Towards action research with trans women sex workers: Policy, space and social challenges.

Trans women face multiple social, economic and health inequalities and the impact of gender oppression and violence is even more profoundly experienced by trans women sex workers, although in culturally specific ways. This paper presents a pilot study conducted to explore and engage with the context of trans women sex workers in Lido Tre Archi, Italy. In line with the community psychology values of social justice, social change and participation, we outline our engagement process and key challenges observed when attempting action research with such a highly marginalised group. Data were collected in the form of ethnographic notes from informal consultations with different stakeholders and participant observations, and of documents (newspaper articles, picture captions) and were analysed using thematic analysis. Findings are organised in three themes relating to the social, policy and physical context in Lido Tre Archi demonstrating evidence of the contextual challenges and how they intertwine to generate a spiral of marginalisation and social exclusion for the participants. We take a collaborative and reflexive stance in our work and conclude with recommended steps and potential limitations to initiate an action research project.


| Trans identities as a site of oppression
Trans is an inclusive umbrella term referring to those people whose gender identity and/or gender expression differs from the sex they were assigned at birth (American Psychological Association, 2015; ILGA-Europe, 2019).
The term trans comprises, although is not limited to, transsexuals, transgender and transvestites. The desire to transition through medical intervention from the sex assigned at birth to the opposite sex (transsexualism) or to live in a gender other than that assigned at birth (transgenderism) are still widely considered a pathological condition in the psychiatric and clinical psychological practice, based on the diagnosis of gender dysphoria provided in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5;American Psychiatric Association, 2013).
Transvestitism is also pathologised, although it is diagnosed as a fetishism under the paraphilia disorders in the DSM-5 (American Psychiatric Association, 2013). The classification of gender dysphoria as a mental disorder is a highly controversial matter and a wealth of critical academic literature highlights how the experience of not matching the sex assigned at birth is far more complex and not necessarily distressful for trans people (Davy, 2015). Moreover, while the trans condition is not intrinsically pathological or distressing, the consequences of multiple forms of social stigma can be far more detrimental to trans people's mental health and wellbeing (Bockting et al., 2013). On the basis of this evidence, the recently revised International Statistical Classification of Diseases and Related Health Problems (ICD-11;World Health Organization, 2018) removed the "gender identity disorder" as a typography of mental disorders. The term was replaced with "gender incongruence" and listed within the sexual health chapter (World Health Organization, 2018). While this has been welcomed as a step towards depathologisation, it is a very recent change that has not yet been adopted by the Italian mental health system. In this study, we aligned the guidelines of the World Professional Association of Transgender Health (Coleman et al., 2012) with critical community approaches to mental distress (Cromby et al., 2013;Fox et al., 2009) by understanding trans identities as a form of gender variance, and psychological distress not as a symptom of the individual, but as a response to the historical and collective pathologisation and stigmatisation of non-normative gender identities.
The mode within which stigmatisation and pathologisation occur can be referred to as transphobia. Transphobia operates within a complex system of social practices, symbolic structures and material/institutional arrangements that link with sociocultural attitudes and behaviour towards gender transgression (Bettcher, 2014). Transphobic attitudes and behaviours are particularly acute when targeted at transgender women (transmisogyny) because transphobia is further supported by culturally rooted sexism and degradation of women (i.e., misogyny; Serano, 2007). It is well documented internationally that trans people are highly exposed to transphobic violence (e.g., Grant et al., 2011;Martínez Guzmán & Johnson, 2020), where transphobic violence is constituted as any verbal, physical and psychological act that specifically target individuals whose gender identity does not conform to cisgender norms (Lyons et al., 2017).
Internationally, there have been notable improvements in recent years in LGBTI rights; however, ILGA-Europe (2019) ranks Italy at 35th place out of 49 countries in Europe with an overall score of only 22% of achieved LGBTI human rights. The legitimacy of trans rights has made less progress and has recently been targeted as part of the rise of a "gender critical" discourse that pitches the rights of trans women and against those of cis women (Pearce et al., 2020).
There is also evidence that the abolitionist approaches to sex work (favoured by some feminist perspectives) have not only overlooked the place of trans women in sex work and trafficking, but also involved debating the existence of trans women in demeaning and transphobic ways (Stabile, 2020).
Some have noted in the local context of Barcelona how the rights of transgender people appear to have improved at the same time as the progressive repression of sex workers, which raises specific challenges for trans sex worker activism and social and political change (Espejo et al., 2020). Given the controversial context of sex work in Italy, and local specificities in relation to trans people, and the Marche region where this study took place, our case study was conceived as a tool to lay the groundwork for conducting AR with trans women sex workers to elevate their unheard voices and to collaborate to enable participant-led change and life improvement. In line with critical Community Psychology praxis (Kagan et al., 2019;Nelson & Prilleltensky, 2010), we assume this can only originate from initiating and sustaining an ongoing dialogue with and between local people, professionals, academics and those affected by the issue, namely trans women sex workers. By understanding and negotiating between the needs of the different stakeholders it is possible to devise a change in local structures (policies, environment and social fabric) that is meaningful to the people affected.

| Sex work and the international policy context
Sex work, sometimes referred to as prostitution, is a highly controversial subject of political, social, and academic debate. Different conceptualisations of the activity of buying and selling sex can lead to different ways of naming, theorising and addressing it. On the one hand, sex work is conceptualised as a form of violence against women (following a patriarchal assumption that this is inherently a female activity). Within this framework, it is usually called prostitution and theorised as a patriarchal legacy that locks women into a seemingly unavoidable cycle of degradation of their own dignity. Here, strategies to tackle prostitution at a policy level are justified on the basis of maintaining public order in the name of human dignity. By relying on positions of moral judgement of prostitution as humiliating or of compassion towards prostitutes as powerless victims, this paradigm advocates for targeting the activity in various forms: full criminalisation, neo-abolitionism, prohibition and regulationism (Sanders et al., 2018).
Full criminalisation is a legal model where everyone involved in the sexual transaction (sex worker, client and third parties) is punishable with jail by law. This applies to several states of the United States, South Africa, Kenya, Uganda, Russia, Iran, Pakistan and China (Smith & Mac, 2018). Neo-abolitionism, sometimes known as the Nordic model, or the Swedish model, aims at "ending the demand" for sex work by punishing buyers and third parties while decriminalising sex workers (Sanders et al., 2018;Self, 2003). This regime is in place in Sweden, Norway, Ireland and Northern Ireland, Canada, France and Iceland (Smith & Mac, 2018). Prohibition or abolitionism postulates that ZOLI ET AL. | 3 buying and selling sex is not illegal, but their support and promotion is; hence, a number of behaviours are punishable (Sanders et al., 2018;Self, 2003). This system applies in the United Kingdom and Italy, and is also referred to as partial criminalisation. Approaches that entail different degrees of criminalisation are problematic as they aim at discouraging prostitution by putting strong limitations on its enactment, rather than targeting the structural inequalities and barriers that lead people into prostitution. Working in a similar vein, although at first sight paradoxically, is the legal model known as legalisation or regulationism. Legalisation is applied in Germany, The Netherlands, Bangladesh, Austria, Senegal, Latvia, Tunisia, Hungary, Peru, Chile, parts of Australia and some counties in Nevada (Smith & Mac, 2018, p. 177). While common sense (and even some academic) discourse positions regulationism as the recognition of sex work as work and, therefore, subject to regulation, it overlooks that sex work conducted outside of designated areas or modalities continues to be criminalised. As such in the legalisation model, criminalisation of sex work is still the default position, and this is why the model is also known as "backdoor criminalisation." Moreover, complying with regulations has costs that are sometimes unaffordable for sex workers, therefore, neglecting the very reason why most people enter sex work, which is out of economic need.
On the other hand, sex work is conceptualised as work, and theorised as the conscious choice of some people to face poverty and structural inequalities by selling sex as a job. As such, sex work is considered like any other job and should be regulated by unions, social welfare, and guarantee its professionals the same benefits and assistance of any other working citizen. Conversely to regulationism though, this position known as full decriminalisation has decriminalisation of sex work as its default position. The most important shift in a decriminalised system is that sex work and related activities are not regulated through punishment under the criminal law, but rather through exchanges under the commercial and labour law (Smith & Mac, 2018). Currently, the closest examples to full decriminalisation are in place in Aotearoa/New Zealand and New South Wales, Australia. Aotearoa/New Zealand, in particular, is considered a virtuous example of decriminalisation of sex work for the active role that sex workers themselves had in shaping the law, and its focus on the safety of sex workers, those primarily affected by the law. However, it has been pointed out how there is still ample margin also in this country for reaching full decriminalisation which extends its benefits to other marginalised groups such as migrant sex workers (Smith & Mac, 2018).
Sex work in Italy follows the abolitionist position and does not criminalise the sale and purchase of sexual services, but forbids indoor "prostitution" (brothels) and prosecutes loitering, kerb crawling and soliciting (Crowhurst, 2012). In other words, according to Law 75/1958 (known as "Merlin Law"), prostitution in Italy is neither illegal or legal, rather as Crowhurst (2012, p. 225) puts it, it is "a non-crime and a non-right". Law 75/1958 is still the only bill regulating sex work in Italy, as two law reform proposals in 2002 and 2008 failed to be approved. However, the governance of prostitution at the local level was modified through the introduction of a national "Security Package" by the third Berlusconi government in September 2008. This set of emergency measures was adopted using the subtle but effective rhetoric equating migrants' prostitution to trafficking. As highlighted elsewhere by Matthieu (2012), this discursive strategy allowed the framing of prostitution as public safety and humanitarian matter, while paradoxically introducing measures to police and punish the very same prostitutes deemed to be victims of an exploitative system. According to Crowhurst (2012), the consequences of introducing the "Security Package" were not beneficial to anyone but the politicians themselves and the Catholic Church which maintains a line of moral condemnation of harlotry. Instead of reducing migration and prostitution, the "Security Package" pushed prostitution into hidden and less safe places, hindering the support of sex workers and potential victims provided by outreach workers and the organisations operating in the field. Another deleterious consequence highlighted by Crowhurst (2012) was the decreased trust in the local authorities, in particular, police forces and health practitioners working with sex workers. Interestingly, sex workers were not consulted at any stage, nor were the experiences of the organisations operating in the field acknowledged or taken into account (Crowhurst, 2012).
Finally, sex work regulation and the governance of non-normative bodies, such as trans bodies, is still a highly controversial issue in Italy. As Di Feliciantonio (2015) argues, the categories of "sex workers," "women" and "trans people" bodies and behaviours are increasingly targeted by punitive local measures in neoliberal and austerity times in Italy. These are enacted through narrative and legislative dispositifs that tackle prostitution by relying on discourses of danger and safety, and on local policies issued under the brand of "urgency" and "necessity" (Di Feliciantonio, 2015).
Furthermore, it is recognised that experiences of social stigma and discrimination and lack of employment opportunities constitute a push factor into the sex work market for trans people in Italy (D'Ippoliti & Botti, 2017).

| Towards AR with trans women sex workers in LTA
In Italy, the prevalence of sex work has increased in the last 15 years, especially among trans women (Di Nicola et al., 2015). The LTA area is a multiethnic neighbourhood in the Province of Fermo and Marche Region, Italy. Here, trans women sex workers from different nationalities suffer enduring marginalisation, stigmatisation and discrimination (i.e., exclusion from basic assistance and social services; D'Ippoliti & Botti, 2017). For example, most of the local services and authorities are not equipped to face the multifaceted reality of sex workers, especially trans (transsexual, transgender, transvestite) women. This is evidenced by the lack of a larger support network in the Marche Region, where there are currently no dedicated projects for trans women sex workers, nor LGBTI + associations (Global Network of Sex Work Projects [GNSWP], n.d.). There is a structural gap in which trans women sex workers struggle for dignity and support (Crowhurst, 2012). They are, therefore, a highly marginalised group whose voice is unheard by local institutions, and whose needs are neglected by local services.
In this paper, we present insights from a pilot study which aimed to engage with this group to develop an AR project that might address this gap. AR is devised as a collaborative project to co-generate local policies and social inclusion for sexual minorities (trans women), with the ambition to enact transformative social change in the long term (Nelson & Prilleltensky, 2010). To start with, it was necessary for the researchers to enter the context and gather information about how "prostitution" was becoming a problematic phenomenon in LTA. G. C. and G. P. noticed from the local news and talking to people in the area that "street prostitution" was recurrently mentioned as an outstanding concern due to its increased visibility. This is in line with the (re)emerging discourses that portray sex work visibility as dangerous and risky (Di Feliciantonio, 2015). Given G. C.'s background working with marginalised neighbourhoods (Cingolani & Fermani, 2014) and G. P.'s established collaboration with a local association reaching out to sex workers in LTA, this was chosen as a topic of further investigation. As researchers, we were interested in understanding the context where such narratives were deployed and the lived experiences of the subjects, namely trans women sex workers. In line with AR theory (Colucci & Colombo, 2018;Lewin, 1946) and Community Psychology approaches (Kagan et al., 2019;Nelson & Prilleltensky, 2010), we compiled a literature review on sex work, which informs the introduction to this paper, and we questioned the premises of our knowledge and assumptions.
This was particularly relevant for reflecting on the social construction of prostitution as intrinsically problematic and as an oversimplification of the complex phenomenon of sex work. The main outcome of this stage was to deconstruct the naïve approach summarised by the expression "we have a prostitution problem," and to reconstruct it into a more critical one "we have a problem with prostitution." This helped us generate two research questions: (1) Why is sex work problematic in Lido Tre Archi?
(2) What are the key challenges with doing action research with trans women sex workers in Lido Tre Archi? 2 | METHODS

| The researchers and their roles
The research is co-designed by the authors who share a background in community psychology, LGBTI + activism and research, and collaborative work with the marginalised. G. C. and G. P. have been living and working in the ZOLI ET AL.

| 5
Marche Region for the past two decades, therefore, ensuring a strong ongoing connection with the territory, the inhabitants and the sociopolitical networks. Moreover, G. C. (since 2017) and G. P. (since 2011) have been in contact with sex workers of the LTA area through a local association which is part of a national charity working for the prevention of sexual exploitation, and for the promotion of physical and mental health among sex workers. In particular, G. P. is a chartered psychologist and psychotherapist, and his outreach work was in the role of clinical psychologist aimed at building trust and supporting sex workers in assessing their needs and strengths. G. C. is an anthropologist and filmmaker, who has worked on participatory filmmaking with marginalised communities in various Marche suburbs. His outreach work was in the role of volunteer aimed at understanding trans women sex workers, their practices and interactions in context. In such roles, G. P. and G. C. did outreach work alongside the other association volunteers, and talked to the people, listened to their stories, and built mutual trust. These are identified in the literature, and by sex workers' associations as the most effective strategies for establishing and maintaining contact with sex workers. Thanks to the trust gained, G. C. and G. P. were able to act as gatekeepers of the sex workers in LTA and be pointed by them to key informants.
A. Z. lived in the Marche Region for 10 years, and given her ongoing work in community psychology and LGBTI rights, she was contacted for a collaboration by G. C. A. Z. joined G. C. for the informal consultations with the key informants, where they had an overt role as researchers interested in gaining a deeper understanding of the challenges for trans women sex workers in LTA. Finally, A. Z. also formed a collaboration with K. J. when working together in the United Kingdom. K. J. is an established researcher in the field of LGBTQ + community psychology and has collaborated on LGBT PAR projects in the United Kingdom, Mexico and Australia. Her role was to supervise the project, and to support framing the context-specific experience of building an AR with trans women sex workers in LTA into the global understanding of doing AR with such population.
All the authors position themselves as scholar activists, that is to say, they engage with structural inequalities by working with marginalised groups outside of academia and in open alignments with the critical community psychology values of social justice and community (Kagan et al., 2019;Murray, 2012). Namely, social justice refers to ways of doing research that explicitly envision the pursuit of a more equitable, fair, and peaceful society; and community highlights the role of collaboration and active participation of the marginalised in such process (Kagan et al., 2019). As scholar-activists, the authors engage with an ongoing reflection on their privileged position, and they critically use it as a gateway for the marginalised to be heard and to assist in the delivery of change through policy, practice and public discourse.

| Ethics and participants
Following institutional procedures for the establishment of AR projects, A. Z. applied for ethical approval. Ethical issues included a consideration of the discussion of sensitive topics, participants' vulnerability, and the cooperation of gatekeepers. The study was reviewed by the Research Ethics Committee of the University of Brighton, and ethical approval was secured. Given the need to minimise risk and maximise trust building at this stage, participants were not recorded, and all names of people, places and organisation were anonymised in the fieldnotes.
Participants were the trans women sex workers that G. C. and G. P. met during their outreach volunteer work, and the key informants that A. Z. and G. C. consulted. The sex workers contacted during the outreach work were aware of the researchers' identity and role, in compliance with the association's practices. However, given the spontaneous nature of the encounters and the focus of the interaction on the participants' needs there was no recording nor signing of consent forms. Participants were not in any way identifiable and their names were not transcribed in the fieldnotes (see also Section 2.3 below). Through this process, G. C. and G. P. were able to identify the key informants for the informal consultations: two trans women sex workers, a local architect and two volunteers of the association. In this case, the role of the consultation was explained and clarified before the conversation started, and verbal consent to proceed was gathered although no recording was effectuated.
Moreover, the researchers assessed that key informants were all of age and not experiencing severe mental distress. The two sex workers identified as trans (transgender, transsexual, transvestite), women (male to female) and sex workers (in activity or not). Participation was in all cases voluntary and free from coercion.

| Data collection and analysis
This pilot drew on ethnographic principles and techniques to gather information on the context of LTA. Ethnography provides a collection of methods which primarily involve observation and informal chats in naturalistic settings to create a sympathetic connection with people in context (Brink & Edgecombe, 2003;Willig & Stainton-Rogers, 2017). Over the period of a year, we accessed open settings with an overt role as partially participating observers (Bryman, 2016). In other words, in line with the ethnographic ethos, we gradually approached and immersed ourselves in the research context, its stakeholders and physical environment to reach an interpretative understanding of trans women sex workers' social world in LTA (Serranò & Fasulo, 2011). interactions. At the end of all the observations, the field-notes were organised, collated and anonymised for discussion with G. P. and A. Z.
As gatekeepers, G. C. and G. P. assisted in identifying key informants for informal consultations (N = 5). These were carried out by A. Z. and G. C. with trans women sex workers (n = 2), a local architect (n = 1) and volunteers who supported sex workers (n = 2). Informal consultations can be considered similar to interviews as they were "conversations with a purpose" (Willig & Stainton-Rogers, 2017, p. 25) aimed at exploring sex work and its context; however, they were informal as they did not have an interview schedule and were not recorded. After every consultation, A. Z. and G. C. wrote down their fieldnotes as close to verbatim as possible. During the informal consultations, the participants shared documents and pictures of LTA. These included six archive newspaper articles from September 1990, and eight pictures of LTA. For each picture, a short caption was generated.
Overall data were collected in the form of ethnographic notes from the participant observations and the informal consultations, captions of the pictures and archive newspapers. These were gathered to form a polyhydric picture of the context as well as develop alliances with other people who were already active in LTA, or who have been working and living there for several years. It is worth highlighting that, in line with qualitative methodologies, data collection was not a defined and linear process. Rather it helped further refining the research questions and focus, as well as expanding the number of people we consulted as well as the places we visited. Importantly, this way of moving between stages was pivotal to frame "problem situations as human systems" (Kagan et al., 2019, p. 142). Secondly, collecting data was the cause and effect of making contact with locals, and gaining entry in, and intersecting with the community (Montero, 2006), while reflecting on how to do this sensitively and respectfully of stakeholders' boundaries. Finally, as data were collected from different sources and from different researchers, there was constant contact, confrontation and discussion both to maintain a logical chain of evidence and to use data and investigator triangulation as research strengths (Yin, 2017). ZOLI ET AL.

| 7
Data analysis was carried out as follows. Ethnographic notes, picture captions and documents were organised separately by A. Z. and G. C. After familiarising themselves with the data through multiple readings, A. Z. and G. C., following guidelines of thematic analysis (Braun & Clarke, 2006), separately coded the data line by line and developed provisional themes. These were presented, discussed and challenged during periodic meetings between A. Z., G. C. and G. P. until all codes were saturated and the final themes were agreed. These themes and possible interpretations were then discussed further with K. J., with interpretations, both thematic and linguistic, refined in terms of improving our reflections on strategies for engagement and challenges to developing AR with trans sex workers in the context of LTA.

| FINDINGS
Through thematization (Braun & Clarke, 2006), we identified three main themes which outline the key contextual forces (social, policy, physical) that generate marginalisation and social exclusion of trans women sex workers. We explore them individually below.

| Social context: "A wild and bloody fight for survival"
According to the data provided by the "Ufficio Anagrafe" (Italian for "General Register Office") of the Council of Fermo based on the latest census (ISTAT, 2011), LTA is a multiethnic neighbourhood of 2011 inhabitants (1055 males, 956 females) of which 766 are foreigners (414 males, 352 females). Of these, almost 40% are foreigners from 43 different countries (229 Europeans, 167 Africans, 13 Americans, 357 Asians). The strong visibility of foreigners is enhanced when considering that 80% of the pupils at the local primary school are from migrant families, and 80% of the foreigners are young people born after 1970. However, further sociodemographic data were not available, and the scientific literature referring to the context is scarce and outdated (Lanzani & Vitali, 2003). In this sense, the lack of updated and detailed statistics is a relevant indicator of the disinvestment in the social development of LTA. Another obstacle to gathering more precise demographic data is that LTA is a neighbourhood with arbitrary borders and does not appear as a codified area on any mapping or cartography.
Nonetheless, people name and describe it as a given entity.
The demographic data on foreign inhabitants is also strongly implicated in considerations of the intersection of migration and sex work in the Italian context. Talking to local Italian people, sex work was conceptualised as problematic due to its association with migrant women. This sits within a broader discourse where migration is perceived as a social threat in current Italian sociopolitical life. We found that Italian sex workers also identified foreign sex workers as a threat to their profession.
Italian sex workers now have the strong competition of foreign sex workers, who have no scruples to sell unprotected sex at a very low price.
[…] They are identified by the people we spoke to as Romanian and Nigerian women mainly. And they are described as mainly drug addicts, selling sex to buy a drug dose.
Prostitution is not a profession anymore (in most cases, and on the street at least), it is a wild and bloody fight for survival. (Fieldnote 12) According to the voices of the sex workers who have been in the job for longer, The 'real' problems of sex workers now are not being paid as much as in the past, poverty, and housing.

(Fieldnote 11)
A concern that was widely shared seemed to be that trans sex workers are growing older, and they are cut off from the sex work market by younger and cheaper foreign sex workers. They cannot do the only job they can do, and they will never get another job. These people need social housing, and probably will not qualify for social and housing benefits even in the case of the regulation of sex work. (Fieldnote 13) These extracts highlight a number of intersecting issues which echo international literature on sex work as well as more pressing problems specific to LTA. First, they reflect the notion that people enter sex work out of economic need, to escape poverty. This positions sex work as a rational survival strategy for many, and trans women, in particular. Second, they show multiple marginalised identities at play. Trans women endured marginalisation as a sexual minority due to their gender identity (trans), and their gender (women). Moreover, some trans women experience additional discrimination due to their ethnicity (Black, eastern European), and citizenship status (migrants, documented or undocumented), as reflected in the views of threat and competition of some Italian sex workers.
Finally, poverty, migration status and sex work also pose another obstacle to gathering a more accurate social picture of the area. Indeed, several trans women sex workers mentioned working in LTA while only temporarily and provisionally sharing a flat or a room with a compatriot, or with other sex workers. As such, they are not registered as residents, and consequently, they are not entitled to access a number of services, creating a spiral of social isolation.

| Policy context: "Treated like a criminal or a threat"
The second theme is the policy context. As outlined in the literature review, the fact that street sex work in Italy is decriminalised yet not legalised, widens the gap between lay citizens and service providers on one side, and sex workers on the other. As sex workers are pushed by the law to work on the edge of legality and in fear of the authorities, they generate a spiral of self-segregation where they feel they cannot trust police forces and healthcare providers as a source of support (D'Ippoliti & Botti, 2017). At the same time, authorities feel they are entitled to pursue an intransigent line given that a number of behaviours can be listed as criminal as a consequence of the "Security Package" (Crowhurst, 2012). This is enhanced by the commitment of authorities to fight migration and prosecute foreign sex workers. If we add another layer of complexity by considering trans women sex workers, we must add to the abovementioned constraints the lack of support for trans people in Italy, and in the Marche region, in particular. Together with the lack of recognition of crimes based on gender identity and sexual orientation at the national level, the support network for LGBTI people in the Marche region is not as developed as in other regions.
Two trans women sex workers recall being in touch with authorities as a shameful and degrading experience. As their appearance does not match with their legal identity, they are stopped, made to sit, or even forced to go on an institutional vehicle. They feel treated like a criminal or a threat. They are scared ZOLI ET AL.
| 9 and fear for their safety; they do not know who they are waiting for and why, or where they are taking them and what they will face. Some staff start mocking them, smiling, winking, and telling jokes.

(Fieldnote 5)
As exemplified by this extract, transgender identity coupled with a job as a sex worker creates an intersectionality that is highly stigmatised and marginalised with little chance of being protected. In this context, trans women sex workers are unlikely to report violence to the police or access healthcare services to avoid further discrimination as sex workers, trans, women and migrants.
Further evidence of the impact of policy barriers includes the historical criminalisation of trans identities as by the Italian penal code until 1982 (Pezzoli, 2006). As some older trans women sex workers recalled At the time in Italy trans people could not have their documents changed unless they undertook complete surgery. But they had many other limitations, for example they could not have a driving licence, and the fascist law punishing "Transvestitism" was still in place in the penal code. This cost trans women sex workers appearing many times in front of a judge, where they would be called by their male name.
[…] In the palace of justice, they had to prove they were not dressing as a woman for a queer performance or to deceive others, including authorities, but they are women and therefore addressing accordingly. (Fieldnotes 3-4) Such experiences highlight how sociohistorical events and transitions continue to impact people's lives, bringing to the forefront the role of the chronosystem when using an ecological approach to community research (Bronfenbrenner, 1979). In line with the findings of D'Ippoliti & Botti (2017, p. 81) enduring social stigma, and a historical lack of employment opportunities are "push factors into the sex market" for trans women. These policies, coupled with the abolitionist model that makes it legal to sell sex but illegal for sex workers to self-organise and form unions, undermine solidarity and sex workers' rights and contribute to the contextual forces which marginalise trans women sex workers.

| Physical context: Dreaming of "a competitive touristic destination"
The third theme is the physical context of LTA. Following the boom in mass tourism of the Adriatic Coast, LTA was elected as a new touristic destination in the 1970s, and was developed according to the wild rules of speculative building at the time. Mimicking the thriving beach destinations of Rimini and Milano Marittima, LTA was conceived as a potentially highly profitable touristic area (Bravo, 2005;Trentin, 2006). As evidenced by the pictures and postcards of the area we gathered, There is a massive skyscraper on the coast, and a number of huge buildings containing hundreds of flats each around it. Nowadays such a plan would be forbidden, but at the time it was considered fashionable, desirable and ultimately highly profitable. In fact, at the time many touristic areas in the Adriatic coast already had such buildings, and in particular a skyscraper as a symbol of touristic power (like Cesenatico, Rimini, Milano Marittima, Porto Recanati, Porto San Giorgio). So Fermo province had to have its own in order to place itself as a competitive touristic destination. (Fieldnotes 14-15) Until the late 1980s/early 1990s, LTA lived up to the speculators' expectations and turned out to be a highly profitable investment. All the flats were quickly sold, and investors were euphoric. People, especially from Northern Italy, bought their second house in LTA precipitating skyrocketing mortgages as they deemed it a profitable buy that would be paid back by high rentals or higher sells in the future. It was a blind optimism of the economic boom, based on the belief of unlimited economic growth. However, they underestimated the impact of forthcoming globalisation and did not consider any sustainability or resiliency plan (Trasatti, 1990).
So, when the economic crisis affected the industries of Northern Italy, many workers were left reliant on social benefits and struggling to afford second houses, or holidays at all, and LTA started to visibly decline. Many holidaymakers opted for closer destinations such as Romagna, while foreign tourists were increasingly attracted to other countries that entered the touristic market with cheaper European fares offering similar sunny beaches. Consequently, the Fermo province that until then had seemed a promising gold mine was cut off from the market.
The failure of the capitalist business plan led to institutional and social abandonment of the area (Urban Device, 2016) with businesses originally planned either never starting, or soon closed, and the area diminished into a segregated ghetto within 10 years. With the increasing decline of the area, the value of the real estate properties decreased making it more affordable to people with lower income such as migrants and sex workers.
The physical degradation of LTA alongside the criminalisation of sex work had the effect of collocating "prostitution" with other illicit activities in hidden and isolated areas. This aligns with others' observations of the impact of abolitionist models of sex work and the detrimental consequences it has for the safety of sex workers (Smith & Mac, 2018).
Only very recently have politicians and the administration shown interest in the regeneration of the area.
However, funds have been delayed and an investigation was put through to (try to) ensure that the planned works are carried out in the promised timeframe. An indicator of the level of ghettoization of the area is that the first and most urgent intervention for the area is to set up street lighting, video surveillance, and enforce local police (CronacheFermane.it, 2019). It is beyond the scope of this paper to theoretically analyse such a choice, but a quick link to the ideas of Foucauldian techno-surveillance and bio-politics can spark the suspicion that the requalification is planned to control rather than to serve the marginalised. As Lyons et al. (2017) point out, police enforcement is instrumental to the gentrification process of neighbourhoods and communities.

| DISCUSSION
The findings showed the key challenges faced by trans women sex workers in LTA revolve around the social, policy and physical contexts. In terms of social context, there is a lack of up-to-date and detailed statistics of a multilayered and magmatic local community that has no representation in the official data. The main points of tension pertain the profound change in the status of sex work apparently due to its intersection with the phenomenon of migration. This is coupled with a general perception by trans women sex workers that the lack of welfare measures and support create a spiral of isolation and exclusion from basic health and social services. The hardship of the social context is exacerbated by the unstable policy framework on sex work in Italy. The partial criminalisation of "prostitution" locks sex workers into a fear of interactions with local inhabitants, police forces and services reducing chances of inclusion and safety. Conversely, the increased physical and social distance exacerbates conflicts with locals and police forces.
Such tensions are ostensibly induced and maintained by the physical setting of the LTA area. The skeletons of an era of auspicated prosperity provide the spaces to sell and buy sex out of the spot light, but with decreased safety for sex workers themselves who have fewer chances to help each other and seek help in situations of danger. Indeed, the same area provides an ideal territory for other illicit activities to thrive.
As part of our ongoing engagement with the stakeholders, findings were fed back during a round table held by the researchers and members of the main charity doing outreach work with sex workers in LTA. Themes were discussed and contrasted against the first-hand experience of the members, and the researchers answered any questions raised. The members considered the findings relevant and collaboratively discussed how they can lead to an AR with trans women sex workers, who will play a key role in devising guidelines for the social change they want to see in the area. AR has developed and taken different turns since its implementation by the social psychologist Kurt Lewin (1946). Nonetheless, its main legacy remains its capability to actively involve participants in the ZOLI ET AL.
| 11 research process to enact a more durable and deeply rooted behavioural change (Dubost & Lévy, 2002). Shifting from individual behavioural change to groups and community behaviours, AR has also been largely applied to tackle group conflicts, minorities' marginalisation, and power asymmetries (Colucci & Colombo, 2018). AR coupled with the values of community psychology and its goal to conscientize (Freire, 1970), liberate (Martín-Baró, 1994) and transform experiences of the oppressed and marginalised, is an effective tool to enact structural social change in local contexts. These suggestions were welcomed during the round table, and the members decided to bring forward to the board their request for the charity to provide: formal support and facilitation to the recruitment of participants, and a safe space within the charity building to carry out the AR.
In the round table, the researchers provided a draft of the AR, which was discussed, and a preliminary design agreed as follows. Data will be collected through in-depth biographical interviews, and an individual body mapping activity. These will further explore in-depth the policy, social and physical contexts through the lenses of participants' subjectivity and how these three elements are interconnected with the participants' lived experiences as trans women sex workers. Subsequently, the researchers will have a first go at analysing the collected data. However, at this stage participants will act as peers and research experts themselves, and their input will shape the analysis. A third face-toface activity will allow the researchers to share codes with participants, see how they make sense of their own data, and what overlaps and differs from their insights. Once data have been co-analysed, a fourth activity will be aimed at participants to design a manifesto with action points for the local services and administration to implement. Although it is difficult to predict how the manifesto will look like and what participants specifically will require, its relevance and originality consist in the usually unheard and unrequested voice of sex workers to implement local policies.
The round table also assessed some of the challenges to the participatory feature of the research and the impact of time, space and economic constraints. Participation in the research is entirely voluntary, but due to the economic need of sex workers to work it was agreed it would not be possible to plan an AR project that requires more than four encounters. This might limit the level of participant engagement in decision making, data collection and analysis. To counterbalance this, we planned an iterative and collaborative research process in which participation is sought at both the data analysis and action stages.
Finally, the round table also identified potential obstacles to the AR. These include the time-consuming work of building trust and networks with participants, associations and institutions so that the AR project meets the participants' needs while ensuring their safety and confidentiality. Another consideration is that contexts are always dynamic.
The physical, policy and social contexts are not a static background, they change with people, as people change with them. Therefore, we constantly need to manage a certain degree of uncertainty and variability in our research design.

| CONCLUSIONS
In this paper, we outlined a case study that lays the ground for an AR project with trans women sex workers in LTA, Italy. Limitations of this pilot study include that it necessarily relies on secondary data and informal fieldnotes.
However, the triangulation of the sources and the triangulation of researchers allow for a more trustworthy outcome (Yin, 2017). Another limitation is that although this process has taken up to a year, it has not yet produced concrete results in social change for the marginalised population, namely trans women sex workers. This is a major aim of AR and will be key to our future research endeavours. Finally, an important limitation is that due to the nature of the research (time and economic constraints, and participants vulnerability), we will only be able to recruit a limited number of trans women sex workers. Therefore, while we continue to consult widely with those affected by the issue, we are aware that the voices we engage will not be representative of the much more heterogeneous group of trans women sex workers. Nonetheless, and to conclude, the peculiar approach from a community psychology perspective where a psychological focus on emotions, thoughts and behaviours is paired with a strong emphasis on social justice for the marginalised has been pivotal in gaining access to the field and building coalitions with the local stakeholders.