté, , 2022Corradi, 2017;Fremlova, 2017Fremlova, , 2019Fremlova, , 2021Eminson, 2021;Sartori, 2022) has demonstrated that the lives of queer Roma entail rich, multiple, intersectional experiences stemming from queer Roma's multi-faceted identities. Simultaneously, queer Roma experience intersectional discrimination, marginalisation and multiple objectifications from both non-Roma and Roma. As such, the lived experiences of queer Roma cannot be fitted into any one representation.

METHODOLOGY, PARTICIPANTS AND PROCESS
In the course of devising and implementing this ESRC-funded project in 2018-2019 4 , I was faced with a methodological question when thinking about how to enhance the impact and accessibility of the knowledge coproduced with queer Romani participants during my doctoral research. My concern was especially with queer Roma's experiences of antigypsyism intersecting with homophobia, transphobia, sexism and other forms of oppression, as well as with an overall lack of visibility in mainstream discourses on Roma and queer people.
The transdisciplinary, collaborative, qualitative, participatory action research project brough together eight queer Romani/Gypsy research participants, including visual activists and artists from the UK and continental Europe, in order to generate a series of visual self-representations. The participation of trans people, including a sex worker activist, also allowed the project to visually speak to and showcase the diversity of experiences among the different national subgroups of Roma and Romani Gypsies in terms of ethnicity/'race', sex/gender, sexual and gender identities. This 'top-up' research re-engaged two research participants and engaged six new ones. The question 'How can we challenge dominant representations of queer Roma in public spaces through queer creative and discursive research-informed interventions?' guided the investigation into stereotypical representations of queer Roma.
The research project employed elements of Participatory Action Research (PAR) as a valuable methodological tool for working with groups and communities described as marginalised or oppressed (Johnson and Guzman, 2012;Jenzen, 2017), including queer Roma, with a view to articulating the principles of knowledge co-production and transformative action. The research also utilised visual methods and methodologies related to knowledge coproduction, 'visualising the everyday' and 'viewing photography as an anti-essentialist opportunity to "counter fixed and stereotypical views"' of Roma (Tremlett, 2017: 5); and photo elicitation interviews (Harper, 2002;Rose, 2016) with the research participants to 'put more of the power into the hands of the researched' (Tremlett, 2017: 7). Methodologically, the benefit of using visual images and visual methods consists in research participants showing to the researcher how they see the world whilst having some control over the image-making as opposed to just talking about it.
Since 2009, visual representations and self-representations of queer Roma circulated in the public domain have helped to document the queer Romani movement's existence and evolution. These visual self-representations are not only the product of the queer Romani movement; they are also 'part of the symbolic practices which constitute the movement and its identity' (Daphi et al., 2013: 76). The resulting visual trail -or visual data -can be viewed, preserved, archived and re/interpreted (Kharroub and Bas, 2016;Highfield and Leaver, 2016).

DATA ANALYSIS: TYPOLOGY OF SELF-REPRESENTATION
At the initial stage of the research project, I examined approximately 70 existing visual representations of queer Roma produced and shared in online and offline space between 2009 (i.e., the year Roma Boys: A Love Story, a film about gay Roma was released) and February 2019. The goal of the preliminary visual data analysis was twofold: 1. to inform the creative process and the analysis of the visual self-representations created by the queer Romani visual activists and artists as part of the research project and 2. to investigate the potential of these visual self-representations to contextualise, critique and challengethrough critical, non-normative (queer) and discursive interventions -dominant stereotypical misrepresentations of Roma and LGBTIQ people and dominant representational canons that lie at the root of antigypsyism and its intersection with homophobia, transphobia, sexism and other forms of oppression.
For each of the 70 existing visual representations (and later the self-representations created by the queer Romani participants), I considered several aspects: genre, composition, the people portrayed, colours, objects, settings, and the positioning of the (intended) spectator. I also considered what each representation depicts (denotation) and alludes to (connotation); the spectator's cultural (this connotation is present in studium) participation in the figures, the faces, the gestures, the settings, the actions; and punctum: 'this element that rises from the scene, shoots out of it like an arrow and pierces me' (Barthes, 1977(Barthes, , 1981. For coding purposes, I used computer-assisted qualitative data analysis software (CAQDAS) NVivo 11. One of its features enables the researcher to process the labelling of the content of each photograph, which can then be generated into broader themes. Initial coding and categorisation helped me to determine key preset, as well as recurrent and re-emergent constructs, patterns and themes, which I revised throughout the research process (Willms et al., 1990;Miles and Huberman, 1994).
The queer Romani visual activists and artists met at two creative methods and methodologies workshops held in Brighton, UK, in March and September 2019. At the first workshop, I shared with them the findings of the initial data analysis, which informed the subsequent creative process. At the second workshop, the research participants and mentors discussed the primary theme established during the first workshop, 'The Present versus the Past'.
Subsequently, the participants created visual self-representations challenging the trope of the 'Gypsy' trailer, the fortune teller, adding other themes ('Everyday life', 'Nudes') and genres such as portrayal photographs with text projected onto the participants' bodies ('On my skin'), or stock photos ('Everyday life', including indoor/outdoor scenes and experiences of surviving antigypsyism and homophobia). As a team, we also discussed the corresponding strategies, or types of visual self-representations -'Direct challenge though self-representation'; 'A queer take on normalisation'; 'Pure self-representation and its limits' -and recorded them in a chart (Figure 1). The participants created over 200 photographic self-representations: of these, they ended up choosing 15 wellexecuted photographs for a final exhibition called 'Visualising the Lives of LGBTIQ Roma' 5 . They also produced two video interviews and a graphic novel: 'queer reading' of their experiences of antigypsyism, homophobia and 5 The exhibition first opened at Aspex Gallery, Portsmouth, UK, in November 2019, and was accompanied by a creative workshop for the general public and an academic symposium. Since then, the exhibition has been on display at the University of Brighton (November 2019), Brighton Museum (January 2020), virtually at Ljubljana Pride, Slovenia (September 2020) and at the Emil Filla Gallery in Ústí nad Labem, Czech Republic (September-October 2020). Several of the photographs feature in Brighton Museum's exhibition Queer the Pier that highlights the rich cultural history of the local LGBTIQ community, including queer Roma. Part of the exhibition can be viewed here: https://qrstock.qrtv.eu. The whole exhibition is available on request. Contact the author. transphobia (Figure 2). In the six months between the two workshops, additional data were generated through four photo elicitation interviews. In the photo elicitation interviews, which I analyse in the empirical part below, the research participants discussed the visual self-representations produced during the first visual methods workshop in March 2019, as well as photographs of their own choice.
When analysing the photo elicitation interviews, I used thematic analysis at a latent level, sensitive to critical, discursive concepts such as queer theory and intersectionality (Erel et al., 2008;Yekani et al., 2010;Rahman, 2010;Fremlova, 2017Fremlova, , 2018Fremlova, , 2019Fremlova, , 2021. Thanks to its flexibility (Braun and Clark, 2006), latent thematic analysis looks beyond the individual themes by examining the underlying ideas and assumptions. Thematic analysis seeks 'to unearth the themes salient in a text at different levels, and thematic networks aim to facilitate the structuring and depiction of these themes' (Attride-Stirling, 2001: 387). This was particularly relevant to the already interpretive and theorised process of developing the concepts, themes and thematic networks that I identified during the photo elicitation interviews and refined in the discussions with the participants during the second workshop.
In this reiterative, abductive process, the participants' ongoing input, which was provided at key stages of the research, played a central role, allowing me, the researcher, to continue to identify, name, and define themes; and to review and interpret them. Both the secondary and the primary visual data, including the visual selfrepresentations created during the project, as well as the textual data generated through the photo elicitation interviews offer a unique insight into queer Roma's lived experiences. Equally, they help us understand which visual strategies and queer research-informed interventions can be used to effectively challenge misrepresentations of Roma, and how and why this can be done.

ANALYSING AND INTERPRETING THE EMPIRICAL EVIDENCE: CONNOTATION AND DENOTATION
Photographs are snapshots of events (Ledin and Machin, 2018: 42), where 'the distinction between photograph and reality -as between spin and policy -can easily evaporate' (Sontag, 2004). As lay spectators, we are often unaware of how our perception of such snapshots and reality itself is shaped by an image and the conventions that the photographer employed, including different artistic techniques such as, for example, the use of black-and-white photography; decisions made by the photographer as to proximity, angle, setting, editing, and photoshopping; and devices such as the use of children. Children are often associated with purity, innocence, vulnerability and are used in images portraying poverty, social unrest, war and disasters. However, Romani children, who tend to prevail in photographs portraying Roma, are routinely associated with a lack of parental concern or safeguarding when shown in 'clan and tribe-type structures' (Tremlett, 2017: 728). Such Western, ethnocentric, white-, cis-and heteronormative perspectives and highly romanticised view of cultures often entail a particular set of ideas and assumptions regarding social organisation, which are used to frame the entire world (Ledin and Machin, 2018: 44).
Historically, dominant social groups have claimed to see with 'universal relevance' (Haraway, 1991: 188), thus establishing a dominant, normative visuality that often denies, oppresses and overrides the validity of minority groups and the ways in which they see and exist. Specific visual representations of social categories and differencewhether constituted on the basis of ethnicity/'race', sexuality or gender identity -are thus produced and made visible; images of difference are thus embedded within the wider social contexts -or social modalities (Rose, 2007). In this binary opposition, marked, 'abnormal' categories such as Romani ethnic identity, non-normative (queer) sexual and gender identities, even womanhood or femininity are constructed and visualised as diverging, or 'deviating' from their unmarked, neutral counterparts: non-Romaniness or whiteness, heterosexuality, cisgender identity, manhood and masculinity. Certain visualities are mobilised, privileged and lifted above others. This results in the construction and/or reification of dominant, normative representations, socio-cultural practices, codes, scripts, norms and normativities through an established visual canon; or, to paraphrase Barthes (1977), a collective, anonymous and authoritative voice presenting a ready-made, universally accepted truth. That is why such visual representations need to be understood as part of the photographic canons and the discourses they tend to carry, e.g., the media misrepresenting Roma (individuals, groups) as a threat/invasion/influx within the context of the deeply ideological discourse of antigypsyism embedded within European societies.
The 70 existing visual representations and self-representations of queer Roma I analysed tended to differ from the established conventions and canon of documentary photography, much of which has (mis)represented Roma through both romanticising and vilifying tropes as poor, uneducated, vulnerable, often uncivilised, backward, primitive lovers of free-roaming and unchained freedom, antithetical to Western modernity. Most often, the existing self-representations were highly individualised personal photographs, including professional portraits and selfies; artistic photographs; symbolic photographs; photographs documenting public events; and videos, films and other multimedia publicly available on the internet (i.e., websites, Instagram accounts, Facebook pages) or in artistic portfolios. The existing self-representations of queer Roma have tended to emphasise Roma agency and empowerment through employing a perspective that focuses on, foregrounds and makes salient either individual queer Roma in shots of a single person, or groups of queer Roma at public events such as conferences, pride marches and music or theatre performances (e.g., Mindj Panther 6 , Roma Armee 7 ). Additionally, many of the protagonists are public figures active in community organising and the entertainment industry, who understand self-representation thanks to the personal, as well as professional experience they have amassed over the years. These qualities of the photographs were achieved through what can be dubbed 'unorthodox visuality', portraying the Romani protagonists as independent, confident-looking, modern-day, relatable individuals who are sex and gender diverse: their looks defy, challenge and subvert the Western ethnocentric, white/hetero/cis-normative gaze associated with the photographic canon. The photographs employ a large range of highly saturated, at times blended colours, connoting energy, vibrancy, fun, playfulness, ambiguity, complexity, hybridity, even intersectionality. All these aspects have the effect of making the spectator curious and drawing them closer to the person/people portrayed. Combined with ways of looking that foreground Roma as self-reliant, self-sufficient, independent agents, this approach contrasts with some of the conventional approaches used in colour photographs, portraying and romanticising Roma and Romani culture through visual tropes focusing on Romani women and children and their colourful clothing and dance; or with some of the photographic approaches used in black-and-white, grim and bleak -though visually beautiful -photographic representations of Roma.
The above considerations informed the creative process, during which the queer Romani visual activist and artists produced the visual self-representation. To do so, the participants had to make a series of artistic, compositional and editorial decisions, some of which I describe in the following sections.

DIRECT CHALLENGE THROUGH SELF-REPRESENTATION
According to Barthes (1977), denotation, or the first level of analysing visual data, involves describing the persons, things and places in an image. Similar to, yet different from language, visual images constitute connotative semiotics (Hjelmslev, 1961), whereby associations and meaning emerge on different levels through devices such as objects, the person/people portrayed, colours, settings, foreground/background, etc. So, what do the objectsand their combination -in the visual self-representations created by the queer Romani visual activist and artist connote in terms of wider meanings? Considering the immediate level of denotation in Figure 4 and Figure 5, when looking at the two colour photographs, we see two people sitting on the floor. They sit on two colourful From the start of the creative process, the setting/background of each of the photographs -or lack thereof discussed in the section Self-representation and its limits -was a key component that contributed to establishing the context, or connotation, within which the meaning of each of the photographs is constructed. While in Figure 4, the protagonists are photoshopped into a famous Van Gogh painting, in Figure 5, they sit in front of a television set that features an iconic, black-and-white photograph of Roma taken by the Czechoslovak-French photographer Josef Koudelka in Romania in 1968. Using this postproduction technique made it possible to explore how the meaning changes when certain elements are removed (the Koudelka photograph in Figure 5) and new ones are introduced (i.e., a romanticising background with Roma, trailers and horses in Figure 4).
The Koudelka photograph and the van Gogh painting in the background change the meaning through connotation (i.e., romanticised notions of poverty and nomadism stereotypically associated with Roma). The visual self-representations contain certain cultural and social references that shape their meaning. For example, the man and the woman in Figure 4 and Figure 5 are most likely Roma: we may infer this because of their colourful, vibrant clothes stereotypically associated with Roma, the book the woman reads (Gypsy Feminism by Laura Corradi) and the Romani flag they sit on. They could be gay or straight Roma: the context is provided by the rainbow flag located next to the Romani flag.
In the photo elicitation interviews, the research participants' most immediate reflections related to the people and objects that can be observed directly in the photographs, including clothes, as illustrated by the following quote from L: I like to claim the right to wear this kind of colourful, maybe feminine clothing. Actually, they are ladies' trousers. When I wear them in Budapest, I get so many mostly negative and hateful looks from people. People's reactions to me are the same in Hungary when I wear the leopard sweater. It's been getting worse in the last few years but it's also very visible if somebody likes this. Unfortunately, especially when I go to the Hungarian countryside, to my parents', for example, the people are very negative about it. Here in Berlin, I receive mainly positive feedback (L).
Clothing is often co-opted into heteronormative definitions of 'genuine', authentic identities, including Romani ethnicity. The clothing items in Figure 4 and Figure 5 function as conversational pieces: 'a good tool to talk to people' (L). In the male protagonist's case, the design of his trousers bears some resemblance to 'Romani motives' (vivid colours, flowers). It is complemented by his flamboyant sweatshirt with a leopard pattern that, according to L, 'may evoke the [sexual] behaviour of a wild animal'. The woman's dress may remind one of traditional Romani clothing (long skirts/dress) or a loose, bohemian dress. Employing these socio-cultural references in a critical, queer way, clothing becomes a political statement of queer Romani belonging: a means of transgressing and transforming social norms. Refusal to comply with these norms is integral to these photos and is a key tenet of 'queerness'. Simultaneously, one can see that the clothes they wear are modern, everyday clothing. This signals that they are not orientalised/romanticised Gypsies stuck in the past. On the contrary, they are modern Roma/Gypsies living in the 21 st century and performing their queer intersectional identities.
The research participants' observations also related to the connotative meaning of these photographic selfrepresentations, as illustrated by the following quote from L: I'm very happy that I am in the photo with D, whom I know as a Gypsy artist from Great Britain. I am happy because she's my idol from the artistic field. I think it also creates a connection. The fact that I am in the photo as a Hungarian Roma person, and she is in the photo as a Gypsy female artist already shows the diversity in Roma communities. She is a Gypsy woman from the UK with her white skin and me as a Hungarian Roma with brown skin: this shows that Roma can be white, creole, brown or even black. Also, it's not only about the skin colour: somebody self-identifies as Roma or Gypsy. I am very happy that the Romani flag and the rainbow flag can be seen on this photo together because I know that this kind of setting can still provoke very negative reactions by some people in the Roma community. (…) I can be very sensitive about these things, and it can hurt me even if I don't know these people. If there are people who are so disturbed by these things, it means that the two flags should be shown together (giggles). Until these people start to understand why this is important for us, we just have to show it because both the Romani flag and the rainbow flag belong to us, to Roma LGBTQI people and it should be accepted. I think it's very important that as a Roma LGBTIQ person, I can use these flags together. It expresses to me that actually a Roma person or a person who has some connection with the Roma people supports me in my gay, queer identity with the nail varnishing. I think it's an important thing to strengthen in the Roma people and also in the non-Roma people, to remind them that it can happen like this.
In the above reflection on Figure 4 and Figure 5, L dwells on the topic of displaying the Romani and rainbow flags together, something that turned out to be contentious in 2015 and 2017 (Fremlova, 2021: 20). He also touches upon the rich diversity and heterogeneity of Roma communities in terms of nationalities (e.g., Hungarian and British), sub-groups (Roma and Gypsies), skin complexions, sexualities and gender identities. Due to the historic and modern experience of antigypsyism, the word 'Gypsy' is still a racial slur and an insult for many Roma in Central and Eastern Europe, though other Roma may identify as Gypsy. At times, this may produce tensions between the different sub-groups, as well as within the Romani rights movement. The word 'Gypsy' also appears on the front page of the book Gypsy Feminism that one of the participants holds in Figure 4 and Figure 5. In this connection, L discusses the importance of feminism, especially Romani intersectional feminism, for Romani gay and feminine-looking men: I also like that the title Gypsy Feminism is very concrete: a person who doesn't know the circumstances of the creation of this photo can already relate to the topic. This way, it's clear that she's reading about Gypsy feminism. D is sitting next to me and she's holding and reading this book. Somehow it expresses me, my connection to feminism and Gypsy feminism. I'm trying to observe what D is reading in this book: this represents my curiosity and my sympathy with the feminist and the Gypsy feminist movements. I find them very important, not because of the Romani women or women in general but because I think it is also very important for us, LGBTIQ people, that this movement exists, and these voices are heard. Because of my gay or queer identity, I received some negative or hurting comments from people. In more cases, they were related to my being feminine. In Roma communities, these comments can come from Roma non-LGBT people and mainly from Roma men because being feminine or being a woman can be not appreciated or not respected enough. That's why I am happy about this setting: in a lucky scenario, we have each other's work, we can help and support each other. I mean the Roma feminists and the Roma LGBT activists. Of course, I see misogyny and sexism are not just in the Roma populations, it's in general. Yet, somehow, I feel that it can be very specific in Roma communities.
The same photographic scene may also be reminiscent of a picnic, where the two flags function as picnic blankets, as expressed by R: I like the idea of the two flags together. I found that the flags are the most powerful symbols. I like it that it's like blankets. I liked that L is a counter-stereotype of a kind of controlling man with his behaviour, paying attention to the lady who is very confident in that picture. Van Gogh has a painting about Romani people with a caravan. Everybody loves van Gogh obviously, but then, I needed to realise that his interpretation of this caravan can be very stereotypical as well, for the viewer. This is the kind of modernist approach which romanticised the Gypsies in ways that are just not valid anymore in the 21 st century -how non-Roma think we are and how they describe us.
Here, R hints at a particular way of portraying Roma as a cultural and ethnic/'racial' monolith or a group where the focus is on generic features. In turn, these features, whether cultural or biological, end up categorising Roma and turning them into generic (stereo)types, suppressing the complex, nuanced nature of their diverse lived experiences. The visual self-representations in Figure 4 and Figure 5 thus challenge a key stereotype about Roma as a homogenous ethnic/'racial' group. Figure 5 contains another element, initially chosen by the researcher (me) and then used by R when he was staging the photographs during the first photo shoot: the Koudelka photograph. It portrays a Romani boy in the foreground, with two traditionally clad Romani women dancing on the road, a horse-drawn carriage and a group of Roma in the background. In terms of its composition, the photograph is split, as it were, into two vertical halves: one half (the asphalt road) may be understood as the present/future and the other (grassy side of the road, on which we can see the carriage) as the past. The participants felt that Koudelka's depiction of Roma is stereotypical in the sense that the Roma portrayed end up representing all Roma. At the same time, they thought it depicted what was and still may be a reality for some Roma. What draws the spectator to Koudelka's photograph is punctum: its visual beauty and high artistic quality, as expressed by R, J and L, respectively, in the following quotes: It's a socio-photograph which paints Romani people in a romanticised style as connected to poverty.
[…] Sitting on a rainbow flag and reading a feminist book, the photo has the instant message that time has passed. It's another world. There's a past behind and there's the present. [The Koudelka] picture is black and white, our picture is in colour, so I think it's beautiful in a way. To have an image in an image where one is an analogue image, which is black and white, and the other one is digital image, which is colourful. I think the message is clear: L and D are sitting in this setting and somehow, they're trying to approach their own identity as presented by the mass media.
It's stereotypical because it shows that Roma are these romantic people who are always dancing and of course they are poor, but they don't care about that. That's what he wanted to show. From a photographical point of view, it's really perfect, though. The problem is that [the media] put this kind of pictures next to everything related to Roma.
I think the image of the Roma boy is stereotypical, but I really like this setting. The only thing that makes me feel a bit bad is that I feel that he has some pain in his eyes while I'm having fun [in Figure 5]. But maybe it's also a good contrast between stereotypical images and our setting because I see that some of the people who work on our visual representations might like to represent us as sad or poor. At the same time, I'm here with my happiness. It's a strong contrast between the stereotypical image and our setting, which shows the life of Roma people in 2019. The black and white photo is also still a reality of some Roma people in our present, but my life wasn't like this. Here in Western Europe, when non-Roma people get to know that I am Roma, they believe that I also live like this.
Thus, through connotation and punctum, Figure 5 can be understood as a visual representation of two parallel 'realities': an encounter between the past and the present, myth and reality. In one of these 'realities', which has become a myth, Roma are poor; some appear sad and some happy; they travel and dance. In the other reality, the reality of self-representation, we see the everyday life of Roma, some of whom are educated, settled, happy, feminist, queer or straight. Yet, within the photographic canon, the former visual representation (i.e., the Koudelka photo) has become lifted over the latter, thus reinforcing the generic, stereotypical image of Roma being poor, nomadic and somehow primitive.

A QUEER TAKE ON NORMALISATION
So far, I have demonstrated that challenges to visual misrepresentations can be posed discursively and contextually, by making socio-cultural references to the wider social discourses. One way of challenging misrepresentations of Roma is by juxtaposing such stereotypical images with something different, something that is not necessarily the opposite. For example, according to one stereotype and the visual trope, a non-Romani person's future is read from tarot cards inside a Roma/Gypsy trailer (Figure 6). Using this misrepresentation and inserting into it the figure of a present-day gay Roma, we can visualise the difference between romanticised, stereotypical 'myth' on the one hand and the queer Roma person's highly heterogenous lived experience in the 21 st century on the other.
Another type of visual self-representations introduces a queer take on normalisation through a strategic deployment of critical queer research-informed interventions in the form of stock photos. There were two main reasons why the participants decided to use this genre: one was the absence of stock photos portraying queer Roma in neutral or positive ways -and the attendant urgent need to create stock photos portraying queer Roma; the other was the nature of stock photos themselves and the possibility to employ it in a subversive manner. Stock photos often resemble photographs used in advertising. According to Ledin and Machin (2018: 46), this type of photography, is idealised and is intended to symbolise a set of ideas and moods. Advertising photographs simplify and beautify the world as part of the process of loading a set of ideas and values onto products. (…) [T]hese images are created precisely to be used in this symbolic way. The archives can be searched with key words (…) Such archives are driven by the need to understand what kinds of images are required at any time.
Utilising the feel-good, optimistic, advert-like character of stock photos, an archival genre of photography associated with generic types and social norms where one can look for key words such as 'happy queer Roma couple', impressions and imitations of normalcy may be achieved by means of both denotation and connotation: for example, a happy household (Figure 7) portraying a cis-and straight-looking couple in a pensive, intimate mood, unwinding and relaxing on a comfy sofa in what looks like the living room of a modern, bright flat; or a happy couple on the beach (Figure 8), depicting a gay couple on the seaside.
Denotation plays an important role also in terms of the setting, showing where the scenes take place. As R pointed out in the photo elicitation interview, knowing and choosing who one speaks to by means of a visual selfrepresentation, including through choosing the setting, has important consequences for what messages are conveyed to what audience and how that audience makes sense of them.
The visual self-representation in Figure 7 was produced indoors, in a modern, bright living room: the bright, soft light and pastel, muted and coordinated colours contribute to the overall optimistic atmosphere. Apart from the protagonists, the setting features only a few objects (a picture on the wall, a small plant) and items of furniture: a coffee table with some décor, a book and a picture frame. Combined with the overall unclutteredness and airiness of the setting, the objects better serve their symbolic role, emphasising the cosiness of this home, which is enhanced by the bright saturated light. The visual self-representation in Figure 8 was produced outdoors, on the British seaside. We see a young gay couple in a pensive mood, sitting on an elevated wall (coastal groyne) on a pebble beach, next to an undercliff walk, both unfolding into the background. One of the protagonists is sitting up, holding his partner who is lying down. The muted, moderate colours of the setting, complemented by the hues of the overcast sky, accentuate the intimate atmosphere of the scene. Through denotation, the two self-representations connote ordinary, mundane, everyday things that people do (hence the title 'Everyday life'). Yet, stigmatised groups such as Roma are rarely portrayed living everyday lives, doing everyday things: in fact, Roma, including queer Roma, are most often portrayed in derelict, dilapidated, squalid, impoverished interiors and exteriors. They are rarely portrayed in cosy homes or relaxing in a leisurely environment by the sea. In Figure 7 and Figure 8, this quality is achieved not only through denotation and connotation, but also through photographic perspective (the positioning of the intended spectator) and individualisation. Contrary to the conventional portrayal of Roma at a vertical angle (literally looking down on Roma) as vulnerable, even inferior (also in gendered and racialised terms), in all the visual self-representations, the visual activists and artists used perspectives looking from the front, at horizontal angles, often taking close shots. This approach also helps achieve a greater degree of engagement by and intimacy with the intended spectator, who gets  to see highly individualised representations of Roma that do not categorise -or essentialise -them in biological or cultural terms.
A 'queer twist' is introduced by revealing to the spectator, through captions or other hints and references, that the people portrayed in these normative-looking photographs are a cis/trans Romani couple (Figure 7) and a gay Romani couple (Figure 8). Simultaneously, these photographs convey powerful messages of survival in the face of experiences of intersectional discrimination and bullying. The 'normalising' essence of stock photos as a genre is thus employed to help queer Roma, a minority within a minority, inhabit spaces that are often associated with social norms and normativities (i.e., heteronormativity, white-normativity, cis normativity). Using insight from Gonzalez-Torres (in Katz, 2015) who spoke of his queer positionality as an HIV positive gay man, the queer Romani bearer of such strategic 'normalisation', may be regarded as a 'spy', strategically deploying sameness -a queer positionality resisting binary orthodoxies (Fremlova, 2021) -to 'infiltrate' social norms and normativities from within in order to overwhelm and unsettle them.

SELF-REPRESENTATION AND ITS LIMITS
A key reason why the participants were interested in producing visual self-representations was to tell a story about themselves, a story that, according to R, has not been told before and can be told visually, verbally or figuratively, where the participants have full control over the process of narration: We are all artists, we all use our stories, our narratives, our memories or our bodies as a tool to express ourselves. We all challenge the stereotypes with our personality, sharing our personal stories.
We have seen that in order to tell these visual stories, the participants made use of a number of denotative and connotative devices such as the people portrayed, colours, objects, settings, the positioning of the (intended) spectator, including proximity. The close-up shots in Figure 9 and Figure 10 help establish a direct contact, a more intimate rapport between the protagonists and the spectator. Additionally, in Figure 10, the spectator's engagement is enhanced by the protagonist's direct gaze ('demand image'), which is avoided in Figure 9 since the protagonist covers his eyes/face with his hands. The ultimate effect may be an increased degree of curiosity on the side of the spectator who can better focus on the text projected onto the body: or what Kress and Van Leeuwen (1996) refer to as 'offer image'. Visual self-representations with white space as the setting came to constitute a specific type of decontextualised photographic representations. In Figure 9 and Figure 10, the protagonists and the writing projected onto their bodies play a greater symbolic role. Through these close-ups, the decontextualised images not only foreground the queer Romani protagonists and their agency as individuals; they are also used to symbolise the idea that these selfrepresentations are highly personal. Therefore, as such, they cannot, and do not represent or define the LGBTIQ Roma 'community' as a whole, despite the frequent assumption that members of non-dominant groups such as ethnic/'racial', sexual and gender identity minorities will be 'everything for everyone'. The limits of selfrepresentation and its use lie precisely in the notion that they do not speak for all queer Roma. Figure 11 is a visual self-representation created by one of the participants as part of a different project entitled 'Owning the Game' that was taking place at the same time. The self-representation is similarly decontextualised as it has a white space as a setting. According to J, this was a conscious choice: My picture was made in a studio and [the Koudelka photo] was taken in real life. They're completely different. I wanted to do it in the studio because I didn't want it to show any of the background, to not imply anything social. From the photo, you don't know that I'm poor or rich or where I'm from. You cannot see this. Somehow to not focus on that; focus more on the person. The main audience intended was the Roma LGBTIQ community, family and friends. And the second was to challenge also these stereotypical things in the LGBTIQ community, in the Roma community.
Through the lens of adulthood, J's visual self-representation tells the intersectional story of a Romani gay man's life: It's really visible that I'm a dark-skinned person. I'm part of an ethnic group and it's clear that I'm part of the LGBTQ community because of some elements of the picture. As you can see in the picture, I'm dressed in something which is like the 80s fashion. We were influenced by the 80s in New York when the Latina and Afro-American LGBT groups created the ball culture. (…) It was an original idea that my mentor and I worked on. We wanted this kind of glamour photo session. We had reasons why because we wanted to have some dignity, some powerful messages. We wanted to show that we are strong and gorgeous mainly because there is a lack of visual representation where Roma people -and especially this intersection [of Roma and queer] -can show that we're having fun and we like how we look. Here, I'm not giving an interview behind a dirty wall or showing a really sad face. J's visual self-representation demonstrates to the spectator that Roma are in full control of their visual representation. One of J's key messages relates to the lack of positive, unbiased visual representations of Roma, and the attendant misrepresentations associated with Roma such as poverty, dirt, unhappiness and lack of dignity. Through his visual self-representation, J also wanted to tell the following childhood story: When I was a small kid, I really liked to dance in my grandmother's place because she had a really nice radio. She also had a collection of old cassettes and there was one that I particularly liked. She was a teenage Hungarian girl, and she was really famous at that time. Because she was in a skirt, I was dancing to this music wearing a skirt. Nobody said or at least I didn't feel anything like 'okay, you cannot do such things'. Just once, when I was dancing in the skirt, I fell and bumped my head on the table. I had a fracture and a laceration. There was a lot of blood, so they had to take me to A&E. It was there that the question 'why did he wear a skirt?' came up. My parents felt ashamed to explain how it had happened. I felt that it was something I shouldn't do. From that time, I didn't dress in skirts my whole life so it's kind of symbolic. I'm not missing the skirt as an object. It was more like my femininity. It didn't stop me from being feminine because I am. But it was long years that I felt ashamed: when I was listening to myself and heard my voice or I was recorded and I looked at how I'm moving, you know? And I felt that it's not good. It's not attractive. I didn't like it in myself for many years. (…) It felt so good that I can dress as a model. It was really a good experience. For example, one guy especially, who was wearing high heels, I felt like he was enjoying himself like a child. He was like in Disneyland.
According to J, rather than focusing on the everyday to challenge stereotypes, self-representation can be a celebration, a playful game. Such self-representation functions as the opposite of the everyday: a platform for the participants to play, to wear high heels, hair extensions and make-up; to not be their everyday themselves. Thus far, we have seen that visual self-representations can be used as powerful tools to challenge harmful stereotypes. Yet, as J's project shows, self-representations do not necessarily have to aspire to challenge stereotypical misrepresentations of Roma. In fact, they can operate at a 'purely' artistic, symbolic level, as expressed by J: How I can truly represent myself, if I need to continuously think about the majority view? In that case, I have a feeling that I'm serving their needs and not mine. If we are against every stereotypical thing, somehow, we exclude the people who may, in a sense, embody a stereotype. If you go to Budapest's 8 th district where Roma live, you can see that they dress a little bit different. And what's the problem with that? Why is it a problem if a woman wears trousers with a leopard pattern? Sometimes I don't like when Roma intellectuals say 'Why show a woman in leopard trousers? It's stereotypical'. Yeah, but there are some women who dress in this way, and you exclude them. I feel that [Roma intellectuals] don't want to be a community with the people who are in the [Koudelka] picture. This sends a message that Roma can be only this or that. If someone wears traditional clothes, if they are dirty, so what? It doesn't mean that they are not Roma, that they are different Roma than me. So that's why I think self-representation and challenging stereotypes do not go together.
The above concerns regarding the intersection of ethnicity/'race' and social status/class, and what constitutes being a 'good' or 'bad' Roma are reminiscent of debates in LGBT studies/queer theory in relation to the 'good/bad' gay binary. In the late 1990s and early 2000s, the proliferation of positive images of lesbian and gay identities in the media resulted in a particular type of visibilities for queer people, lesbian women and gay men in particular. Just like both heteronormative and LGBTIQ society became more clued-in on the increasingly visible and easy to read 'signifiers of lesbian and gay identities' (Valentine, 2002), similar, albeit non-identical societal dynamics have led to the emergence of a 'good/bad' Roma binary within both non-Romani and Romani societies.
So, paraphrasing the questions asked above by J, what are the societal costs for those ethnicised/'racialised', sexualised, gendered, poor and excluded ('bad') Roma who refuse to play by the majority's 'rules'? It is precisely in answering this question that the salience of the visual self-representations lies. The visual self-representations created by the queer Romani visual activist and artists were produced in order to reclaim Romani ethnic identity, queer sexuality and gender identity from both the ethnic majority and the custodians of Roma conservatism on queer Roma's own terms: this was a key point made by the queer Romani research participants, and by Romani lesbian and trans women in particular. The self-representations treat the issue of visually representing Roma in all its complexity. Simultaneously, by opening up the category 'Roma' through accommodating Roma of all possible identities, they transcend identity politics whilst acknowledging the role of dominant representational canons and conventional visual representations of Roma, including the more problematic ones, in an all-encompassing way.

CONCLUSION
The visual and empirical findings presented in this article challenge dominant accounts and misrepresentations of Roma, including those that portray Roma as anachronistic and antithetical to modernity and Europeanness. Using a range of photographic genres, strategies and devices, the visual self-representations of queer Roma not only capture the lived experiences of queer Roma. They also represent a symbolic trajectory from indoors to outdoors; from the personal/private to the public/political; from the past to the present; from the 'Gypsy' myth to the reality of Roma's lives; from extraordinary, essentialising stereotypes to the ordinary, everyday lived experiences of queer Roma in the 21 st century that disrupt and transcend the established dominant visual paradigms and binary representations of Roma.
The visual self-representations of queer Roma show that if sexuality, gender identity and other categories of identification are foregrounded as an intersection, ethnicity/'race' can be visualised in ways that enable 'new', alternative readings of 'Roma', as well as 'queer'. Just like the lived experiences of queer Roma, these visual selfrepresentations pose a fundamental, strategic challenge to stereotypical, one-dimensional, often negative misrepresentations of and misconceptions about both Roma and queer people. The visual self-representations increase the visibility of queer Roma and demonstrate the ways in which queer Roma wish to be represented. At the same time, projects like this one point to the important role of impactful transdisciplinary research facilitating knowledge co-production at the interface between community organising, policy, academia and art.